Our criminal justice system rests on the widespread belief that human intuition allows us to know when a witness is telling the truth or lying. As recently as 1998, the U.S. Supreme Court stated that it is a “fundamental premise” of our system that the “jury is the lie detector,” and that “determining . . . the credibility of witness testimony” is the jury’s job because they are “fitted for it” due to their “practical knowledge of men and the ways of men.”1 Accordingly, jurors are formally instructed in nearly every courtroom in America that detecting the truthfulness of witnesses is a primary—if not the primary—reason they are there. They are routinely told to analyze the “demeanor” of each witness on the stand and their “manner of testifying” to resolve who is lying and who is telling the truth, and to render a verdict accordingly.2
This belief that humans are lie detectors is not limited to the criminal justice system. Popular culture is saturated with supposed experts who analyze the demeanor of celebrities or politicians caught in scandals, and who opine with confidence as to whether the public statements of these notables demonstrate truthfulness or deceit. The American Amanda Knox, for example, gained international celebrity when she was wrongfully convicted and imprisoned for several years in Italy for allegedly murdering her roommate, only to be eventually exonerated by the Italian Supreme Court. After her first interviews on national television in the United States, which aired prior to her exoneration, “experts” lined up to analyze her facial movements and voice, and render opinions as to whether she was innocent or guilty.3 When these supposed experts published their opinions in articles on the Internet, members of the public filled the comment boxes with their own strong opinions about whether or not Knox was telling the truth. These lay experts seemed supremely confident in their abilities to divine the truth from gestures like Knox’s eye movements and the way she held her mouth when listening to a question. It was almost comical.
In the Steven Avery case, depicted in the Netflix docu-series Making a Murderer, a local Wisconsin politician decried the public support for Avery’s possible innocence after the show aired, writing, “I will never forget the look in Steven Avery’s eyes when he was being brought into the courthouse. . . . I know a person cannot be convicted based on that, but I made up my mind that day. Avery was guilty.”4 Wow.
Although the public and our judicial system have yet to recognize it, we now know that this confidence in human lie detection is misplaced. In reality, we are pretty bad at measuring truthfulness from watching another person’s demeanor or listening to their voice as they recount their story. The things we accept as barometers of truthfulness, such as steady eye contact and confidence, are simply not good barometers. And things we accept as indicators of dishonesty—appearing nervous, fidgeting, a lack of eye contact, or a cracking voice—are not great indicators of lying. These are just notions we have been taught and have accepted as they have been handed down for centuries—folklore—that are now refuted both by new understandings of human psychology and by the simple facts about human error brought out through the innocence movement. While human intuition should not be completely ignored, it is not something that should ever be taken to the bank. We must fight hard against coming to conclusions based solely on “gut feelings” or intuition, or placing so much confidence in our intuition that it clouds our view of the objective facts.
I learned this lesson the hard way. One of my first cases as a prosecutor involved a bank robbery in which the FBI agents investigating the case believed that a particular teller had been involved. This teller had taken an extra $10,000 from the vault and placed it in her cash drawer shortly before the robbery occurred. When the robbers approached her teller station with guns drawn, she gave them all the money in her drawer, including the extra $10,000. When questioned after the robbery, she claimed that she had retrieved the $10,000 from the vault because a gentleman had come in a few minutes earlier to cash a $10,000 check but had forgotten his ID and said he would return a few minutes later to complete the transaction. The FBI agents found this answer suspicious, and from this fact alone concluded that this was an “inside job” and that the teller must have been involved in planning the robbery.
I was assigned the case, and for months the agents and I couldn’t crack it. The teller was our only lead, as the robbers had been wearing masks, and so for months the agents dogged her. They brought her to the FBI headquarters over and over and aggressively interrogated her for hours each time. But she stuck to her story and never wavered. And she was convincing. I was present for one of these interviews, along with two interns who I brought along with me, where she started crying and pleading with the FBI agents to leave her alone and stop ruining her life through their constant harassment. She explained how the stress from their harassment had carried over to her family, and how her young son was now struggling in school because of his mother’s stress and depression. Her gaze was steady. Her voice was sincere and emotional. She was adamant and confident. She seemed to me to be a woman clearly suffering badly from false accusations. I started feeling uneasy, and before long the two interns were staring at me as if to say, “I hope they leave this poor innocent woman alone.” I was convinced by the end that she had nothing to do with the robbery.
The investigation went cold for several more months, until another woman was arrested on an unrelated charge. In order to work a deal, this woman told the arresting officers that she had information about my unsolved bank robbery case. When the FBI agents and I interviewed her, she said that her good friend—the bank teller—had been involved in the robbery. And, what’s more, this woman had been present when the teller created the demand note used by the robbers by cutting letters out of a magazine and pasting them on a piece of paper. She claimed to be in the same circle of friends with the teller and the men who robbed the bank, and named all of the robbers. Sure enough, when we got the phone records of the teller and the men whom the woman had named as the robbers, we learned that the teller had called several of the men in succession on the morning of the robbery. We then arrested the men, and each of them told us the same story. They admitted their involvement, and said that it had been an “inside job” led by the teller, who had promised them that she would get a large amount of cash out of the vault before the robbery went down. She even told them the best time of day to rob the bank and where the guards would be standing. Eventually, the teller pled guilty to being an accomplice in the bank robbery and admitted her complete involvement.
I was stunned that I had been so convinced of her innocence. Her performance was Oscar-worthy.
My work as an innocence lawyer has taught me the same lesson, again and again. I have had inmates apply for help from the Ohio Innocence Project (OIP) who came across very convincingly in our prison interviews with them, only to be confirmed guilty later by DNA testing. In some cases, I have had inmates who were later confirmed guilty talk about their feelings of being wrongfully convicted, and the stages they have had to go through psychologically to cope with being in prison for crimes they didn’t commit. Their stories sometimes mirrored the psychological reactions of the truly innocent I have represented, and were exceptionally convincing. But the DNA testing confirmed their guilt.
It goes the other way as well. I have had clients who appeared nervous and shifty during the interviews, and who had trouble making eye contact with me when telling me their stories of innocence. They acted guilty, or sometimes even “creepy” to the students accompanying me on the prison visits. They seemed like they were lying. Their voices cracked and wavered. But DNA testing proved them innocent.
I now tell my students working in the OIP that while their instincts about whether a witness is lying or telling the truth should not be totally disregarded, their gut reactions should never be viewed as very important factors in the overall investigation. Judge on the hard evidence, I tell them, not on demeanor. But this is a hard lesson for the students to learn. Our societal belief in our lie-detection abilities is so ingrained that I still sometimes find students slacking on a case when they have not been moved by the demeanor of the inmate claiming innocence, or telling me that they are going to keep working hard on a case that appears dead in the water because they met the inmate and “know” from his “vibe” that he is innocent. I remind them that you can’t take your instincts to the bank. Humans are just not good lie detectors.
I routinely see stubborn arrogance from police and prosecutors about their lie-detection abilities. It is not uncommon for prosecutors or police to react with great indignation when they learn that the Ohio Innocence Project is investigating one of their old convictions. I can’t count the number of times I have called prosecutors or police officers to ask for files or evidence in an old case, only to be told something like, “You’ve got to be kidding me. I was the one who met with the witnesses. I’m the one that looked into their eyes and heard their stories. They were telling the truth and were not mistaken. There is no doubt about it. You’re barking up the wrong tree with this case, sir.”
I recall attending a conference on innocence work in New Orleans and meeting a board member from an innocence organization in another state. She had been a prosecutor her entire career, and decided to volunteer for this organization after her recent retirement. We struck up a conversation over lunch, as I was interested in why she decided to “switch sides.” I asked her if someone she convicted as a prosecutor had turned out to be innocent and whether that is what caused her to take up innocence work. She said no, quite the opposite. She reported that she believes in innocence work because she has come to accept that some people are wrongfully convicted in this country, but the one inmate from her prosecutor’s office who had been exonerated and released by the innocence organization she now volunteered for was most definitely guilty despite being exonerated and released!
Curious, I asked her why she felt so strongly that this exoneree was actually guilty. She replied, “Because my friends who prosecuted the case—and even the judge—later told me after this woman was exonerated that they knew she was lying when she testified in her defense back in her original trial.” She went on to express absolute confidence in the lie-detection abilities of her friends and the judge in the case, and made clear that this confidence outweighed all other evidence of innocence that surfaced later. I just listened, of course, and didn’t bring up the research and facts demonstrating her faulty reasoning. Nor did I mention how confirmation bias—stemming from a belief from the beginning of the trial that the woman was guilty—likely clouded the perceptions of her prosecutor friends and the judge as they watched the woman’s testimony that day in court many years earlier.
OIP client David Ayers was exonerated in 2011 and released after spending twelve years in prison for a murder he didn’t commit. Later, in his civil suit for monetary damages against the Cleveland detectives who had arrested him, a detective took the stand and explained why she initially focused on Ayers as a suspect and eventually brought charges against him. Her testimony, which took place in 2013, was as startling as it was illuminating. Throughout her testimony, she made clear that the primary justification for her decision to pursue and charge Ayers was her unwavering belief, from her years of experience, that she can tell which suspects are lying and which ones are telling the truth. And she was certain that Ayers was lying and being evasive about nearly everything he said in her interviews with him. Her instincts told her this and apparently spoke to her loudly and clearly. And based on these infallible instincts, honed through years on the street as a detective, she claimed that she had never arrested an innocent person. She can tell from her intuition right away if a person is innocent, she claimed, and wouldn’t waste her time on an innocent person. She believed she was a human lie detector with no rate of error.5 It was that simple to her. Indeed, at one point in her testimony, when asked why she thought Ayers was guilty, she stated: “I’d bet my life on it. And take my word for it, my partner would bet his life on it. We’ve been doing homicides for thirteen years, thirteen years. I can pick out a killer in five minutes. I can tell talking to a person within two minutes whether they’re lying to me. After thirteen years of doing hundreds and hundreds of homicides, I think I know what I’m doing.” Armed with strong evidence of Ayers’s innocence, including DNA test results and extensive evidence of police misconduct, Ayers’s lawyers obtained a jury verdict in excess of $13 million against this detective and her partner. She promptly filed for bankruptcy and her partner died, and to date (as of 2017) Ayers hasn’t collected a dime from either one.
• • •
In the cases of the first 330 DNA exonerations in this country, approximately half the defendants had testified in their own defense at trial, proclaimed their innocence, and were flatly disbelieved by the jury. Instead, the jurors believed witnesses such as jailhouse snitches who were later proven to be liars, forensic scientists who provided exaggerated or false testimony, or witnesses who were lying to protect their own involvement in the crime. My clients Clarence Ekins, Robert McClendon, Raymond Towler, and Ricky Jackson all testified in their defense at their original trials and told the jury they had nothing to do with the crimes. The jury thought they were lying. Yet all four were later proven innocent. Elkins had served seven and a half years in prison, McClendon eighteen years, Towler twenty-nine years, and Jackson a whopping thirty-nine years, making him the longest-serving person in U.S. history to later be declared innocent and exonerated.
In Steven Avery’s trial for the rape of Penny Beernsten, twenty different alibi witnesses testified for Avery that he was nowhere near the scene of the attack. But the police and the jury disbelieved every single one of them. Avery was later proven innocent by DNA testing and exonerated and released.6
In some cases, cops and juries are double-duped: they not only fail to believe the innocent party, but believe the false statements of the true perpetrator who is trying to pin the crime on an innocent person. In the Kansas case of Floyd Bledsoe, the police believed Floyd’s brother Tom, who told them that Floyd had murdered the victim and dumped the body on their parents’ property. Based on Tom’s testimony, Floyd—who adamantly protested his innocence but was disbelieved by both the police and the jury—was convicted and sent to prison. Years later, DNA testing proved that Tom was the true perpetrator and that Floyd was entirely innocent, as he had claimed all along.7
My Ohio Innocence Project case of Walter Zimmer and Thomas Siller described earlier had a similar outcome. After an elderly woman was murdered in her Cleveland home, the police arrested Jason Smith. But Smith soon convinced the police that Zimmer and Siller were the true culprits, and that he had been waiting outside in a car when Zimmer and Siller began beating the victim, then gone inside the house and witnessed the attack from a distance for a few seconds, then left. DNA testing later proved that Smith was the real culprit, and Zimmer and Siller were exonerated after each spending more than a decade in prison. Eventually, they each won a sizeable settlement against the city of Cleveland for their wrongful convictions. The police and the jury believed Smith—a career criminal who was lying to save his own skin—and disbelieved Zimmer and Siller, who were innocent.8
Psychologist Paul Ekman has performed groundbreaking research over a period of decades demonstrating that, contrary to popular opinion, human beings are not very good at detecting truthfulness in others. In fact, Ekman’s studies show that we are accurate at detecting lies at a rate that is only slightly better than a coin toss. In one study recounted in his influential book Telling Lies, Ekman divided a group of nursing students into two groups.9 One group watched a video depicting a pleasant ocean scene and were then asked to accurately and honestly describe what they had just watched while being videotaped by Ekman. A second group of nursing students were shown a video of gory medical scenes but were asked to tell a fabricated story afterward that they had watched a pleasant video about flowers. The lies of this second group were also videotaped. Dr. Ekman set the stakes high. The nursing students were told that lying was a necessary part of being a nurse, because sometimes a nurse must hide bad facts from patients while appearing happy and relaxed until the doctor determines it is the appropriate time to tell the patients the bad news. The nursing students were told that they were being judged on how well they lied and how well they controlled and hid their emotions when lying. Thus, the nursing students in the “lying group” had a personal incentive to lie as effectively as possible to show that they could perform well on the job.
Dr. Ekman then had a group of subjects watch both tapes and render verdicts on which nursing students were lying and which were telling the truth. He divided the subjects into two groups. Ekman intentionally made the first group suspicious—he told them to be on the lookout for liars in the videotapes, as half of them would be lying. The second group was told nothing. But both groups performed barely above chance at detecting which group of nursing students were lying and which were telling the truth. The “suspicious group” did no better than the other group.
This study has been replicated in a number of settings with similar results. Psychologists Kraut and Vrij conducted independent studies that placed the accuracy rate of lie detection at about 57 percent—a strikingly low figure given that guessing or a coin toss should result in a correct answer 50 percent of the time.10 A meta-analysis of more than two hundred studies on this topic involving more than twenty-four thousand different subjects shows an accuracy rate in the general public of lie detection at about 54 percent.11 Many of these studies have also revealed what is called a “truth bias.” That is, subjects correctly identify truthful statements about 67 percent of the time, while accurately identifying lies only 44 percent of the time—below chance. This result reflects the societal bias that we generally want to believe others; our default position, absent proof to the contrary, is to believe.
Furthermore, several studies have demonstrated that police officers are no better than lay subjects at ferreting out the truth. In various studies, police officers achieve between 50 percent and 57 percent accuracy. Secret-service agents, however, performed slightly better (67 percent accuracy rates), as did CIA agents (73 percent accuracy rates). Not surprisingly, those in law enforcement, while not particularly good at detecting lies, did not suffer from the same “truth bias” as the general public. Police officers are of course looking for lies and expect lies, and thus are presumably less likely to give others the benefit of the doubt than are members of the general public.
In 2004, a group of leading psychologists evaluated the ability of police officers to detect lies in a setting particularly true to life.12 They took videotaped interrogations from fourteen real cases where the suspects at first lied (told a story that was later proven to be false) and then later gave a truthful version after more evidence of guilt surfaced. Cases were carefully selected to include only those where the eventual evidence of guilt was clear, such as DNA test results or an eventual confession by the suspect that included facts that corroborated the confession (such as where the murder weapon had been hidden).
The psychologists played both video segments from the cases to ninety-nine police officers of varying age and experience. When watching the videos, the officers could hear the suspects’ voices well, and could pick up facial movements such as eye blinks. The group correctly identified the truthful statements 64 percent of the time and the false statements 66 percent of the time. Not surprisingly, more experienced police officers fared slightly better than those who were newer to the police force. While these success rates are higher than those of the studies discussed above, they are still a far cry from demonstrating that police officers are accurate lie detectors whose intuitions are rock solid. Notably, the police officers were incorrect approximately one-third of the time in this real-life setting.
This study had another important result. Objective factors that the public commonly believes indicate dishonesty, such as fidgety body movements or gaze aversion, had little to do with whether or not the suspect was telling the truth or lying. This supports the findings of numerous other studies undermining the commonly held belief that a person who acts fidgety or avoids eye contact is likely lying, or that a person who is calm and makes confident eye contact is telling the truth. It is simply not supported by the research. Indeed, police officers in the study who cited amounts of eye contact or nervous bodily movements as the reasons for their decisions were less accurate than officers who avoided such temptations and instead based their decisions on more objective factors, like whether the stories made sense internally or contradicted other evidence.
Given that police training manuals typically claim that police officers are excellent lie detectors, and instruct officers to strongly consider factors such as eye contact and fidgetiness, those officers who followed the lessons of well-known training techniques such as the Reid technique were far less accurate than those who ignored such training.13 A supreme irony indeed. Yet this training seems to give police officers a false sense of infallibility about their lie-detection abilities.
But it is not only police officers and jurors who fall prey to inflated confidence in their lie-detection abilities. The federal government’s Transportation Security Administration has spent nearly $1 billion in recent years training its agents at airports to detect passengers who are allegedly lying during screening interviews, and thus are potential terrorism threats, using the same indicators of dishonesty—fidgety behavior, minimal eye contact, and so on—that we now know are not accurate indicators.14 The United States Government Accountability Office issued a scathing report to Congress on this practice in 2013 stating:
Available evidence does not support whether behavioral indicators, which are used in the Transportation Security Administration’s (TSA) Screening of Passengers by Observation Techniques (SPOT) program, can be used to identify persons who may pose a risk to aviation security. GAO reviewed four meta-analyses (reviews that analyze other studies and synthesize their findings) that included over 400 studies from the past 60 years and found that the human ability to accurately identify deceptive behavior based on behavioral indicators is the same as or slightly better than chance.15
Rep. Mike McCaul, R-Texas, the chairman of the House Homeland Security Committee, said that the GAO report is “concerning, particularly in light of the fact that TSA has spent almost $1 billion on the program.” He added, “While I believe that there is value in utilizing behavioral detection and analysis in the aviation environment, we can only support programs that are proven effective. The terrorist threats to our aviation system require us to constantly re-evaluate and evolve our security procedure, and if this program isn’t working, we need to find something that will.”16
In other words, when it came to wasting nearly $1 billion training agents to use these supposed deception-detecting techniques, the TSA was publicly criticized, and Congress began evaluating alternatives. Yet, when it comes to determining the guilt or innocence of our own citizens—and possibly sending them to prison for life, or worse, to death row—based on the same type of faulty reasoning, nothing changes.
Any attorney who has engaged in innocence work for any length of time can cite numerous cases where the police claimed that when they arrived on the crime scene, the defendant appeared too “calm and collected” for someone who, for example, had allegedly just arrived home and found his wife dead with a stab wound to the chest. Or, conversely, “too upset and over-the-top.” In both situations, the police’s perception was that the defendant’s demeanor was being staged. Police who arrive on a possible crime scene are, naturally, often somewhat suspicious of the person on the scene, and confirmation bias can kick in. If the person is calm, they may see this as suspicious. If the person is out of control with emotion, they may perceive it as a predesigned act. It’s hard to win with something so subjective. Or, given the malleability of human memory, once some alleged evidence of guilt is later uncovered, the police officers may reflect, “Now that I think of it, he was a little too calm under the circumstances.” This thought later becomes trial testimony that the defendant acted unnaturally calm, and this can be powerful evidence of guilt in the eyes of the jury. The following cases are just a few of the many that illustrate this point.
In 1991, Cameron Todd Willingham’s house caught fire and burned down, killing his three daughters.17 At first it was believed that the fire originated from a nonhuman source, but later it was ruled an arson and Willingham was charged with murder. Before it was ruled an arson, Willingham’s grief was perceived by his friends and neighbors to be genuine. But once he was accused of setting the fire, Willingham’s devastation was perceived as a sign of a guilty conscience.
Indeed, the arson ruling caused a dramatic shift in the public’s perception of the grieving father. Willingham’s priest, who had initially described him as devastated, suddenly became suspicious of him, saying that “things were not as they seemed. I had the feeling that [Willingham] was in complete control.” According to one neighbor, although Willingham was dramatically crying out for his children as the fire progressed, he “did not appear to be excited or concerned,” implying that his emotion was an act. Another neighbor told the police that Willingham seemed more concerned about the fate of his car than the fate of his children, reporting that while he had moved the car out of the fire’s path, he had made no attempt to enter the house and rescue his daughters until the firefighters arrived.
With no motive to explain why Willingham would murder his daughters, the prosecution pieced together the theory that Willingham was a sociopath—that his daughters interfered with his desired fast and loose lifestyle. But while Willingham was being painted as a sociopath in the media, the majority of those who knew him well did not believe he was capable of setting the fire.
After a two-day trial, Willingham was convicted of capital murder and sentenced to death. His demeanor during and after the fire and the interpretations of his demeanor by various witnesses were used heavily by the prosecution to suggest guilt. After several unsuccessful attempts to prove his innocence, Willingham was executed by lethal injection on February 17, 2004.18
After Willingham’s death, the “science” used in the investigation of the fire that killed his daughters was deemed to be invalid by a panel of the nation’s leading arson experts.19 New understandings in fire science confirmed that the fire could not have been an act of arson, but rather supported the initial beliefs that the fire was of nonhuman origin. It is now widely believed that Willingham was innocent; several independent investigations have come to that conclusion.20 Indeed, Willingham’s conviction was the impetus for Texas passing a new law allowing those who were convicted on bad science to apply for a new trial and seek exoneration. And the prosecutor who convicted Willingham has recently faced disciplinary charges for hiding evidence that pointed to Willingham’s innocence. At the time of this writing, those charges are still pending.
This case, like many others, shows how confirmation bias works in tandem with demeanor evidence. Willingham’s demeanor was not considered suspicious until a fire investigator, using a method later determined to be bogus, ruled the fire an arson. Then, and only then, did witnesses come forward with “memories” of Willingham acting in feigned ways that they claimed were suggestive of guilt. Former homicide detective and now innocence reformer Jim Trainum said on the famous podcast Serial, when asked how much weight should be given to people’s interpretations of the demeanor of the defendant in that case, Adnan Syed, “Toss it out all out because it’s subjective, it’s hindsight, and also, people tend to bend their memories to what they think the police want to hear.”21 I couldn’t agree more.
Michael Morton spent twenty-five years in prison for the murder of his wife, Christine, before being exonerated in 2011 with DNA evidence. Morton’s demeanor and actions before and after his wife’s death were heavily scrutinized at trial and played a huge role in his conviction.22 The police developed a motive based on a note that Michael had left his wife the morning of her death in which he complained about her refusal to have sex with him the night before. The prosecution fleshed out the motive at trial by claiming, falsely and without evidentiary support, that Michael was a sexual deviant who had masturbated over his wife’s corpse after violently murdering her (sometimes the truth really is stranger than fiction).
The prosecution argued that Michael’s stoicism in the aftermath of the murder demonstrated indifference toward his wife, and convinced the jury that he was a disturbed human being capable of murdering her. The most damning testimony against Michael came from the Mortons’ neighbor and friend, Elizabeth Gee. She described the Mortons’ marriage as unhappy, and recalled instances in which Michael had said unkind things to Christine. Gee also testified that Michael’s stoic demeanor following the murder had bothered her. Moreover, the fact that Michael had cut down his wife’s marigolds two days before her funeral seemed callous and suspicious to Gee, given the timing.
While it was true that Michael had showed little emotion when told of his wife’s death and had made some odd decisions in the aftermath of the murder—like choosing to return home where the murder took place instead of staying in a hotel, and cutting down the marigolds—the perception that these behaviors somehow revealed his guilt was dead wrong. The DNA testing that proved Michael innocent also identified the individual actually responsible for the crime. Two years after Michael’s release from prison, Mark Norwood was convicted of Christine Morton’s murder and sentenced to life in prison.23
On December 27, 2011, Russ Faria placed a hysterical call to 911 after discovering his wife’s body covered in blood.24 He told the operator that he believed his terminally ill wife had committed suicide. Although he had an alibi and there was little evidence implicating him, Faria was arrested and charged with his wife’s murder a week later.
At trial, the prosecution argued that Faria’s hysteria during the 911 call was staged. The prosecution played the 911 tape to the jury and called the supervisor of the 911 operator who took the call to testify as to his interpretation of Faria’s demeanor. Amid his hysteria, Faria had lucid moments of relative composure. The supervisor testified that in his experience Faria’s fluctuation between these extremes was unusual and suspicious. Faria was convicted.
Afterward, however, the 911 operator who actually took Faria’s call told reporters that she found his demeanor during the call to be genuine. The operator explained that the fluctuation in his mood reflected the effectiveness of the common techniques used by all 911 operators to calm down hysterical callers. On the basis of this and other new evidence of innocence, Faria was granted a new trial. He was acquitted of all charges in November 2015 and exonerated.
Here again the reliance on demeanor evidence, and the belief that someone can be deemed guilty or innocent based on how they reacted in an unusual, unprecedented moment, is troubling. The fact that the operator who took the 911 call and the supervisor of the operator had such widely divergent views about Faria’s reaction reflects the highly subjective and unreliable nature of this evidence. And while Willingham and Morton were accused of reacting too calmly to shocking news, Faria was criticized for allegedly overreacting.
It’s damned if you do, damned if you don’t.
Like Willingham, Han Tak Lee was convicted of intentionally setting a fire that resulted in the death of his daughter.25 After spending twenty-four years incarcerated, Lee was exonerated through the invalidation of the arson science used to determine the origins of the fire.
At his original trial, Lee’s demeanor in the wake of the tragedy was heavily scrutinized. The first officer who arrived at the scene described Lee as “nonchalant,” impassively observing the fire. Other witnesses described him as “calm” and “cold.”26 A firefighter noted that Lee seemed “very depressed,” as if he were “mad at himself.”27 In contrast, Lee’s wife nearly collapsed when she learned of her daughter’s death. Witnesses reported that Lee made no attempt to console his grieving wife; instead, he “walked right by [her] like nothing happened.”28 Lee’s surviving daughter responded to the fire like Lee’s wife, becoming so distraught that she had to be removed from the area.
The defense argued that Lee’s demeanor was conventional by Korean cultural standards and not out of the ordinary within that context. However, the jury rejected this argument.
By the time of his exoneration and release, Lee was nearly eighty years old. The state appealed Lee’s exoneration, turning to the same arguments about Lee’s demeanor that were raised at the original trial:
The accounts of Appellee’s demeanor are significant in light of what a great majority of people would consider the apparent and immediate tragedy of losing a child in such a horrific way. Though there was some question raised in testimony that Appellee’s demeanor was consistent with a cultural norm, it is significant that neither his wife nor his daughter shared his stoicism. . . . The jury could have interpreted this evidence as Appellee’s callous disregard for the fatality in light of his hand in Ji Yun Lee’s death.29
The prosecution’s appeal was rejected and Lee’s exoneration remains intact.
Ashley and Albert Debelbot were jointly tried and convicted of murdering their newborn daughter just hours after bringing her home from the hospital.30 The child’s cause of death was alleged to be intentional blunt-force trauma. With no alternative explanation for the child’s death, police believed the couple must have done it.
The prosecution relied heavily on circumstantial evidence of demeanor to advance the theory that Ashley was responsible for the child’s death, while Albert played a minor role. Ashley was described as appearing calm and reserved, while Albert was visibly distraught.31 The prosecution tactically exploited the contrast in the couple’s demeanors to portray Ashley as especially callous:
PROSECUTOR DAILY: |
Did you have an opportunity to observe both Albert Debelbot and Ashley Debelbot’s demeanor? |
OFFICER TYNER: |
Yes. |
PROSECUTOR DAILY: |
Can you describe Albert Debelbot’s demeanor? |
OFFICER TYNER: |
. . . Albert appeared to be very distraught . . . And he was very shaken by, appeared to be very shaken by what had happened. |
PROSECUTOR DAILY: |
|
OFFICER TYNER: |
Ashley was a little bit more alert, at least she appeared to be. She appeared to be a little bit more rational in her thinking and just appeared to be all around more alert. |
PROSECUTOR DAILY: |
Alright. Did she appear distraught? |
OFFICER TYNER: |
She was not near as distraught as her husband. |
PROSECUTOR DAILY: |
Was she crying? |
OFFICER TYNER: |
I don’t recall her crying, no. |
PROSECUTOR DAILY: |
Was Albert Debelbot crying? |
OFFICER TYNER: |
Yes. Mr. Debelbot cried several times.32 |
Calling Ashley “emotionless,” the prosecution argued that Ashley’s failure to cry while writing her witness statement and the fact that she did not refer to the baby by name were additional proof of her indifference toward the child. But, as other testimony revealed, Ashley was anything but indifferent to her daughter’s death.
OFFICER GARDNER: |
[Ashley] went through a few different changes in emotion as would be expected. At the hospital she was quite stressed . . . which is quite natural having just lost a baby. By the time we got back to the house she was more talkative and seemed a lot nicer. . . . We were investigating her child’s death as a suspicious death. . . . |
PROSECUTOR DAILY: |
. . . When you saw her at the hospital and at the residence, was she crying? |
OFFICER GARDNER: |
She was a little teary at the residence. She wasn’t flat-out crying, but she was teary. . . . I believe it was a mother or grandmother had called her while she was at the residence and she cried a little bit there when she spoke to her. But after she got off the phone with the person, she regained her composure.33 |
Both Ashley and Albert Debelbot were convicted of murder and sentenced to life in prison. In 2015, the Wisconsin Innocence Project and the Georgia Public Defender filed a motion on Albert Debelbot’s behalf requesting a new trial based on newly discovered evidence showing that the death of his daughter was the result of brain and skull abnormalities formed while the child was in utero.34 The litigation is currently ongoing with Ashley expected to proceed with similar filings soon.35
At the age of sixteen, Jeffrey Deskovic was found guilty of raping and murdering a fifteen-year-old classmate.36 Deskovic’s curiosity and fascination with the investigation of his classmate’s death was mistaken by the police as a sign of his guilt. Eager to help with the investigation, Deskovic voluntarily met with the police on several occasions to put forward his own theories about the case. He was misled to believe that he would be allowed to participate in the investigation if he could just prove his worth to the investigators. Taking advantage of Deskovic’s eagerness to please, the detectives manipulated the teenager into falsely confessing after a six-hour interrogation.37
Deskovic was exonerated with DNA evidence sixteen years later. After performing an internal review of the investigation that led to Deskovic’s wrongful conviction, the local district attorney issued a report that made the following observation: “As the police theorized, Deskovic’s actions could have been those of a killer inserting himself in the investigation and working up the gumption to confess. By the same token, Deskovic’s actions could have been—and, indeed, apparently were—those of a troubled sixteen year old, who having experienced the death of a peer from the first time, became obsessed with it and naively believed that the detectives, who fed him pizza and spent time with him discussing the case, would invite him into their investigation if he could prove himself worthy.”38
• • •
A few years ago, I heard a loud noise in the bathroom and found my wife in the midst of a seizure on the floor of the bathtub while the water from the shower ran over her. Her eyes were unnaturally frozen open, foamy vomit dribbled from the side of her mouth, and her body was stiff like a board and twitching. She could not respond to anything I said. She just stared blankly and twitched from the seizure. I called 911 and reported what I was seeing, and answered questions about her age, health, the fact that she had no history of seizures, and so on. I had never seen anything like this before, and I was understandably shocked and quite worried. For all I knew, she could be dying.
But for some reason, my natural reaction was to be very calm. I spoke very slowly and deliberately to the 911 dispatcher and pronounced my words very carefully. There was no emotion in my voice.
Fortunately, my wife turned out to be fine, and there was no investigation, as it was clearly a medical condition rather than a crime. But I couldn’t help thinking afterward that, for some unexplainable reason, I had become very calm and focused. And because of my line of work, it occurred to me that if for some reason my demeanor were later analyzed by suspicious police or prosecutors because they believed a crime had occurred, I would have been accused of being far too collected and calm. I was unnaturally emotionless, they would say. And that would mean I was guilty.
My client Ryan Widmer was not so lucky. When he found his wife dead in the bathtub and called 911, the sound of his voice during the call and his reactions at the scene once the paramedics and police arrived were aggressively used against him at trial. The police believed he’d drowned his wife, while the defense put on evidence that she suffered from a medical condition that caused her to lose consciousness and drown. When I listened to the 911 call, I heard nothing either way that told me if Widmer was innocent or guilty. It just seemed like a typical 911 call, devoid of clues. But law enforcement somehow knew that his voice was that of a murderer staging the scene as an accident, and they argued this over and over again to the jury. The jury found this convincing and they convicted with other fairly weak evidence.
Later, when I got involved in the case after Widmer was convicted, I met with the elected district attorney to try to get him to agree to release the deceased’s bodily tissue samples so I could perform genetic testing to see if she suffered from a disorder like Long QT syndrome that could have caused her drowning. During her life, she demonstrated many of the symptoms of Long QT syndrome, such as a cleft palate, short stature, a heart murmur, and narcolepsy-like symptoms throughout her lifetime. In fact, her nickname was “Sleepy” because of her strange tendency to fall asleep in the sitting position at bizarre times, such as when the family was gathered together opening Christmas presents.
In rejecting my request for access to her tissue, the prosecutor cited the sound of Widmer’s voice on the 911 tapes and the way he behaved on the scene once the police arrived: too calm. Also, the case had recently been on Dateline NBC shortly before my meeting with the prosecutor, and Widmer had decided to be interviewed on camera from prison for the show. The prosecutor said, “After that Dateline interview, no one around here believes he is innocent anymore. You could tell from the way he came across that he did it.” I sure wish I possessed such amazing wizardry at divining the truth from one’s demeanor.
To date, the prosecutor has still not released the tissue for biological testing, and the courts have said that he doesn’t have to because there is no law in Ohio that expressly requires him to release it. Widmer remains in prison despite the fact that genetic testing could be done that could shed great light on what happened in his case.
While I don’t know whether or not Widmer is innocent, as I was not there when his wife died, I find the supreme confidence of others in his guilt—based in large part on Widmer’s demeanor—to be quite troubling, though it reflects our system’s overreliance on demeanor evidence and our disturbing overconfidence in our lie-detection abilities.