The Scientific Review of Mental Health Practice

Objective Investigations of Controversial and Unorthodox Claims in Clinical Psychology, Psychiatry, and Social Work


Freud’s Theories of Repression and Memory

A Critique of Freud and False Memory Syndrome by Phil Mollon

Phil Mollon’s Freud and False Memory Syndrome (2000) gives a succinct account of Freud’s early clinical experiences, and of his theories of repression and the nature of memory. Mollon’s discussion of some of Freud’s early cases enable the reader to gain an idea of the roots from which Freud’s theories developed, and afford the present writer the opportunity to consider these in some detail. It will be evident that what follows is not a conventional review of the book. Though Mollon’s expositions will be critically examined, they will predominantly be used as a convenient base for a discussion of Freud’s clinical claims and his theories of repression and memory in relation to childhood experiences.

Mollon opens with a brief discussion of the bearing of Freud’s early clinical experiences on current concerns as to the apparent recovery of memories of childhood sexual abuse, mostly in the course of psychotherapy (Mollon, 2000, pp. 3–8). However, comment on this is postponed until Freud’s ideas on memories of early childhood have been considered and his clinical claims at the beginning of his psychoanalytic career have been critically examined. Before discussing details of Freud’s early case histories in Studies on Hysteria (1895b), Mollon addresses “certain misconceptions of this book [that] have been propagated and recycled in the current debates about recovered memory,” and he selects a passage from Daniel Schacter’s book Searching for Memory (1996) as typifying what he calls “the attempt to assimilate Freud into the category of ‘recovered memory therapist’ ” (Mollon, 2000, pp. 9–10). As Mollon reports, Schacter writes that “Sigmund Freud and Josef Breuer’s classic studies of hysteria described patients who could not explicitly remember childhood sexual abuse, but experienced disabling fears, nagging anxieties, intrusive thoughts, or disturbing images that reflected implicit memory for the trauma. However, these cases proved difficult to interpret because independent corroboration was often lacking. (Schacter, 1996, p. 274)” (Mollon, 2000, p. 9).

Mollon describes this passage as “typical of the rhetoric of the false memory debate” and says that “it can only be based on a fantasy of what Freud said rather than an actual reading of the text, for the five case histories do not concern problems resulting from childhood sexual abuse” (p. 10). From this passage and what follows (pp. 10–17) it is evident that he has taken Schacter to be referring only to the cases reported in Studies on Hysteria (1895b). However, a closer reading of the passage in question indicates that what Schacter had in mind was the whole period in the late 1800s during which Breuer and (especially) Freud investigated cases they diagnosed as hysteria. Schacter’s mentioning childhood sexual abuse is misleading to the extent that it seems to imply that such a trauma was frequently implicated throughout these years, rather than during the brief seduction theory episode (1895–1897). Also, many of these cases involved specific physical symptoms in addition to the symptomatology listed by Schacter. Otherwise he provides a reasonably accurate overview of the cases reported by Breuer and Freud in this period, with the caveat that their contention that the symptoms reflected implicit memories of traumas was assumed rather than demonstrated, as is implied by Schacter’s noting the general lack of independent corroboration.

In his introductory remarks on the cases in Studies on Hysteria, Mollon (p. 9) alludes to what he describes as a “discovery” that he partly ascribes to Breuer, who provided the first case history in the book (that of Anna O.). He writes that when a patient was encouraged to talk freely it was often possible to trace the origin of a patient’s symptom, the psychodynamic understanding of which would result in its removal. However, Mollon makes no mention of the doubts that have been expressed in recent times about the traditional accounts of the Anna O. case, which have been shown to be misleading in several important respects. For instance, some of the patient’s symptoms diminished spontaneously (Hirschmüller, 1989, pp. 286, 289), the patient was far from cured when the treatment terminated (Ellenberger, 1972; Hirschmüller, 1989, pp. 113–116, 201), and it is evident from the discovery of his original notes that the case history written by Breuer (partly from memory) some 13 years after the events is a retrospectively idealized account of the treatment (Webster, 1995, pp. 122–135; Borch-Jacobsen, 1996b, pp. 49–61; Macmillan, 1997, pp. 3–24).

Mollon goes on to provide a brief summary of Freud’s 4 main cases in Studies on Hysteria, and concludes: “Contrary to Schacter’s assertion, the cases in the Studies do not concern repression of memory, nor recovered memory, and are not related to childhood sexual abuse.” Rather, “the four cases treated by Freud are all presented as examples of emotional conflict and trauma, resulting in undischarged and dammed up affect or libido.” His final conclusion is that “[n]one of these cases, as described by Freud, involved repression of memory. The objects of repression were thoughts, feelings and desires.” (pp. 16–17)

Mollon is certainly correct in saying that sexual abuse plays little role in the cases in Studies (though, in one of the main cases, Katharina reported an instance of attempted sexual assault in her early teens), but his final conclusion is questionable. In Breuer and Freud’s theory the undischarged affect associated with repressed memories of traumatic events give rise to ideas, and the goal of the therapy is to enable these unconscious ideas to reach consciousness so that the patient can abreact the affect. This aim can be achieved only if the original event is recalled, and Studies contains several examples of patients supposedly recovering forgotten memories of incidents associated with disturbing ideas (e.g., Freud, 1895b, pp. 120, 178, 179, 180, 273, 274). Breuer and Freud (1895b) claimed that patients’ symptoms disappeared “when we had succeeded in bringing clearly to light the memory of the event by which it was provoked and in arousing its accompanying affect . . .” (pp. 6, 255), and it was their contention that this happened on occasion with their patients.

In his brief accounts of the 4 main cases described by Freud in Studies, Mollon (pp. 10–17) follows Freud’s versions of events, writing that (with the exception of the case of Emmy von N.) “painful emotional conflicts and dilemmas involving desire, guilt and shame” (p. 17) were at the root of the patients’ symptoms. But for all Freud’s confident assurances to the reader, it remains a fact that these are, for the most part, assumptions on his part that he has by no means demonstrated. [1] As Slater (1965) has pointed out, “trouble, discord, anxiety and frustration are so prevalent at all stages of life that their mere occurrence near to the time of onset of an illness does not mean very much” (p. 1399).

Mollon’s propensity to accept uncritically whatever Freud reported is illustrated by his writing in relation to Katharina that the patient “agreed with [Freud’s] reconstruction” of the supposed traumatic idea he had inferred (p. 14), when according to the case history she actually said, rather unconvincingly: “It may well be that that was what I was disgusted at and that that was what I thought” (Freud, 1895b, p. 131). A more critical commentator might observe that Katharina’s words suggest that Freud’s patients may have been inclined to “confirm” whatever he insisted was at the root of their symptoms. (For a perceptive discussion of the Katharina case, see Swales, 1988, pp. 125–129.)

That it was not unusual in this period for Freud, rather than the patient, to provide the supposed traumatic idea or memory is evident from his writing in Studies (1895b) that “the principal point is that I should guess the [traumatic] secret and tell it to the patient straight out,” and that “it is of use if we can guess the way in which things are connected up and tell the patient before we have uncovered it” (pp. 281, 295). Freud’s early cases are usually represented as demonstrating his facilitating patients’ recall of forgotten traumatic incidents, thereby relieving the patients’ symptoms (Freud, 1910, pp. 22–23), but a critical examination of the relevant case histories reveals that this is an idealized story, originating not only from the impression created by Freud’s contemporary accounts, but also from misleading retrospective reports in his general expositions. For instance, in the “Five Lectures on Psychoanalysis” that Freud delivered at Clark University in 1909 he alluded to the case of Elisabeth von R. in terms that must have seemed immensely impressive to his audience. He reported that the patient had “completely forgotten” a specific “scene by her sister’s [death]bedside” and “the odious egoistic impulse” that she experienced at that moment, but “remembered it during the treatment and reproduced the pathogenic moment with signs of the most violent emotion, and, as a result of the treatment, she became healthy once more” (1910, pp. 24–25). However, it is evident from the case history in Studies that Elisabeth had not forgotten the scene by her sister’s deathbed (1895b, pp. 142–143). Moreover, close reading of the relevant paragraph in the case history reveals that Freud did not explicitly state that Elisabeth recalled the traumatic “impulse” (though his readers may well be left with that impression from the artfully composed passage in question) (pp. 157–158); rather, he inferred that it was present as an unconscious memory from her strenuous “resistance” when he “put the situation [i.e., his own surmise] drily before her.” Furthermore, his claim that her leg pains were cured as a result of his supposedly uncovering the “pathogenic moment” (1910, p. 25) does not withstand critical scrutiny (1895b, pp. 158–160; see Webster, 1995, pp. 163–164). [2]

The last part of this case history (1895b, pp. 145–160) exemplifies the way in which Freud’s excessive confidence in his own analytic inferences results in his entwining them with factual information in such a way that readers are almost insidiously led to view the patient’s experiences through Freud’s interpretative lens. Mahony (1984) observes that he is “ever struck by Freud’s early written endorsement of hypnosis and enthusiastic description of its cures as compared with his subsequent scepticism about its means and achievements,” and notes more generally the “assured tone” and “persuasive force” of Freud’s expositions that contrast with his “subsequent retractions” when he has moved on to new procedures and theories (p. 160). These observations are especially pertinent to the next stage of Freud’s career, the period immediately following the publication of Studies on Hysteria.

The Seduction Theory

Following his discussion of the cases in Studies, Mollon (pp. 17–18) turns to Freud’s 1896 papers, in which he claimed that for all his patients he had uncovered repressed memories of sexual abuse in early childhood. As Mollon notes later (p. 40), during the period in question Freud was using an early version of his technique of free association, starting from a symptom, supplemented as necessary by his pressure procedure. This involved his pressing on the patient’s forehead and insisting that something would come to mind, with the expectation that the resulting ideas or images would eventually lead to material relating to an event or idea that could be identified as the determining trauma (Freud, 1895b, pp. 268–272). In the seduction theory papers, Freud claimed that the sexual abuse he had supposedly uncovered most commonly occurred at age 3 or 4, and for 2 cases at half these ages (Freud, 1896a, p. 152; 1896b, p. 165). Furthermore, he reported that for most of his patients he had found that there had been 2 or more infantile abuse experiences, and that in a few instances “the accumulation of sexual experiences coming from different quarters was truly amazing” (1896c, p. 208). Although he had not reported having uncovered any cases of sexual abuse in infancy before he alighted on the seduction theory in early October 1895 (Masson, 1985, pp. 144, 145), only 4 months later he claimed to have analytically “traced back” from his patients’ symptoms to repressed memories of multiple childhood sexual experiences, mostly in infancy, for no fewer than 16 patients (13 “hysterics” and 3 “pure” obsessionals: Freud, 1896a, pp. 142, 152, 155; 1896b, pp. 168–169). But, as he himself acknowledged, he did not in these papers provide “the evidence needful to support [his] assertions” (1896a, p. 162; see also 1896c, p. 203); as Eissler (2001) observes, he “spread his clinical evidence surprisingly thin” (p. 113). So Freud’s readers were effectively left to take on trust his claim that he had succeeded “in awakening the psychical trace of a precocious sexual event” (1896a, p. 153) for all his patients.

This raises the issue of what exactly his evidence comprised. Eissler (2001, p. 137) asks pertinently: “Was a period of not quite four months really all that was needed” to uncover the data for all the cases alluded to in the first seduction theory papers? [3] In the “Aetiology” paper Freud claimed that the patients “reproduced” the infantile sexual experiences that he “warn[ed] them” would emerge (1896c, p. 204), but it is far from clear what he meant by “reproductions.” Considering his 1896 papers and his later sketchy reports to Fliess about individual cases, all one can say is that they seem to have ranged from “violent sensations,” which he interpreted as indicating the emerging of repressed memories, to the production of fragmentary ideas and images, which he took to be representations of the anticipated “sexual scenes” (1896c, p. 204; 1896a, p. 153). It seems likely that in some cases Freud relied almost entirely on the symbolic interpretation of symptoms that lay at the heart of his etiological determinations. [4]

This is apparent in the one case of hysteria or obsessional neurosis in the 1896 papers for which he gives specific details of his interpretive procedure (1896b, pp. 172, 172 n.1), and in the case of Miss G. de B. reported to Wilhelm Fliess in January 1897 (Masson, 1985, pp. 220–221). (In the latter case, having inferred an infantile scene of fellatio from eczema around the mouth and other indications, Freud wrote that he “thrust the explanation” at the patient, and then “threatened to send her away” if she persisted in her skepticism.) Given also the extraordinarily short time between Freud’s postulating the theory and his claim of 100% corroboration for 16 patients, the situation is best summed up by Schimek’s (1987) conclusion that “the knowledge of the original trauma, whether considered as unconscious memory or fantasy, was based on Freud’s interpretation and reconstruction; it was not directly revealed by the patient” (p. 960), a view originally put forward by Cioffi in the early 1970s (Cioffi, 1972, 1974; see also Esterson, 1993 [pp. 11–31], 1998; Israëls and Schatzman, 1993; Scharnberg, 1993; Schatzman, 1992).

Mollon’s propensity to take Freud’s clinical claims at face value is apparent in his discussion of the instances of supposedly “objective” corroborations adduced in “The Aetiology of Hysteria” (1896c, p. 206). He notes that Freud argued that there would be “unassailable” proof if there were confirmation from someone other than the person being analysed, praises him for having put forward the “standard of proof requested by those who support today’s false memory societies,” and goes on to report Freud’s examples. In the first of these, Mollon writes, “a brother confirmed his childhood sexual activity with his sister” (p. 35). If we turn to what Freud actually wrote, however, we find that he did not deliver what he had promised, for the brother “of his own accord confirmed—not, it is true, his earliest sexual experiences with his sister (who was the patient)—but at least scenes of that kind from later childhood, and the fact that there had been sexual relations dating further back” (1896c, p. 206). So the crucial infantile “sexual scene” was not confirmed by the brother. Moreover, Freud failed to provide some important details: Did he obtain the information directly from the brother, or was it conveyed to him secondhand? What, precisely, was the sexual behavior allegedly confirmed? How can the brother have confirmed that there had been sexual relations dating further back, i.e., before those incidents he actually recalled? These are all elementary questions that need answering before we attain the “standard of proof” alluded to by Mollon, even to provide satisfactory corroboration of the later sexual activities. In addition, Freud is not consistent in his claims relating to such activities between siblings. In “Heredity and the Aetiology of the Neuroses” (1896a) he stated that where there had been sexual relations between a slightly older boy and his patient (mostly brother and sister), this had “sometimes continued . . . until the little guilty parties reach puberty” (p. 152). Yet in the “Aetiology” (1896c) paper, he asserted that such practices had been “often prolonged beyond puberty” (p. 208).

In the face of several such “inconsistencies and incongruities” (Eissler, 2001, pp. 107–117), it should be evident that Freud’s clinical claims in the “Aetiology” paper need to be treated with considerable caution, as is clear from the second instance of corroboration cited by Mollon. This involved 2 female patients who, in infancy, had supposedly had sexual relations with the same man. As Freud (1896c) expressed it, “A particular symptom, which was derived from these childhood events [sexual scenes à trois], had developed in both women, as evidence of what they had experienced in common” (p. 206). This hardly lives up to the claim of “objective confirmation” asserted by Freud; as Smith (1991) observes, “One would need to know how it was that Freud reached his conclusions in order to even begin to assess his claim” (p. 13). And there is another question to be considered here. Freud (1896c) stated that “out of eighteen cases,” he had obtained “an objective confirmation . . . in two” (p. 206). Yet, as Scharnberg (1993) has pointed out (pp. 156–157), the second instance involves the supposed mutual experiences of 2 patients, making a total of 3 of the 18. (Note that Freud did not state that either of these patients explicitly implicated the other in the alleged events; and in any case, such a circumstance would not have met the “objective” standard claimed, since it could have arisen as a product of his flawed clinical procedure.) This anomaly casts further doubt on the second “confirmation.” In short, the information adduced by Freud does not provide the objective confirmatory evidence that he claimed. It is no easy task to obtain reliable corroboration of early childhood events the memories of which are supposedly recovered 2 or more decades later (Pope, 1997; Pope, Oliva, & Hudson, 2002), and Freud’s report does not come close to providing such evidence.

Equally problematic are Freud’s claims about his having uncovered material concerning sexual relations between very young children. Mollon writes that “Freud provides the further interesting observation that where there had been a relationship between two children, he sometimes found that the boy who initiated the sexual activity had previously been seduced by an adult female and that he had subsequently tried to repeat with the little girl exactly the same activities that the adult had performed on him” (p. 37; see Freud, 1896a, p. 152; 1896b, pp. 164–165; 1896c, p. 208). The first question to ask here is: What was the source of the information that the boys in question (the “assailants” of some of his female patients) had themselves been previously seduced by adult females? And, even more pertinently, how could Freud have discovered that the supposed abusers in these cases had in early childhood been subjected to exactly the same sexual practices as they had perpetrated on the infant girls? (The problem is, of course, compounded by the fact that the supposed infantile experiences of his own patients were themselves doubtful analytic findings, purportedly of deeply repressed unconscious memories.) Freud made no attempt to provide any information about how he acquired this unlikely knowledge, and Mollon evidently requires none. That Freud (1895b) asserted that he was able to obtain this knowledge about the abusers “with certainty” in a few cases (pp. 164–165) tells us more about his reporting practices than about the reliability of his claims. It is of more than passing interest to note that he never again mentioned these remarkable findings once he had abandoned the seduction theory.

After a lengthy passage (pp. 28–40) in which he presents a highly favorable account of the arguments adduced in “The Aetiology of Hysteria,” Mollon alludes (pp. 40–41) to Freud’s use of the “pressure” procedure, and writes of the danger of its generating confabulations rather than authentic memories. However, he is mistaken in saying (p. 41) that it requires “today’s knowledge” to appreciate this. That patients might confabulate as a result of suggestions emanating from the physician, even unconsciously, was widely recognized in the last decades of the 19th century (Borch-Jacobsen, 1996a, pp. 21–25), and it was precisely on this issue that much contemporary criticism of Freud’s clinical findings was based (see Kiell, 1988, pp. 68, 74, 82; Israëls and Schatzman, 1993, p. 43–44). However, Freud’s contemporaries would not have appreciated that when he asserted that no physician who did not use his new clinical methodology was in a position to reject his claims (1896c, p. 220), he was alluding to his technique of analytic reconstruction, which did not necessitate patients actually reporting the crucial unconscious “memories” he believed he had uncovered (1896c, pp. 191–192; 1924, p. 196). Those who are inclined to accept the “false memories” version of the seduction theory episode should consider whether it is conceivable that Freud’s 6 obsessional patients reported to him that they had recovered repressed memories of both having been sexually abused in infancy and also having themselves, a few years later, sexually abused an infant sister (1896a, p. 155; 1896b, pp. 168–169). Similarly, those who believe that Freud’s seduction theory claims were for the most part based on authentic reports by patients need to consider if it is likely that the 6 obsessionals should have had these remarkable multiple experiences in childhood, so conveniently in accord with his theoretical requirements (Masson, 1985, pp. 144, 154, 209; Freud, 1896b, pp. 168–169). (It should be noted here that in a paper [1895a] published only 8 months before he announced the seduction theory to Fliess, in not a single one of 11 cases of obsessional neurosis that he subjected to “psychological analysis” [pp. 75–79] did Freud report that he had uncovered childhood sexual abuse.)

Mollon’s attempt to maintain a precarious balance between his favorable exposition of Freud’s arguments in the “Aetiology” paper and an acknowledgment of serious deficiencies in his clinical procedures leads him into inconsistencies. In two separate passages (pp. 34, 38) he commends Freud for being “highly alert” to the dangers of patients’ confabulating, yet he implicitly exculpates Freud’s supposed failure to distinguish confabulations from authentic memories by reason of his not having “today’s knowledge” of “the propensity for hypnotism and related methods to generate confabulation rather than authentic memories” (pp. 34, 41). Mollon’s suggestion that such an awareness requires current knowledge is also contradicted by his earlier observation (p. 34) that in a passage in the “Aetiology” paper Freud (1896c) was anticipating colleagues’ objections that the patients were likely to confabulate under the influence of his clinical procedures (pp. 203–205).

Freud’s confident rejection of the “confabulations” challenge to his 1896 clinical claims can be seen in a more revealing light, namely, that what he regarded as his main “proof” (1896c, p. 205) for his purported findings was based on his new analytic technique for reconstructing repressed memories, not on patients’ reports. As he wrote 2 years later: “By a curious circuitous path [i.e., psychoanalysis], . . . it is possible to arrive at a knowledge of this aetiology [of the psychoneuroses] and to understand why the patient was unable to tell us anything about it” (Freud, 1898, p. 267). [5] In his retrospective accounts, when he needed an explanation for his seduction theory “error” that did not raise doubts about the reliability of his analytic technique, he disingenuously asserted that his erroneous claims resulted from his being misled by patients’ reports (1914, p. 74; 1925, pp. 33–34). This is in blatant contradiction to his writing at the time that he would have “to accuse [himself] of blameworthy credulity” if he did not possess more conclusive evidence than reports from patients (1896a, p. 153).

At the end of his section on the seduction theory episode Mollon takes issue with certain critics of Freud’s accounts of the seduction theory in terms that need to be quoted in full:

Some recent commentators have argued that Freud dishonestly distorted the true situation regarding his change of mind. They argue that Freud first coerced his patients into producing false scenes of childhood sexual abuse, which were really his own inventions, and then, realising his error, he subsequently claimed that his patients had spontaneously told him of their sexual scenes, and in this way had given him the idea of childhood sexuality and the Oedipus complex.

They further point out that Freud originally claimed the sexual abusers were nursemaids, governesses and so on, but later claimed that most of his patients had said they were abused by their fathers—a claim, it is alleged, that would have been convenient for his account of how he discovered the Oedipus complex whereby the child desires the father or mother in phantasy.

These points have little substance. They are all covered by Freud’s actual texts and his later footnotes. He freely admits that in early papers he played down the role of seduction by fathers because the idea would seem too disturbing to his readers (as well as perhaps to himself). His change of mind about how crucial sexual abuse was can be seen as part of a wider and consistent pattern whereby Freud continually revised his views in the light of his further clinical experience and thought. (pp. 45–46)

Mollon references Crews et al.’s The Memory Wars (1995) and my Seductive Mirage (1993) in relation to this passage, but neither of us argues that the patients produced “false memories” of infantile sexual abuse. Our position is that the patients did not report recollections of infantile sexual abuse experiences at all, and that the “sexual scenes” were essentially Freud’s own analytic reconstructions made on the basis of the symbolic interpretation of symptoms, patients’ associations, and fragmentary ideas and images produced in the course of his applying his pressure procedure (Crews, 1995, pp. 57–59; Esterson, 1993 [pp. 11–31], 1998). Nor are Mollon’s other assertions in this passage valid. The problem is not that Freud revised his views, but that he made retrospective changes regarding factual details. Mollon purports to refute the contention that there are inconsistencies in Freud’s reports and that he later tendentiously misrepresented his original claims, but he can do so only by selective quotation from Freud’s later writings (pp. 43–45), and, in one instance, an allusion to a nonexistent statement (see below). Numerous inconsistencies in Freud’s accounts have been cited in several articles, [6] the most comprehensive of which is largely devoted to documenting them in some detail (Esterson, 2001). One crucial inconsistency can be seen from Freud’s report of the episode in New Introductory Lectures on Psychoanalysis (1933), in relation to what he describes as “an interesting episode in the history of analytic research”:

In the period in which the main interest was directed to discovering infantile sexual traumas, almost all my women patients told me that they had been seduced by their father. I was driven to recognize in the end that these reports were untrue and so came to understand that hysterical symptoms are derived from phantasies and not from real occurrences. It was only later that I was able to recognize in this phantasy of being seduced by the father the expression of the typical Oedipus complex in women. (p. 120)

One has only to compare the first sentence in this passage with Freud’s statements in “Aetiology” (1896c) that the patients “have no feeling of remembering the [sexual] scenes,” and that they assured him “emphatically of their unbelief” (p. 204), to recognize the misleading nature of the quoted paragraph. [7]

Another anomaly in the 1933 account is Freud’s implication of fathers as the abusers of his female patients during the period in question. As Mollon indicates (p. 46), in one of the earlier of the 1896 papers Freud (1896b) implicated “nursemaids, governesses and domestic servants,” while “teachers . . . figure with regrettable frequency”; in addition, children (mostly brothers) slightly older than their infant girl victims were also prominent among the culprits (p. 164). Mollon claims to explain the stark contradiction between this report and later accounts with his assertion that Freud “freely admits” that in early papers he had played down the role of seduction by fathers because the idea would seem too disturbing (p. 46). He gives no reference at this point, and in fact no such statement is to be found in Freud’s writings. As is well known, in footnotes added in 1924 to Studies in Hysteria, Freud acknowledged that in 2 of the cases discussed in that book he had replaced “father” with “uncle.” However, as Mollon himself points out, none of the cases in Studies were relevant to the seduction theory. (The 2 cases in question concern attempted sexual assaults after puberty, and the memories had not been repressed [1895, pp. 134, n.2, 170, n.1].) Freud reported in 1924 that he had engaged in the subterfuge in the Studies cases for reasons of “discretion,” i.e., to conceal the identities of the individuals concerned, as was customary at that time (as now). As Swales (1988, pp. 94–96) has shown in the case of Katharina, there is no reason to doubt that this is why Freud represented the father as “uncle.” (A century later, Swales was able to trace the identities of the individuals in the Katharina case history with the help of the later information.) The fact is that nowhere in his writings did Freud indicate that he played down the role of fathers in the seduction theory papers; he simply changed his story in his later accounts without reference to his original reports (1925, pp. 33–34; 1933, p. 120).

There is in fact no evidence for the commonly held view that Freud predominantly implicated fathers and that he concealed this in the 1896 papers. The seduction theory required only that there be unconscious memories of infantile sexual abuse; the identities of the abusers were irrelevant, and this is reflected in the supposed culprits listed by Freud at the time, which were in accord with the typical culprits cited in the contemporary literature on child sexual abuse (Makari, 1998, p. 642). It was only after December 1896, when on theoretical grounds he alighted on the notion that fathers were likely to be the abusers (Eissler, 2001, p. 153; Masson, 1985, p. 212), that he first mentioned fathers as the supposed culprits in letters to Fliess (and then only in a minority of cases) (Esterson, 1998, p. 9; Masson, 1985, pp. 220, 224, 238). Several writers have shown that there are cogent grounds for rejecting the notion that the categories cited in the seduction theory papers concealed a preponderance of fathers among the supposed culprits (Eissler, 2001, pp. 151–159, 195–196; Esterson, 1998, pp. 9–10; McCullough, 2001, pp. 6–7; Schimek, 1987, pp. 950–951). As for Mollon’s suggestion that Freud may “perhaps” have found the idea of fathers as abusers too disturbing, one has only to turn to Freud’s letters to Fliess to see that after December 1896 Freud was actively seeking to implicate fathers (Masson, 1985, pp. 212, 220, 249, 286). [8]

Mollon contends that Crews and I unjustly maligned Freud when we stated that he retrospectively tailored his seduction theory claims to make his later explanations more plausible. Now in his “Aetiology” (1896c) paper, Freud described the “sexual scenes” that “were shown in [his] analyses” to be at the root of “exceedingly common hysterical phenomena”: they were, he wrote,

very repellent to the feelings of a sexually normal individual; they include all the abuses known to debauched and impotent persons, among whom the buccal cavity and the rectum are misused for sexual purposes. . . . People who have no hesitation in satisfying their sexual desires upon children cannot be expected to jib at finer shades in the methods of obtaining that satisfaction. . . . Where the relation is between two children, the character of the sexual scenes is none the less of the same repulsive sort. . . . (pp. 214–215)

Clearly such horrendous experiences were not in accord with the later story that the patients were, for the most part, conjuring up unconscious wishful phantasies [9] to cover up disturbing memories of infantile masturbation or (in its final form) of infantile Oedipal desires. After his abandonment of the infantile sexual trauma theory, Freud never again referred to the brutal nature of many of his original abuse claims, retrospectively bringing them into closer accord with his new theory by alluding to patients’ wishfulfilling phantasies of seduction. Then there are the tendentious changes, alluded to above, that occurred with regard to the identities of the supposed abusers. There was no mention of fathers in the first 2 retrospective reports, in which Freud maintained that the putative phantasies of seduction were concealing memories of objectless infantile autoerotic activities (1906, p. 274; 1914, p. 17–18), in line with his theories of psychosexual development expounded in Three Essays on the Theory of Sexuality (1905b, p. 189). The first appearance of fathers in his public reports of the episode was in 1925, when he asserted that they had been implicated in the case of his female patients. And it was then, for the first time in relation to the seduction theory patients, that he used his Oedipal desires explanation for their supposed “wishful phantasies” (1925, p. 34).

There is a more basic problem with the account endorsed by Mollon. He cites (p. 45) Freud’s writing that “[i]f hysterical subjects trace back their symptoms to traumas that are fictitious, then the new fact which emerges is precisely that they create such scenes in phantasy . . .” (1914, p. 17), and comments that this led to Freud’s fundamental discoveries concerning childhood sexuality, the Oedipus complex, and the pervasive role of phantasy. But Mollon acknowledges that Freud’s “coercive” clinical procedure at the time was not “advisable” (pp. 41, 63), and that the “Aetiology” paper was “fundamentally flawed” (p. 42). Even Eissler, notoriously reluctant to criticize Freud, writes disparagingly of the seduction theory papers, documenting the “contradictions they contain,” and concluding that Freud’s then clinical procedures “reduce the probability of gaining reliable data to zero” (Eissler, 2001, pp. 107, 115). Yet Mollon treats Freud’s alleged findings (now mostly considered to be patients’ phantasies) as genuine products of the patients’ unconscious, i.e., he apparently accepts that many of the highly dubious infantile “sexual scenes” that Freud claimed to have uncovered were false memories that have a psychodynamic function. But such a fundamentally flawed clinical procedure cannot provide material that can validly be used as evidence for anything, and certainly not for epoch-making discoveries as Freud later claimed (1914, p. 17–18; 1925, p. 34–35).

Truth and Falsehood

Some of the widespread misconceptions in relation to Freud’s retrospective reports of his clinical experiences in 1895–1897 result from the ambiguity in his use of the word Phantasie. As already indicated (n.9), in the Standard Edition this is translated as “phantasy,” and generally refers to unconscious ideas that Freud reconstructed by analytic inference (see Jones, 1955, p. 272; Esterson, 2001, pp. 343–344). In his first retrospective account of the seduction theory episode he alluded to “these unconscious phantasies” that supposedly covered up memories of infantile sexual activities, and a short time afterward he wrote that “hysterical symptoms are nothing other than unconscious phantasies brought into view through ‘conversion’” (Freud, 1906, p. 274; 1908, p.162, emphasis added). It is significant that in the second of these papers he wrote that “the technique of psychoanalysis” enabled him “in the first place to infer from the symptoms what these unconscious phantasies are” so that he could “make them conscious to the patient” (1908, p. 162, emphasis added). This is consistent with a historical account in which he reported, with regard to the early period immediately after he gave up using direct hypnosis in favor of free association, that “this material [obtained from patients] did not bring up what had actually been forgotten, but it brought up such plain and numerous hints at it that, with the help of a certain amount of supplementing and interpreting, the doctor was able to guess (to reconstruct) the forgotten material from it” (1924, p. 196, emphasis added). It is in this light that one should view Freud’s accounts of his patients’ productions, not only in the seduction theory period but also in relation to the material discussed below.

When he moves on to discuss Freud’s general views about memory, Mollon quotes from a passage in Introductory Lectures on Psychoanalysis in which Freud wrote about the difficulties in deciding whether infantile experiences uncovered in analysis are authentic memories or phantasies (p. 48; Freud, 1916–1917, p. 367). Mollon’s discussion treats the issue as one of interpreting material reported by patients, and he observes that Freud “continued to see recovered memories as likely to comprise truth and falsehood.” But if we turn to the section in question in Introductory Lectures we find that the matter is not so straightforward (pp. 367–371). A close reading of the section indicates that the “truth or falsehood” that Freud was concerned about does not generally relate to actual memories recounted by the patient, but to analytically reconstructed infantile material he inferred to be in the patient’s unconscious. Early in the section he referred to “infantile experiences brought to light by analysis” (p. 367), and the ambiguity inherent in this formulation is also evident in his using the word “phantasy” in a way that often creates the impression that the crucial infantile material had come from the patients themselves, rather than being analytically reconstructed by Freud, as is certainly the case for the bulk of the material alluded to in this section. Consider the first of the examples he chose shortly after describing the material in question as patients’ “invented stories” (p. 368), the “observation of parental intercourse.” An example of how he “uncovered” the repressed “memory” of such an incident is provided in the Dora case history. He reported in relation to Dora’s asthma that

Dora’s symptomatic acts and certain other signs gave me good reasons for supposing that the child . . . had overheard her father in his wife’s room at night and had heard him . . . breathing hard while they had intercourse. . . . I maintained years ago that the dyspnoea and palpitations that occur in hysteria and anxiety neurosis are only detached fragments of the act of copulation; and in many cases, as in Dora’s, I have been able to trace back the symptom of dyspnoea or nervous asthma to the same exciting cause—to the patient’s having overheard sexual intercourse taking place between adults. (1905a, pp. 79–80)

Now all this sounds very definite, but when we examine how Freud “traced back” to this cause in Dora’s case, we find that his reported procedure involved nothing but the detailing of associations made by himself, augmented by analytic conjecture (1905a, p. 80; see Esterson, 1993, p. 43). Nowhere did Freud report that Dora recovered a memory from early childhood of having overheard her parents engaging in sexual intercourse. Another example of the traumatic observation of parental sexual intercourse was claimed in the Wolf Man case history. In this celebrated analysis, Freud interpreted a memory of a dream recollected by the patient from age 4. In the dream the boy is staring at some wolves in a tree, and this was interpreted in minute detail to represent a disguised memory of the patient, at age one and a half, observing his parents engaging in “coitus a tergo [from behind], three times repeated.” Once again the supposed occurrence was reconstructed by Freud himself on the basis of analytic inference (1918, pp. 29–47, 50–53).

In the section in question in Introductory Lectures (1916–1917) Freud referred to “all the things that are told to us in phantasy” (p. 371, emphasis added), and in what is clearly an allusion to the “primal scene” reconstruction in the Wolf Man case history, he wrote of its being “described with the most minute details” (p. 369, emphasis added; see 1918, pp. 57–59). By such means he created the impression that the material he was discussing emanated from patients, giving it an illusory substantiality it did not generally possess. He also provided another example of a patient’s “achievement,” that of “a phantasy of observing parental intercourse while one is still an unborn baby in the womb” (pp. 369–370), and by mixing inextricably infantile “scenes” that are clearly reconstructions with others that might plausibly have been reported by patients, Freud leaves the reader with the sense that the point at issue is how one distinguishes between authentic and false memories. But, as one can observe from a close reading of Freud’s cases histories and expositions, overwhelmingly these supposed infantile occurrences were not memories or phantasies in the sense in which these words are usually understood, they were reconstructed “scenes” that Freud believed he had uncovered in the patient’s unconscious by means of his analytic technique of interpretation (see Esterson, 1993, pp. 166–168; 2001, pp. 343–344).

Just how nebulous these early childhood “memories” were is evident from a footnote added to Three Essays in 1920. With reference to phantasies of the kind he had alluded to in the passage from Introductory Lectures cited above, Freud wrote that they are produced in “the pubertal period,” and are “of very general occurrence and . . . to a great extent independent of individual experience” (1905b, p. 226, n.1). He had expanded on the latter notion in Introductory Lectures (1916–1917), arguing that “these primal phantasies . . . are a phylogenetic endowment” (pp. 370–371, italics in original). In other words, he posited that they frequently do not relate to actual childhood impressions stored in the individual’s unconscious mind, but to experiences that “were once real occurrences in the privaeval times of the human family,” so that “children in their phantasies are simply filling in gaps in individual truth with prehistoric truth” (p. 371). That in his comments on a passage from this section of Introductory Lectures Mollon refers to such speculative analytic reconstructions as “recovered memories” (p. 48) suggests that he has been misled by Freud’s ambiguously worded exposition.

In “Beyond the Pleasure Principle” (1920), Freud reported that in the early days of psychoanalysis, “the analysing physician could do no more than discover the unconscious material that was concealed from the patient, put it together, and, at the right moment, communicate it to him” (p. 18). As psychoanalytic treatment progressed it remained the case that generally the patient “cannot remember . . . the essential part” of what has been repressed, and so “acquires no sense of conviction of the correctness of the construction that has been communicated to him.” The repressed material, however, emerges as “reproductions” in the form of transference phenomena, and these “always have as their subject some portion of infantile sexual life—of the Oedipus complex, that is, and its derivatives” (ibid.). This passage provides another indication that the infantile sexual experiences Freud claimed to have uncovered were analytic reconstructions rather than memory reports from patients. This accounts for the rather extraordinary assertion, in the section of Introductory Lectures (1916–1917) discussed above, that whether childhood sexual abuse experiences had “occurred in reality,” or had been “put together from hints and supplemented by phantasy”: “The outcome is the same, and up to the present we have not succeeded in pointing to any difference in the consequences . . .” (p. 370). Since he had no way of deciding whether his analytic reconstructions represented unconscious phantasies or authentic memories, he had little option but to take the position that it is of no practical consequence one way or the other. This, to reiterate, is why discussion of whether this kind of material represents true or false memories relating to early childhood is beside the point: there were generally no patients’ recollections of the “memories” or phantasies adduced by Freud in such instances.

Freud’s Views on Memory

As an introduction to his discussion of Freud’s view of memory, Mollon writes: “It is sometimes asserted that Freud held that events of a person’s life were all recorded accurately somewhere in the mind, like a continual video-recording, ready to be accessed if only the memories could be released from repression.” He then asks: “Is there any evidence that he held this view?” (p. 50). As I hope to show, a more accurate version of Freud’s view is that the original impressions of disturbing events, ideas, or instinctual impulses, especially those from early childhood, remain in an individual’s unconscious available for retrieval (Freud, 1937, p. 260), rather than the one that Mollon’s videorecording analogy suggests. His ensuing discussion in response to his question opens with Freud’s contention that long-forgotten material may appear in dreams, and he reports (p. 51) an example from one of Freud’s own dreams. According to Freud’s account in The Interpretation of Dreams (1900):

I had a dream of someone who I knew in my dream was the doctor in my native town. His face was indistinct, but was confused with a picture of one of the masters at my secondary school, whom I still meet occasionally. When I woke up I could not discover what connection there was between these two men. I made some enquiries from my mother, however, about this doctor who dated back to the earliest years of my childhood, and learnt that he only had one eye. The schoolmaster whose figure had covered that of the doctor in the dream was also one-eyed. It was thirty-eight years since I had seen the doctor, and so far as I know I had never thought of him in my waking life, though a scar on my chin might have reminded me of his attentions.” (p. 17)

This particular dream was first recorded in a letter to Wilhelm Fliess, in which the information is expressed a little differently (Masson, 1985, p. 271). We have no knowledge of the details of the actual dream, only Freud’s original account, in which these are inextricably mixed up with his own associations and interpretations. Without more precise information about the dream itself and the conversation Freud had with his mother, it is not possible to conclude that there is unequivocal evidence here of the recovery in a dream of forgotten knowledge from childhood.

It is revealing to note a later allusion to this dream in Introductory Lectures on Psychoanalysis (1916–1917):

I once dreamt . . . of a person . . . whom I saw clearly before me. He was a one-eyed man of small stature, stout, and with his head sunk deep in his shoulders. I concluded from the context that he was a doctor. Luckily I was able to enquire from mother . . . what the doctor at my birthplace (which I left when I was three) had looked like; and I learnt from her that he was one-eyed, short, stout and with his head sunk deep in his shoulders. (p. 201)

Compare this with Freud’s earlier report, in which no mention of the physical details of the man in the dream was provided other than that his face was indistinct. Moreover, in his original report to Fliess he did not state that the “dream figure” in question had only one eye or give any other physical details. In Introductory Lectures Freud presents his mistaken notion that he had dreamed of a one-eyed man, with all the above physical details that he “clearly saw” before him, as demonstrating that dreams “have at their disposal the forgotten material of the first years of childhood,” thereby, ironically, illustrating the unreliability of such claims. [10]

Mollon next discusses the “Screen Memories” paper of 1899, which focuses on what Freud calls “retrogressive” screen memories, i.e., apparently innocuous recollections from childhood that supposedly serve to hide a later disturbing experience. Freud sought to demonstrate his thesis by analyzing early memories of a subject (supposedly an acquaintance, but actually himself [p. 302]), purportedly showing that conspicuous elements in the memories conceal unconscious phantasies that incorporate later ideas or incidents of a troubling nature. This he accomplished using his technique of analytic inference to relate these elements to experiences that occurred in his late teenage years that supposedly gave rise to the formation of phantasies that were repressed from consciousness because of their disquieting nature. The unconscious phantasy was projected back onto an appropriate early childhood event (i.e., one that contained an “associative” relationship with the later event) thereby producing what the subject experienced as a memory. (pp. 309–320; 1901, p. 43)

There are, however, several problems with Freud’s thesis. The first is that his contention that a process of repression is necessary for a theoretical account of the forming of memories of early childhood is supported by analogy with hysterics, who supposedly show “amnesia for some or all the experiences which led to the onset of [their] illness” (1899, pp. 303–304, 320–321). This latter claim is based on the assertion that he had, by means of his psychoanalytic procedure, “often succeeded . . . in uncovering the [significant] missing portions of a childhood experience” (p. 306). However, it is precisely during this period that Freud repudiated the supposed “missing portions” (i.e., early childhood sexual molestations) that he claimed his analytic technique had uncovered in his patients diagnosed as hysterics on whom he reported in 1896. Furthermore, he argued his case for the screen memory thesis on the basis of his questionable analytic technique of interpretation (for instance, an early childhood memory of “taking flowers from a girl” supposedly conceals an unconscious phantasy of deflowering a girl he knew in his late teenage years [p. 316]). Then again, in his discussion of one memory he made some assertions about certain of his childhood experiences that led one commentator to describe it as “an implausible and self-undermining piece of analysis” (Smith, 2000, pp. 12–14). This further illustrates that Freud’s method of analytic inference lends itself to the spurious “validation” of whatever thesis he is proposing. Taking all these points into consideration, this purported demonstration that his own fragmentary early memories disguise unconscious wishful phantasies produced at a later date is unconvincing.

Mollon reiterates that some of Freud’s comments on the analyst’s methods of reconstructing repressed material may have led people to believe that “video-like representations of early events were preserved in memory” (p. 57–58), and he cites Freud’s archaeological metaphor: “All the essentials are preserved; even things that seem completely forgotten are present somehow and somewhere, and have merely been buried and made inaccessible to the subject” (1937, p. 260). Mollon writes that “on first reading” it could be assumed that Freud meant that “the psychoanalyst’s search was for a buried ‘psychical object,’ such as a memory of an event.” However, he continues, what Freud actually gives as a hypothetical example is “the reconstruction not of external but of internal events, essentially the data of introspection and empathy” (pp. 58–60). But citing Freud’s describing the reconstruction of a patient’s feelings in childhood does not refute the contention that he held that repressed impressions of actual occurrences in childhood can also be uncovered in analysis, as Mollon seems to be suggesting in this passage. In Introductory Lectures (1916–1917) Freud alluded to the reconstructing of memories of various events that had been repressed by patients (e.g., pp. 283–284, 368–369), and in the Dora and Wolf Man case histories, he claimed to have analytically reconstructed specific childhood occurrences that were retained in the patients’ unconscious minds. More generally, he asserted that “[t]he material of analysis of some patients has enabled us to reconstruct certain external happenings, certain impressive events of their childhood years, of which they have preserved no conscious memory” (1926b, p. 216). It is difficult not to conclude that Mollon’s concern to play down the role that the uncovering of specific traumas played in Freud’s theory and practice, both here and in relation to the cases in Studies on Hysteria, is an attempt to distance Freud from present-day therapies that place the recovery of repressed memories of childhood sexual abuse high on their agenda.

Mollon claims that “Freud’s observation of dreams” revealed that all kinds of impressions from experiences throughout life may be stored in some way within the brain and are available for use by dreams (pp. 51–52). However, he writes, “[T]his does not mean that it is possible to recover those memories in a conscious, coherent and accurate form.” Since Freud’s view is that dreams scavenge elements “promiscuously from diverse experiences and impressions,” Mollon continues, they “can hardly be viewed as a route to the discovery of historical reality” (ibid.) It is certainly true that Freud viewed dreams as an amalgam of diverse elements, but he also contended that they reproduce fragments (and sometimes more) of actual experiences (1900, pp. 20–21, 198–199), and, more generally, that “the dream-life knows how to find access to . . . latent, infantile experiences” (1916–1917, p. 201), as exemplified by the supposed early childhood experience of the Wolf Man cited above. In a later passage Mollon reiterates that

Freud did find evidence, for example in dreams, that knowledge and memory not available to the conscious mind appeared to be stored in some form unconsciously. However, he also—and this is absolutely crucial—emphasised the ways in which memories of experiences, especially those of long ago, are subject to all manner of distortion, mixing elements from different sources, and are rarely available to conscious recall in coherent and accurate form. In this way, Freud’s views of memory, as well as his understanding of the impact of childhood sexual abuse, seem remarkably in tune with today’s perspectives. (pp. 64–65)

It is true that Freud’s view was that memories of early childhood contain falsifications. However, rather than seeing inaccurate memories as the product of a reconstructive process over time, as is generally held to be the case in modern theories, he asserted that they were formed at a specific later date in the course of a process of repression (1899, p. 322; 1906, p. 274; 1909, p. 206, n.1). This applies both to the retrogressive screen memories alluded to above (which ceased to play much role in his writings after 1901) and to memories of early childhood in general. Freud held that the reason that most memories of early childhood are of commonplace events is that the recollections of significant experiences (e.g., infantile autoerotic activities) are effaced around puberty by a repressive process involving the production of unconscious phantasies (1901, p. 43–44; 1906, p. 274; 1909, p. 206, n.1). Leaving aside reservations about the interpretative procedure used to support his schema, it is difficult to extract from Freud’s expositions any coherent explanation of how these phantasies are produced.

Freud’s conception that inaccurate memories of early childhood are formed at a specific age is not “remarkably in tune” with current perspectives, as Mollon contends (p. 65). For a 19th-century view that better fits this description we should rather turn to Cobbe, who wrote in 1867 that “memory [is] a finger-mark traced on shifting sand, ever exposed to obliteration when left unrenewed; and if renewed, then modified, and made, not the same, but a fresh and different mark.” Noting among the several factors that distort memory those of “passion and prejudice,” she argued that “what we call a remembrance” is “not the event that it assumes to record, but the idea of it which, after twenty modifying repetitions, has left for the moment the uppermost trace in our minds” (Taylor & Shuttleworth, 1998, p. 151). A similar view was expressed by William James (1890/1950) in a passage discussing “alterations of memory” in The Principles of Psychology concerning “alterations of memory” in which he observed that “[f]alse memories are by no means rare occurrences in most of us” (pp. 373–374).

Freud’s Concept of Repression

Mollon (pp. 60–61) cites Freud as defining the basic idea of repression in his 1915 paper “Repression” (1915a) as follows: “The essence of repression lies simply in turning something away, and keeping it at a distance, from the conscious” (1915a, p. 147). He notes that what he calls this “definition” (it is not called that by Freud) [11] “does not imply that the process of repression is necessarily completely unconscious to begin with, nor that it is completely successful. It could well start with a person’s deliberate effort to put something painful out of their mind, or not to think about. This is not unusual or mysterious. Cognitive therapists call this ‘cognitive avoidance.’” However, this attempt to align Freud’s theories with modern ideas is too simplistic, as I show below. Freud certainly wrote about the conscious suppression of memories, thoughts and impulses (especially in his early psychoanalytic publications), but in later writings the emphasis tended to be on their repression at an unconscious level (e.g., 1914, p. 93; 1915a, p. 153; 1926a, p. 91; 1939, pp. 97–98; 1940, p. 187).

Some idea of Freud’s view of the processes involved in repression can be gained from an example provided in his paper on “Repression” (1915a). In the first part of the paper he gave a general exposition of these processes, frequently alluding to his “clinical observations” or to “what psychoanalysis is able to show us” as validating his account (pp. 147, 149, 151, 152). This all sounds very impressive, until one examines the first of the clinical examples he used to illustrate the process. For this purpose he chose “a well-analysed example of an animal phobia,” that of the Wolf Man:

The instinctual impulse subjected to repression here is a libidinal attitude towards the father, coupled with a fear of him. After repression, this impulse vanishes out of consciousness: the father does not appear in it as an object of libido. As a substitute for him we find in a corresponding place some animal which is more or less fitted to be an object of anxiety. The formation of the substitute for the ideational portion [of the instinctual representative] has come about by displacement along a chain of connections which is determined in a particular way. The quantitative portion has not vanished, but been transformed into anxiety. The result is fear of a wolf, instead of a demand for love from the father. (pp. 155; see also p. 155, n.1)

Summing up, Freud is saying that the little boy’s libidinal desire for his father was repressed into his unconscious, and that the sexual impulse was transformed into an animal phobia. It sounds vaguely plausible—if one can accept the idea of infant boys experiencing homosexual desires for their father and an outdated theory of psychical processes analogous to a hydraulic model of energy transformation. It is obviously crucial to discover how Freud actually uncovered the repressed idea or instinctual impulse that is central to his exposition. If we turn to the Wolf Man paper, we find that the clinical evidence is the celebrated wolf-dream in which the 4-year-old boy looks out of a window and sees a tree in the branches of which are seated a number of motionless wolves. According to Freud’s ideas, this dream to which the Wolf Man (or Freud) kept returning represents some memory or unconscious phantasy that is so disturbing it has to be concealed by a seemingly innocuous conscious idea (the wolf-dream). Freud interpreted it as a disguised representation of a memory of the boy’s parents copulating, on which occasion the infant was seized with horror at the sight of his mother sans penis. This activated severe castration anxiety, which suppressed his negative Oedipus complex, i.e., his homosexual libidinal desires for his father, what Freud described in the case history as his “wish to be copulated with by his father” (1918, pp. 29–38, 46). Following the sight of his parents copulating, “both his father and mother became wolves” and the infant boy later “identified himself with his castrated mother during the dream” (p. 47). By some mechanism not entirely clear, under the pressure of his castration anxiety, the libidinal desire for his father was repressed and the boy developed a wolf phobia. (For a fuller examination of this part of Freud’s analysis, see Esterson, 1993, pp. 72–75, 156–157.)

One looks in vain for any empirical ground on which Freud’s analysis can be supported. His argument for the repression process is circular. The presumption of repression is used to justify the search for, and uncovering of, the traumatic experience, and the inferred repressed ideas purportedly explain the repression process. The only firm evidence provided is the wolf-dream and subsidiary information—and even this turns out to be unreliable. We now know that the wolves in the dream were actually dogs, and there are serious doubts as to whether the patient even had a wolf phobia (Esterson, 1993, pp. 70, 156–157). A theory of repression based on analyses like this (not untypical) example can scarcely be said to have much of an empirical foundation.

As Mollon indicates (p. 62), by 1915 Freud’s expositions on repression were concerned primarily with the repression of instinctual impulses and the ideas arising from them, rather than of specific memories. He also notes that Freud postulated “further subtle distinctions between ‘primal repression’ and ‘repression proper’” (p. 61), but as he himself does not find that these ideas have held any clinical meaning he does not explore them. This is just as well, since Freud’s deeper explorations into the psychoanalytic processes involved in repression became increasingly incoherent (Macmillan, 1997, pp. 325–327, 477–484). Freud himself confessed: “A number of simple formulas which to begin with seemed to meet our needs have later turned out to be inadequate” (1933, p. 92). In fact he had earlier acknowledged that his exposition of the process of repression in Introductory Lectures (1916–1917) was “both crude and fantastic and quite impermissible in a scientific account” (p. 296). Nor were his subsequent attempts to overcome the theoretical problems he faced any more successful. It was no exaggeration when, in the course of attempting to make sense of Freud’s later expositions on the subject, McDougall (1936) described them as “a great tangle in which Freud lashes about like a great whale caught in a net of his own contriving” (1936, p. 60; see also Esterson, 1993, pp. 227–232).

Freud and Recovered Memories

Mollon’s concluding remarks open with the assertion that “Freud discovered that psychodynamic conflict . . . can result in neurotic (hysterical) symptoms” that “often involved sexuality in one form or another.” He goes on to allude to the “brief period of a year or so” during which Freud held his seduction theory, observing that his method of analysis at that time involved “quite coercive efforts to persuade patients to follow their chains of associations.” “Today,” Mollon notes, “such methods would be considered likely to contribute to confabulated images, falsely perceived as memories” (pp. 63–64). But Freud began using his pressure technique around 1892 or 1893, alongside the direct hypnotic procedures he had used previously (Freud, 1895b, pp. 110, n.1, 135, 145). This means that the serious deficiencies in his clinical procedures in this period undermine not only the seduction theory claims, but also his earlier clinical findings that buttressed the “discoveries” that Mollon cites.

Mollon reiterates that the renunciation of the seduction theory led to Freud’s investigating “psychic reality,” and to “the inner world of instincts, wishes and phantasies” (pp. 63–64). He goes on to relate more of Freud’s supposed discoveries, including that “even the most firmly and fervently held beliefs, such as those of religion, may rest upon unconscious and repressed infantile foundations.” In fact the “infantile foundations” in question purportedly have an even more wide-ranging influence; as Freud put it: “The reactions against the instinctual demands of the Oedipus complex are the source of the most precious and socially important achievements of the human mind” (1926c, p. 268). His supposed discoveries about infantile life also led Freud to conclude that “fear of castration is one of the commonest and strongest motives for repression and thus for the formation of neuroses” (1933, p. 87). As is often the case, Mollon’s generalizations about what Freud accomplished sound impressive—until one examines the actual content of Freud’s expositions (see Esterson, 1993, pp. 153–173).

At the beginning of his book Mollon takes to task commentators who have associated Freud with the development of therapies that may induce people to produce false memories of childhood sexual abuse, writing that Frederick Crews “argues that Freud was the originator of recovered memory therapy.” He asks, “So was Freud the bad father who impregnated society with illegitimate fantasies, giving rise to the monster of false memory syndrome a hundred years later?” and goes on to suggest that the contents of his book will show that such claims are so at odds with the truth that their assertion itself requires a psychoanalytic explanation (pp. 7–8). In fact Crews (1995) acknowledges “the chasm that yawns between the principles of the recovered memory movement and psychoanalysis in Freud’s most familiar articulation of it” (pp. 214–216). However, he also notes that there remains a “core of shared assumptions” (pp. 216–218, 274, n.3). These include one of the achievements claimed for Freud by Mollon at the end of his book (pp. 64–65): namely, that knowledge of occurrences in early childhood can be obtained from the analysis of dreams. Others are that adult emotional disorders originate from repressed memories of early traumatic experiences predominantly of a sexual nature, that these unconscious impressions remain essentially unchanged and can be uncovered by psychotherapeutic procedures, that symptoms are symbolic representations of early traumas, and that it is essential to uncover and work through repressed material from early childhood if current emotional problems are to be surmounted.

Mollon gives the orthodox psychoanalytic view that Freud should be commended for recognizing that “phantasies about childhood can be misperceived as true recollections” (p. 44). It is true that according to the traditional story Freud repudiated his seduction theory findings on the grounds that he had mistakenly taken fantasies (i.e., conscious ideas) of sexual abuse told him by patients to be authentic memories. However, there are important caveats about representing his experiences as providing insights relevant to the current false memory debate, the primary one being that the traditional account is erroneous: Freud did not postulate his seduction theory as a result of hearing frequent reports from patients of childhood sexual abuse experiences that he mistakenly thought were authentic. What that episode actually exemplifies is not the danger of mistaking patients’ false memories for authentic events, but of taking internal properties of a reconstruction, such as its supposed symbolic relationship with symptoms and its self-consistency within a narrative in part fashioned by Freud himself, to determine its validity; i.e., the “jigsaw puzzle” contention that “the contents of the infantile scenes turn out to be indispensable supplements to the associative and logical framework of the neurosis” (Freud, 1896c, p. 205). [12]

At the end of his concluding remarks on Freud’s theories Mollon writes that although it is over 100 years since Freud first began writing about psychoanalysis, the study of the mind’s capacity to deceive itself “still he seems to bother us” (p. 65). Presumably he is suggesting that those who take serious issue with Freudian theories find them disturbing, an echo of Freud’s (1933) notorious dictum that “opponents of psychoanalysis” are manifesting “the same resistance which I had to struggle against in individual patients” (pp. 137–138). In the context of the subject matter of Mollon’s book, however, the real issue is that Freud’s modern followers are still trying to convince us that his theories of memory and repression constitute a valuable contribution to current explorations of the workings of the mind. Unfortunately, many people who only read the bowdlerised versions of Freud’s theories often presented nowadays are being misled into believing that this is the case.


  1. For critical accounts of Freud’s early cases, see Webster, 1995, pp. 136–167, and Macmillan, 1997, pp. 69–114.
  2. See Esterson, 1993, p. 8, n.2, for another illustration of the unreliability of Freud’s reporting of this case.
  3. There are indications that some of the patients in question started treatment with Freud after October 1895, leaving an even shorter period for the acquisition of this data in their cases (Masson, 1985, p. 154).
  4. . . . the symptoms of hysteria . . . are determined by certain experiences of the patient’s which have operated in a traumatic fashion and which are being reproduced in his psychical life in the form of mnemic symbols” (Freud, 1896c, pp. 192–193).
  5. In the “Aetiology” (1896c) paper he wrote that his methodology was analogous to that of a forensic physician “who can arrive at the cause of an injury even if he has to do without any information from the injured person” (pp. 191–192).
  6. Cioffi, 1972, pp. 147–148; Schimek, 1987, pp. 955–960; Israëls & Schatzman, 1993, pp. 39–47; Esterson, 1993 (pp, 21–25), 1998 (pp. 13–15).
  7. It is not generally appreciated that in his first retrospective comments on the episode published some 8 years after his abandonment of the seduction theory Freud was still maintaining that it was “not open to doubt” that in “a disproportionately large number of cases” there had been early childhood sexual abuse (1906, p. 74), yet in his next account there was no indication that any of these purported findings were authentic (1914, p. 17).
  8. The widely held notion that Freud’s seduction theory claims outraged his colleagues, and that he was shunned by them on that account, is refuted in Esterson, 2002b.
  9. In the Standard Edition, the word Phantasie is translated as “phantasy,” and was generally used by Freud to denote an unconscious idea or image that he had analytically inferred (Esterson, 2002a). Its frequent translation as “fantasy” has exacerbated the tendency to misconstrue the ideas or images in question as conscious experiences of Freud’s patients that they had reported to him (Esterson, 1993 [pp. 166–168], 2001 [pp. 343–344]).
  10. It is interesting to note Egger’s remarks, published in 1895, on the reconstructive nature of recollections of dreams: “. . . if we then proceed to give an account [of a dream] we are liable to fill in from our imagination the incoherent and disjointed fragments furnished by memory. . . . We unwittingly become creative artists” (quoted in Freud, 1900, p. 46, n.1).
  11. Elsewhere Freud (1915b) wrote that “the essence of the process of repression lies, not in putting an end to, in annihilating, the idea which represents an instinct, but in preventing it from becoming conscious”(p. 166).
  12. The flaws in such an argument are exemplified by the fact that when Freud changed his theory he tacitly dropped items he had originally adduced as part of his infantile sexual abuse narrative that no longer fit his revised schema.


Borch-Jacobsen, M. (1996a). Neurotica: Freud and the seduction theory. October, 76, 15–43.

Borch-Jacobsen, M. (1996b). Remembering Anna O.: A century of mystification. New York & London: Routledge.

Cobbe, F. P. (1898). The fallacies of memory. In J. B. Taylor & S. Shuttleworth (Eds.), Embodied selves: An anthology of psychological texts 1830–1890. Oxford, England: Clarendon Press. (Original work published 1867)

Cioffi, F. (1972). Wollheim on Freud. Inquiry, 15, 171–186.

Cioffi, F. (1974). Was Freud a liar? The Listener, 91, 172–174.

Crews, F., et al. (1995). The memory wars: Freud’s legacy in dispute. New York: New York Review of Books.

Eissler, K. R. (2001). Freud and the seduction theory: A brief love affair. Madison, CT: International Universities Press.

Ellenberger, H. (1972). The story of ‘Anna O.’: a critical review with new data. Journal of the History of the Behavioral Sciences, 8, 267–279.

Esterson, A. (1993). Seductive mirage: An exploration of the work of Sigmund Freud. Chicago: Open Court.

Esterson, A. (1998). Jeffrey Masson and Freud’s seduction theory: a new fable based on old myths. History of the Human Sciences, 11(1), 1–21.

Esterson, A. (2001). The mythologizing of psychoanalytic history: Deception and self-deception in Freud’s accounts of the seduction theory episode. History of Psychiatry, 12(3), 329–352.

Esterson, A. (2002a). Fantasy (phantasy). In E. Erwin (Ed.), The Freud encyclopedia: Theory, therapy, and culture (pp. 188–192). New York & London: Routledge.

Esterson, A. (2002b). The myth of Freud’s ostracism by the medical community in 1896–1905: Jeffrey Masson’s assault on truth. History of Psychology, 5(2), 115–134.

Freud, S. (1953–1974). The standard edition of the complete psychological works of Sigmund Freud (J. Strachey, Ed. & Trans.). London: Hogarth Press.

Freud, S. (1895a). Obsessions and phobias: Their psychical mechanism and their aetiology. In Standard edition (Vol. 3, pp. 74–82).

Freud, S. (with J. Breuer) (1895b). Studies on hysteria. In Standard edition (Vol. 2, pp. 1–335).

Freud, S. (1896a). Heredity and the aetiology of the neuroses. In Standard edition (Vol. 3, pp. 142–156).

Freud, S. (1896b). Further remarks on the neuro-psychoses of defence. In Standard edition (Vol. 3, pp. 159–185).

Freud, S. (1896c). The aetiology of hysteria. In Standard edition (Vol. 3, pp. 189–221).

Freud, S. (1898). Sexuality in the aetiology of the neuroses. In Standard edition (Vol. 3, pp. 263–285).

Freud, S. (1899). Screen memories. In Standard edition (Vol. 3, pp. 301–322).

Freud, S. (1900). The interpretation of dreams. In Standard edition (Vols. 4 & 5, pp. 1–627).

Freud, S. (1901). The psychopathology of everyday life. In Standard edition (Vol. 6, pp. 1–310).

Freud, S. (1905a). Fragment of an analysis of a case of hysteria. In Standard edition (Vol. 7, pp. 3–122).

Freud, S. (1905b). Three essays on the theory of sexuality. In Standard edition (Vol. 7, pp. 130–243).

Freud, S. (1906). My views on the part played by sexuality in the aetiology of the neuroses In Standard edition (Vol. 7, pp. 271–279).

Freud, S. (1908). Hysterical phantasies and their relation to bisexuality In Standard edition (Vol. 9, pp. 157–166).

Freud, S. (1909). Notes upon a case of obsessional neurosis. In Standard edition (Vol. 10, pp. 153–249).

Freud, S. (1910). Five lectures on psychoanalysis. In Standard edition (Vol. 11, pp. 9–55).

Freud, S. (1913). On beginning the treatment: Further recommendations on the technique of psychoanalysis. In Standard edition (Vol. 12, pp. 122–144).

Freud, S. (1914). On the history of the psychoanalytic movement. In Standard edition (Vol. 14, pp. 7–66).

Freud, S. (1915a). Repression. In Standard edition (Vol. 14, pp. 143–158).

Freud, S. (1915b). The unconscious. In Standard edition (Vol. 14, pp. 161–215).

Freud, S. (1916–1917). Introductory lectures on psychoanalysis. In Standard edition (Vols. 15 &16, pp. 13–496).

Freud, S. (1918). From the history of an infantile neurosis. In Standard edition (Vol. 17, pp. 7–122).

Freud, S. (1920). Beyond the pleasure principle. In Standard edition (Vol. 20, pp. 3–64).

Freud, S. (1924). A short account of psychoanalysis. In Standard edition (Vol. 19, pp. 191–209).

Freud, S. (1925). An autobiographical study. In Standard edition (Vol. 20, pp. 7–74).

Freud, S. (1926a). Inhibitions, symptoms and anxiety. In Standard edition (Vol. 20, pp. 77–174).

Freud, S. (1926b). The question of lay analysis. In Standard edition (Vol. 20, pp. 179–217).

Freud, S. (1926c). Psycho-analysis. In Standard edition (Vol. 20, pp. 261–270).

Freud, S. (1932). The acquisition and control of fire. In Standard Edition (Vol. 22, pp. 185–193).

Freud, S. (1933). New introductory lectures on psychoanalysis. In Standard edition (Vol. 22, pp. 5–185).

Freud, S. (1937). Constructions in analysis. In Standard edition (Vol. 23, pp. 256–269).

Freud, S. (1939). Moses and monotheism. In Standard edition (Vol. 23, pp. 3–137).

Freud, S. (1940). An outline of psychoanalysis. In Standard edition (Vol. 23, pp. 141–207.

Hirschmüller, A. (1989). The life and work of Josef Breuer. New York: New York University Press.

Israëls, H. and Schatzman, M. (1993). The seduction theory. History of Psychiatry, 4, 23–59.

James, W. (1950). The Principles of Psychology (Vol. 1). New York: Dover. (Original work published 1890)

Jones, E. (1955). Sigmund Freud: Life and work (Vol. 2). London: Hogarth Press.

Kiell, N. (1988). Freud without hindsight: Reviews of his work. Madison, CT: International Universities Press.

Macmillan, M. (1997). Freud evaluated: The completed arc. (2nd ed.). Cambridge, MA: MIT Press.

Mahony, P. J. (1984). Cries of the Wolf Man. New York: International Universities Press.

Makari, G. M. (1998). Between seduction and libido: Sigmund Freud’s masturbation hypotheses and the realignment of his etiologic thinking, 1897–1905. Bulletin of the History of Medicine, 72, 638–662.

Masson, J. M. (Ed. & Trans.) (1985). The Complete Letters of Sigmund Freud to Wilhelm Fliess, 1887–1904. Cambridge, MA: Harvard University Press.

McCullough, M. (2001). Freud’s seduction theory and its rehabilitation: a saga of one mistake after another. Review of General Psychology, 5(1), 3–22.

McDougall, W. (1936). Psycho-analysis and social psychology. London: Methuen.

Mollon, P. (2000). Freud and false memory syndrome. New York: Totem Books.

Pope, H. G. (1997). Psychology astray: Fallacies in studies of “repressed memory” and childhood trauma. Boca Raton, FL: Upton Books.

Pope, H. G., Jr., Oliva, P. S., & Hudson, J. I. (2002). Repressed memories: Scientific status. In D. L. Faigman,, D. H. Kaye, M. J. Saks, & J. Sanders (Eds.), Science in the law: Social and behavioral science issues (pp. 487–526). St. Paul, MN: West Group.

Scharnberg, M. (1993), The non-authentic nature of Freud’s observations, Vol. 1, The seduction theory. Stockholm: Almqvist & Wiksell International.

Schatzman M. (1992, March 21). Freud: Who seduced whom? New Scientist, 34–37.

Schimek, J. (1987). Fact and fantasy in the seduction theory: A historical review. Journal of the American Pyschoanalytic Association, 35, 937–965.

Schacter, D. L. (1996). Searching for memory. New York: Basic Books.

Smith, D. L. (1991). Hidden conversations: An introduction to communicative psychoanalysis. London & New York: Routledge.

Slater, E. (1965, May 29). Diagnosis of “hysteria.” British Medical Journal, 1395–1399.

Smith, D. L. (2000). The mirror image of the present: Freud’s theory of retrogressive screen memories. Psychoanalytische Perspectieven, 39, 7–28.

Swales, P. (1988). Freud, Katharina, and the first “wild analysis.” In P. Stepansky (Ed.), Freud: Appraisals and reappraisals. Hillsdale, NJ: Analytic Press.

Taylor, J. B., & Shuttleworth, S. (Eds.) (1998). Embodied selves: An anthology of psychological texts 1830–1890. Oxford, England: Clarendon Press.

Webster, R. (1995). Why Freud was wrong: Sin, science, and psychoanalysis. London: HarperCollins.

Allen Esterson
London, England

You can read this review in
The Scientific Review of Mental Health Practice, vol. 2, no. 2 (Fall/Winter 2003).
Subscribe now!

  ©2004 Center for Inquiry    | SRMHP Home | About SRMHP | Contact Us |