False Confessions
Not all confessions can be taken at face value. False confessions—in which individuals admit to having committed crimes that they did not commit—have been recognized for many years. False confessions can be classified in several ways, but the most generally accepted and most useful system was delineated by Kassin and Wrightsman (1985). They group false confessions into three general categories: (a) spontaneous-voluntary false confessions, in which individuals—on their own accord—come to authorities and admit to having committed crimes that they did not commit; (b) coerced-compliant false confessions, in which suspects confess to crimes in order to achieve a specific goal, such as bringing a lengthy and stressful police interrogation to an end; and (c) coerced-internalized false confessions, in which suspects come to believe—during the interrogation—that they are guilty of a crime that they did not commit and did not initially remember committing (a phenomenon similar to false-memory syndrome, see Pope, 1996).
Freud (1916) discussed false confessions and believed that unconscious guilt feelings might induce certain individuals to commit crimes in order to be punished, and some false confessions may be connected to similar psychodynamics. Some believe that false confessions, which were rampant during the 16th and the 17th-century witch hunts, can be explained by fear or the effect of torture, mass hysteria, delusions, and implanted beliefs. But even without cruelty and torture, false confessions that are later recanted—or denial of guilt with a later confession—seem to be more common than was previously thought.
Because a confession may have more impact on a jury verdict than any other type of evidence (Driver, 1968), correctly identifying a false confession is important. In fact, it has been known for years that the likelihood of gaining a conviction in a case in which the defendant has confessed is extremely high (Sutherland, 1965) since it is hard for a jury to understand, as a matter of common sense and human nature, why people would admit to committing crimes unless they were guilty.
For the past 25 years, there has been an explosion of research on false confessions (e.g., Gudjonsson, 2018; Kassin, 1997, 2008; Lassiter and Meissner, 2010). The findings have shown that there are internal and external factors that can contribute to a false confession. Internal factors involve psychological vulnerabilities of the suspect such as juvenile status, presence of mental illness, or borderline or lower intelligence. The common denominator among the internal psychological factors is that such vulnerable individuals may be susceptible to becoming confused, as well as to being easily manipulated, especially by authority figures such as the police. And because of their vulnerability, they cannot stand up for themselves and may yield to police pressure or may shift their answers during an interrogation. External factors involve the process of the interrogation itself (e.g., length, trickery, deception, threats, and coercion) and other tactics that might be used to get the subject to confess.
One of the most controversial methods of eliciting a confession is the well-known Reid Technique (Inbau, Reid, Buckley, and Jayne, 2004) which involves a nine-step method of interrogation. Given the controversy surrounding the Reid Technique—specifically the possibility of inducing false confessions—it has been prohibited in several countries but is still allowed in the United States. This technique, as well as other methods of interrogation, are sometimes based on police officers believing that they know the truth of what happened, and they increase pressure on the defendant to confess to what the authorities believe actually occurred. Many police officers believe that they can detect lying in a suspect by such behaviors as eye blinks and turning one’s head right and left, although these methods of detection are not generally taught today.
False confessions, in general, are complicated because the suspect can have all the above-noted vulnerabilities—and might even be an intellectually disabled (mentally retarded) psychotic adolescent—and still give a true confession; an individual without any such vulnerabilities can give a false confession, and even a known and proven false confessor can give a true confession. False confessions are only beginning to be understood. The research on false confession is relatively recent (since the 1990s) compared with the research on other psycholegal processes such as the reliability of eyewitnesses (which began around the early 1900s). And the various research studies of false confession are complex, nuanced, and involve issues of methodology, sampling, inferential statistics, and varying interpretations and generalizability of the findings.
There is no generally accepted estimate of the prevalence of false confessions; its prevalence, in some respects, may even be unknowable. Those individuals critical of police interrogations believe false confessions are common, while others have a much different point of view. It is not known exactly how many documented false confessions there have been in homicide cases (well over several hundred), but there are approximately 25,000 homicides a year in the United States, and many of the documented cases of false confession extend back to the 1970s.
What exactly constitutes internal psychological vulnerability also lacks consensus and clarity. For example, an individual who is overtly psychotic and becomes confused during an interrogation is different from an individual who is depressed and makes a spontaneous-voluntary false confession possibly to relieve guilt feelings. Ongoing research in this area has studied the relationship of false confessions to various internal psychological factors such as anxiety, impulsivity, social desirability, assertiveness, and self-esteem, but none of these findings and conclusions are settled or generally accepted. And a close scrutiny of much of this research—such as the relationship of false confession to a personality disorder or depression—reveals that it is unknown and has never been determined (by the researchers themselves) whether their subjects’ reported false confessions were really false or just a claim of false confession (and the subject actually committed the crime).
Given the amount of research on false confessions, it is surprising that there is relatively little research on the confession process itself. Deslauriers-Varin, Lussier, and St. Yves (2011) are an exception. These researchers studied how individuals decide to confess, and they found that those who had no prior experience in the criminal justice system, lacked social support, and believed that evidence against them was strong were most likely to confess to the authorities. Wetmore, Neuschatz, and Gronlund (2014) found that secondary confessions—an admission of guilt by the subject to a third party such as a jailhouse informant—often had as much weight with juries as primary confessions, notwithstanding their frequent unreliability. Other studies have researched confessions; however, most of these studies are in some way related to false confessions (e.g., DeClue, 2005), the effect of confessions on judgments of guilt, and the like.
A recent study (Feliciano, Robins, Fletouris, Felps, Schlesinger, and Craun, in press) examined confessions in intimate partner homicide. The purpose was to study the confession process itself; specifically, the different ways individuals admit their guilt because a confession is not always made to the police. In their nonrandom national sample of 70 intimate partner homicides, the most common method of confession was to a non-law- enforcement individual (e.g., a family member or an emergency medical technician); the second most common method was a voluntary-spontaneous confession to a member of law enforcement without being challenged; the third most common was an evidence-based confession during an interrogation. Seven of the intimate partner homicide offenders confessed in a suicide note. Some offenders confessed but, at the same time, provided a defense, such as acting in self-defense or in response to psychotic delusions. The researchers are currently studying confessions in sexual murders and in other types of homicides because it seems clear that the way individuals confess depends on the type of offense they committed. Preliminary findings indicate that over one-third of sexual murderers do not confess at all. It is hoped that these findings will help us to gain a better understanding of the way people confess in general, in order to place false confession in context.
Many researchers have reported interesting and illuminating cases of false confessions. For instance, Kennedy (1961) reported the case of a false confession, which was later retracted but that nevertheless resulted in the execution of an innocent man. The man was convicted of murdering his wife and daughter, who had actually been killed by a sexual murderer. The accused—illiterate and confused—explained that he had confessed because he was upset and fearful of the interrogation process. This case illustrates how, during an interrogation, perhaps because of fatigue or anxiety, a suspect may suddenly blurt out a confession, and investigators (also fatigued) may convince themselves that the confession is genuine. In another case, reported by MacDonald (1961), 30 men spontaneously falsely confessed to the murder of a young woman, whose nude body had been cut in half at the waist and mutilated with multiple stab wounds. Revitch and Schlesinger (1981) reported three cases of false confession; one was described as coerced, one as internalized, and one as spontaneous—descriptions that coincided with the Kassin and Wrightsman (1985) classification system published several years later.
Just as subjects may temporarily delude themselves into believing they are guilty when they are not, they may also delude themselves into believing they are not guilty when they actually are, particularly if the homicide was ego dystonic and explosive. It is, therefore, critical that the forensic examiner be aware of the various forces, both within the person and in the interrogation process itself, that can lead to false confessions.
Case 2.7: Coerced-Compliant and Internalized False Confession
B.H., 24 years old, had been living for about a year with a young woman and her two children (a boy and girl) aged 7 and 8. It was a tumultuous relationship: B.FI. was described as having a violent temper directed toward the woman and sometimes toward the children. The defendant was known to the police for fighting and other types of disorderly conduct, and he also had served a prison sentence for burglary.
The bodies of both children were found in the basement of their apartment building by a tenant. The 8-year-old girl was found bound, naked, with her underwear shoved in her mouth. The little boy was killed by the offender banging five long nails into his head. The children’s mother told the authorities that she thought B.FI. was capable of carrying out the homicides. During the course of the investigation, B.H. gave three statements to the police. The first statement was a detailed denial of guilt. The next day, he gave a second statement and changed some of what he had previously told the police but continued to deny involvement in the murders. Later that same day, B.H. gave a third statement implicating himself in the two murders and the sexual assault on the female child. B.H. then denied guilt when he was brought to the county jail within hours after he implicated himself.
The defendant was found guilty by a jury of aggravated manslaughter although the trial began as a capital murder case. It became clear that members of the jury found problems in the defendant’s statements and the investigation itself, but they nevertheless convicted him of the lesser charge of aggravated manslaughter.
B.H. underwent extended periods of interrogation. He was interrogated on the first day for about 14 hours after not having slept the prior night. On the second day of questioning,
B.H. was with the police for about 16 hours and was interrogated for about 12 hours. His five-page typed statement, in which he implicated himself, took 5 1/2 hours for the officers to complete. The specific questions B.H. was asked are not known, and neither are the answers he gave over the lengthy period of interrogation, as his statement and the police reports are not a verbatim question-and-answer account of what was said.
It seems that the detectives believed that the defendant was guilty because of what they considered to be signs of deception including eye blinks, grinning, and “lack of appropriate emotional response.” Although B.H. did not know certain facts about the crime, the interviewers concluded that the suspect was being “deceptive” and not forthcoming. Rather than concluding that B.H. may not have known the facts because he was not involved in the murders, the interviewers incorrectly concluded that the suspect was simply lying or “playing games.”
The police were certain they knew exactly what happened. For example, when asked how the nails were driven into the child’s head, the suspect first said, “Crowbar?”, “Hammer?” and then said “chair.” When the police pointed to a rock they had placed on a table, B.H. said a rock was used to bang in the nails. The detectives assumed that B.H. knew the correct answers because he committed the crime. When B.H. asked the officers what would happen if their investigation determined that someone else committed the crime, the police told him no one else was involved. Many individuals, while giving a statement that turns out to be a false confession, often try to elicit the “correct” answers from the police; in retrospect, B.H. seems to have been attempting to elicit answers. But the detectives thought “he was testing our knowledge of the crime.”
In many false confession cases, detectives believe that the suspect is guilty, and findings that point toward innocence (or absence of guilt) are often minimized or rationalized. This minimization occurred on a number of occasions with B.H. For example, B.H. gave two different versions of how the children were killed. This discrepancy was not considered a red flag to be further investigated. Only the “facts” linking him to the murders were the focus of the questions. In addition, the suspect said he did not use any electrical wire, while the autopsy showed electrical wire was used. B.H. also said that he vomited and left a Canadian Club bottle close to the crime scene, but neither was ever found.
And when B.H. asked the detectives what would happen if the FBI analysis of forensic evidence showed that he did not commit the crime, he was told that the analysis would show that he did commit the crime. B.H. was supposed to have known information that only the offender would have known—specifically, the child’s sexual assault and that her underwear was placed in her mouth. When asked at first, he said he did not know if or how she was assaulted. He then said the assault took place on the couch and her underwear was placed in her mouth. But the specific questions asked of him were not recorded, so it is unknown whether the police mentioned the couch and underwear to him at some prior point.
A neighbor who knew that the children were alone the day they were killed was not extensively questioned because the detectives believed that B.H. committed the crime. Moreover, B.H. (in his first statement) lied to the police about being in a fight, failed the polygraph (even though the polygrapher said the results should not be relied on given B.H.’s lack of sleep), and gave times (regarding his whereabouts and when phone calls were made) different from the times supplied by other individuals. B.H. also said he ate hotdogs and did not immediately call the police when the children were discovered missing, behavior the detectives believed was inconsistent with how an innocent person in his position should have behaved.
In truthful confessions, there are frequently false statements; in false confessions, the offenders’ statements are often accurate in details because many times they are getting the details (inadvertently) from the police. It is not uncommon for individuals not to know exactly when they carried out certain activities. Because B.H.’s timeline was different from those of others does not mean B.H. was lying; it is equally possible that the other individuals were incorrect in their times. B.H. said he did not call the police because the children’s mother told him that she would call. His eating hotdogs during the time the children were missing was interpreted by the detectives to be an indication of his callousness.
According to various records, B.H. was in a number of institutions as a child and adolescent, in addition to having lived in foster care. While serving his sentence for the crime for 23 years, B.H. always proclaimed his innocence and had a number of DNA tests done, which were inconclusive. Finally, an advanced analysis of DNA was carried out, and it was determined that B.H.’s DNA was not at the crime scene. Instead, his neighbor’s DNA was at the scene, including on cigarette butts as well as on the victims’ bodies. When the police went to arrest the offender, he was already in prison for a number of rapes, which were committed in a manner similar to the way the children were killed. He was interviewed by the police and revealed his sexual arousal to the 8-year-old female victim. Unfortunately, this man died in the county jail just prior to his trial, so many of the questions that were raised in the initial case were not fully resolved.
This seems to have been a case of a coerced-compliant false confession. However, the defendant’s attorney believed that during the interrogation itself B.H. may have come to believe that he actually did carry out the murders. His doing so would possibly indicate a coerced-internalized false confession. This case is interesting because both types of false confession seem to be present.
Case 2.8: Spontaneous-Voluntary False Confession
A 29-year-old woman stayed at a welfare hotel with a 32-year-old male. The male paid the rent, and she had sex with him as her contribution. They got into an argument, which resulted in the female fatally stabbing the male. Following the homicide, she left the hotel and, after a few days, checked into an alcohol rehabilitation program in another town.
The victim’s body was found by his homeless brother, who came to visit periodically. He called the police and spontaneously confessed to them that it was he who killed his brother as well as the woman who also resided there. He said that he stabbed his brother about nine times and reported being the victim of physical and mental cruelty by his brother. He also said that he stabbed the woman four times and discarded her body in a nearby dumpster. When the authorities searched the dumpster, they did not find the remains of a female’s body.
As the investigation proceeded, police eventually located the woman, arrested her, and took her to the county jail. They said to her, “There’s your co-defendant,” the victim’s brother. At this point, the woman told the police she had no co-defendant and that she killed the man herself during an argument.
When the police confronted the brother with the fact that the woman was alive and said she acted alone, he told the police he concocted the entire confession because “I was telling you what you wanted to hear.” It was later determined that the victim’s brother lied to the police in order to get a place to stay, as the weather was turning colder and he was tired of being homeless. This false confession indicates that individuals have many motives for falsely confessing, and many cases do not neatly fit into one classification category or another, as confession cases—like all criminal cases—are complicated, with often unique sets of facts and circumstances.
