Throughout most of this century, public response to the sexual abuse of children had been to disregard, minimize, or deny allegations of child sexual abuse (Brownmiller, 1975; Finkelhor, 1979; Hechler, 1988). Incest was said to be a very rare event, involving fewer than two offenders per million population (Weinberg, 1955). In the 1970s reports in the clinical and popular literature indicated that child sexual abuse was more extensive than hitherto accepted (for example, Browning ex Boatman, 1977; Butler, 1978; Giaretto, 1978; Herman & Hirschman, 1977; Kempe, 1978; Meiselman, 1978). Subsequent survey studies showed the wide extent of child sexual abuse (for example, Finkelhor, 1979, 1984; Fritz, Stoll, & Wagner, 1981; Russell, 1983, 1984).
In the late 1980s reports of a previously undiscussed sexual abuse phenomenon appeared. Some therapists reported that adult clients had regained formerly lost memories of childhood sexual abuse (Blume, 1990; Courtois, 1988; Olio, 1989). Later survey studies indicated that some individuals forgot being sexually abused (Briere & Conte, 1993; Gold, Hawes, & Hohnecker, 1994; Williams, 1994, 1995). In later studies, Briere and Conte (1993) surveyed 450 adults in therapy who reported sexual abuse histories. Of these, 59 percent reported some period before age 18 when they could not remember their abuse, although there was no independent corroboration of the abuse reports. Gold et al. (1994) surveyed 105 adults in treatment for child sexual abuse issues. About 30 percent of their respondents reported a time period when they completely blocked any recollection of abuse. As was the case previously, there was no independent corroboration of the abuse incident.
Williams (1994, 1995) conducted a longitudinal follow-up study of 206 women who had been treated as girls at a hospital emergency room for sexual abuse. This study, with its prospective design, was an improvement over the previous retrospective studies. Of the 206 women, 129 (63 percent) were interviewed. Forty-nine (38 percent) did not report recall of the specific sexual abuse incident (although 33 of these 49 did recall other incidents of sexual abuse). These reports led many mental health workers to conclude that childhood sexual abuse memories were often repressed. For those clients whom they suspected of being sexually molested in childhood, attempts were made to recover abuse memories - "recovered memory therapy." A recovered memory therapist presumably can identify symptoms of forgotten childhood sexual abuse. The damaging nature of these unavailable memories makes it necessary that they be recovered and addressed in therapy. This is accomplished using memory recovery techniques (Bass &Davis, 1988; Blume, 1990;Courtois, 1988; Frederickson, 1992; Olio, 1989).
However, other clinicians and researchers questioned the validity of some recovered memories (Ceci, Huffman, & Smith, 1994; Ceci & Loftus, 1994; Goldstein & Farmer, 1992; Loftus, 1993; Loftus & Ketcham, 1994; Yapko, 1994). Also, many former memory recovery clients retracted childhood sexual abuse allegations, saying that memory recovery techniques had led to distorted or confabulated recollections (Lief & Fetkewicz, 1995; Merskey, 1996; Seltzer, 1994; Wakefield & Underwager, 1994).
In light of societal ambivalence toward child maltreatment issues, it is not surprising that many workers equate doubts about memory recovery with doubts about the reality of sexual abuse. Some memory recovery advocates characterized these concerns as denial or minimization or even as "backlash" (Fowler, 1994; Hechler, 1988; Herman & Harvey, 1993). However, skepticism about many claims of recovered memory is not equivalent to skepticism about the reality of sexual abuse. Although child maltreatment and sexual abuse are authentic social and personal problems, it is not clear that memories obtained through memory work are equally authentic. There are grounds for concern about recovered memory therapy that do not involve denial, minimization, or backlash.
This article addresses two areas of research into recovered memory therapy. The first covers research on the validity of memories derived from specific memory recovery techniques. The second entails an examination of recent information on the effects of recovered memory therapy on client outcome.
Memory Recovery Techniques
Sexual Abuse Symptoms
Several authors have suggested a post-sexual abuse syndrome. Briere (1984) proposed a trait cluster associated with sexual abuse including fear, self-injurious feelings, anger problems, chronic muscle tension, and symptoms of dissociation and withdrawal. Bass and Davis (1988) cited eating disorders, drug or alcohol addiction, suicidal feelings, and sexual problems as symptoms of sexual abuse. Blume (1990) described a "post-incest syndrome" (p. vi) and developed an Incest Survivors' Aftereffects Checklist. The checklist includes physical symptoms (for example, arthritis, eating disorders, gynecological disorders, headaches) and psychological signs (for example, phobias, depression, suicidal ideation, low self-esteem, memory gaps).
There is a prevailing belief in a post-sexual abuse syndrome among psychotherapists. Poole, Lindsay, Memon, and Bull (1995) conducted two surveys of U.S. and one of British psychotherapists about recovered memory issues. They found that 36 percent of U.S. respondents and 37 percent of British respondents interpreted physical symptoms in the process of recovering memories of sexual abuse. The Blume (1990) checklist is still widely used and has the endorsement of a prominent therapist who called it a useful "guide for the therapist" (Walker, 1994, p. 114). Despite the belief, there is no empirical evidence for a post-sexual abuse syndrome.
In their extensive review of the sexual abuse literature, Beitchman, Zucker, Hood, daCosta, and Cassavia (1992) found evidence that sexual abuse sequelae may involve a wide array of psychological disorders in adulthood. However, they found no evidence for a post-sexual abuse syndrome. Another review on sexual abuse sequelae for children by Kendall-Tackett, Williams, and Finkelhor (1993) reported similar findings. Although sexually abused children experienced more difficulties than nonabused children, no symptom characterized a majority. The authors concluded that there was no specific syndrome and no single traumatizing process.
Bulimia and other eating disorders are frequently cited as symptoms of childhood sexual abuse (Bass & Davis, 1988; Blume, 1990; Goldfarb, 1987; Poston & Lison, 1989). Pope and Hudson (1992) reviewed the scientific literature on bulimia nervosa and reported that controlled studies generally did not find that bulimic patients showed significantly higher prevalence of childhood sexual abuse than control groups. Neither controlled nor uncontrolled studies found higher rates of childhood sexual abuse than were found in studies of the general population that used comparable methods.
The issue of memory gaps in childhood bears special attention. Blume (1990) alleged that "many incest survivors 'lose' years of their childhood, most frequently ages 1 through 12. In fact, it is a serious 'red flag' to me when a client can't remember much of her childhood. This common occurrence, which psychotherapy has failed to attribute to any specific common theme, generally indicates severe physical or emotional abuse (such as incest)" (p. 108).
Siegel (1995) reviewed the literature on memory development and found that many adults who experienced a nonclinical, normal development reported that they did not recall details from childhood: "Clinical implications of these findings are that therapists should not overzealously interpret lack of recall as a pathognomic indicator of 'repressed' trauma. Also, an increased tendency to recall childhood in midlife may be a normal developmental event and not a sign that something in childhood is 'hidden' and now is intruding on consciousness" (p. 108).
A variation on the symptom notion is that of the "body memory" of sexual abuse (Bass & Davis, 1988; Blume, 1990; Courtois, 1992; Frederickson, 1992): "The body stores the memories of incest, and I have heard of dramatic uncovering and recovery of feelings and experiences through body work. This type of therapy includes massage therapy and other traditional forms of body work, as well as newer types or adaptations specifically designed to unlock memories of such childhood traumas as incest" (Blume, 1990, p. 279). In its most common form, the body memory notion is that the body stores memories at the cellular level outside the brain. The body then independently attempts to communicate about abuse through particular somatic illnesses, signs, or stigmata (see, for example, Smith & Pazder, 1980).
There is no evidence that memories are "stored" anywhere but in the brain. However, there are studies suggesting that the brain processes and stores information in more than one way. Two proposed processes are implicit and explicit memory. Implicit memory develops earlier and refers to behavioral memory processes. It also involves emotional and sensory recall. Explicit memory develops later and refers to declarative, narrative memory (Schacter, 1987; Schacter, Chiu, & Ochsner, 1993; Squire, 1992a, 1992b).
"Body work" based in this less extreme notion assumes that implicit memories may be translated into explicit, narrative form. There is no empirical evidence that it can make this translation. Indeed, there is no evidence that such memories are reliable or even that implicit memories themselves are necessarily reliable.
Studies of the development of autobiographical memory indicate that subsequent discussion of early (implicit) memories can change recall of an event (Destun & Kuiper, 1996; Fivush & Hudson, 1990; Nelson, 1993). This opens the door for error and distortion of the implicit memory in the explicit memory system. There is no...