The Success and Failure of Primal Therapy:
A Critical Review By Stephen Khamsi, Ph.D.

Primal Therapy was introduced in 1970 as "the cure for neurosis" (Janov, 1970). Since that time it has been praised (e.g., Ebel, 1975; Rosen, 1977; Schaef, Kirkman & Ungashick, 1981; and Verny, 1978) and panned (e.g., Kaufmann, 1974; Lonsbury, 1978; Strupp, 1973; and Torrey, 1976). The "primal revolution" has generated much controversy and dispute, but little in the way of empirical research, even though empirical findings might settle some of the wrangling and debate. The process and outcomes of Primal Therapy have yet to be adequately researched, so responsible work in this area warrants our careful consideration. Tomas Videgard's (1984a) recent The Success and Failure of Primal Therapy (Stockholm, Sweden; Almqvist & Wiskell International, 313 pp.) is an important book. It is a sincere and ambitious attempt to evaluate object relations theory and primal theory in relation to psychoanalysis, and includes the first independent research on Primal Therapy outcomes. As such, Videgard's undertaking deserves the attention of the therapeutic and scientific communities. That this important dissertation-like work has been published only in Sweden is further justification for an in-depth critical review.

There are really two main aspects of this book: one theoretical and one empirical. Videgard's (1984a) primary purpose was to disassemble the orthodox psychoanalytic paradigm and to replace it with "trauma-relations oriented psychoanalytic thinking" (p. 4). This latter perspective has coalesced from elements of primal theory, object relations theory, Bowlby's attachment theory, and Kohut's self-psychology.

Videgard's interest in replacing the orthodox psychoanalytic paradigm prompted him to conduct an empirical inquiry of the result of Primal Therapy. His basic objective here was to determine whether and how Primal Therapy outcomes might contribute to the trauma-relations perspective. Secondarily, Videgard wanted to know if Primal Therapy was suitable for specific types of patients and, if so, whether they could be selected in advance. He has presented 32 systematic case studies, 21 in his book (cf. Videgard, 1984a, Part VI, pp. 103-240) and 11 more (containing "sensitive information") in a supplement for professionals (cf. Videgard, 1984b, pp. 1-61).

Videgard's Critique of Psychoanalysis

Videgard is intrigued by psychoanalysis, but more by the innovations of object relations theory and Primal Therapy than the orthodox "drive-oriented" Freudian approach. His book, in fact, represents an attempt to overthrow the psychoanalytic paradigm by drawing firepower from object relations theory and Primal Therapy.

According to Videgard (1984a), Freud made critical mistakes with respect to developmental theory (pp. 20-21), basic notions concerning human nature (p. 15), and therapeutic technique (p. 16). Freud is specifically indicted for valuing human relationships only for their tension-reduction properties, for overlooking the effects of actual traumata while concentrating on instinctual drives, and for limiting therapy primarily to an analysis of drive impulses and infantile wishes. These shortcomings, however, are considered to be surface manifestations of a flawed theoretical foundation. Videgard is calling for a paradigm shift--a profound change in the thoughts, perceptions, and values that form the psychoanalytic vision of reality: "Psychoanalysis needs a new paradigm which will once and for all leave the nineteenth century natural scientific, reductionistic perspective in which psychoanalysis was born" (p. 11).

The Trauma-Relations Perspective

Videgard first became aware of Primal Therapy while training to become an analyst at the Swedish Psychoanalytic Institute in Stockholm. He was deeply moved by the films of Primal Therapy, and later by Janov's (1970) The Primal Scream. Already influenced by Guntrip, Fairbairn, and Winnicott, Videgard quickly recognized close parallels between primal theory and object relations theory. Videgard decided to further his study in this area, and secured financial support from three Swedish foundations to research Primal Therapy and to undergo the therapy himself.

Primal theory has been characterized by Videgard (1984a) as a "pure trauma theory." Primal theory is considered similar to early Freudian theory, except that childhood trauma can result from "microtraumas" as well as gross dramatic events (p. 4).

Videgard (1984a) also considers object relations theory as "fundamentally a trauma theory" (p. 4). In making this connection between the centrality of trauma in primal and object relations theories, Videgard identified what be believes is an essential difference between object relations and primal theories, on the one hand, and Freudian psychoanalytic theory on the other. Primal theory and object relations theory are in agreement that infants become neurotic because of "defective relationships," not because of asocial drives (p. 4). This is the basis for Videgard's dichotomy between the drive-oriented and the trauma-relations oriented perspectives (p. 9).

In discussing the various disciplines that comprise the trauma-relations perspective, Videgard (1984a) notes that "they all seem to participate in a common cause against traditional psychoanalytic drive-explanations and for the uncovering of the more or less subtle traumatization processes that, according to them are the real agents in creating neuroses" (p. 4). Neuroses, then, results more from environmental influences, especially human relations and psychological traumas, than from the frustration of instinctual drives.

Videgard's Critique of Primal Therapy

Theory and Therapy. Videgard (1984a) credits Janov with sensitizing the psychotherapeutic community to the subtleties of childhood trauma and with developing effective expressive techniques, but hastens to add that the consistent application of Janov's "discharge model" leads to therapy failures. Videgard argues that primal theory is insufficient and needs to be replaced (p. 296).

Primal theory, according to Videgard (1984a), is misguided with respect to human relationships. Primal theory suffers from an inadequate and underdeveloped model of personality development (p. 293), and Primal Therapy lacks an appropriate therapist-patient relationship Videgard, 1984a (p. 287).

The therapeutic relationship, in Janov's opinion, is of minor importance (p. 287). The role of primal therapist consists primarily of being a catalyst for the emergence of stored historic pain. From the object relations perspective, however, the reliving of traumas is therapeutic only in the relational context of being heard and accepted (p. 294).

Research. Videgard (1984a) notes a serious discrepancy between Janov's claims of success and his own actual research findings. Videgard believes that "the published data are likely to represent an overestimation of the positive effect of Primal Therapy" (p. 254). Except for the Goodman-Sobel biochemical study (reported in Janov, 1980), in fact, Videgard has serious objections to all the empirical research from the Primal Institute.

The Primal Therapy Outcome Study

Primal Therapy was ostensibly chosen for empirical study because Videgard (1984a) considered it the "simplest and purest" example of a trauma theory (p. 5). Primal Therapy is considered inextricably bound to other representatives of the trauma-relations perspective.

Primal Therapy may appear to be a small island miles apart from the psychoanalytic-psychodynamic mainland. I want to show you that this impression is mistaken. Primal Therapy is rather the most radical representative of a broad trend of psychological thinking that has gained considerable momentum since the late sixties, with names such as Laing, Guntrip, Winnicott, Bowlby, Kuhot, Alice Miller etc. In fact, it seems to me that psychoanalysis has become divided into two opposing directions which can be termed trauma-relations oriented psychoanalysis and drive-oriented psychoanalysis. ORT [Object Relations Theory], Kuhut, Bowlby, and Miller belong to the former, main-stream psychoanalysis and the Kleinians to the latter. (pp. 8-9)

Videgard (1984a) argues that case studies are needed to understand what happens to subjects in psychotherapy outcome research (p. 7). He insists that the effects of psychotherapy should be judged not on simple symptomatic outcome criteria alone but on dynamic criteria, taking the patients' individual development into consideration (p. 76). Based on Malan's (1976) Tavistock model, Videgard's 32 systematic case studies incorporate quantitative test data, interviews, multiple outcome judgments, and case commentaries.

Subjects. Two methods of sampling were used. The first was random selection, which was used to secure 25 subjects from the population of incoming Primal Institute patients. The second method was purposive sampling, in which all the Scandinavian patients were selected in order to secure subjects as compatible as possible to previous studies by Hessle (1975) and Sigrell (1977); this method provided 9 additional subjects.

A total of 34 subjects were reportedly tested before therapy; one was lost due to technical failure and another due to an "unmanageable" research relationship (Videgard, 1984a, pp. 91-92). Videgard writes that only 25 completed the whole test program, including 18 men and 7 women (p. 92). Four of those failing to complete the entire test program either moved away or could not be reached, one had an incomplete posttest, and one committed suicide. [Note: the author's information regarding subjects may be incorrect. According to my calculations a total of 26 subjects--including 1 additional male subject--completed the entire testing program.]

Research Instrumentation. Within the context of qualitative case studies, Videgard used two basic methods of data collection: interviews and projective tests. The interviews were nonstandardized and semi-structured; some specific questions were formulated mid-way through the post-testing, but were asked only of those interviewed in the latter stages of post-testing.

Three projective measures were used: (1) the Defense Mechanism Test (DMT), (2) the Self Test, and (3) the Relation Test (the last two are variants of the first). The DMT was originally designed by Kragh (1955) to assist the Swedish Air Force in selecting pilots. Since the DMT is not listed in the encyclopedic Mental Measurements Yearbook (Buros, 1978), and since it comprised the core of the projective testing program, it warrants further discussion.

The DMT arose from "extensive perception-personality research" that has been conducted for over 30 years at Lund University in Sweden. According to this research "it is possible to detect specific personality disturbances by examining how people perceive pictures at very short exposure times" (Videgard, 1984a, p. 78).

The DMT is a projective personality test in which TAT-like pictures are flashed in a tachistoscope at very short exposure times, from 10 to 2000 milliseconds; after each exposure the subject is instructed to describe what was seen, both verbally and in the form of a simple drawing (p. 79). Each picture is shown repeatedly for up to 20 times, unless the subject reaches a correct perception of the picture motif earlier. The time of administration is 40 to 90 minutes.

The DMT pictures were designed to be provocative with respect to anxiety and defense mechanisms. Distortions in perception and reporting are assumed to result from psychological defenses (e.g., isolation, repression, denial); approximately 50 variants have been coded. Through using this method, Videgard hoped to get a fairly direct picture of the unconscious working of the psyches of his subjects--and a valid picture of the changes that can result from Primal Therapy.

One noteworthy aspect of the DMT is the "parallel hypothesis." It is believed that "with the tachistoscopic technique it is possible to trace when in an individual's life specific traumas have occurred" (Videgard, 1984a, p. 8). Although this fascinating proposition shows promise for future research, its use in this study was primarily heuristic.

Procedure. Videgard tested subjects before and after Primal Therapy. Pre-testing began in December 1975 and consisted of two 90-minute appointments; the Self and Relational Tests were administered in the first appointment, and the DMT and a nonstandard pretest interview were administered in the second. Post-testing included the same measures, except that the posttest interview focused primarily on the effects of the therapy. Most post-testing took place 18 to 20 months later, although it took place considerably later for some of the Scandinavian subjects. In one case post-testing took place seven years after the beginning of treatment (cf. "Cindy," in Videgard, 1984b, p. 39).

Treatment of the data. In order to understand how the data were treated, one must learn of two key events that lead to a turning point in this study.

First, Videgard became aware in mid-project that his primary research objective--that of comparing his (quantitative projective) test results with those of Hessle (1975) and Sigrell (1977)--was in jeopardy and might have to be abandoned.

"The anticipated comparison with Hessle's data from the group-psychotherapy project--originally the highlight of the work, proved unexpectedly difficult because the coders in the two projects had used different coding rules. . . . The differences were not great but enough to make a direct comparison between the two sets of data meaningless." (Videgard, 1984a, p. 95)

Hessle's codes had followed Kragh's original manual, whereas Videgard's scores used an updated military version of the manual. Videgard's preferred alternative at this point had been to recode Hessle's tests, but this proved impossible because Hessle had burnt his protocols for ethical reasons. Rather than recode his own protocols in accordance with an outmoded coding manual, Videgard abandoned the comparison (p. 95).

Second, Videgard learned of Malan's (1976) Tavistock model of psychotherapy outcome research, which is based on a qualitative case study approach. Videgard revised his research plan at this point and decided to proceed with qualitative case studies, noting that "my interest in the case study approach increased pari passu with the decreasing usefulness of the statistical data" (p. 95). He decided to use all available data--interviews, quantitative and qualitative projective findings--to reach a deep understanding of each individual's development in Primal Therapy (p. 96).

After this turning point, the pretest projective findings were carefully read and then subjected both to quantitative and qualitative content analysis (cf. Videgard, 1984a, p. 82). Notes from the pretest interviews were then examined and "psychodynamic hypotheses" (concerned with elucidating a subjects basic neurotic conflicts, symptoms, and their possible origins; cf. p. 98), "criteria of ideal therapeutic result," and a "background summary" were formulated for each subject. A psychiatric diagnosis and list of disturbances were also formulated for each subject.

The posttest measures were later treated as the pretest data had been, and then the following were formulated for each subject: summary of the course of therapy, insights, changes, assessment of therapeutic involvement, evaluation of pretest and posttest scores, evaluation of outcome, "prognosis," and a summary (including specific external stress, specific predisposition, symptomatic criteria, dynamic criteria, therapeutic involvement, and outcome prognosis) (Videgard, 1984a, p. 98). Also included were statements regarding the degree to which the criteria of ideal therapeutic result had been achieved at the time of post-testing, and the total follow-up time each subject had spent in the therapy (15 to 32 months).

The Results

Is Primal Therapy effective? Three different estimations of therapeutic outcome were formulated for each subject: (1) "symptomatic improvement," (2) "dynamic improvement," and (3) "prognosis for future development."

Regarding "symptomatic improvement," 6 were judged much improved, 4 improved, 11 slightly improved, 4 unchanged, 1 worse. Regarding "dynamic improvement" 8 were judged much improved, 7 improved, 7 slightly improved, and 4 unimproved. [Each of these listings included a total of 26 subjects since, as noted above, one had committed suicide and five could not be post-tested.]

Of the three estimations of outcome, however, Videgard (1984a) considers the prognosis for future development "the most relevant outcome evaluation since it is an estimation of the long-term effects of Primal Therapy" (p. 244). Prognoses for future development were judged as very good and good ("satisfactory" outcomes), medium and bad ("unsatisfactory") (pp. 100-101). The prognoses for future development included 4 that were deemed very good, 9 good, 8 medium, 6 bad (including the suicide), and 5 that left therapy prematurely and hence were unable to be post-tested (Videgard, 1984a, p. 295). In other words, 13 subjects were considered to have had satisfactory outcomes, while between 14 (p. 295) and 19 (p. 303) had unsatisfactory outcomes.

How does Videgard (1984a) interpret these findings? "The main result is that about 40% of the primal patients achieve a satisfactory result within 15 to 25 months" (p. 249). This statistic may be figured in many different ways, and various calculations of satisfactory results range form 39% for the random sample to 50% of the total sample (exclusive of subjects who could not be post-tested).

Which subjects benefitted most from Primal Therapy? Videgard correlated prognostic outcomes with several subject variables including (1) age, (2) sex, (3) marital status, (4) nationality, and (5) initial prognosis (i.e., the diagnosis). Several correlations were clearly evident between outcome categories and subject background variables. Although the small sample size allowed for only one statistically significant finding, some of the others are clinically significant and should be considered in future research.

1. An obvious correlation, suggesting a curvilinear relationship, exists with respect to age and outcome: Subjects between 30 and 40 (n=11) were more successful that subjects under 30 (n=15) [64% and 40% respectively], and no subject over 40 (n=5) was successful (Videgard, 1984a, p. 247). Perhaps there is an optimal age range for success in Primal Therapy (i.e., 30-40); in any event, the outcomes of the therapy were different for different age groups.

2. Are the results of Primal Therapy different for men and women? This study suggests not. Men (n=21) and women (n=10) showed very similar rates of success [43% and 40% respectively] (cf. Videgard, 1984a, p. 246).

3. The relationship between marital status and outcome revealed a highly significant finding (p < .01): married subjects (n=9) were more successful than divorced (n=5) or unmarried (n=17) subjects [67%, 60%, and 24% satisfactory prognostic outcomes, respectively] (Videgard, 1984a, p. 247). It should be added, however, that unmarried and divorced subjects actually came to Primal Therapy with more severe disturbances than married subjects: none of the married subjects (n=8) entered therapy with a "severe disturbance, although 71% of unmarried (n=17) and 67% (n=6) subjects did so. [Note: there is a discrepancy between the number of married (9, 8) and divorced (5, 6) subjects in the original Tables 4 and 5, respectively (cf. pp. 246-248)].

4. The correlation between nationality and outcome revealed that (1) Europeans (n=15) and (2) Scandinavians (n=12) were more successful than (3) American and Canadians combined (n=16) [53%, 50% and 31% respectively] (Videgard, 1984a, p. 247). As Videgard correctly notes, this may be a chance finding. Another plausible hypothesis is that (a) Europeans and Scandinavians were somehow different than North Americans--perhaps more highly motivated, since they had to overcome greater obstacles to obtain treatment, or (b) Primal Therapy is somehow more effective on the character structures of Europeans and Scandinavians than Americans and Canadians.

5. Subjects entering therapy with a "severe disturbance" (n=16) are less successful than those entering with a "moderately severe disturbance" (n=15) [25% and 60% satisfactory outcomes, respectively] (Videgard, 1984a, p. 248). "The difference in outcome between the severe and the moderately severe groups is significant (p < .05). . . . Patients with deep disturbances are less likely to benefit from Primal Therapy than those with milder disturbances" (p. 249).

6. An important but unreported correlation is that subjects with no previous therapy (n=11) were more successful than those with previous therapy (n=17) [55% and 41%, respectively] (cf. Videgard, 1984a, p. 246).

The following were reported by Videgard, but seem to have been derived in a less systematic fashion.

7. New memories (i.e., a reappearance of previously forgotten scenes) were reported by only a few subjects. This finding, however, should be qualified in at least the following ways. First, well-known childhood scenes reportedly took on new meanings. Second, many subjects reported strong feelings and bodily sensations which they associated with very early events but were unaccompanied by images or memories "in the common meaning of that word" (Videgard, 1984a, pp. 274-275). In light of findings from Primal Therapy and other experiential psychotherapies, perhaps we need to expand the common meaning of the word "memory" (cf. Khamsi, 1985, regarding "memory" and the primal process). Third, it is probable that repression intervened between the time of memories experienced in therapy and the time of the posttest interviews. As a therapist, I have often seen the pernicious effects of repression; it is not at all unusual for a person to have deep experiential memories one moment, and literally be unable to recall them the next.

8. Videgard (1984a) reports that only three subjects had "connected" birth feelings (p. 275), and that "integrated experiences," an undefined term, were rare before the ages of three or four. The reader may receive a false impression, however, since many more of these subjects reported having had numerous nonverbal, perhaps ineffable experiences--many of which were explicitly reported to be birth-related (cf. Khamsi, 1985, for a qualitative account of birth feelings in Primal Therapy). Videgard, then, appears to be extremely conservative in accepting nonverbal and birth feelings as "connected" or "integrated," and thus as legitimate objects of scientific inquiry.

Videgard (1984a) believes that most, but not all, of his subjects may have done as well or better in more conventional therapies. "Except for the . . . three to four subjects who had been both 'desperate and integrated' and who had relived their birth-traumas, it seems the rest of the successful patients, at least in principle, could have achieved their results in good insight psychotherapy" (p. 280). This is speculation, and I disagree. Videgard's own book, in fact, has innumerable statements of subjects that contradict the notion that other approaches would have worked as well for them. But it does indicate Videgard's unstated respect for the power of birth feelings--for certain subjects.

9. None of the subjects considered themselves cured (Videgard, 1984a, p. 275; cf. the "Failure Rate" section in the "Discussion" below).

10. Thirteen subjects reported improved relations with the opposite sex, and two reported improved relations with the same sex (Videgard, 1984a, p. 276). One-third reported a better sex life, one-third were unchanged, and three or four were "aggravated."

In addition to the above findings, Videgard reported the following impressions from his research.

11. "A great majority of the patients found the Primal Therapy slower and much more difficult than expected" (Videgard, 1984a, p. 273).

12. "Most of the successful patients had positive feelings for at least one or two therapists [while] none of the five least successful patients had developed strong positive feelings for any therapist" (Videgard, 1984a, p. 273). Related to this is the impression that "about 50% of the patients wanted more individual contact with the therapists" (p. 277). These findings provide strong support for Videgard's argument to rethink the therapeutic relationship in Primal Therapy.

13. Those patients whose childhoods had been characterized by a general lack of emotional contact with both parents seem to have very small chances of benefitting from Primal Therapy (Videgard, 1984a, p. 282). This finding underscores Videgard's concern about the importance of human relationships, both in development and in therapy.

14. No patient claimed to have experienced the complete sequence of events by which Holden (1976) describes a "primal" (Videgard, 1984a, p. 274). So-called "primal screams" were reportedly rare, and even screaming was an exception. According to Videgard, "most patients preferred to talk about feelings instead of primals." (p. 274)

There appears to be a chasm between Primal Therapy experiences as they are (a) lived and (b) described in the literature. Moreover, "primals" and their neurophysiological correlates are apparently most intriguing to patients before their therapy, not after.

15. Most subjects had one or two key scenes, but crying was more often about generalized feelings than about specific scenes (Videgard, 1984a, p. 275).

Generalized feelings have been denigrated by Janov because his theory is based on the importance of reliving specific "primal scenes." According to these reports from Janov's own clientele, however, most feelings in Primal Therapy are in fact generalized and not specific. This finding can be interpreted to support my own thesis (Khamsi, 1984) that the "reliving" of traumas ("specific scenes") takes place within a broader, more encompassing context--the primal sense dimension. According to this thesis, an awareness of the inner body sense--previously termed the orgonotic, existential, and felt sense--is more basic to Primal Therapy than even the reliving of traumas. The primal sense is a literal sixth Aristotelean sense, related to but different than the vestibular, kinesthetic and cutaneous "body senses."

16. Improved work capacity was the only area of perceived progress for several subjects (Videgard, 1984a, p. 276). Obviously, these were subjects with unsatisfactory outcomes.

17. Videgard (1984a) discovered that "patients who did not perceive the threat in the last exposure of the DMT before therapy fail significantly more often than those who perceive the threat" (p. 282). This is an important finding, and is deserving of future research attention.


Videgard is clearly in favor of replacing the psychoanalytic paradigm. He has leveled a massive attack on orthodox "drive-oriented" psychoanalytic theory and, drawing firepower from ORT and Primal Therapy, has proposed as alternative--the "trauma-relations" orientation. In resurrecting the psychoanalytic paradigm, Videgard argues in favor of incorporating the theory and relational therapeutic context of ORT with many of the therapeutic techniques of Primal Therapy. In order to strengthen and further elucidate the trauma-relations position, Videgard has conducted an empirical outcome study of Primal Therapy.

Videgard's Primal Therapy Outcome Study

Videgard (1984a) attempted to assess the therapeutic outcomes of subjects approximately two years after beginning Primal Therapy, and to compare these findings with other psychotherapy outcome studies.

Critique. The completion of a critical, independent outcome study of Primal Therapy from within the Primal Institute must be considered a monumental success. This book provides the only systematic, documented case histories of patients at the Primal Institute. Both the case studies and the comparisons to other studies are responsible and effective.

The primary problems with Videgard's study resulted from poor planning (cf. "Treatment of the Data" above). Videgard should have foreseen the problem concerning the coding of his data. This oversight weakened the study in two ways. First, the Scandinavian subjects had been selected in order to compare the findings of Hessle (1975) and Sigrell (1977). In retrospect, since these comparisons were never completed, the study would have been substantially stronger if it had employed a random sample of 34 subjects, rather than 25.

The second weakening effect of Videgard's oversight is that the original "highlight" of the study seems to have been almost readily abandoned. He should have recoded his data, or at least offered a more convincing argument for abandoning the comparisons. It is not too late for Videgard to recode his data, perform the comparisons with the date of Hessle (1975) and Sigrell (1977), and publish the findings. Such a report would be of interest to the psychotherapeutic community.

Videgard (1984a) might also conduct one or more follow-up studies of these same primal subjects sometime in the future. This meritorious undertaking would begin to chart the long-term effects of Primal Therapy. "Cindy" (pp. 39-43), interviewed seven years after beginning the therapy, shows that Primal Therapy outcomes may appear very different when evaluated over a longer period. Videgard could fairly easily expand his original project into a longitudinal study, retesting and/or reinterviewing the subjects periodically, perhaps once every five or ten years. Given the huge amount of work already invested, and the importance of the findings thus far, it may well be advisable to expand the research in this way.

Comparison of primal subjects with related studies. Videgard compared his findings with those of the Tavistock clinic and the Menninger Foundation. In relation to Malan's (1976) findings for comparable subjects, he concluded that the success rate of Primal Therapy is at least as favorable as that of the Tavistock clinic [60% and 48% respectively] (Videgard, 1984a, p. 263). Also, noting that the Menninger subjects were more disturbed but received longer treatment, Videgard concluded that the results of the psychoanalysis and pychoanalytically-oriented therapies at the Menninger clinic seemed less successful than his primal sample (p. 266). The Primal Institute, then, was considered to have slightly better outcomes than either the Tavistock or Menninger clinic.

The failure rate of Primal Therapy. Carlini and Bernfeld ("Questionnaire," 1979) conducted a pilot study of 200 Primal Institute patients. They estimated an overall "failure rate" of Primal Therapy from the following: (1) 21% of their sample claimed to be unable to primal as described in the literature, (2) 24% claimed they had not reexperienced a repressed feeling or event, and (3) 19% stated that they were unable to feel previously-repressed feeling (p. 5). From their large sample, Carlini and Bernfeld estimated the failure rate of Primal Therapy to be approximately 21% (p. 5).

In Videgard's (1984a) study, however, "almost one-third (9 out of 31 patients) had either left the therapy prematurely (including one suicide) and/or expressed strong dissatisfaction with their own development and follow-up" (p. 267). In all, 40% of Videgard's subjects were judged to have reached a satisfactory result (p.267); this contrasts sharply with Carlini and Bernfeld's explicit success rate of 79% and with Janov's implicit success rate of 90-98%.

How can such major discrepancies exist between these postulated rates of success in Primal Therapy? Clearly, part of the answer is that Videgard employed more exacting methods to determine therapeutic success and failure than either Carlini and Bernfeld or Janov; Janov's estimates have been impressionistic, while Carlini and Bernfeld's was based on data from self-reports. In any event, the primal community must continue to examine its therapy outcomes. There is a need to examine what "success" and "failure" mean in Primal Therapy, as well as how and why they do or do not occur. This is a deep and complex issue that deserves much future attention.

One important aspect of any therapy success or failure concerns the issue of therapeutic technique. Videgard (1984a) believes that "at least some of the failures in Primal Therapy can be attributed to specific shortcomings in the way the therapy is done" (p. 284). This charge deserves comment.

The technique of Primal Therapy. Videgard (1984a) approves of the sensitive use of primal techniques (p. 288). He believes that the physical setting and focusing technique of Primal Therapy allow maximum freedom of emotional expression. "The primal technique," in fact, "may be a help to follow the patient's natural pace" (p. 288).

Videgard (1984a) objects, however, to the lack of an on-going therapeutic relationship. Most Primal Therapy failures can be attributed to the lack of individual therapy sessions for most patients. The discharge-model is simply insufficient, says Videgard, so primal theory needs to be replaced and "continuous, individual therapeutic contact" should contextualize the primal therapeutic process (p. 296).

Videgard is correct. In 1969, after leaving the Primal Institute, therapists at the Marin Center for Intensive Therapy began offering Primal Therapy that included an explicit and defined client-centered relationship. Since commencing therapy there in 1973 I have observed Primal Therapy practiced in many ways. Primal clients in general have much better outcomes when their therapy includes an explicit therapeutic relationship; they often get "stuck" without it. This is an important issue for future research. One might also study whether there are differential outcomes from nondirective (i.e. Rogerian) and directive (i.e. early Janovian) primal therapeutic relationships (cf. Ristad, 1977; Wadler, Morris, Khamsi & Evan, 1979)

What is Primal Therapy?

Certain facts about Primal Therapy have been established by Videgard. Following from these facts, he has provided an interpretation of the findings and has offered a well-reasoned argument in favor of an alternative paradigm--the trauma relations perspective.

Videgard's facts have been established with proper scientific rigor and reporting, and thus represent an important new source of information for the empirical data base of Primal Therapy. These data may, however, be interpreted in other ways. While I generally accept his data as facts, my own interpretations sometimes differ.

I find Videgard's findings plausible and his arguments fairly persuasive. I agree, for instance, that ORT is superior to primal theory with respect to developmental theory and client-therapist relationships. In a prior article, however, I interpreted these same facts in a slightly different way--my argument was in favor of a "humanistic" or "client-centered" Primal Therapy (Khamsi, 1981). With respect to this fundamental deficiency in primal therapeutic relationships, Videgard and I have offered similar, constructive alternatives. Primal theory needs to be reworked in light of its own failings, taking into account the strengths of ORT and person-centered theory.

In attempting to get at the core of Primal Therapy, Videgard (1984a) has attempted to determine if it is a unique method, i.e., if it is the only approach that is able to help certain people (p. 279). He believes that it may have been for some of his subjects. Unfortunately, Videgard here has pursued the "primal-is-the-only-cure-for-mental-illness" myth, which is dated both as an honest misconception and as a sales campaign. Primal is one viable approach, preferred by many; but viewing Primal Therapy as discontinuous with and/or better than other approaches keeps us from understanding and researching exactly what it is and how it relates to any larger scheme of things.

The critical and defining factors of the primal process do not lie with the properties of Primal Therapy. Through imaginative variation we can see that the essence of Primal Therapy lies neither with primal screaming, nor three-week Intensives, nor particular techniques. Primal Therapy is a way of feeling and being real. This insight has been central to my own thesis that the primal sense dimension is most essential to the primal therapeutic process.

Primal is, in essence, a way of being real or authentic. It emerges from an individual's "decision" to open to what is, to feel-change-grow in spite of pain or difficulty. Nothing done to a person--such as "therapy"--facilitates real change. Being real come from within. Therapist and client/patient/person can work together, sharing experiences, ideas and feelings, so that both may live with greater feeling, meaning and authenticity. Being real can never be forced.

Videgard's primary purpose has been to argue in favor of a trauma-relations perspective. This argument, however, has only been partially successful. The importance of "trauma" and "relations" were derived from the overlap of theories--object relations and primal--and were formulated before this empirical study was conducted. To continue to argue in favor of a trauma-relations theory, the empirical data base should have supported Videgard's contentions about both relations and traumas. Videgard's data clearly support the importance of relations, with respect both to personality development and to Primal Therapy. They generally refute the centrality of trauma, however, to Primal Therapy. A solid trauma theory needs to be supported by empirical data that show a prevalence of new memories and specific feelings--which clearly was not the case in this study.

With respect to trauma, we must distinguish between theories of etiology and theories of therapy. A trauma theory of etiology needs to show that personality development is significantly affected by traumatic incidents, and this idea has been widely accepted. A trauma theory of therapy, on the other hand, would need to demonstrate a prevalence of subjects reliving specific traumatic scenes--which clearly was not the case in this study. There were fewer reports of new memories and specific feelings--hallmarks of reliving traumatic scenes--than of general feelings about such scenes. In general, then, the data support a relational Primal Therapy and a trauma theory of etiology but refute a trauma theory of therapy.

Theory can open or close our eyes. Just as Janov opened eyes when his insights were fresh, we can continue to open eyes and hearts and minds by researching and reporting human experience as it is lived--not theorized. Janov helped us see beyond the bounds of psychological theories extant. Now we must see beyond his.


Buros, O.K. (Ed.). (1978). The eighth mental measurements yearbook (Vols. 1-2). Highland Park, NJ: Gryphon Press.
Ebel, H. (1975). Primal therapy and psychohistory. History of Childhood Quarterly: The Journal of Psychohistory, 2, 563-570.
Hessle, S. (1975). Sjalvidentitet och varseblivning. (BIG-rapport nr 20) Department of Education, University of Stockholm.
Holden, E. M. (1976). Primals versus non-primals. Journal of Primal Therapy, 3, 50-62.
Janov, A. (1970). The primal scream. New York: G.P. Putnam's Sons.
Janov, A. (1980). Prisoners of pain. Garden City, NY: Anchor Press/Doubleday.
Kaufmann, W. (1974). An anatomy of the primal revolution. Journal of Humanistic Psychology, 14, 849-862.
Khamsi, S. (1981). The primal paradigm. Humanistic Psychology Institute Review, 3, 51-60.
Khamsi, S. (1984). Primal perspectives in psychotherapy. Aesthema, 4, 7-24.
Khamsi. S. (1985). An exploration of birth feelings in primal therapy: A human science approach. Dissertation Abstracts International, 46. (University Microfilms No. 8523664).
Kragh, U. (1955). The actual-genetic model of perception-personality. Lund, Sweden: Gleerup.
Lonsbury, J. (1978). Inside primal therapy. Journal of Humanistic Psychology, 18, 19-28.
Malan, D. M. (1976). Toward the validation of dynamic psychotherapy. New York: Plenum.
Questionnaire survey results, part II. (1979). Primal Institute Newsletter, 1(9), 1-5.
Ristad, H. (1977). Natural vs. directive primal therapy. Primal Community, 2(1), 36-38.
Rosen, D. (1977). A primal primer for psychiatrists. American Journal of Psychiatry, 134, 445-446.
Schaef, R.F.A., Kirkman, D.O., & Ungashick, B. (1981). Primal therapy. In R.J. Corsini (Ed.), Handbook of innovative psychotherapies (pp. 650-664). New York: John Wiley & Sons.
Sigrell, B. (1977). Effektkriterier i psykoterapiforskning. (BIG-rapport nr 32) Department of Education, University of Stockholm, Sweden.
Strupp, H.H. (1973). A new messiah [review of The anatomy of mental illness by A. Janov]. Contemporary Psychology, 18, 117-119.
Torrey, E.F. (1976, December). The primal therapy trip: Medicine or religion? Psychology Today, 62-64; 67-68.
Verny, T. (1978). An introduction to holistic primal therapy. Canadian Psychiatric Association Journal, 23, 381-387.
Videgard, T. (1984a). The success and failure of primal therapy: 32 patients treated at The Primal Institute (Janov) viewed in the perspective of object relations theory. Stockholm, Sweden: Almqvist & Wiskell International.
Videgard, T. (1984b). The success and failure of primal therapy: Supplement (for professionals). Available from T. Videgard, Heleneborgsgatan 44, S-117 32 Stockholm, Sweden.
Wadler, R., Morris, G., Khamsi, S., & Evan, L. (1979, June-August). The myth of natural vs. directive primal therapy. Newsletter of the International Primal Association, 9.

[This article was originally published in Aesthema, Journal of the International Primal Association No. 8, "A New Look At Primal Therapy" 1988, pp. 11-23.]

Stephen Khamsi, a primal psychotherapist since 1977, is currently in private practice in San Francisco. In addition, he teaches psychology and human sexuality at the collegiate level. Dr. Khamsi received his doctorate from the Saybook Institute in San Francisco, in part for his original research on primal therapy.

Return to the Int. Primal Assn. List of Articles on the Primal Psychotherapy Page