During the 1990's a public debate erupted over widespread accusations of child sexual abuse and the credibility of those allegations. Psychologists who were drawn into this debate were often unaware of the broader context of sexual politics in which the issue had arisen, the ways in which their discipline had historically been compromised by these politics, and the unwitting role that they themselves had occasionally played. As a consequence, they often managed to give the public only a partial and distorted understanding of the complex issues involved in rape, trauma and memory. To consider these issues in more detail it will be helpful to begin with some history.
In the 1960s and 70s a number of feminist writers began to expose the longstanding and widespread existence of sexual abuse, sexual harassment and rape in the United States. Encouraged by this new openness, individuals began to reveal their own encounters with these traumatic events. In the 1970s, in particular, psychotherapists observed a dramatic increase in patients reporting experiences of past sexual victimization. Some of these patients, suffering from years of anger, depression, or self-blame, focused their therapy quite intentionally on their past traumatic experiences. Others focused initially on different issues but found that these issues led, inevitably, to memories of past sexual traumas; and in some cases memories of past abuse seemed to emerge spontaneously in patients who had previously not recalled them. This phenomenon of spontaneous recollection of previously unremembered trauma - known as "recovered memory" - became a focus of special attention for some therapists, who began actively looking for previously unremembered traumatic experiences, and, in some cases, using a variety of techniques to elicit recovered memories, including imaging, suggestion and hypnosis.
Over the next two decades, as many people went public with their experiences of sexual abuse, a number of cases involving recovered memories gained a great deal of public attention. Some of these were cases that had gone to court as alleged perpetrators of sexual abuse were charged criminally or became involved in lawsuits, and a few of them involved recovered memories that strained credibility. In one widely publicized case, for example, a young woman named Beth Rutherford had gradually recalled in therapy that her father, a clergyman, had raped her repeatedly as a child, resulting in her becoming pregnant and undergoing a coat-hanger abortion. During the legal process, it was determined that these events were medically impossible--she was still a virgin and her father had had a vasectomy when his daughter was 4 years old. Moreover, Rutherford revealed that from the start of her treatment she had been actively encouraged by her therapist to recall sexual abuse. Several other cases of recovered memories were similarly problematic and were eventually discredited and/or recanted. In the early 1990's the phenomenon of "false memory" began to gain widespread recognition, and critics of recovered memory began to allege that careless therapists had caused an epidemic of false memories and false accusations of child abuse.
Psychologists now became increasingly entangled in the public debate about recovered memories and false memories, sometimes in the role of therapists and sometimes in the role of legal experts. What the public did not appreciate, however, and what the legal system was soon to find out, was that the discipline of psychology was itself seriously divided on the status of recovered memories. Furthermore, the division ran along the oldest and bitterest fault line in the field of psychology, dating back to the earliest days of the discipline and involving deep differences over theory, evidence and methodology.
On one side, the concept of recovered memory had been endorsed primarily (though not exclusively) by clinical psychologists, particularly those who had practiced in the psychodynamic tradition going back to Sigmund Freud and psychoanalysis. This tradition had drawn heavily on the notion of intrapsychic conflict and the ability of the mind, under certain stressful conditions, to disown some of its own thoughts, feelings or memories, a mechanism known as "repression." The repressed aspects of experience might be reclaimed at a later time by a process of self-exploration like psychotherapy. This psychodynamic perspective and the accompanying idea of repression had had a strong intuitive resonance for many people and had come to influence thinking about psychopathology and psychotherapy for nearly a century. Even schools of psychotherapy that eventually broke with Freud and the psychoanalytic tradition generally retained some commitment to the idea that parts of experience could be lost to consciousness and later regained. And the notion seemed to be supported not only by numerous case studies but also by the personal accounts of many individuals, including those who had revealed their own encounters with sexual trauma, self-alienation and, eventually, healing.
On the other side of the divide were many academic research psychologists, ever-skeptical of Freud and psychoanalysis, and determined to make psychology into a hard-nosed discipline modeled after the natural sciences by emphasizing the need for controlled experimentation. This tradition had begun in studies of animal behavior and had more recently moved on to the experimental study of cognitive processes, including the acquisition, storage and retrieval of memories. Just as clinical psychologists had aligned themselves with the psychodynamic tradition and the popular notion of psychotherapy, so academic research psychologists had aligned themselves with the popular notion of "laboratory science" and had gained uncontested dominance in departments of psychology in colleges and universities. Many academic psychologists regarded the Freudian construct of repression - and, by implication, the idea of recovered memory - as fundamentally unscientific since this phenomenon could not be reliably reproduced in the laboratory. For example, after conducting an extensive review of the research literature, one widely cited academic psychologist flatly stated that sixty years of experimental investigation had produced "no evidence for repression" (Holmes, 1990, p. 97). Others argued that traumatic events are remembered more vividly than ordinary ones, making repression implausible (e.g., Roedigger & Bergman, 1998).
Furthermore, academic researchers had been gathering experimental evidence supporting the existence of false memory. Laboratory investigations had already demonstrated that memory could be manipulated rather easily by using suggestive or leading questions (e.g. Loftus and Palmer, 1974 ). When false recollections of child abuse became public, additional experiments were done demonstrating that it was possible to implant completely false memories by using suggestive techniques much like those used by some therapists. In one experiment, for example, researchers told participants that one of their parents had supplied a description of a childhood experience in which the participant had been lost in a mall (in reality, however, the parents had confirmed that no such event had ever occurred). When actively encouraged to try to remember the event, 25% of the participants eventually reported remembering the fictitious episode (Loftus and Pickrell, 1995). Even more remarkably, the participants who internalized such false memories often embellished them and came to be quite certain that the memories were real. Evidence like this strongly supported the widely shared suspicion that recovered memories of child sexual abuse were being regularly manufactured in psychotherapy sessions by overzealous therapists using inappropriately suggestive techniques.
For many people, the results of these experiments, along with the highly publicized cases of discredited recovered memories, sealed the case. After the mid-1990s, media coverage became decidedly skeptical toward reports of recovered memories and soon after this the entire issue faded from the news. The public was left with the impression that recovered memories were at best questionable and at worst a dangerous pseudoscientific fiction, now thankfully disproven by science. This, however, was far from accurate. In reality, the original concerns about false memories subsequently generated a great deal of new research, and the results of that research have uncovered a much more complicated picture.
To begin with, a number of researchers have focused on the relationship between trauma and memory using prospective research designs. These designs start by locating individuals for whom there is documented evidence of some kind of trauma (e.g., hospital or legal reports) and then examining the functioning of their memory in relation to that trauma. Since they start with documented evidence of sexual abuse or some other trauma, prospective studies are particularly useful in that they eliminate the question of whether or not the trauma actually occurred. The results of these prospective studies have consistently indicated that while trauma is, indeed, remembered vividly by most people, there are many people for whom this is not the case. For example, two prospective studies have looked at recently traumatized adults and both have found subgroups of trauma survivors who developed significant - sometimes severe - gaps in recollection, that were, however, actually or potentially reversible at a later time (Mechanic et al.,1998; Yovell et al., 2003). Consistent with this, three other prospective studies in which survivors of child sexual abuse have been interviewed 13-20 years after the trauma have each found that a significant proportion of those who were interviewed (16-38%) denied any memory of the (documented) childhood sexual abuse (Goodman et al., 2003; Widom et al., 1997; Williams, 1994). Moreover, in one of these studies (Williams, 1994) participants who did remember their sexual abuse were asked if there had been a period of time during which they had not remembered it, and 16% responded that they had gone through such a period.
The findings of these prospective studies have also been supported by findings from a much larger number of retrospective studies (in which participants' own reports of trauma serve as the focus). Dozens of retrospective studies have surveyed persons reporting various kinds of traumas, and these surveys have, again, repeatedly found that a significant subgroup of respondents report having gone through a period in which they could not remember the trauma (e.g., Brewerton, Dansky, Kilpatrick & O'Neil, 1990; Dorado, 1999; Feldman-Summers & Pope, 1994; Dalenberg, 1996; Elliott, 1997; Mechanic, Resnick & Griffin, 1998; Melchert, 1999; Wilsnack et al., 2002). The size of this subgroup ranges from 10-50%, depending on the study; and even committed critics of recovered memory have had to acknowledge the ubiquity of such reports (e.g., Loftus, Garry & Feldman, 1994), often encountering them in their own research (e.g. Means & Loftus, 1991). Moreover, in contrast to the widely publicized descriptions of therapists' influence, individuals in these studies typically report recovering memories of sexual trauma spontaneously or in response to an unexpected cue not necessarily associated with psychotherapy - for example, returning to a place where abuse had occurred or seeing a person or place associated with or similar to one that was involved in the abuse.
Given the retrospective design of these studies, it might be questioned whether the memories reported by the participants represent events that had actually occurred. It is often possible to address this question, at least partially. When past sexual abuse is alleged, a variety of different kinds of corroboration are potentially possible, including identification of witnesses or other victims, location of documentary evidence such as medical records, and even obtaining confessions by perpetrators. A number of studies have attempted to obtain such corroboration and in a large proportion of cases it has been possible to do so. In several studies, researchers have reported obtaining corroboration for anywhere from 24-74% of reported recovered memories (Cameron, 1996; Dalenberg, 1996; Feldman-Summers & Pope, 1994; Herman & Harvey, 1997; Herman & Schatzow, 1987; Schooler, 2001; van der Kolk & Fisler, 1995). In these retrospective studies, and in a variety of other contexts, corroboration for recovered memories has often been quite strong. For example, Ross Cheit, a professor of political science at Brown University and himself a survivor of childhood sexual abuse, has collected an archive of over 100 cases in which claims of recovered memories were subsequently substantiated by such evidence as contemporary documents, eyewitness accounts, material evidence, legal verdicts and confessions (Cheit, 2005).
The evidence from these studies cumulatively suggests that for many people trauma is, indeed, followed by a period of memory loss and subsequent memory recovery. It should be noted that this does not contradict the evidence that false memories can also be produced by strong suggestive pressures, either inside or outside of psychotherapy. In fact, there are reasons to believe that both recovered memories and false memories are produced by some of the same underlying mechanisms of memory; and most psychologists now accept that both phenomena - recovered memories and false memories - can and do occur (Dalenberg, 2006). Nevertheless, a small group of academic psychologists remains skeptical about all recovered memories, and more specifically about the mechanism of repression.
For example, Pope and Hudson (1995) have argued that the denial of past sexual abuse by participants in prospective studies may be nothing more than a simple preference not to disclose the information. They cite a prospective study of survivors of physical abuse (della Femina, Yaeger & Lewis, 1990) in which participants who denied remembering the abuse were later questioned in a "clarification interview" and appeared (in Pope and Hudson's view) to admit that they had actually remembered but not reported the abuse. On closer examination, however, Pope and Hudson's analysis appears deficient in some important respects. While some victims of past abuse may simply withhold the information, there are reasons to believe that genuine forgetting also occurs in a significant number of these cases. In the della Femina study, for example, participants were asked to clarify the discrepancy between their negative accounts and the documented abuse but not about their memory. Even so, their explanations for their previous negative responses included reports of deeply troubled feelings from which the participants had actively attempted to escape psychologically. For example, one of the participants burst into tears, saying "I wanted to forget. I wanted it to be private. I only cry when I think about it" (p. 229). The researchers themselves conclude that such reports are consistent with previous clinical accounts of motivated forgetting - in this case involving "an effort to use 'selective inattention' by suppressing the awareness of abuse and attending only to positive aspects of an experience" (p. 230). Other prospective studies of childhood sexual abuse have avoided clarification interviews for ethical reasons, but when participants in retrospective studies have been asked about forgetting and remembering in relation to trauma, they often give responses that, consistent with della Femina et al.'s conclusion, indicate that intentional or unintentional exclusion from consciousness was involved in their failure to recall (for example, see Dalenberg, 2006, pp. 280 & 288).
Other critics of recovered memory take a different tack, arguing that the loss and recovery of traumatic memories is due not to repression and recovery but to "ordinary" forgetting and remembering. For example, Kihlstrom (2005) states that "the recovery of a forgotten trauma may be no different in kind from the recovery of one's memory for where one put the car keys, or the name of one's third grade teacher" (p. 235). But one can hardly maintain such an interpretation in the face of actual accounts of persons who have recovered memories, which frequently include descriptions of a sudden and/or piecemeal return of memories accompanied by disturbing and disequilibrating emotions. To give one example, Ross Cheit's (2005) archive includes an account by a 35-year-old woman who, while driving, heard a radio report about the war in Bosnia that included a young girl screaming. She reports that the next thing she knew she had pulled over and was crying uncontrollably. She returned home and "dashed into the attic to get my old doll out of the loft" and started to bathe and wash "because I felt dirty." She goes on to describe how she feared she was going insane and saw a doctor who referred her to a therapist. After six months of further flashbacks she located social service records that confirmed that she had been sexually abused and tortured before she was adopted at age five. Presumably, even Kihlstrom would agree that a recollection like this is qualitatively different from remembering "where one put the car keys."
The critics of recovered memory are resistant to acknowledging it, above all, because to do so might be seen as legitimizing the Freudian concept of repression. When one examines the grounds for their concerns, however, the rejection of the concept of repression begins to look less like a judgment based on evidence and more like a determination to disregard any and all evidence that does not fit a preconceived and untenably narrow notion of science. An example is the earlier quoted statement by Holmes (1990) that previous research had produced "no evidence" for repression. In reality, Holmes's review identifies a number of studies that are either consistent with or supportive of repression - including studies by Holmes himself. But what Holmes is referring to is the fact that no researcher has succeeded in identifying any technique that can reliably produce a completely unambiguous specimen of repression in a controlled laboratory setting. In his review, Holmes disregards many previous studies that suggest the existence of motivated forgetting because their results cannot be definitively identified as repression, as distinct from other psychological processes, including defense mechanisms like denial, suppression, or negation. Yet Freud repeatedly emphasized the need to be flexible about theoretical constructions and he often used the word "repression" as a generic term to cover a variety of related processes (Rabin, 2010). Other researchers who have remained more faithful to Freud's actual conception of repression have discovered substantial evidence for the process (Bornstein, 2003; Weston, 1998; Weston & Gabbard, 1999). Most psychologists who have researched memory and its distortions currently believe that a variety of psychological mechanisms may be involved, some of which are similar to Freud's notion of repression in the narrowest sense and others of which encompass various neurological and cognitive processes, including psychodynamic mechanisms of motivated forgetting (i.e., of repression in the broader sense) (e.g., see Dalenberg, 2006). The issue is not the existence of repression in a narrow and technical sense but whether memory is distorted, transformed, and sometimes lost altogether in reaction to overwhelming trauma, and whether it can subsequently be retrieved; and for this there is abundant evidence.
The resistance of some psychologists to recognizing the disruption and recovery of memory appears to be driven by a complex of factors, including legitimate questions about the phenomenon, methodological problems, professional egotism, and academic territoriality. But it also seems to be part of a larger pathological dynamic in the field of psychology. While academic researchers played an important and at times courageous role in exposing the excesses of therapists whose simplistic conceptions of repression led them to implant false memories of sexual abuse in their patients, many of the same researchers have been reluctant to confront the equally simplistic dismissal of the complex dynamics of memory and trauma by some of their colleagues. I am aware of no contemporary psychologist who would complacently ignore the abundant research of the past twenty years supporting the possibility of creating of false memories; but it is not difficult to find critics of recovered memory who display a willingness to disregard equally abundant evidence supporting the disruption and restoration of traumatic memory, often using sweeping and extreme rhetoric, such as that recovered memory is a "myth" (Loftus, 1996) or "folklore" (McNally, 2004), or that "there just isn't any evidence" (Myers, 1996; quoted in Letourneau & Lewis, 1999, p. 257).
Most academic research psychologists know that there is, in fact, substantial evidence for the loss and recovery of traumatic memory. Yet in some cases they or their colleagues display a minimizing and evasive attitude toward the data. I am not speaking here about an appropriate scientific prudence or caution, but rather about an acquiescence to a pervasive and disdainful culture of dismissiveness within which complex social issues that elude complete experimental control are treated as if they were nonexistent, or at best irrelevant, even in the face of obvious evidence to the contrary. This is nowhere more evident than in introductory psychology textbooks, a fact which was pointed out in paper by Letourneau and Lewis (1999). The authors surveyed 24 of the most commonly used such textbooks and found them to be systematically and seriously biased in their treatments of sexual abuse, false memory, and recovered memory. The survey found numerous factual errors, some of them egregious (including the statement by Myers, above). But far more significant in the textbooks was a persistent minimizing of the problem of childhood sexual abuse. Most of the books treated this topic, if at all, only in relation to memory, and then typically in narratives that focused primarily on the danger of creating false memories leading to the prosecution of innocent individuals. Letourneau and Lewis cite statistics documenting that even at the height of the false memory era unfounded accusations of child abuse constituted a miniscule proportion of reported cases, while most actual child abuse cases went unreported. Noting that these facts were inadequately covered in all of the textbooks, they conclude that the books might "lead readers to believe that prosecution of alleged child molesters occurs too frequently (or too rashly), when the data suggest exactly the opposite" (p. 257).
How did academic psychologists become so dismissive toward a problem like sexual abuse? As noted above, part of the answer lies in the adherence of research psychology to the model of investigation by controlled experimentation and the inability of such experiments to fully encompass complex dynamics like those among abuse, trauma and memory. To some degree this is an unfortunate but inevitable concomitant of the attempt to gather objective knowledge in social science. But researchers' adherence to the experimental model is so relentless and, at times, so pervaded by denial and openly disparaging of the social issues that are bypassed in the process, that a more complete explanation seems necessary. I can only infer that many researchers are drawn to the model of controlled experimentation partly because it provides a safe and insulated arena in which they need not confront the difficult social issues that lie behind the phenomena they are studying and that being directly confronted with nonexperimental but important data relevant to these issues is experienced as an unwanted intrusion by emotional people who are preoccupied with unseemly matters and disturbing the business of science.
The sexual abuse of children is a particularly painful issue for a culture like ours to face. It threatens our self-image as a civilized society, especially when the abuse of children so often occurs in socially well-positioned families. For this reason, data and arguments supporting the existence of widespread sexual abuse are understandably met with disbelief, resistance and anger by a large part of the public, particularly those who feel threatened by acknowledgement of this problem; conversely, data and arguments that support denial of the problem are readily welcomed by the same segment of public opinion. In the controversy about recovered memory and false memory, therefore, it is not entirely surprising that both the media and the public have come to focus disproportionately on false memories of child sexual abuse. What is more surprising, though, is that research psychology has so readily acquiesced in this trend when it is well known that both false memories and recovered memories are supported by substantial evidence, although of somewhat different kinds.
The struggle to confront child sexual abuse has been a long and difficult one. Herman and Hirschman (2000) have pointed out that Western society has gone through periodic cycles in which widespread sexual abuse has been discovered and then subsequently denied. The first such cycle occurred when physician Ambrose Tardieu documented extensive child sexual abuse in France in 1873. Shortly after Tardieu's death a few years later some of his successors began to express the opinion that the children had been lying, and the matter faded from public attention. A generation later, Sigmund Freud publicly presented his "seduction theory" which attributed widespread psychopathology in girls to child sexual abuse. Freud's theory was met with sharp criticism and a short time later he reinterpreted the accounts of his patients as evidence of unconscious oedipal fantasies. In the 1940s Alfred Kinsey discovered that a large proportion of adult women reported childhood sexual abuse. When he subsequently published this finding, he minimized its significance, interpreting the abuse as normal sexual behavior. Finally, as noted at the beginning of this article, in the 1960s and 1970s a large number of women came forth in the context of the feminist movement to share their own personal experiences with sexual abuse. It now appears that another move toward denial has followed these reports. How active a role psychologists will play in perpetuating this denial remains to be determined.