All right, y'all, welcome to the Scott Horton Show.
I am the Director of the Libertarian Institute, Editorial Director of Antiwar.com, author of the book Fool's Errand, Time to End the War in Afghanistan, and I've recorded more than 5,000 interviews going back to 2003, all of which are available at scotthorton.org.
You can also sign up for the podcast feed.
The full archive is also available at youtube.com slash scotthortonshow.
All right, you guys, on the line, I've got the great Jordan Smith, and you may or may not know that she's a legend for years here in Austin, Texas, writing as an investigative reporter for the Austin Chronicle, and then now for, I don't know, the past five or ten years or so, she's been at The Intercept, and just killing it there, of course.
This one is called The Junk Science Cops Use to Decide You're Lying.
Welcome back to the show, Jordan.
How are you doing?
I'm doing great.
Thanks for having me.
So happy to have you on the show, and I can't get too much into it, but I just had this great argument with a very important reporter about reporting with anonymous sources on national security issues, and he said to me, well, based on your version of the way it has to be done, you couldn't even cover crime unless the reporter could independently verify that the person was guilty before, you know, which is basically a completely hyperbolic take on my take that you should have independent cooperation of what you're talking about before you go spouting off nonsense, right?
But then he brought up local police crime coverage as the exemplar that this is the model that national security reporters should use when this is one of the absolute worst atrocities in American society, is the way that the local news always speaks from the point of view of the cops and the prosecutors.
They never use the word claims for when the cops or prosecutors say something, and if a cop shoots someone, well, it's an officer-involved shooting.
We're not even sure in which way he might have been involved, but he was involved somehow, maybe.
And if he actually crosses the line so badly that they have to charge him with murder, then the local media will call him an ex-cop the whole time that they cover the trouble he's in, as though he was already fired before he did the bad thing that he did.
And local news is nothing but a PR front for the local cops in all 18,000 counties or what ever it is in this country.
And I hate it.
And I just thought it was hilarious that this was brought up as the example of how you're supposed to do it.
When the government says something, then you repeat that to the people.
That's the job of a reporter.
Yeah.
You know, that is a long-term problem in crime coverage.
You know, there's just a lot of local reporters are very beholden to their institutional sources.
And what ends up happening, you know, I've actually had a situation where I was talking to, I can't remember if it was the district attorney's office or it might have been the U.S. attorney's office here.
And I was questioning them about something.
And they basically told me, stop biting the hand that feeds you.
And I said, excuse me, you work for me, like, you know, I'm doing my job.
But you know, they do take it at face value and precisely because of threats of access.
Right.
I would rather not have the access and say, look, you know, the D.A. won't talk to me and that you should know that.
Right.
Because I'm asking questions they don't like.
You should know that.
That is that is the role you should have, you know, precisely different than what you're saying, what you're saying in the example they gave for sure.
In fact, one of the things I should have included in my rant there is along the lines of your article here is that they will always repeat the prosecution's expert testimony as though it's the voice of the Gospels or whatever that you must have blind faith in, even when it's completely discredited nonsense like the bite mark analysis or the hair matching and all this, where even as we've talked about before, even the FBI admits that, yeah, we got up there for 40 years pretending we could match hairs by looking at them.
And yeah, we're lying.
Yeah.
I mean, it's it's that's yeah.
You know, I have a passion for junk science.
So, you know, in this scenario, when the blue leaks hack came out, I just like, oh, I'm going to like poke around and see what's in there.
Right.
It might be, you know, sort of next to my heart.
And I first started searching for Reed technique because not that I was going to be shocked like there's gambling in Casablanca.
I'm shocked because, of course, there's going to be a Reed technique stuff in there.
It's been ubiquitous over the years.
It's a very flawed interrogation process.
But I just wanted to see how widespread it was.
And, of course, it was like sigh.
It's still pretty predominant.
But then all these other things started coming up, which are just completely junk science.
And I was a little bit surprised about how widespread those kinds of trainings were.
I mean, so that's really important problematic.
Go back to read for a second real quick.
And then because a lot of this other junk is supposed to be an improvement upon read.
Right.
So what's read first?
So the Reed technique is like sort of this ubiquitous interviewing and interrogation technique used by police in the criminal sort of investigative process.
It's extremely accusatory.
It teaches basically you.
It's very confrontational.
It's guilt presumptive.
And it also sort of relies on the interpretation of nonverbal cues as indications of deception.
So basically, you interrogate someone and you come at them saying, look, we know you did this.
And that is the tone of the interrogation.
And so then you look at like, oh, are there eye movements or are you crossing your arms or is your leg twitching?
And they're routinely those things, though they could mean literally anything, are part of that guilt presumptive process.
So if they come at you and say, I know you did this thing and they can lie to you about evidence.
Right.
Most states you can lie about evidence and the Reed technique encourages that.
So look, I know you did this.
I have this big stack of papers here and it's all this evidence and there's DNA and and now you're twitching your leg, which could be entirely because you're like, holy hell, I didn't do this.
What's going on?
But they look at it and go, aha, see, I've hit a nerve.
And so that is precisely how the Reed technique works.
And the problem, of course, is that people it has been tied to a number of false confessions because a high number.
When you say a number, you mean huge numbers, huge amounts of humans in cages.
Yeah.
And the one thing that of all the things that we've learned about how a lot of the forensic sciences that we use are incredibly well, are unhinged a little bit.
Right.
They're not pinned to a solid scientific foundation.
The one that people still really have a problem with is is the notion of false confessions.
People just do not believe that they would ever say they did something that they didn't do.
And we know that that happens all the time.
But but confessions are still looked at as sort of the gold standard of evidence despite their inherent flaws.
So that's why this kind of guilt, presumptive approach, confrontational approach and looking towards non nonverbal cues is incredibly problematic.
Right.
So for a while, people have been trying to look at new ways of of of interviewing.
And you know, not surprisingly, what what turns out is backed by science is that you just have these really sort of rapport building, sort of open ended discussions with people, you know, akin, a lot akin to the way you would interview as a reporter or this kind of conversation that we're having now, it'd be far more like this.
That is what science shows us actually works.
And then you you get actionable intelligence from that kind of conversation that you can then go and follow up on and see if a person was actually telling you the truth or not.
And that is what science tells us is the way to go.
But what we can see from the Blue Leagues hack is that that is not the predominant sort of way in which trainings are being offered to law enforcement in this more open style.
There are a few flyers in the hack that I've been able to find that do, you know, kind of use this more science based conversational approach and offer that training.
The majority of them are really read style or are or are completely related to to nonverbal behaviors.
And that's really disturbing to people who who, you know, the sort of social psychologists who are leaders in this field, you know, because you're peddling junk to a group of people, police and other kinds of law enforcement who obviously wield a tremendous amount of power over individual liberty.
Now, if they're being told that if you look left when they ask you a question that you're lying and they believe that we're in trouble, you know what I mean?
And this is just total voodoo nonsense.
I mean, this means nothing.
Right.
I mean, you say, for example, crossing your arms.
Well, someone's falsely accusing me of something and then I feel complete contempt for them for that.
Then I just might cross my arms as a sign of contempt for them.
Right.
And then they could say, oh, no.
Maybe he's hugging himself because he's nervous, because he's lying.
But this is like saying we're in the fourth phase of Jupiter or where this means nothing.
No, it doesn't.
It means nothing.
And, you know, I talked to some of the the practitioners of these methods, both from Reid and then these other, you know, sort of subconscious communications.
And this one woman eyes for lies, who says she is a human lie detector and can teach you how to be a human lie detector, which, by the way, is garbage.
And they really do believe that that you can tell these things and they believe this is validated by science and it just is not.
I think one of the funny things to me, it's funny, although it shouldn't be amusing, is that they really have a feedback loop here.
So, for example, they say, well, you know, you're right.
I should not look at nonverbal cues in isolation to decide someone's lying and have to be sort of context based.
But their context is that they can tell from talking to you before they accuse you of something of what your normal baseline of behavior is, which is just absurd because you and I put it to them and said, look, if if you're asking someone, you say you can develop a baseline on my behavior from asking me some non confrontational questions.
You don't know me, though.
You can't know me.
You don't know what my history with law enforcement is, if I have had bad experiences in my past, which would make me, you know, feel a certain way towards you regardless of how you're speaking to me.
Or there may be sort of cultural norms, right, the way people behave and you can't possibly know that.
So the idea that you can get context right for someone's you can set sort of a baseline and have a context for interpreting these nonverbal cues.
It's just bogus.
It's just bogus.
And it's very dangerous sort of situation, because with science, I think one of the things that is most important to remember, this is that the actual people who have studied this stuff for years have have shown that the individual's ability to determine deception in another person is about chance.
Right.
Right.
It's about 50 50.
Right.
But when people are trained to believe that they can do that, then they get more confident in their ability to do it.
But they actually are no better at doing it.
Right.
So they feel about chance, but they feel stronger in their conviction of their conclusions.
And that's dangerous.
Right.
So you have a cop who's been taught that he's a human lie detector.
And so he still is no better than chance in determining whether you're actually lying to him.
But he's certain that you're lying to him.
And that, you know, can set up a domino effect that's just winds you up in jail.
Right.
I like how you have a quote in here of one of these college professors and researchers saying that, look, they took twelve thousand supposed lie detecting people and then they took the 50 the 50 of them who had the very best results.
And then they asked them, well, how do you do it since you're the 50 best out of the twelve thousand?
He says, yeah, but this is like flipping a quarter ten times, twelve thousand times.
A few of those times, maybe 50 times, you'll get 10 heads in a row or 10 tails in a row.
But it's still just chance.
Still doesn't mean anything.
You're just dealing with a high enough set of probabilities and numbers that it's going to play out that time.
That way.
Twelve thousand people.
Yeah.
You're going to have the 50 best out of the twelve thousand.
But so what?
It doesn't really mean anything at all.
It's an optical illusion.
Exactly.
It's an artifact of the way probability works.
Right.
So that's the wizard project.
And they basically just were like, OK, we're going to, you know, give these thousands of people a test.
And these were people who worked in law enforcement down to regular folks like you and me.
And 50 of them, yes, apparently showed exceptional ability to determine deception.
And then the idea was they're going to reverse engineer that.
Right.
So they're going to look at what do these people have.
And then, you know, this woman, Renee Ellery, was one of those 50 people.
And she is certain.
She is absolutely 100 percent certain that that she is exceptional at determining deception in others.
I mean, she truly believes it and was very offended by my inquiries about it.
And the thing is, she goes around making a lot of money teaching cops these things.
But it's almost like a psychic, you know, hey, I'm talking to your dead grandmother right now.
Was there a concern about the money or like this cold reading stuff?
No one ever puts them to the test.
No one ever says, OK, you're so smart.
We're paying you fifty thousand dollars for this seminar.
Let's see you do it.
My dog just died.
Am I lying or not?
Right.
Tell me.
Right.
Right.
And that's really the point.
You know, because like the researchers, you know, they they the people who believe in this stuff have a million reasons why the researchers disagree with them.
And it's you know, they're mostly like, well, these are done in controlled studies of, you know, just like a lot of hocus pocus.
Right.
But the point is, is that if somebody could actually do lie detection in a way that's reproducible.
Right.
And prove it over and over again and prove that it is not only scientifically valid, but also scientifically reliable, then, you know, money would be thrown at you.
I think it's one of my my sources, but money would be thrown at you.
And all the researchers who have spent decades looking at this would be the first to say, wow, you were right.
We were wrong.
But they cannot do that.
You know, it's just there's in sort of the eyes for lies lady.
She said, I don't need to reprove this.
Well, yes, you do.
That's how you determine science.
Right.
It has to be repeatable.
It has to be reliable and valid.
And so, yes, it does need to be reproven over and over.
You know, there's a huge reproducibility problem, particularly in the field of social in social science fields and in psychology, where, you know, you take a study that is published and if you try to reproduce it, it fails.
So we need to be challenging these things all the time.
Right.
And challenging them against themselves and refining them and honing, especially when you're talking about bullying people into admitting that they committed some crime, that now you get to lock them in prison.
Absolutely.
And, you know, that's the basis of science, science is doing something over and over again and tweaking it.
And so that's what we need to be doing.
And that's kind of what researchers have been doing for years.
And going back to what I've seen before and what they've been able to show over and over again is that this non-confrontational fact finding, rapport building conversations are far more valuable in in sort of developing actionable intelligence that you can then, you know, back up with evidence.
And you know, what's interesting about that is how obvious that is.
Right.
Everybody knows that.
That's like the Sam Jackson movie from the mid 90s, The Negotiator, where there's some guy has hostages and Sam Jackson's like, look, pal, I'll get you some doughnuts where he's as you say in here.
Oh, it sounds like coddling, but it's a means to an end.
Are you trying to free the people from the bank or not?
Are you trying to get the guy to admit that he did the crime or not?
And then obviously the best way to get somebody in a lie is to catch him in the lie, not to just sit there and beat him over the head and say, admit it, admit it, admit it, but to get them to try to explain it.
And as you write in here, quoting one of these experts, lies are hard to remember and they just don't have the richness of detail that the truth does.
And so when you get them talking, they'll end up hanging up on themselves.
And then you can say, ah, but you said a minute ago, this, what about that?
And that's how you really get them.
Right.
Yeah.
You know, one of the interesting techniques around that, which I think is really interesting and how you can, cause cause a lie is a, you know, to tell a lie, a specific sort of story, fixed story.
Right.
So the idea is then to try to, you know, sort of like topple people off of that.
One of the ways that really works to do this, apparently according to research, which I kind of find it fascinating.
One of my sources tried it on me and I actually, it was like, he was like, make something up now do it like this.
And I'm like, Ooh, can't, which is this.
So if you think someone's lying and they tell you a story, then you, you do things that they wouldn't have anticipated.
Like for example, if you say, well, I was in the restaurant at the time this thing happened.
Okay.
Well, draw me a picture of the layout of that restaurant.
Right.
So it's like, oh crap.
Or another one is, okay, tell me this story.
Now I want you to tell me this story in reverse order and go back and tell it, you know, contemporaneously back to the past.
And apparently that just really trips people up because you have to stop and think.
And also people who are telling the truth are sort of tangential, you know, and don't necessarily have, um, you know, they'll, the little things will come in or they'll correct themselves.
Oh wait, no, that happened actually before that, right.
Those things that some people might think are hallmarks of deception are actually hallmarks of people telling the truth because, you know, the mind is not a camera.
Um, we learned that with, um, eyewitness identification, right?
For a long time, people think like, oh, the mind's like a camera, it captures a perfect picture of a point in time and of a person, uh, in that point in time.
And that's just not true.
The mind does not work that way.
You know, it's, it's, um, the lessons that we learned from, from like how the mind works with eyewitness identification are, are definitely applicable to this.
Um, but you know, it's very hard.
I am still sort of overwhelmed by the fact that, you know, all these years into really having a collective conscience that we know fall like wrongful convictions happen.
Like it's, it's demonstrable.
We see this, but that people still really have a hard time believing that people confess to things they didn't do.
Uh, it's still, I find it astonishing and it's probably because I'm very much baked into this world, you know?
Um, but I think no story like this, you can understand, maybe easier for people to understand why someone, you know, some of the reasons that people confess to things they didn't do are, are precisely because of these, um, techniques like read and these other guilt presumptive processes.
Yeah.
No, there's nothing quirky about you there.
It's a basic interest in justice.
Just what's weird about you is you work so hard on it, which is great for the rest of us.
Um, but yeah, I mean, and this happens all the time.
I remember when they first started doing the DNA tests and I don't know how well these exact percentages held up over time.
When they first started doing the DNA tests, they had anywhere from a third to a half of rape convicts were being set free.
And that's with the woman on the stand saying, that's the guy.
And it's because a lot of times they're going off the guy that they picked from the photo lineup or that they picked from, you know, some other thing.
They're remembering him from the cops or, you know, showing them the picture rather than the actual original memory.
There's a million little tricks of air there.
And if, if there's one book that people ought to ever read, man, it is a textbook on social psychology.
I mean, that is the most hilarious stuff to see just how impressionable people are for their most, their most basic and sincere understandings of things can be so easily manipulated.
It's just tragic, but it's, it's, it's as interesting as it is crazy.
You know, it's to see it's nuts.
And this is what the read method is really about, right?
Two big fat, stinky cops in a broom closet with a bright light, trying to shorten your time preference so that you're willing to say anything to get out of this room.
In fact, they say torture doesn't work.
Torture works.
And that's the lightest torture of all.
Two big fat, stinky cops in a small room in a hot light.
But that's essentially what they're doing.
They're trying to get you to feel like, you know what?
I'll just have to deal with the trial later.
Right now I gotta get out of here.
And there you go.
Yeah.
And, you know, I would note that there, there's sort of the, an interesting way in which what we know works, um, has come out of this, this, this sort of government project, um, called the high value detainee interrogation group.
I think that's right, but it's known as the HIG, H I G, the HIG.
Um, and this was basically an Obama administration, uh, program that combined a lot of, it's an interagency program that was overseen by the Department of Justice, um, by the FBI.
And it was basically to get away from torture, to reverse Bush era torture tactics, and to try to figure out what really worked if we're talking about, you know, high value detainees, terrorism suspects.
It was developed with that in mind, but they really looked to, um, social science research and social psychology research around, you know, police interrogations.
And that's how this, um, you know, non-confrontational sort of approach has been validated over and over again, thanks to this HIG group, um, the HIG.
And, you know, I guess what's sort of, you know, um, perhaps unsurprising, but unfortunate is that basically after Trump came into office, the work of the HIG, um, was just abandoned.
I mean, completely abandoned.
Um, I guess that's not terribly surprising since they, you know, also sort of shuttered everything that was going on under the DOJ that was trying to look at forensics reform, um, that's just all collapsed under Trump, which is dangerous.
Um, regardless of how you feel about government, I do think that when government is trying to make something more scientific and some, to develop practices and techniques that actually safeguard individual liberty, I think that's a program I can fully support.
And it's, um, unfortunate to see a lot of years of work, just be sort of like sidelined.
Yeah.
Which is typically when you come back to the idea that we know people are being trained in this bull crap, right?
That, you know, the re read and, and these other things that is still going on right now.
And you quote the experts in your article saying, look, you're talking about a bunch of old cops having to admit that they've been wrong and they don't want to do that.
So we're going to have to wait till they all just get old and retire, which, hey, lucky for us government employees retire at 50.
So I guess we'll be rid of them soon enough.
You know, we'll still have to be paying for them for the rest of our lives anyway at gunpoint.
But um, you know, that's, that's like the only solution is well, eventually the boomers will die.
Oh, okay.
Good.
You know, I will say there's actually been some good stories.
There was a real, a real, a good story about, about this in the Marshall project from a couple of years ago and where they were, you know, the HIG, the HIG folks were like trying to get police departments to, to try out this stuff.
And they went to LAPD and had essentially what you're saying, some grizzled veterans, you know, who had been, you know, baked to the read technique, baked into them.
Then they thought it was a good technique and they were asked to sort of, you know, there were a couple of squads within LAPD who are asked to, you know, let's try to adopt this other approach.
And of course they were all skeptical at first, as you would imagine, but came away from the experience realizing just how valuable it is and how much more valuable it is precisely because it does not induce false confessions.
Right?
Like, so the idea is to get to a point where we can have like the greatest level of success, i.e. obtaining the information that helps us close a case, but while also not creating wrongful convictions.
Right.
And so it's trying to find that, that balance point.
And to date, you know, this sort of open ended interviewing style of interrogation and interviewing just, it works and that's what we need to be doing.
Right on.
Listen, I'm so sorry that we're out of time, but I love interviewing you so much and you do such great journalism, Jordan, and I appreciate your time very much on the show here.
Oh, I always love coming on.
Thank you so much for having me.
All right, you guys, that is Jordan Smith, investigative reporter at The Intercept.
And this one is called The Junk Science Cops Use to Decide You're Lying.
The Scott Horton Show, Antiwar Radio, can be heard on KPFK 90.7 FM in LA, APSradio.com, Antiwar.com, ScottHorton.org, and LibertarianInstitute.org.
