In 1990, a landmark case went to trial in Redwood City, California. The defendant, George Franklin Sr., 51 years old, stood trial for a murder that had occurred over 20 years earlier. The victim, eight-year old Susan Kay Nason, was murdered on September 22, 1969. Franklin's daughter, Eileen, only eight-years old herself at the time of the murder, provided the major evidence against her father. What was unusual about the case is that Eileen's memory of witnessing the murder had been repressed for over 20 years. Eileen's memory did not come back all at once. She claimed that her first flashback came one afternoon in January 1989 when she was playing with her two-year old son, Aaron, and her five-year old daughter, Jessica. At one particular moment, Jessica looked up and asked her mother a question like "Isn't that right, Mommy?" A memory of Susan Nason suddenly came back. Eileen recalled the look of betrayal in the eyes of Susie just before the murder. Later, more fragments would return, until Eileen had a rich and detailed memory. She remembered her father sexually assaulting Susie in the back of the VW van. She remembered that Susie was struggling as she said "No don't" and "Stop." She remembered her father saying "Now Susie" and she even mimicked his precise intonation. Next, her memory took the three of them outside the van, where she now saw her father with his hands raised above his head with a rock in them. She remembered screaming. She remembered walking back to where Susie lay, covered with blood, the silver ring on her finger smashed. Eileen's memory report was believed by her therapist, by several members of her family, and by the San Mateo County District Attorney's office which chose to prosecute her father. It was also believed by the jury, which convicted George Franklin, Sr. of murder. The jury began its deliberations on November 29, 1990, and returned a verdict the next day. Impressed by Eileen's detailed and confident memory, they found her father guilty of murder in the first degree. Eileen's detailed and confident memory impressed a number of people. But is her memory authentic? Did she really witness the murder of her best friend 20 years earlier?

Popular articles. Long-repressed memories that return after decades, often while a person is in therapy, have become highly publicized via popular articles. In 1991, Roseanne Barr Arnold's story was on the cover of People Magazine. Her memories were of her mother abusing her from the time she was an infant until she was 6 or 7 years old - memories that returned in therapy ("A star cries incest", 1991; Darton, 1991). Barr Arnold was not the first case to capture the cover of People Magazine that year. Just three months earlier, People had also reported a story about former Miss America Marilyn Van Derbur, who repressed any knowledge of sexual violation by her father until she was 24 years old and told the world about it after her father died ("The darkest secret", l99l; Darton, 1991). Highly publicized cases involving memories that recently sprung into consciousness were told repeatedly in scores of popular articles (eg., in the Washington Post, Oldenberg, 1991; in the Los Angeles Times, Ritter, 1991; in Seventeen Magazine, Dormen, 1991; In Glamour Magazine, Edmiston, 1990; in Newsweek, Kantrowitz, 1991; in Time Magazine, Toufexis, 1991).

Letters. Scores of spontaneously-written letters from strangers also describe the emergence of memories. Some letters were written (to this writer) by people who had been accused of abuse by their children. A 75-year old physician from Florida wrote desperate to understand why his 49-year old daughter was suddenly claiming he had abused her during her early childhood and teen years. A woman from Canada wrote about the nightmare of being "falsely accused of sexual abuse by our 30-year old daughter." A woman from Michigan wrote about her 38-year old daughter who "after a year of counseling now accuses us of abuse... very much like Roseanne Barr and the former Miss America, Marilyn Van Derbur."

A couple from Texas wrote to tell about their youngest son who had accused the couple of abusing them long ago. One letter from a mother in California well expresses the pain:

Legal cases. Another development after the Franklin conviction was that lawyers started calling to obtain assistance with a puzzling new type of legal case. For example, one case involved a 27-year old San Diego woman (KL) who began to have recollections of molestation by her father (DL), recollections that were repressed but then were later brought out through "counseling and therapeutic intervention" (Lofft v Lofft, 1989). The daughter claimed that her father routinely and continuously molested and sexually abused her, performing "lewd and lascivious acts, including but not limited to touching and fondling the genital areas, fornication and oral copulation." Her earliest memories were of her father fondling her in the master bedroom when she was three years old. Most of her memories appeared to date back between the ages of 3 and 8. She sued her father for damages for emotional and physical distress, medical expenses, and lost earnings. She claimed that due to the trauma of the experience she had no recollection or knowledge of the sexual abuse until her repression was lifted, shortly before she filed suit.

A few years ago plaintiffs like KL who claimed they were survivors of childhood sexual abuse would have been barred from suing by the statute of limitations. Statutes of limitations, which force plaintiffs to initiate claims promptly, exist for good reason; they protect people from having to defend themselves against stale claims. They exist in recognition that with the passage of time, memories fade and evidence becomes more difficult to obtain. Succinctly and articulately put, the primary purpose of statutes of limitation is to prevent "surprises through the revival of claims that have been allowed to slumber until evidence has been lost, memories have faded, and witnesses have disappeared" (Telegraphers v. Railway Express Agency, 1944). When much time has passed, defendants find it hard to mount an effective defense. While a statute of limitations on child sexual abuse might be suspended until a victim reaches the age of majority or a few years beyond, it previously would not typically have extended until the age of 27, for example, when KL first recalled her abuse.

Then, in 1989, things changed for plaintiffs in the State of Washington. Legislation went into effect that permitted people to sue for recovery of damages for injury suffered as a result of childhood sexual abuse any time within three years of the time they remembered the abuse (Washington, 1989; See also Petersen v. Bruen, 1990). The legislature invoked a novel application of the "delayed discovery doctrine", that essentially says that the statute of limitations does not begin to run until the plaintiff has discovered the facts that are essential to the cause of action. Traditionally the delayed discovery doctrine has been used in the area of medical malpractice. A patient who discovers during a physical exam that his abdominal discomfort is caused by a surgical instrument left after an appendectomy performed 20 years earlier can sue since he could not have discovered the facts essential to his harm until he had the physical exam. Analogizing to the surgical instrument that was hidden from the patient until an exam makes its presence known, so the memory for abuse was hidden away until it too is discovered and the plaintiff now possesses the facts that are essential to the cause of action.

Within three years of enactment of the Washington statute, 18 other states3 enacted similar legislation allowing for the tolling of the statute of limitations. Many other states introduced bills in the 1991- 92 legislative sessions that would achieve the same result, or they have began studying similar legislation. As a consequence, repressed memories now form the basis for growing number of civil law suits. As one writer put it, "Such wholesale forgetting - or more precisely, the eventual remembering - is forcing society to grapple in unaccustomed ways with the old problem of child molestation" (Davis, 1991, p 81). Increasing numbers of women, and also some men, are coming out of therapy with freshly retrieved memories of abuse. They sue for damages rather than file criminal complaints, because criminal charges are often too difficult to prove (Davis, 1991). In a few states (eg. Wyoming) they can a bring criminal charges, and moves are afoot to change laws in more states to permit criminal prosecutions to go forward too.

As a consequence, juries are now hearing cases in which plaintiffs are suing their parents, relatives, neighbors, teachers, Church members, and others for acts of childhood sexual abuse that allegedly occurred 10, 20, 30 even 40 years earlier. Juries and judges are learning about repression of memory. Juries and judges are learning about newly- emerged memories of molestation not only in the United States, but also in Canada, Great Britain, and other parts of the world. Many interesting questions leap to mind about repressed memories. Chief among them is how common are claims of repressed memory? How do people in general and jurors in particular react to claims of recently-unburied repressed memories? What are the memories like? How authentic are they?

How common are claims of repressed memory? There is little doubt that actual childhood sexual abuse is tragically common. Even those who claim that the statistics are exaggerated, still agree that child abuse constitutes a serious social problem (Kutchinsky, 1992). I do not question the commonness of childhood sexual abuse itself, but ask here about how the abuse is recalled in the mind of grown adults. Specifically, how common is it to repress memories of childhood sexual abuse? Claims about the commonness of repressed memories are freely made; It is typical to read estimates such as "most incest survivors have limited recall about their abuse" or "half of all incest survivors do not remember that the abuse occurred." (Blume, 1990, p. 81). One psychotherapist with 18 years of experience claims that "Millions of people have blocked out frightening episodes of abuse, years of their life, or their entire childhood." (Fredrickson, 1992, p. 15). Later she tells readers that "sexual abuse is particularly susceptible to memory repression." (p. 23). Beliefs about the commonness of repressed memories are expressed not only by those in the therapeutic community but also by legal scholars who have used these beliefs to argue for changes in legislation. For example Lamm (1991) argued in favor of legislation that would ease access to the courts for victims of childhood sexual abuse. She applauded legislation such as that enacted in California in 1991 that allows victims, no matter how old they are, to sue within three years after discovering their injuries, or eight years after reaching majority, whichever date occurs later. As part of her argument that victims should have more time to file claims against their abusers, she expressed a view about repression: "total repression of memories of abuse is common" (p. 2198).

Despite the confidence with which these assertions are made, there are few studies that provide evidence of the extent to which repression occurs. One study (Briere and Conte, in press) sampled 450 adult clinical clients who had reported sexual abuse histories. Therapists approached their individual clients or group clients with this question: "During the period of time between when the first forced sexual experience happened and your l8th birthday was there ever a time when you could not remember the forced sexual experience?" The main result obtained in this largely female (93%) largely Caucasian (90%) sample was that 59% said "yes". A "yes" response was more likely in cases involving violent abuse (physical injury, multiple perpetrators, fears of death if abuse was disclosed) as opposed to nonviolent. Reported amnesia was more likely with early molestation onset, and longer abuse, and greater current symptomatology. Based on these findings, the authors concluded that amnesia for abuse was a common phenomenon. The Briere and Conte result has been taken by others as evidence for the widespread extent of repression. For example, Summit (1992) interpreted the 59% "yes" rate as evidence that this proportion of people "went through periods of amnesia when they were not aware of their prior abuse." He used the finding to support the commonness of childhood dissociation. (p. 22).

One problem with the Briere and Conte estimate is that it obviously depends on how the respondent interprets the eliciting question. A "yes" response to the "...was there ever a time..." question could be interpreted in a variety of ways other than "I repressed my memory for abuse." For example, a "yes" response could mean "Sometimes I found it too unpleasant to remember, so I tried not too." or "There were times when I could not remember without feeling terrible" or "There were times I could not bring myself to remember the abuse because I would rather not think about it." While no question is free of the possibility of multiple interpretations, the great potential for idiosyncratic interpretation by respondents to the particular wording used here warrants a further examination of the issue with a different eliciting question.

A further problem with the Briere and Conte study is that the respondents were all in therapy. If some of these clinicians were under the belief that repression of memory is common they may have communicated this belief to their clients. A client could readily infer that if repression of memory is so common, it is likely to have happened to me, thus the answer to the question is probably yes. This would, of course, inflate the estimates of the prevalence of repression. Other studies have given much lower estimates for the existence of repression. One study (Herman and Schatzow, 1987) gathered data from 53 women in therapy groups for incest survivors in the Boston area. Of the 53 cases, 15 (28%) reported severe memory deficits (which included both women who could recall very little from childhood and women who showed a recently unearthed repressed memory). Severe memory problems were most likely in cases of abuse that began early in childhood and ended before adolescence. Cases of violent or sadistic abuse were most likely to be associated with "massive repression as a defense." (p. 5).5

An even lower estimate was obtained in a study of 100 women in out- patient treatment for substance abuse in a New York City hospital (Fullilove et al, 1992). Over half of the women in this sample reported memories of childhood sexual abuse. The vast majority of them remembered their abuse their whole lives. Only l8% claimed that they forgot the abuse for a period of time and later regained the memory. Whether the women remembered the abuse their whole lives or forgot it for a period was completely unrelated to the violence of the abuse. Of course the data obtained from the New York sample may include an underestimation factor since there could be many more women in the sample who were sexually abused, repressed the memory, and have not yet regained it. In support of this hypothesis, one could point to the research of Williams (1992) who interviewed 100 women, mostly African- American, known to have been abused 17 years earlier in their lives. Of these, 38% were amnestic for the abuse or chose not to report it. Perhaps there are women in the New York sample who deny sexual abuse but who are still repressing it. Possibly there are women who were actually abused but d not remember it, however, it is misleading to assume that simple failure to remember means that "repression" has occurred. If an event happened so early in life, prior to the offset of childhood amnesia, then a woman would not be expected to remember it as an adult, whether it was abuse or not. This would not imply the mechanism of repression. Moreover, ordinary forgetting of all sorts of events is a fact of life, but is not thought to involve some special repression mechanism. For example, studies have shown that people routinely fail to remember significant life events even a year after they have occurred. One study consisted of interviews with 590 persons known to have been in injury-producing motor vehicle accidents during the previous year. Approximately 14% did not remember the accident a year later. Another study consisted of interviews with 1500 people who had been discharged from a hospital within the previous year. Over a quarter did not remember the hospitalization a year later (Government studies, cited in Loftus, 1982).

How common are repressed memories of childhood abuse? There is no absolute answer available. There are few satisfying ways to discover the answer, since we are in the odd position of asking people about a memory for forgetting a memory. But, for the moment we have some figures ranging from 18% to 59%. The range is disturbingly great suggesting serious scholarly exploration is warranted in order to know how to interpret claims about the commonness of repression, or what abuse characteristics the repression might be related to.

Jurors reaction to repressed memory cases. How do people in general and jurors in particular react to repressed memory cases? Are memories that were once previously repressed as credible as memories that were never repressed? Understanding layperson's reactions and credibility judgments is important not only for theoretical reasons, but for practical ones as well. Theoretically speaking, layperson's implicit or intuitive theories about repressed memories guide our society's thinking on this topic6 Such implicit theories can also illuminate how therapists' theories of repression are formed; in part they derive from the therapist's own implicit theories.

On a more practical level, understanding implicit theories of repression is important. Plaintiff's lawyers who are deciding whether to file repressed memory cases are eager to know in advance their likelihood of a successful outcome. Defense lawyers also care, since such subjective probabilities affect their decisions about whether to proceed to trial, or to settle a case early. And, perhaps most importantly, the plaintiff herself or himself should care. Plaintiffs bring lawsuits for myriad reasons. Some therapists encourage their clients to sue as "hope for emotional justice" (Forward & Buck, 1988). One therapist who had treated more than 1500 incest victims argued that the lawsuit, despite how grueling, is "a very important step towards devictimization"..."a further source of validation"..."the personal satisfaction can be significant" (p. 159). If the lawsuit is good for the plaintiff's mental health, what happens to mental health if a jury does not find the notion of repressed memories tenable and the plaintiff, consequently, does not prevail? Actual cases

We might start by examining actual cases that have gone to trial in recent years. We would find that a wide range of outcomes. Some trials ended in defense verdicts (e.g., Lofft v Lofft, 1989, in San Diego; Collier v Collier, 1991,in Santa Clara County). Others ended in a plaintiff verdicts. For example, a 39 year old woman sued her father in Los Angeles, and the jury awarded $500,000 (McMillan, 1992). A 33-year-old woman sued her uncle in Akron, Ohio, and the jury awarded $5.15 million, $150,000 in compensatory damages and $5 million in punitive damages (Fields, 1992). Since the laws are new, and most cases have settled, there are too few actual trials from which to gather data about reactions to repressed memory claims. Until more cases are tried to verdict, we may need to rely on simulated jury research to gather information on this issue.

Simulations

Several juror simulation studies explore how people are likely to react to repressed memory cases (Loftus, Weingardt & Hoffman, 1992). In these studies, mock jurors learned about a legal case that arose out of allegations of sexual assault. Subjects considered the case of a daughter (Roberta) and her father (Jim), a case modeled loosely after an actual case tried in the State of Washington, in 1991. Roberta, they learned, accused her father of raping her on several occasions when she was approximately 10 years old. She claimed she repressed all memory for these incidents. At about age 20, Roberta's memory returned while she was in therapy. She filed charges against her father a year after her memory came back. Roberta and her therapist blamed her current problems of depression, anxiety, and sexual dysfunction on the sexual abuse that happened when she was ten. The father (Jim) denied the allegations, claiming that Roberta was influenced by her therapist's suggestive questioning. She was looking for someone or something to blame for her troubles. How did people react to Roberta's claim? Did their reactions differ from reactions to a case that was identical except for the repression of memory? A different set of subjects reacted to a modified scenario involving a different woman (Nancy) whose memory was not repressed. Nancy's factual situation was identical to Roberta's except, subjects were told, Nancy never told anyone until age 20, when she went into therapy and told her therapist. Who is believed more, Roberta or Nancy? Several consistent findings emerged from these studies. First, people tended to be slightly more skeptical about Roberta's case (the repressed memory) than they were about Nancy's case (the nonrepressed memory). Both male and female subjects reacted this way, with males being overall more skeptical.

When subjects disbelieved the claims, they were more likely to think that the false claims were due to an honest mistake than a deliberate lie. One small difference emerged: Repressed and nonrepressed memory cases appear to bring slightly different thoughts to mind. When subjects consider Nancy's case thoughts of lying were slightly more likely to be evoked than when they consider Roberta's case. But one of the clearest results is that in general, the majority of subjects believed that the claims of both Roberta and Nancy were true and accurate.

What are the memories like?

The quality of the memories that filter back vary tremendously. They are sometimes detailed and vivid, and sometimes very vague. Sometimes they pertain to events that allegedly happened in early childhood, and sometimes in adolescence. Sometimes they pertain to events that allegedly happened 5 years ago, and sometimes 40 years ago. Sometimes they include fondling, sometimes rape, and sometimes ritualism of an unimaginable sort.

Highly detailed memories have been reported even for events that allegedly happened over 25 years earlier and during the first year of life. One father/daughter case recently tried in Santa Clara County, California illustrates this pattern (Collier v. Collier). The daughter DC, a college graduate who worked as a technical writer, claimed that her father sexually abused her from the time she was 6 months old until she was 18. She repressed the memories until the age of approximately 26, while she was in individual and group therapy. Other cases involve richly detailed allegations of a more bizarre, ritualistic type, as in a case reported by Rogers (1992). The plaintiff Bonnie, in her late 40s at the time of trial, accused her parents of physically, sexually, and emotionally abusing her from birth to approximately age 25. A sister Patti, in her mid-30s at the time of trial, said she was abused from infancy to age 15. The allegations involved torture by drugs, electric shock, rape, sodomy, forced oral sex and ritualistic killing of babies born to or aborted by the daughters. The events were first recalled when the plaintiffs went into therapy in the late 80's.

In short, reports of memories after years of repression are as varied as they can be. One important way that they differ in terms of the age at which the events being remembered allegedly happened. In many instances, repressed memory claims refer back to events that occurred when the child was a year old, or less. This observation invites an examination of the literature on childhood amnesia. It is well known that humans experience a poverty of recollections of their first several years in life. Freud identified the phenomenon is some of his earliest writings:

What I have in mind is the peculiar amnesia which, ....hides the earliest beginnings of the childhood up to their sixth or eighth year." (1905/1953, p. 174).

Contemporary cognitive psychologists place the offset of childhood amnesia at a somewhat earlier age: "past the age of ten, or thereabouts, most of us find it impossible to recall anything that happened before the age of four or five" (Morton, 1990, p. 3). Most empirical studies of childhood amnesia suggest that people's earliest recollection does not date back before the age of about 3 or 4 (Kihlstrom & Harackiewicz, 1982; Howe & Courage, 1992; Pillemer & White, 1989). One study showed that few subjects who were younger than three recalled any information about where they were when they heard about the Kennedy assassination, although most subjects who were over eight at the time had some recall (Winograd & Killinger, 1983). Although one recent study suggests that some people might have a memory for a hospitalization or the birth of a sibling that occurred at age two (Usher & Neisser, in press), these data do not completely rule out the possibility that the memories are not true memories but remembrances of things told to us by others (Loftus, in press). Still, the literature on childhood amnesia ought to figure in some way into our thinking about recollections of child molestation that supposedly occurred in infancy. Are the memories authentic?

Therapists beliefs about authenticity

Many therapists believe in the authenticity of the recovered memories that they hear from their clients. Two empirical studies reveal this high degree of faith. One large scale survey was conducted of clinicians who had come across, in their practice, ritualistic and religion related abuse cases (Bottoms, Shaver & Goodman, 1991). The satanic ritualistic abuse (SRA) cases are those involving allegations of highly bizarre and heinous criminal ritual abuse in the context of an alleged vast covert network of highly organized transgenerational satanic cults (Braun and Sachs, 1988; Ganaway, 1989, 1991). Clients with SRA memories have reported vividly detailed memories of cannibalistic revels, and experiences such as being used by cults during adolescence as serial baby breeders to provide untraceable infants for ritual sacrifices (Ganaway, 1989; Rogers, 1992b). If therapists believe these types of claims, it seems likely that they would be even more likely to believe the less aggravated claims involving ordinary childhood sexual abuse. The Bottoms et al. (1991) analysis revealed that 30% of clinicians responding had seen at least one case. A detailed analysis of 200 clinicians' experiences revealed that a substantial number involving amnesic periods (44% of adult survivor cases). 

Overall, 93% of clinicians believed the alleged harm was actually done, and that the ritualistic aspects were actually experienced by the clients. The conclusion, in the investigators' own words: "The clinical psychologists in our sample believe their clients' claims." (p. 10). A different approach to the issue of therapist belief was taken by (Loftus & Herzog, 1991). This study involved in-depth interviews with 16 clinicians who had seen at least one repressed memory case. In this small, non-random sample, 13 (81%) said they invariably believe their clients. One therapist said "if a woman said it happened, it happened." Another said "I have no reason not to believe them". The most common basis for belief was symptomatology (low self-esteem, sexual dysfunction; self destructive behavior) or body memories (voice frozen at young age, rash on body matching inflicted injury). Over two-thirds of the clinicians reacted emotionally to any use of the term "authentic", feeling that determining what is authentic and what is not authentic is not the job of a therapist. The conclusion from this small study: Therapists believe their clients, and often use symptomatology as evidence. These and other data suggest that therapists believe in their clients' memories. They point to symptomatology as their evidence. They are impressed with the emotional pain that accompanies the expression of the memories. Dawes (1992) has argued that this "epidemic of belief" is based in large part on authority and social consensus.

There are those with extreme positions who would like to deny the authenticity of all repressed memories and those who would accept them all as true. But as Van Benschoten (1990) has pointed out, these extreme positions will exacerbate our problems: "Denial fosters overdetermination, and overdetermination invites denial." (p. 25). If we assume, then, that some of the memories might be authentic and some might not be, we can then raise this question: If a memory is recovered that is not authentic, where would it come from? Ganaway (1989) proposes several hypotheses to explain SRA memories, and these same ideas are relevant to memories of a repressed past. If not authentic, the memories could be due to fantasy, illusion, hallucination- mediated screen memories, internally derived as a defense mechanism. Further paraphrasing Ganaway, the SRA memories combine a mixture of borrowed ideas, characters, myths, and accounts from exogenous sources with idiosyncratic internal beliefs. Once activated, the manufactured memories are indistinguishable from factual memories. Not authentic memories could also be externally derived as a result of unintentional implantation of suggestion by a therapist or other perceived authority figure with whom the client desires a special relationship, interest or approval.

The memories are authentic.

Childhood sexual abuse is tragically common, no question about it (Daro, 1988). Surveys reveal a large range in the estimated rates (10 to 50%), but as Freyd (1991) has argued, even the most conservative of them are high enough to support the enormity of child abuse. A sizeable number of people who enter therapy were abused as children, and they have always remembered their abuse. Even when they have severe emotional problems, they can provide rich recollections of abuse, often with many unique, peripheral details (Rogers, 1992). Occasionally the abuse is corroborated, sometimes with very cogent corroboration, such a pornographic photographs. If confirmed abuse is prevalent, why couldn't many instances of repressed memory abuse cases also be potentially authentic.

Unfortunately, in the repressed memory cases, particularly when memories do not return for 20 or 30 years, there is little in the way of documented corroboration. This of course does not mean that they are false. Claims of corroborated repressed memories occasionally appear in the published literature. For example, Mack (1980) reports on a 1955 case involving a 27 year old borderline man who, during therapy, recovered memories of witnessing his mother attempting to kill herself by hanging. The man's father later confirmed that the mother had attempted suicide several times, and the boy witnessed one attempt when he was three years old. The father's confirmation apparently led to a relief of symptoms in the son.

It is hard to know what to make of examples such as these. Did the son really remember back to age 3 or did he hear discussions of his mother's suicide attempts later in life? The memories could be real, that is, genuine instances of repressed memories that accurately returned much later. If true, this would only prove that some memory reports are authentic, but obviously not that all reports are authentic. Analogously, examples of repressed memories that were later retracted, later proven to be false, or later proven to be the result of suggestion would only prove that some memory reports are not authentic, but obviously not that all such reports are illusory.

Some who question the authenticity of the memories of abuse do so in part because of the intensity and sincerity of the accused persons who are denying the abuse. Many of the thousands of people who have been accused flatly deny the allegations and the cry of "witch hunt" is often heard (Baker, 1992, p. 48; Gardner, 1991). Witch hunt is, of course, a term that has been loosely used by virtually anyone faced by a pack of accusers (Watson, 1992). Analogies have been drawn between the current allegations and the witch-craze of the l6th and l7th centuries, when an estimated half million people were convicted of witchcraft and burned to death in Europe alone (Harris, 1974; Trott, 1991b). Although the denials during the witchcraze are now seen as authentic in the light of hindsight, the current denials of those accused of sex abuse are not proof that the allegations are false. Research with known rapists, pedophiles and incest offenders has illustrated that they often exhibit a "cognitive distortion" - a tendency to justify, minimize, or rationalize their behavior (Gudjonsson, 1992). Since accused persons are motivated to verbally and even mentally deny an abusive past, simple denials cannot constitute cogent evidence that the victim's memories are not authentic.

The memories are not authentic or false.

To say that memory might be false does not mean that the client is deliberately lying. Although lying is always possible, even psychotherapists who question the authenticity of reports have been impressed with the honesty and intensity of the terror, rage, guilt, depression, and overall behavioral dysfunction accompanying the awareness of abuse (Ganaway, 1989, p. 211). There are at least two ways that false memories could come about. Honestly believed, but false memories could come about, according to Ganaway, because of internal or external sources. The internal drive to manufacture an abuse memory may come about as a way to provide a screen for perhaps more prosaic, but ironically less tolerable, painful experiences of childhood. Creating a fantasy of abuse with its relatively clear-cut distinction between good and evil may provide the needed logical explanation for confusing experiences and feelings. The core material for the false memories can be borrowed from the accounts of others who are either personally known, or encountered in literature, movies, and TV.7

Sources of details that could affect memory

There are at least two important sources that could potentially feed into the construction of false memories. These include popular writings and therapists' suggestions. Popular writings. All roads on the search for popular writings inevitably lead to one, The Courage to Heal (Bass & Davis, 1988), often referred to as the "Bible" of the incest book industry. The Courage to Heal advertises itself as a guide for women survivors of child sexual abuse. While the book is undoubtedly a great comfort to the sexual abuse survivors who have been living with their private and painful memories, one cannot help but wonder about its impact on those who have no memories. The reader who is wondering if she might be a victim of child sexual abuse is provided with a list of possible activities ranging from the relatively benign (eg. held in a way that made you uncomfortable) to the unequivocally abusive (eg., raped or otherwise penetrated). The reader is then told "If you are unable to remember any specific instances like the ones mentioned above but still have a feeling that something abusive happened to you, it probably did." (p 21). On the next page, the reader is told: "You may think you don't have memories, but often as you begin to talk about what you do remember, there emerges a constellation of feelings, reactions and recollections that add up to substantial information. To say, "I was abused," you don't need the kind of recall that would stand up in a court of law. Often the knowledge that you were abused starts with a tiny feeling, an intuition...Assume your feelings are valid. So far, no one we've talked to thought she might have been abused, and then later discovered that she hadn't been. The progression always goes the other way, from suspicion to confirmation. If you think you were abused and your life shows the symptoms, then you were. " (p. 22). What symptoms?: Low self esteem, suicidal or self-destructive thoughts, depression, sexual dysfunction, among others.8

Others have worried about the role played by The Courage to Heal. A recent survey of several hundred families accused by unrepressed memories reveals that the book was implicated "in almost all cases" (Wakefield & Underwager, 1992, p. 486). Complaint about the book range from its repeated suggestion that abuse probably happened even if you have no memories and that demands for corroboration are not reasonable to its overt encouragement of "revenge, anger, fantasies of murder or castration, and deathbed confrontations." (Wakefield & Underwager, 1992, p. 485). In all fairness, however, it should be mentioned that the book is long (495 pages), and sentences taken out of context may distort their intended meaning. Nonetheless, readers without any memories of their own cannot escape the message that there is a strong likelihood that abuse occurred even in the absence of any memories.

The recent incest book industry published not only stories of abuse, but also suggestions to readers that they were likely abused even if there are no memories, that repressed memories of abuse undoubtedly underlie one's troubles, or that benefits derive from uncovering repressed memories and believing them.9 One popular incest book is the paperback by E. Sue Blume (1990) whose book jacket itemizes one of the author's chief credentials as the "Creator of the Incest Survivors' Aftereffects Checklist".10 Blume, a private practice therapist, tells readers that she has "found that most incest survivors have limited recall about their abuse." (p 81). She goes on to say that "Indeed, so few incest survivors in my experience have identified themselves as abused in the beginning of therapy that I have concluded that perhaps half of all incest survivors do not remember that the abuse occurred." (p. 81).

Some of the volumes provide exercises to help the reader lift the repression. Farmer (1989), for example, tells readers to try one particular exercise "whether or not you have any conscious recollection of the abuse you suffered." (p. 91). The reader is to sit down, relax, and mentally return to childhood. The next step is to choose a particular memory, whether fuzzy or clear, and "bring that memory to your full attention". Details about what to do with the memory are provided, along with an example from the life of "Danielle" who thought about how verbally abusive her father had been, and "Hazel" who remembered anger at her mother's treating her like a rag doll. This exercise allegedly helped to "lift the lid of repression", and unbury the Hurting Child.

Do these examples lift the lid of repression? Perhaps. But another equally viable hypothesis is that the examples influence the creation of memories, or at the very least, "direct" the search through memory, that the reader will ultimately take.

Therapists' Suggestions. Blume's observation that so many individuals enter therapy without memories of abuse, but acquire memories during therapy, naturally makes us wonder about what might be happening in therapy. According to Ganaway (1989), honestly believed but false memories could come about in another way, namely via unintentional suggestion from therapists. Ganaway notes a growing trend towards the facile acceptance and expressed validation of uncorroborated trauma memories, perhaps in part due to sensitization from years of accusations that the memories are purely fantasy. Herman (1992, p. 180) made a similar point: whereas an earlier generation of therapists might have been discounting or minimizing their patients' traumatic experiences, the recent rediscovery of psychological trauma has let to errors of the opposite kind. Some contemporary therapists have been known to tell patients, merely on the basis of a suggestive history or "symptom profile" that they definitely had a traumatic experience. Even if there is no memory, but merely some vague symptoms, certain therapists will inform a patient after a single session that they were very likely the victim of a Satanic cult. Once "diagnosed", the therapist urges the patient to pursue the recalcitrant memories.

While some therapists recommend against persistent, intrusive probing to uncover early traumatic memories (eg., Bruhn, 1990), others enthusiastically engage in these therapeutic strategies. Evidence for this claim comes in a variety of forms: 1) therapist accounts of what is appropriate to do with clients, 2) patient/client accounts of what happened during therapy, 3) sworn statements of clients and therapists during litigation, and 4) taped interviews of therapy sessions.

Therapist Accounts. One therapist, who has treated more than 1500 incest victims (Forward & Buck, 1988), openly discusses her method of approaching clients. "You know, in my experience, a lot of people who are struggling with many of the same problems you are, have often had some kind of really painful things happen to them as kids - maybe they were beaten or molested. And I wonder if anything like that ever happened to you?" (p. 161). Other clinicians claim to know of therapists who say "Your symptoms sound like you've been abused when you were a child. What can you tell me about that?" (Trott, 1991, p. 18), or worse, "You sound to me like the sort of person who must have been sexually abused. Tell me what that bastard did to you" (Davis, 1991, p. 82). At least one clinician advocates "It is crucial...that clinicians ask about sexual abuse during every intake" (Frawley, 1990). The rationale for this prescription is that the clinician who asks conveys to the client that the client will be believed, and that the clinician will join with the client in working through the memories and emotions linked with childhood sexual abuse.

Asking about sex abuse along with a list of other past life events makes sense given the high instance of actual abuse, but the question is how the issue is raised, and what therapists do in cases where clients initially deny an abusive past. Evidence exists that some therapists do not take "no" for an answer. One therapist (who otherwise seems sensitive to problems of memory tampering) still recommends: "When the client does not remember what happened to her, the therapist's encouragement to "guess" or "tell a story" will help the survivor regain access to the lost material" (Olio, 1989, p. 6). She goes on to provide the example of a client who suspected sexual abuse, but had no memories. The client had become extremely anxious at a social gathering in the presence of a 3 year old girl. She had no idea why she was upset except she wanted the little girl to keep her dress down. When encouraged in therapy to tell a story about what was going to happen to the little girl, the client ultimately related with tears and trembling one of the first memories of her own abuse. She used the story to "bypass her cognitive inhibitions and express the content of the memory". Later she "integrated the awareness that she was indeed the little girl in the story." One cannot help but wonder about these mental fantasy exercises in light of known research showing that the simple act of imagination makes an event subjectively more likely (e.g., Sherman, Cialdini, Schwartzman, & Reynolds, 1985).

Even if the therapist does not encourage the client to guess or tell a story, stories sometimes get told in the form of client dreams. If discussions of incest go on during the day, and day residue gets into the dreams at night, it would not be surprising to see that dreams of incest might result. Poston and Lison (1990) describe a woman with "repressed memories" of incest who reported a dream about watching a little girl ice skate on a frozen river. In her dream, the woman tried desperately to warn the child that monsters and snakes were making their way through the ice to devour the little girl. Although frightened the woman was powerless and could not warn the innocent child. A few days later the client began remembering incest from her childhood. Knowing she had "a trusted relationship with a therapist and a survivor's group that would understand and accept her" (p. 197), the memories began to flow.

Examples of therapists interpreting dreams as signs of memory for abuse can be found throughout the literature. One clinician describes with pride how she communicated to her male patient the basis for her suspicions that he had been abused: "On many occasions, I explained that these dreams had preserved experiences and impressions of an indelible nature." (Williams, 1987, p.. 152). Frederickson (1992), who has worked with many incest survivors, also talks in detail about her methods of getting patients to remember. She recommends that the therapist guide the patient "to expand on or explore images that have broken through to the conscious mind, allowing related images of the abuse to surface. The process lets the survivor complete the picture of what happened, using a current image or flash as a jumping-off point." (p. 97). She also suggests that the therapist help the patient expand on the images and sensations evoked by dreams "to shed light on or recover our repressed memories." (p. 98). She extols the virtues of hypnosis to "retrieve buried memories" (p. 98). She recommends that patients "jot down suspected memories of abuse you would like to explore. Include your own felt sense of how you think you were abused." (p. 102).

Even if the clinician is not the first to bring up sexual abuse, they will often reinforce what begins as a mere suspicion. One client developed the idea that she might have been sexually abused, tried hypnosis to help her recover memories, and obsessed for years. Only after her therapist stated that she believed sexual assault was "indeed possible" (p. 569), and cited nightmares, phobia of men, and other symptoms as evidence, did the client come up with some specific memories (Schuker, 1979).

Before leaving the examples of therapists accounts of what goes on in therapy, it is important to add a word of caution. Psychiatric anthropologist, Sherrill Mulhern, has documented the alarming discrepancies that often exist between therapists' accounts of what they have done in therapy, and what is revealed in video or audio tapes of those same sessions (Mulhern, 1991a, b). If memories are uncovered - whether after repeated probing, after telling stories, after dreams or seemingly spontaneously - or even if the memories remain buried, therapists often send their clients to support groups. In one study of clients who, in the course of therapy, verbalized their victimization through ritualistic abuse, the majority reported that they had participated in these types of groups (Shaffer & Cozolino, 1992). One group, Survivors of Incest Anonymous (SIA), publishes extensive reading materials intended to aid the recovery of incest survivors. (SIA merged with Sexual Abuse Anonymous in 1987). The criteria for "admission" make it clear that entry is fine for those with no memories of sexual abuse. "Do you have blocks of your childhood you can't remember? Do you have a sense that "something happened?"" (SIA, 1985). These and other questions (such as Do you have problems with self-confidence and self esteem? Do you feel easily intimidated by authority figures?) are among the set of 20 questions that help a potential survivor decide whether SIA can be of assistance. It's OK not to remember at first because "Many survivors have 'repressed' actual abuse memories in order to survive." But, the goal is to remember: "..participating in SIA helps us to remember what happened to us so we can stop being controlled by incest." (SIA, 1990, p. 1). Although the groups are undoubtedly invaluable for genuine survivors of sex abuse, as they are for other survivors of extreme situations such as combat and political persecution (Herman, 1992, p. 215), concerns about the incest survivor groups have been expressed.

Do these groups foster the development of constructed memories? An investigative journalist attending a four-day workshop watched the construction of memory at work (Nathan, 1992). With members recounting graphic details of SRA abuse, how long will they listen to the person who can only say "I think I was abused, but I don't have any memories." Others have worried in the literature that such groups may induce "proto-extension", that is they actually encourage a troubled person to remember details from other survivor stories as having happened to them as well (Ellis, 1992).

Patient/Client Accounts. Another source for suggestions in therapy can be found in client accounts of what happened to them. Recently clients have been reporting that a therapist has suggested that childhood abuse was the cause of their current distress. But they have no memories. One woman from Oregon entered therapy to deal with depression and anxiety, and within a few months her therapist suggested the cause could be childhood sexual abuse. She wrote asking for help in remembering:

"... Since that time, he has become more and more certain of his diagnosis... I have no direct memories of this abuse....The question I can't get past is how something so terrible could have happened to me without me remembering anything. For the past two years I have done little else but try to remember. I've tried self-hypnosis and light trance work with my therapist. And I even travelled to childhood homes...in an attempt to trigger memories..."

One client revealed the suggestive nature of his therapist's questioning on ABC's PrimeTime Live (1992). Attorney Greg Zimmerman went to a psychotherapist in Boulder, Colorado to deal with his father's suicide. He told ABC "I would try to talk to her about the things that were very painful in my life and she kept saying that there was something else." Zimmerman grew more and more depressed as the mystery of that "something else" would not unravel, and then, during a therapy session, his therapist stunned him with her diagnosis: "I don't know how to tell you this, but you display the same kinds of characteristics as some of my patients who are victims of Satanic ritualistic abuse." Zimmerman had said nothing whatsoever to her to provoke this diagnosis, apparently her standard.

It is easy to find published accounts that describe the emergence of memories in therapy, and that describe the techniques that therapists have used to uncover those memories (eg. Bass & Thornton, 1991). One account, written under the pseudonym of Jill Morgan, told of a series of positively horrifying memories of abuse by her father. He raped her when she was four years old, again at age 9, once again at age 13, for 7 straight days and nights at age 15, and for the final time at age 18. For the next several years, all misery was withheld from conscious memory, and then, at age 29, she was helped to remember in therapy: "Through hypnosis and age regression, a skilled therapist gave me back my memory." (p. 111). The involvement of hypnosis and age regression prompts the natural inquiry into whether these techniques produce authentic memories?

Unfortunately the evidence is discouraging: there is an extensive literature seriously questioning the reliability of hypnotically enhanced memory in general (Smith, 1983), and hypnotic age regression in particular (Nash, 1987). Hypnotic attempts to improve memory increase the confidence in what is recalled more than the accuracy (Bowers, 1992). Even more worrisome is the impossibility of reversing the process; the hypnotically induced memory becomes the person's reality (Orne, 1979). With hypnotic regression, men and women have been known to recall being abducted by aliens aboard exotic spacecraft and other forgotten events (Gordon, 1991). A more detailed client account is that of Betsy Petersen (1991), as described in an autobiographical account, Dancing with Daddy. Petersen, a Harvard graduate and accomplished writer reveals in her first book that she repressed memory for sexual abuse by her father until she was 45-years old. She now remembers sexual abuse from the time she was 3 and 1/2 until she was 18. Betsy entered therapy (with "Kris") for problems relating to her children, and almost a year after starting therapy she started worrying: "I'm afraid my father did something to me." She tried hard to recall putting "together a scenario of what might have happened." (p. 65). When she told her therapist about this, she said "I don't know if I made it up or if it's real." Kris replied: "It feels like a story to you, because when something like that happens, everybody acts like it didn't." Betsy: "You mean it might really have happened!" Kris told her there was a good chance it had happened. Kris told her, in Betsy's words, "It was consistent with what I remembered about my father and my relationship with him, and with the dreams I had been having, and with the difficulties I had being close to my children, and also, she said, with the feelings I had during and after sex with my husband." (p 65). Betsy worked hard to retrieve incest memories: "I had no memory of what my father had done to me, so I tried to reconstruct it. I put all my skill- as a reporter, novelist, scholar - to work making that reconstruction as accurate and vivid as possible. I used the memories I had to get to the memories I didn't have" (p. 66).12 If accurate, this account tells us something about one therapist's approach. The therapist convinces the patient with no memories that abuse is likely, and the patient obligingly uses reconstructive strategies to generate memories that would support that conviction.

In addition to the first person accounts, more formal studies of incest survivors provide clues as to what might be happening in therapy. One study of 20 adults who uncovered ritualistic abuse memories stemming from childhood revealed that the majority sought psychotherapy because of symptoms (eg depression, anxiety, Shaffer & Cozolino, 1992)). The primary focus of their therapy was "the uncovering of memories" (p. 189). The majority participated in 12 step programs (e.g., Incest Survivors Anonymous) as "necessary adjuncts to their psychotherapy." (p. 190). These groups provided substitute families for the clients who had severed ties with their families of origin. Other similar studies of ritualistic abuse rememberers have revealed that most of the victims have no memory of the abuse before therapy (e.g. Driscoll & Wright, 1991) but that techniques such as hypnosis (Driscoll & Wright, 1991) or dreams and artwork (e..g., Young et al, 1991) were used by therapists to unlock those recalcitrant memories.

Litigation accounts. Information gathered during litigation is another source of knowledge about the emergence of memories in therapy. Take the case of Patti Barton against her father, John Peters, a successful businessman.13 Depositions taken in the case of Barton v. Peters (1990), reveal that Patti Barton began therapy with a Dr. CD, a Doctor of Divinity in July 1986. Dr. CD's notes indicate that during the 32nd session of therapy she expressed "fear her father has sexually tampered with her." (Deposition of CD, 4/21/91, p. 39). This was the first time that anything like that had come up in any of the sessions. Shortly thereafter, Patti related a dream that a man was after her.14 Dr. CD apparently then used the technique of "visualization," wherein Patti would try to visualize her past. He got her to remember eye surgery at the age of 7 months. As for the abuse, one of the earliest acts of abuse he managed to dredge up with this method occurred when Patti was 15 months of age. "I visualized that my father stuck his tongue in my mouth." (Deposition of PB, May 1991). "..after he stuck his tongue in my mouth -- Well, it seemed to last for hours and hours even though I know it didn't. But it was awful to me and an event that seemed to last for hours. I started crying, and I crawled over to the wall. And I started banging my head on the wall. And my mother came into the room, and she picked me up. And I tried to tell her in baby talk what had happened. I said "Ma, ma, ma, ma," and I said, "Da,Da,Da,Da" and I said, "Me-e-e-." And that's all that I can remember." (p. 193). Later Patti would remember that her father touched her in her crotch and put his penis in her mouth at age 3, and that she stroked his penis over and over at age 4. Rape would come later. Patti's father eventually agreed to give his daughter the deed to a piece of land he owned, but he continues to deny the charges. Her brother, a Baptist minister in Alaska claims that Satan's wicked spirits planted untruths in Patti's head (Laker, 1992). Did it take 30 some sessions for the therapist to uncover actual memories of abuse, or 30 some sessions for false memories of abuse to begin to be visualized and constructed? Taped interviews. Often there is a confidentiality considerations prevent access to interactions between therapists and clients. But when cases get into litigation, special interviewing is frequently done, and occasionally it is recorded. Recordings were done in a case implicating a man named Paul Ingram from Olympia, Washington (Watters, 1991). Ingram was arrested for child abuse in 1988, amid expressions of shock from his community. At the time he was chair of the county Republican committee, and was chief civil deputy in the sheriff's office. He had worked in law enforcement for more than a decade. The Ingram case began at a time when waves of rumor and media hype over satanic ritualistic abuse were rampant. At first Ingram denied everything, and detectives told him he was in denial. With the help of a psychologist who exerted enormous pressure over endless hours of interrogation, Ingram's memories of abusing his daughter began to appear. Then the psychologist, with the help of a detective, "interviewed" Ingram's son. In that interview, the son reports on his dreams, and the therapist and detective convince him the dreams are real.15

In another case, a father (Mr. K) hired a private investigator after his 26 year old daughter reported a recently uncovered repressed memory and accused him of incest. The investigator, acting undercover, went to see the daughter's therapist complaining that she had a nightmares and had trouble sleeping. On the third visit, the therapist told undercover agent that she was an incest survivor. According to the investigator's report the therapist said this to her pseudo-patient: "She then told me that she was certain I was experiencing body memory from a trauma, earlier in life, that I could not remember. I could not remember because my brain had blocked the memory that was too painful to deal with." When the patient said she didn't remember any trauma, the therapist told her "that is the case and many people at far later times in their lives go through this when the memory starts to surface." The therapist told her that many people go through this experience such as "Viet Nam Vets, Earthquake Survivors and Incest Survivors". When the patient said that she had never been in Viet Nam nor in an earthquake, the therapist nodded her head yes and said "Yes, I know." The therapist then said she should read "Courage to Heal," a book she recommends to all survivors. After that there was the "Courage to Heal Workbook," tells survivors how to cope with the fears and memories. She pulled "Secret Survivors" by E.S. Blume from the shelf, opened the cover and read the list of symptoms of incest survivors. With two-thirds of the symptoms, she would look at the pseudo-patient and shake her head yes as this was confirmation of her diagnosis. She recommended incest survivor groups. (Monesi, 1992).

In the fourth session, the diagnosis of probable incest victim was confirmed based on the "classic symptoms" of body memory and sleep disorders. When the patient insisted that she had no memory of such events, the therapist assured her this was often the case. Why would therapists suggest things to their patients?

The core of treatment, it is widely believed, is to help the client reclaim their "traumatic past" (Rieker & Carmen, 1986, p. 369). Therapists routinely dig deliberately into the ugly underbelly of mental life. They dig for memories purposefully because they believe that in order to get well, to become survivors not just victims, their clients must overcome the protective denial that was used to tolerate the abuse during childhood (Sgroi, 1989, p. 112). Memory blocks can be protective in many ways, but they come at a cost; they cut off the survivor from a significant part of past history, and leave her without a good explanation for her negative self-image, low self-esteem, and other mental problems. These memories must be brought into consciousness, not as an end in itself, but only insofar as it helps the survivor acknowledge reality and overcome denial processes that are now dysfunctional (p. 115). Another reason why therapists may be unwittingly suggesting ideas to their clients is that they have fallen prey to a bias that affects all of us, known as the "confirmatory bias" (Baron, Beattie, & Hershey, 1988). People in general, therapists included, have a tendency to search for evidence that confirms their hunches rather than searching for evidence that disconfirms. It is not easy for us to discard our long-held or cherished beliefs, in part because we are eager to verify those believes and not inclined to seek evidence that might disprove them. The beliefs that individuals hold can create their own social reality, a notion that is the essence of the self-fulfilling prophecy (Snyder, 1984). How does "reality" get constructed? One way this can happen is through the interview strategies that the person chooses to use. Interviewers are known to choose questions that inquire about behaviors and experiences thought to be characteristic, rather than those thought to be uncharacteristic, of some particular classification. If therapists ask questions that tend to elicit behaviors and experiences thought to be characteristic of someone who had been a victim of childhood trauma, might they too be creating this social reality? Whatever the good intentions of therapists, the documented examples of rampant suggestion should force us to at least ponder whether the therapists might be suggesting illusory memories to their clients, rather than unlocking distant memories that are authentic. Or, paraphrasing Gardner (1992), what is considered to be present in the client's unconscious mind, might actually be present solely in the therapist's conscious mind (p. 689). Ganaway (1989) worries that once seeded by the therapist, false memories could be developing that replace previously unsatisfactory internal explanations for intolerable but more prosaic childhood trauma. The creation of false memories

The hypothesis that false memories could be created invites an inquiry into the important question of what is known about false memories. Since the mid-l970's at least, investigations have been done into the creation of false memories through exposure to misinformation. Now, nearly two decades later, we have hundreds of studies to support a high degree of memory distortion. People have recalled nonexistent broken glass and tape recorders, a clean-shaven man as having a mustache, straight hair as curly, and even something as large and conspicuous as a barn in a bucolic scene that contained no buildings at all (Loftus & Ketcham, 1991). This growing body of research shows that new, post- event information often becomes incorporated into memory, supplementing and altering a person's recollection. The new information invades us, like a Trojan horse, precisely because we do not detect its influence. Understanding how we can become tricked by revised data about our past is central to understanding the hypothesis that suggestions from popular writings and therapy sessions can affect our autobiographical recall. One frequently-heard comment about the research on memory distortion is that all that changes induced by misinformation are about trivial details (Darton, 1991; Franklin & Wright, 1991). There is no evidence, the critics allege, that you can tinker with memories of real traumatic events or that you can inject whole events into the human mind for things that never happened.

Can real traumatic memories be changed?

In the category of "documented anecdotes" there is the example of one of the worst public/personal tragedies in the history of baseball (Anderson, 1990; described in Loftus & Kaufman, 1992). Baseball aficionados will recall that pitcher Jack Hamilton, then with the California Angels, crushed the outfielder, Tony Conigliaro, in the face with a first-pitch fastball. Although Hamilton thought he remembered this horrible event perfectly, he misremembered it as occurring during a day game when it was actually at night, and misremembered it in other critical ways.

Another example will be appreciated by history buffs, particularly those with an interest in the second world war. An American general, Brig. Gen Elliot Thorpe, recalled the day after the bombing of Pearl Harbor one way in a memoir and completely differently in an oral history taken on his retirement. Both accounts, in fact, were riddled with errors (Weintraub, 1991).

Evidence of a less anecdotal, more experimental, nature supports the imperfections of personally experienced traumatic memories. For example, one study examined people's recollections of how they heard the news of the l986 explosion of the space shuttle Challenger (Harsch & Neisser, 1989; Neisser & Harsch, 1992). Subjects were questioned on the morning after the explosion, and again nearly three years later. Most individuals described their memories as "vivid" but none of them were entirely correct and over a third were "wildly inaccurate." One subject, for example, was on the telephone having a business discussion when her best friend interrupted the call with the news. Later she would remember that she heard the news in class, and at first thought it was a joke. She later walked into a T.V. lounge and saw the news, and then reacted to the disaster.

Another study demonstrates the malleability of memory for a serious life and death situation (Abhold, 1992). The subjects were individuals who had attended an important high-school football game at which a player on the field went into cardiac arrest. Paramedics tried to resuscitate the player, and apparently failed. The audience reactions ranged from complete silence, to sobbing, to screaming. (Ultimately, fortunately, the player was revived at the hospital). Six years later many of these individuals were interviewed. Errors of recollection were common. Moreover, when exposed to misleading information about this life and death event, many individuals absorbed the misinformation into their recollections. For example, over a quarter of the subjects were persuaded that they had seen blood on the player's jersey after receiving a false suggestion to this effect.

These anecdotes and experimental examples suggest that even details of genuinely experienced traumatic events are, as Christianson (1992) put it, "by no means, completely accurate." (p. 207). Can you inject a complete memory for something that never happened? It is one thing to discover that memory for an actual traumatic event is changed over time, but quite another to show that you can inject a whole event into someone's mind for something that never happened. There are numerous anecdotes and experimental studies that show it is indeed possible to lead people to construct entire events.

Piaget's memory. Whole memories can also be implanted into a person's real-life autobiography, as is best shown by Piaget's classic childhood memory of an attempted kidnapping (Piaget, 1962; described in Loftus & Ketcham, 1991). The false memories were with him for at least decade. The memory was of an attempted kidnapping that occurred when he was an infant. He found out it was false when his nanny confessed years later that she had made up the entire story and felt guilty about keeping the watch she had received as a reward. In "explaining" this false memory, Piaget assumed: "I, therefore, must have heard, as a child, the account of this story, which my parents believed, and projected into the past in the form of a visual memory."

Loud noises at night. Although widely disseminated and impressive at first glance, Piaget's false memory is still but a single anecdote and subject to other interpretations. Was this really a memory, or an interesting story? Could it be that the assault actually happened and the nurse, for some inexplicable reason, lied later? For these reasons it would be nice to find stronger evidence that a false memory for a complete event was genuinely implanted.

An apparently genuine l9th century memory implantation was reported by Laurence and Perry (1983): Bernheim, during hypnosis, suggested to a female subject that she had awakened four times during the previous night to go to the toilet and had fallen on her nose on the 4th occasion. After hypnosis, the woman insisted that the suggested events had actually occurred, despite the hypnotist's insistence that she had dreamed them. Impressed by Bernheim's success, and explorations by Orne (1979), Laurence and Perry asked 27 highly hypnotizable individuals during hypnosis to choose a night of previous week, and to describe their activities during the half hour before going to sleep. Then, the subjects were instructed to relive that night while a suggestion was implanted that they had heard some loud noises and had awakened. Almost half (13) of the 27 subjects accepted the suggestion and stated after hypnosis that the suggested event had actually taken place. Of the 13, six were unequivocal in their certainty. The remainder came to the conclusion on basis of reconstruction. Even when told that the hypnotist had actually suggested the noises, these subjects still maintained that the noises had occurred. One said "I'm pretty certain I heard them. As a matter of fact, I'm pretty damned certain. I'm positive I heard these noises."

The paradigm of inducing pseudomemories of being awakened by loud noises has now been used extensively by other researchers who readily replicate the basic findings. Moreover, the pseudomemories are not limited to hypnotic conditions. Simply inducing subjects to imagine and describe the loud noises resulted in later "memories" for noises that had never occurred (Weekes, Lynn, Green & Brentar, 1992). Other false memories. Other evidence shows that people can be tricked into believing that they experienced an event even in the absence of specific hypnotic suggestions. For example, numerous studies have shown that people misremember that they voted in a particular election when they actually had not (Abelson, Loftus & Greenwald, 1992). One interpretation of these findings is that people fill in the gaps in their memory with socially desirable constructions, thus creating for themselves a false memory of voting.

In other studies people have been led to believe they witnessed assaultive behavior when in fact they did not (eg. Haugaard et al, 1991). In this study subjects (children ages 4-7) were led to believe that they saw a man hit a girl when he hadn't after hearing the girl lie about the assault. Not only did they misrecall the nonexistent hitting, but they added their own details: Of 41 false claims, 39 children said it happened near a pond, one said it was at the girl's house, and one could not specify where exactly the girl was when the man hit her.

Violent false Memories People hold completely false memories for something far more traumatic than awakening at night, voting in a particular election, or a simulation involving a man and a girl. Pynoos and Nader (1989) studied childrens' recollections of a sniper attack at an elementary school playground. Some of those children who were interviewed were not at the school during the shooting, including those already on the way home and those on vacation. Yet, even the non witnesses had memories: "One girl initially said that she was at the school gate nearest the sniper when the shooting began. In truth she was not only out of the line of fire, she was half a block away. A boy who had been away on vacation said that he had been on his way to the school, had seen someone lying on the ground, had heard the shots, and then turned back. In actuality, a police barricade prevented anyone from approaching the block around the school." (p. 238). The memories apparently were created by exposure to the stories of those who truly experienced the trauma.

Memories for being lost. A question arises as to whether one could experimentally implant memories for nonexistent events which, if they had occurred, would have been traumatic. Given the need to protect human subjects, devising a means of accomplishing this was not an easy task. Loftus and Coan (in press) however developed a paradigm for instilling a specific childhood memory for being lost on a particular occasion at the age of five. They chose "getting lost" since it is clearly a great fear of both parents and children. Their initial observations shows how subjects can be readily induced to believe this kind of false memory. The technique involves a subject and a trusted family member who play a variation of "Remember the time that....?" To appreciate the methodology, consider the implanted memory of 14-year old, Chris. Chris was convinced by his older brother Jim that he had been lost in a shopping mall when he was five-years old. Jim told Chris this story as if it were the truth: "It was 1981 or 1982. I remember that Chris was five. We had gone shopping at the University City shopping mall in Spokane. After some panic, we found Chris being led down the mall by a tall, oldish man (I think he was wearing a flannel shirt). Chris was crying and holding the man's hand. The man explained that he had found Chris walking around crying his eyes out just a few moments before and was trying to help him find his parents." Just two days later, Chris recalled his feelings about being lost: "That day I was so scared that I would never see my family again. I knew that I was in trouble" On day 3, he recalled a conversation with his mother: "I remember mom telling me never to do that agin." On day 4: "I also remember that old man's flannel shirt." On day 5 he started remembering the mall itself: "I sort of remember the stores." In his last recollection, he could even remember a conversation with the man who found him: "I remember the man asking me if I was lost." It would be natural to wonder whether perhaps Chris had really gotten lost that day. Maybe it happened, but his brother forgot. But Chris's mother was subjected to the same procedure and was never able to remember the false memory. After five days of trying she said "I feel very badly about it, but I just cannot remember anything like this ever happening."

A couple of weeks later, Chris described his false memory and he greatly expanded on it. "I was with you guys for a second and I think I went over the look at the toy store, the Kay-bee toy and uh, we got lost and I was looking around and I thought, "Uh-oh. I'm in trouble now." You know. And then I..I thought I was never going to see my family again. I was really scared you know. And then this old man, I think he was wearing a blue flannel, came up to me.... he was kind of old. He was kind of bald on top....he had like a ring of gray hair.... and he had glasses. " Thus, in two short weeks, Chris now could even remember the balding head and the glasses worn by the man who rescued him. He characterized his memory and reasonably clear and vivid. Finally Chris was debriefed. He was told that one of the memories presented to him earlier had been false. When asked to guess, he guessed one of the genuine memories. When told that it was the "getting lost" memory, he said: "Really???? I thought I remembered being lost... and looking around for you guys. I do remember that. And then crying. and mom coming up and saying "Where were you. Don't you...Don't you ever do that again."

A false memory for abuse. The lost-in-a-shopping-mall examples show that memory for an entire event can be created, one that is mildly traumatic. But it is still natural to wonder whether one could go even further and implant a memory for abuse. Ethically, of course, it would not be possible, but anecdotally, as it happens, it was done. It is one of the most dramatic cases of false memory for abuse ever to be documented: The case of Paul Ingram from Olympia, Washington (Ofshe, 1992; Watters, 1991). Ingram, recall, was arrested for child abuse in 1988 at the time he was chair of the county Republican committee. At first Ingram denied everything, and detectives told him he was in denial. After five months of interrogation, suggestions from a psychologist, continuing pressure from detectives and advisors, Ingram began to confess to rapes, assaults, child sexual abuse, and participation in a Satan worshiping cult alleged to have murdered 25 babies (Ofshe, 1992). To elicit specific memories, the psychologist or detectives would suggest some act of abuse (eg. on one occasion, Ingram and several other men raped his daughter). Ingram would at first not remember these fragments, but after a concerted effort on his part, Ingram would later would come up with a detailed memory. A social psychologist, Richard Ofshe, hired by the prosecution to interview Ingram and his family members decided to test Ingram's credibility. Ofshe had made up a completely fabricated scenario. He told Ingram that two of his children (a daughter and a son) had reported that Ingram had forced them to have sex in front of him. As with the earlier suggestions, Ingram at first could not remember this. But Ofshe urged Ingram to try to "think" about the scene and try to "see" it happening, just as the interrogators had done to him earlier. Ingram began to get some visual images. Ingram then followed Ofshe's instructions to "pray on" the scene and try to remember more over the next few hours. Several hours later, Ingram had developed detailed memories and wrote a three page statement confessing in graphic detail to the scene that Ofshe had invented (Ofshe, 1992; Watters, 1991). Ofshe (1989, 1992) notes that this is not the first time that a vulnerable individual has been made to believe that he committed a crime for which he originally had no memory and which evidence proves he could not have committed.

What is crucial about the Ingram case is that some of the same methods that are used in repressed memory cases were used with Ingram. These include the use of protracted imagining of events, and authority figures establishing the authenticity of these events. These examples provide further insights into the malleable nature of memory. They suggest that memories for personally experience traumatic events can be altered by new experiences. Moreover, they reveal that entire events that never happened can be injected into memory. The false memories range from the relatively trivial (like remembering that you voted when you didn't) to the bizarre (like remembering that you forced your daughter and son to have sex in front of you, when in fact you didn't). These false memories, with more or less detail, of course do not prove that repressed memories of abuse that return are false. But they do demonstrate a mechanism by which false memories can be created by a small suggestion from a trusted family member, by hearing someone lie, by suggestion from a social psychologist, or by incorporation of the experiences of others into one's own autobiography. Of course the fact that false memories can be planted tells us nothing about whether a given memory of child sex abuse is false or not; nor does it tell us how we might distinguish the real cases from the false ones. These findings on the malleability of memory do, however, raise questions about the wisdom of certain recommendations being promoted in self-help workbooks, in handbooks for therapists, and by some therapists themselves. The false memories created in the above examples were accomplished with techniques that are not all that different from what some therapists are doing: suggesting that the client was probably abused because of some vague symptoms, labeling a client's ambiguous recollections as "evidence" of abuse, and encouraging mental exercises that involve fantasy merging with reality.

Several implications of these observations follow. First, we need a renewed effort at research on the problem of repressed memories.. This should encompass, in part, a reexamination of some of the widely- cherished beliefs of psychotherapists. Is it true that repression of extremely traumatic experiences is common? Do these experiences invade us despite the fact that "all the good juice of consciousness has drained out" (Dennett, 1991, p. 325). Is is common to see analogies drawn between the Vietnam veterans and the incest survivors (eg Herman, 1992; Rieker & CArmen, 1986). Do they share in common the use of "massive repression" (Wolf & Alpert, 1991, p. 314) as a mechanism for coping? If so, how do we explain findings obtained with children who witness parental murder (Malmquist, 1986), and other atrocities. In one study, not a single child aged 5 to 10 years who had witnessed the murder of a parent repressed the memory. Rather they were continually flooded with pangs of emotion about the murder and preoccupation with it.

Is it true that repressed material, like radioactive waste, "lies there is leaky canisters, never losing potency, eternally dangerous" (Hornstein, 1992, p. 260) and constantly threatening to erupt into consciousness? Psychotherapists have assumed for years that repressed memories are powerful influences because they are not accessible to consciousness (Bowers, 1992). Is there evidence for this assumption? Is it necessarily true that all people who display symptoms of severe mental distress have had some early childhood trauma (probably abuse) that is responsible for the distress? With cutting edge research now showing that mental distress involves neuronal and hormonal systems of a much wider scope that previously realized (Chrousos & Gold, 1992; Gershon & Rieder, 1992), should not other potential causes be at least considered?

Questions must also be examined about the well-intentioned treatment strategies of some clinicians. Is it possible that "the therapist's interpretation is the cause of the patient's disorder rather than the effect of the disorder", to paraphrase Guze (1992, p. 78). Is it necessarily true that people who cannot remember an abusive childhood are repressing the memory? Is it necessarily true that people who dream about or visualize abuse are actually getting in touch with true memories? Good scientific research needs to be done to support these assumptions or they should be challenged. Challenging these core assumptions will not be an easy thing to do, anymore than it was for the psychologists of the 1930's to challenge the radical subjectivity of psychoanalysis (Hornstein, 1992), or psychologists of the 1980's to challenge the reliability of the clinical judgments made by psychologists and psychiatrists (Faust & Ziskin, 1988; Fowler & Matarazzo, 1988)16. Nonetheless, when we move from the privacy of the therapy session, where the client's reality may be the only reality that is important, into the courtroom where there can now be but a single reality, then more solid evidence should be a right to which we as citizens in a democratic society are entitled. Until we have better empirical answers, therapists might want to consider whether it is wise to "suggest" that childhood trauma happened, to probe relentlessly for recalcitrant memories, and then to uncritically accept them as fact. Uncritical acceptance of uncorroborated trauma memories by therapists, social agencies and law enforcement personal has been used to promote public accusations by alleged "survivors". If the memories are fabricated, this will of course lead to irreparable damage to the reputations of potentially innocent people, according to Ganaway (1989), who discusses the problem in the context of SRA memories.

Uncritical acceptance of uncorroborated trauma memories poses other potentially dangerous problems for our society. According to Ganaway (1991), reinforcing the validity of unverifiable memories within the therapeutic setting may lead to diversionary paths in the patient's therapy away from actual childhood trauma. This could lead to interminable therapy and a total draining of the patient's financial resources as the therapist and patient collaborate in a mutual deception to pursue a bottomless pit of memories. Worse, the patients initial wonderings supported by therapist affirmations could then become fixed beliefs, precipitating suicidal thoughts and behaviors based on the new belief system, since the patient would no longer challenge the veracity of the new memories. Like Betsy Ross sewing the first American flag, the abuse becomes a myth that was never true, but always will be (Frishholz, 1992). Patients who are reinforced into a new belief system could develop newer, larger problems. If actual childhood sexual abuse is associated with numerous negative long-term effects (eg, severe sexual dysfunction, Ambrose-Bienkowski, Stahly, & Wideman, 1991), what might be the consequence of implanted childhood sexual abuse? If the memories are ultimately shown to be false, therapists may then become the targets of future ethics violations and lawsuits. They will be charged with a grave form of mind abuse, charges that have already been initiated in several states.

What should therapists do instead? As a first step, it is worth recognizing that we do not yet have the tools for reliably distinguishing the signal of true repressed memories from the noise of false ones. Until we gain these tools, it seems prudent to consider some combination of Herman's (1992) advice about probing for traumatic memories and Ganaway's (1991) advice about SRA memories. Zealous conviction is a dangerous substitute for an open mind. Psychotherapists, counselors, social service agencies and law enforcement personnel would be wise to be careful how they probe for horrors on the other side of some presumed amnesic barrier. They need to be circumspect regarding uncorroborated repressed memories that return. Techniques that are less potentially dangerous would involve clarification, compassion, and gentle confrontation along with a demonstration of empathy for the painful struggles these patients must endure as they come to terms with their personal truth. There is one last tragic risk of suggestive probing and uncritical acceptance of all allegations made by clients, no matter how dubious. These activities are bound to lead to an increased likelihood that society in general will disbelieve the genuine cases of childhood sexual abuse that truly deserve our sustained attention. 

1 This manuscript is an expanded version of an invited address, the Psi Chi/Frederick Howell Lewis Distinguished Lecture, presented at the Centennial meeting of the American Psychological Association, Washington DC, August, 1992. I thank Geoffrey Loftus, Ilene Bernstein, Lucy Berliner, Robert Koscielny, and Richard Ofshe for very helpful comments on earlier drafts. I thank many others, especially Ellen Bass, Mark Demos, Marsha Linehan and Denise Park for illuminating discussions of the issues. My gratitude for the vast efforts of the members of the Repressed Memory Research Group at the University of Washington is beyond measure. NIMH and NSF have generously supported the underlying research on memory.

2 Please address all correspondence to Elizabeth Loftus, Dept of Psychology, University of Washington, Seattle, Washington, 98195.

3 Alaska, California, Colorado, Connecticut, Florida, Idaho, Illinois, Iowa, Maine, Minnesota, Missouri, Montana, Nevada, New Hampshire (revising an earlier law), Oregon, South Dakota, Vermont, Virginia (See, for example, Napier, 1990).

4 In Stubbings v. Webb and Another, 1991, in which a British plaintiff claimed she was raped by her adoptive father and brother while she was a child. She sued 12 years after entering adulthood. The court permitted her suit, even though she knew she had been raped, arguing that she might not have associated the mental impairment that she was experiencing with the past rapes until she gained that knowledge as an adult.

5 One curiosity about this report is that the chief investigator published a landmark book on incest six years before this study (Herman, 1981). In the earlier book, the word repression does not even appear in the index. The 40 incest victims interviewed in depth appeared to have persisting, intrusive memories.

6 Many cognitive psychologists have argued that implicit theories about any topic, for example, giftedness, guide a given society's thinking on that topic (eg. Sternberg, 1992). More generally, intuitive theories are considered constructing working models of the world that people use in the service of understanding their world (Medin & Ross, 1992).

7For those who think it is unlikely that one would ever borrow episodes from movies and popular literature and misremember them as actual events, one only has to examine Lou Cannon's (1991) biography of former President Reagan. A curious journalist who tried to verify Reagan's most famous mismemory of heroism found two that were suspiciously similar - one in the movie A Wing and A Prayer, and the other in a Reader's Digest story.

8 Since the publication of Courage to Heal, a number of cases have emerged in which women were led to believe they were abused, and now realize their memories are false (Watters, 1993). Lynn Gondolf is a case in point. During more than a year of therapy she discovered repressed memories of her father raping her. After she stopped therapy, she realized that her therapist had "coerced her and the other members of her group into imagining memories of abuse." (Watters, 1993, p. 26).

9Consider a brief sampling: Poston & Lison (1989, 1990): "Women usually do not make an immediate incest connection. They may not recall for years that the incest occurred: memories have an uncanny way of coming only when the survivor can deal with them" (p. 193), and "Many women do not remember the incest; how are they then to connect adult problems with childhood pain?" (p. l96). Farmer (1989); "You may have even repressed the memories of the abuse. The more severe the abuse, the more likely you were to repress any conscious recollection of it." (p. 52).

10 This book proudly displays an endorsement by Gloria Steinem: "This book, like the truth it helps uncover, can set millions free".

11Popular writings might also be the source of some questionable lay beliefs about early memories. Bradshaw (1990, 1992), a leading figure in the field of recovery and dysfunctional families, invites readers to consult his "index of suspicion." Do you have trouble knowing what you want? Are you afraid to try new experiences? If someone gives you a suggestion, do you feel you ought to follow it? According to Bradshaw, if you answered even one of these questions "yes", then you "can count on some damage having been done to you ..between the 9th and 18th months of your life..." (1992, p. 49). How many Bradshaw aficionados have struggled through their memories trying to find that childhood trauma?

12Dancing with Daddy was reviewed in the New York Times by Culhane (1991) The reviewer called the book "as much a story about our desperate search for one dimensional solutions to multidimensional problems as it is a story about incest and its consequences." p 18.

13 The laws in Washington and other states were changed in part due to the efforts of Kelly and Patti Barton of Seattle. Patti's allegations were described in the local Seattle newspapers, in Newsweek magazine (Darnton, 1991), and on Sally Jesse Raphael. According to the Newsweek account, Patti remembered the alleged abuse when she was in her 30's - too late to sue under the old Washington law. So, with the help of a lawyer, she lobbied to extend the statute of limitations which would now allow victims to bring suit for up to three years after their memory returns. Patti's father, who denied all charges, moved to Alaska, whereupon Patti turned her efforts on the Alaska legislature, and introduced the same legislation there.

14 In the dream, there was "a fellow who was out to hurt and perhaps rape her. In the dream she could not get rid of him. In her primal, she fought him with words until I nudged her on the shoulder with a batacka (a padded bat), and what I did was, I nudged her on the shoulder and she was there....She came unglued at that. She took the batacka, grabbed it, and began striking out. She worked and worked until she got rid of him at the door. She opened the door of the room and kicked him out and locked the door. The man looked like her manic-depressive brother but had mannerisms like Kelly (her husband). " (p. 43).

15 Here is one segment:

Son: " ...I would have dreams of uh little people ...short people coming and walking on me...walking on my bed..." ...

Psychologist: "What you saw was real." ...

Son: Well, this is a different dream...everytime a train came by, a whistle would blow and...witch would come in my window...I would wake up, but I couldn't move. It was like the blankets were tucked under and....I couldn't move my arms."

Psychologist: You were being restrained?"

Son: "right and there was somebody on top of me..." ...

Psychologist: "(Son's name) these things happened to you."... It's real. It's not an hallucination."

16British historian, Hugh Trevor-Roper (1967) pointed out that the skeptics during the 16th century witchcraze did not make much dent in the frequency of bonfires and burnings until they figured out that they had to challenge the "core" belief - that is, the belief in Satan.

Elizabeth Loftus
Psychology Department
Box 351525
University of Washington
Seattle, Washington
98195-1525    USA


e-mail:  eloftus@u.washington
http://weber.u.washington.edu/~eloftus/

Tel:   (206) 543 7184
Fax:  (206)685-3157