In The Magic Lantern, Ingmar Bergman reports that when he read Arthur Janov’s The Primal Scream he was ‘extremely stimulated and started developing a television film in four parts along Janov lines’.¹ Bergman also notes that in 1975 he visited Los Angeles and had his agent arrange a meeting with Janov. Bergman remarks that he and Janov were ‘immediately on the same wavelength’ and ‘swiftly tried to get down to essentials’. In light of such facts, it would be unreasonable to doubt that Janov had an influence on Bergman at this stage in his career. The specific nature and extent of this influence remains difficult to identify, however. Michael Tapper offers the following conjecture on this topic in his book on Face to Face (1976):

Tapper also makes the following observation:

This chapter investigates the Bergman–Janov connection with regard to Höstsonaten (Autumn Sonata, 1978). Three main approaches to the interpretation of this film will be considered. The first reads Autumn Sonata as a work that was, like Face to Face, conceived ‘along Janov lines’, and that consequently resonates positively with the tenets of Janov’s psychology. The second and third interpretations both deny that Autumn Sonata is consistently Janovian. According to a first kind of ‘non-Janovian’ interpretation, Bergman duly accepted and worked with significant Janovian premises as he conceived of the story and characterizations for Autumn Sonata, just as the first interpretation holds, but adds that for various reasons, the director did not, finally, go on to make a thoroughly Janovian work. This interpretation purports, then, to identify unintentional (and indeed serendipitous) non-Janovian elements in the story conveyed by the finished audio-visual display. This is a kind of interpretation that is compatible with the idea that Autumn Sonata is to a significant degree ambiguous or ambivalent when it comes to Janovian doctrine. A third type of interpretation contends that even though Bergman planned to make a work ‘along Janov lines’, he had placed some critical distance between himself and at least some of the main tenets of Janov’s psychological theory and successfully expressed these reservations in his film. According to that interpretation, then, Bergman was not thoroughly or consistently persuaded of the truth of Janov’s theoretical contentions, either at the time of his initial, enthusiastic reading of The Primal Scream or upon subsequent reflection. Tapper could well be right that Bergman never ‘dissociated’ from Janov; but it does not follow that he did not become aware of shortcomings in Janov’s doctrines or that he did not in one way or another distance himself from certain tenets of Janov’s, at the very least when it was a matter of settling on what was meant to be fictionally true regarding the characterizations in Autumn Sonata.

The contest between the three interpretations of Autumn Sonata evoked in the previous paragraph can only be taken up once we have a sufficient understanding of what Janov’s most basic and characteristic doctrines were. Very briefly, then, Janov asserts quite clearly that all neurotic individuals are troubled by repressed pain, which their various neurotic past-times prevent them from confronting, the result being any number of neurotic symptoms, physical disorders, and an overall inability to have authentic feelings. The source of this destructive repression of pain is the childhood trauma of not receiving sufficient parental love and care. As John Lennon aptly put it in a song inspired by Janov, ‘Working Class Hero’: ‘when the pain is so great you feel nothing at all’. The parents really are the disturbed parties—as Tapper proposed above in his brief characterization of what Janov and Bergman had in common—and they pass their affliction along to their children.

As for Janov’s promised remedy for neurosis, the therapy begins by setting aside all neurotic distractions and comforts so as to confront the anguish and pain that well up spontaneously when these neurotic defences are down. The therapist directs the distressed patient’s thoughts towards his or her childhood and, more pointedly, to unrequited desires for parental attention and unconditional approval. Asking for mummy and daddy, the patient starts to feel her own pain, and the primal screaming begins. After many such sessions, there is supposed to be a crescendo, the moment of ‘abreaction’ in which the pain is finally gone and the healed patient is no longer neurotic. Once the repressed pain has been released, the patient can have authentic feelings. Janov claims that such persons find the company of neurotic individuals unpleasant, but co-exist peacefully with others who have finally confronted the infantile trauma and found emotional release from it.

Janov acknowledges that his doctrines recapitulate some very familiar psychoanalytic tenets. He was after all a practising, ‘talking-cure’ therapist, and one who made exorbitant and unjustifiable claims about the benefits to be had from his treatment. Unlike several of the more radical anti-psychiatric figures, he did not espouse the idea that psychotic and neurotic disorders are valuable or insightful ways of being that actually require no cure. There is no claim, for example, to the effect that in a deeply alienating and irrational society, it is the misfit, or somebody who is deemed to be abnormal, who might have a correct perspective on the way things are really going. For example, Janov notoriously deemed homosexuality a psychic disorder that was to be cured by successful primal therapy.

I turn now to the interpretations of Autumn Sonata, my main question being whether this is a work the psychological underpinnings of which are consistently Janovian. One might think so, first of all, because Bergman tells us that he took Janov’s theory very seriously. Since he reported explicitly that Face to Face was conceived ‘along Janov lines’, it is plausible to conjecture that this could also have been true of Autumn Sonata, especially given that Janovian themes obviously resonate with important aspects of the story conveyed by the film. In order to assess such an interpretation, we must provide at least a sketchy account of how central elements of the story and characterizations might be understood along those lines. I shall be fairly brief, and assume that my readers have seen the film at least once.

The character played by Liv Ullmann, Eva, invites her mother Charlotte, a famous concert pianist, portrayed by Ingrid Bergman, to visit her home in Norway. They have not seen each other for seven years. Mother and daughter greet each other warmly. Yet soon tension builds. It is obvious that Eva finds her mother difficult and irritating. She is theatrical, self-absorbed, vain, at times obviously false. Eva gushes fervently to her mother about her mystical ideas; Eva believes in God and is convinced that there are countless realities. She believes that her son, who drowned when he was three years old, still lives in another world, yet is also in close contact with Eva. Charlotte is visibly sceptical and would appear to find her daughter’s mystical thoughts distasteful. Later that evening, Charlotte makes the mistake of asking Eva whether she likes her, and Eva unfolds a bitter litany of complaints and accusations—what Jan Holmberg aptly calls a ‘vomit of lamentations’ (see Chapter 6 in this volume). She tasks her mother with having been an egotistical, cold, hypercritical, and systematically neglectful parent. Scylla was neglect, Charybdis was active disapproval and torment. Eva suffered horribly during the long periods of her mother’s absence; yet when her mother returned, things only got worse, as Charlotte was overbearing and expressed constant disapproval. Eva’s complaints are illustrated in flashbacks which, on this pro-Janov reading of the film, offer veridical depictions of the actual events in an implicit narrational confirmation of Eva’s side of the story.

Eva becomes more and more angry as she lists what she describes as her mother’s selfish, unfeeling, and hypocritical behaviours and the terrible effects they had on her. The mother’s injuries, failures, and unhappiness, she repeats again and again, are passed down to the daughter. The crescendo in this lengthy string of accusations comes when Eva blames her mother for the miserable condition of Eva’s younger sister, Helena, who is severely disabled and bedridden, and who must struggle to articulate even the simplest words. In a sort of primal crisis that emerges in parallel to the dispute between Eva and her mother, the wretched Helena manages to tumble out of her bed, hopelessly crying out for her mother.

Initially, the talented, world-famous pianist tries to defend herself by characterizing Eva’s mounting accusations as so many wild exaggerations. Broken down finally by her daughter’s emotional outbursts, Charlotte responds by explaining her very real emotional shortcomings as the product of her own miserable childhood. Without admitting that all of Eva’s accusations are well-founded, she asks Eva to forgive her.

To sum up, there are reasons to think that Eva’s psychologizing is consistent with Janovian doctrine, and that significant aspects of the work are designed to bring the spectator into sympathetic and empathic alignment with Eva, who effectively functions as Bergman’s porte parole. A critic who has defended these sorts of claims in print is Robin Wood, who reads the film as implicitly endorsing Eva’s accusations along with the psychological and ethical assumptions that subtend her vitriolic attack on her mother. Wood takes Bergman, or at least the film’s implicit authorial persona, to be endorsing the harsh statements that Eva makes in anger to her mother, such as: ‘People like you are a menace. You should be locked away so you can’t do any harm.’⁴ As Wood puts it, the film ‘degenerates into what amounts to a hysterical diatribe against a woman who neglected her children for her career as a concert pianist’.⁵ Wood contends that the film is designed to encourage the spectators to go along with the proposition that even Helena’s severe physical handicap was caused by maternal neglect. It is certainly true that Eva presses this accusation, insisting with ferocity that her mother’s selfish behaviour caused Helena’s condition. There is a flashback illustrating an episode where Charlotte’s abrupt departure, along with the related departure of her companion Leonardo, supposedly precipitates an irreversible crisis in Helena’s condition.

Charlotte appears to be genuinely surprised by this accusation and asks Eva how it could be true. Eva’s response, presented by her as decisive, is to ask Charlotte whether she can prove otherwise. This is a crucial moment in the dialogue, as Eva’s question shifts a heavy burden of proof onto her mother’s shoulders. Why, some spectators may well wonder, should Charlotte have to prove that her behaviour was not the cause of her daughter’s severe affliction?

For those spectators who are not Janovians and who interpret the story as a matter of make-believe or imaginary events that are consistent with real-world physiological (and other) constraints, Charlotte’s failures as a mother can hardly have been the sole cause of Helena’s severe illness and disability. Only in an otherworldly allegory of primal scream theory could this be the aetiology of Helena’s condition. Thus Robin Wood states that her condition is a ‘surely physiological’ degenerative disease.⁶ If that is the right way to understand the story, Eva’s accusation is analogous to Janov’s faulty claim that poor eyesight is a product of neurosis that could be remedied by primal scream therapy. It would be reasonable, then, to protest that Charlotte does not have to shoulder the unfair burden of proof that Eva tries to impose on her; it is Eva who should have to provide evidence to support her problematic accusation that Helena’s dreadful physical disability is all Charlotte’s fault.

Given these strong objections to a fully fledged Janovian interpretation of the work, there is good reason to entertain the two aforementioned alternative interpretations: either Bergman had thoroughly Janovian intentions, but somehow failed to realize them in the work, or, in one way or another, his intentions were not so thoroughly Janovian, and the work was successfully designed to express that ambivalence. I turn first to some evidence that supports either of these interpretations, and then I address the question of which of these alternatives is best supported by additional evidence.

One good reason why Eva should not be taken as Bergman’s porte parole is that when she launches her attack on her mother, she has consumed a great deal of red wine, is obviously intoxicated, and is not expressing attitudes that even she herself would endorse upon further reflection. And indeed, once Charlotte has left the house and Eva has had time to calm down, she writes her a letter in which she apologizes to her mother for having tormented her with what she herself calls an ‘old soured hatred that is no longer real’. Everything she did was wrong, she adds, asking for forgiveness. That would obviously include the unjustifiable accusation regarding Charlotte’s responsibility for Helena’s severe disability.

Given this major inconsistency in Eva’s behaviour in the course of the film, her extremely violent outbursts and hyperbolic, drunken accusations should not be read as following from anything like an accurate and comprehensive understanding of her childhood experience. These outbursts may be better understood as the expressions of a desire to assert herself and to demand her mother’s attention. The film, then, has no ‘moral’ to the effect that the path to well-being passes through the activation of ‘primal’ childhood memories and longings. Nor is it the case that an unrequited infantile demand for unconditional love and care is the avoidable cause of a very wide range of mental and physical disorders. Eva learns about the difficulties her mother faced as a child; and this helps her gain a broader perspective that makes it possible for her to place herself at some distance from her emotive impulses and, most importantly, from her angry judgements. After all, if the key thesis were that the symptoms are the disturbed parent’s fault, this leads to a regress, since the parents’ symptoms were in turn caused by the neurotic behaviour of the disturbed grandparents, and so on, ab ovo. Nor is there any evidence in the story of Autumn Sonata to support the optimistic thesis that some powerful abreactive event can bring about a full and long-lasting release from neurotic symptoms. It is at least symbolically relevant to observe that Helena’s convulsive screaming at the end of film signals only more pain and confusion, not an abreactive release. On the basis of the evidence presented in the film, it would be hard to be optimistic about the prospect of Helena’s being cured by some form of talking therapy.

Janov’s advocate could reply that Eva is not in therapy and that a clinical setting and actual primal treatment, and not an angry encounter with the actual mother, is what she needs. Bergman’s family drama cannot be expected to illustrate the more optimistic tenets of Janovian psychotherapy, and that it does not do so hardly means that the director was assuming an anti-Janov position on that topic. Yet Bergman’s scenario does explore relevant factors that go uncovered in The Primal Scream. Eva may well have had a difficult childhood, but she has also been seriously afflicted by the loss of her child, and it is hard to see how any of the tenets of primal psychology can help her with that. The idea that Eva’s mourning is pathologically prolonged or intensified by her parent-induced neurosis is hard to square with the proposition that her neurosis prevents her from having real feelings. Her mourning for her lost child certainly looks sincere and heartfelt.

More evidence running counter to a thoroughly Janovian interpretation can be found in one of the most interesting scenes in Autumn Sonata—the part where Eva plays a Chopin Prelude for her mother, who in turn comments on the piece and then performs her own interpretation of it. In her discussion of this sequence, Anyssa Neumann (in Chapter 8, this volume) points to shortcomings in what Charlotte has to say about this Chopin composition. Neumann argues that the gap between Käbi Laretei’s two performances of the composition is not as great as the story prescribes. As Neumann admits, these facts do not falsify the basic story premise to the effect that Charlotte’s performance is that of a world-class pianist, whereas Eva’s is not. The narrative context and use of reaction shots no doubt lead the average spectator to magnify perceived differences in skill across the two performances (where the actual differences, as Neumann argues, ought to have been even greater for the sake of the story). There is, in any case, no doubt that Eva is crestfallen once her mother has finished playing. For Eva this is yet another episode in a life as a victim of a selfish, unappreciative, and overbearing parent. However, such a response on her part is a mistake. Charlotte does not expect her daughter to be a brilliant pianist. She only plays and talks about the Chopin piece when Eva insists. In so doing Eva has put Charlotte in a double bind, for Charlotte must either speak dishonestly about the music or speak sincerely and disappoint her daughter’s unrealistic desire. At no point does Eva express any sort of positive attitude towards the fact that her mother has a rare and precious musical talent. That Charlotte indeed has such a talent is a basic premise of this story. Nor does Eva allow that Charlotte’s devotion to the art has any kind of legitimate place in the world. Instead, the brilliance of her mother’s performance makes her jealous. When Charlotte speaks about how long she has worked to try to understand Chopin’s preludes, Eva replies that as a child she was quickly very tired of her mother and her pianos.

I believe the evidence surveyed here suffices to establish that a fully fledged pro-Janov reading of the film is unsatisfactory, as there are too many elements in the audio-visual display that point to story propositions that are inconsistent with the pro-Janov interpretation. That leads us to the question that divides the two other alternative readings under consideration here: are the film’s non-Janovian psychological insights serendipitous, or were they instead the fruit of Bergman’s better intentions?

How might one argue for a reading whereby it was never Bergman’s intention in the conception and making of Autumn Sonata to fashion a story that would serve as a consistent illustration of the tenets of Janovian psychology? One might start by appealing to a relevant counterfactual: surely if had he really wanted to, an author as talented as Bergman could have devised a story that fully meshed with Janov’s theory. However, to this it might be replied that Bergman admitted to having struggled unsuccessfully with the making of Face to Face. He confessed that the work suffered from the ‘ill-digested fruit’ of his reading of Janov. A similar problem, it could be conjectured, arose in the making of Autumn Sonata. Whence the serendipity reading: trying to bring out a story along Janovian lines, Bergman ended up with one that manifests ways in which that doctrine is incomplete and problematic.

Some external evidence lends some tentative support to the interpretation whereby the non-Janovian elements were actually intended by the work’s author. In her investigations into the Bergman–Janov connection, the Canadian musiologist Alexis Luko engaged in email communication with Janov. Luko reports that in one email to her, Janov stated that ‘Bergman’s chief interest in using primal therapy techniques was as a means to coach his actors on how to feel.’⁷ Janov also related to Luko that ‘Bergman says my ideas influenced him but I am not sure how.’⁸ Both of these statements can be interpreted as casting doubt on the thesis that Bergman and Janov had a deeply theoretical meeting of minds. Had they done so, why was Janov still in the dark with regard to how Bergman understood his psychological views? Bergman, it seems, was primarily interested in ways of stirring up feelings, first in his actors and then in his audience. As Tapper put it in the passage quoted at the beginning of this chapter, it could well be the case that Bergman was at no point trying to make a work of fiction that would serve as ‘propaganda’ for primal scream theory.

It could be replied, however, that even if Bergman had always been somewhat agnostic about Janov’s theory and was not finally interested in making a cinematic work that would take a stance on it as a hypothesis in scientific psychology, he might still have sought to recruit some of its tenets as interesting premises for an engaging psychological drama ‘along Janov lines’. Here we need to draw a distinction between (1) the author’s effective and final story intentions pertaining to what would be fictionally true in a film’s story, and (2) whatever fervent psychological-theoretical beliefs the director might have held and would have been willing to assert. It could be a mistake to think that the items in (1) were all consistent with or informed by those in (2). And even if, with regard to (2), Bergman had a fervent belief in some Janovian proposition in psychology, it might not have been his aim to use any particular cinematic fiction, with its corresponding items in (1), to make an assertion along those lines. With this distinction in mind, it might be added that the object of the interpretations of Autumn Sonata under consideration here are the items in (1) in their relation to the finished audio-visual display. Items in (2) are relevant to such interpretations but are not their primary object.

Some of Bergman’s remarks about Autumn Sonata in Images: My Life in Film appear to lend some support to the serendipity interpretation, whereby there is at least some inconsistency between Bergman’s final story intentions and the meanings of the audio-visual display.⁹ Apparently Bergman initially had the idea that this would be a story in which ‘the child gives birth to the parent’. The thought here would be that the talented and successful concert pianist is in some sense not truly alive until her daughter’s tirades bring her to acknowledge her own emotional repression and neurosis. Eva’s accusatory diatribes and screams would, then, be therapeutic, not only or primarily for herself, but for her mother. Yet consider Bergman’s further comment in Images: My Life in Film on this topic: ‘There is something close to an enigma in the concept of the daughter giving birth to the mother. Therein lies an emotion that I was not able to realize and carry through to its conclusion. On the surface, the finished film resembles the outline, but actually that is not the case.’¹⁰ Bergman goes on to say that perhaps he did not ‘drill deeply’ enough. It is unclear how this metaphor is to be understood: is the thought that this schematic ‘giving birth’ motif is psychologically superficial? The implication, then, would be that some of the key premises based on primal psychology (such as the proposition that parentally induced childhood trauma causes severe physiological disorders and degenerative disease) did not survive the process of developing a sufficiently plausible dramatization. One odd irony here is that there is nothing in Janov about a child’s providing needed primal therapy for a parent, so in this regard Bergman’s reasoning ‘along Janov lines’ is strangely non-Janovian.

Some observations about the acting in the film are relevant to the serendipity interpretation. Liv Ullmann’s remarkable body language and intonations help portray an Eva who is immature and weak, at times annoyingly so. She struggles to read a letter aloud to her husband. Some of her pronouncements have a decidedly dim-witted quality. When she gets drunk and lets herself go, she obviously speaks from her heart, but she also becomes blunt and ugly. The spectator is no doubt meant to feel sorry for her, to find her somewhat sympathetic; but this does not mean that her actions and attitudes carry conviction, especially when she aggressively presses what is a highly dubious accusation against her mother. Ingrid Bergman’s performance certainly brings out some of the famous pianist’s vanity and self-absorption, partially re-enforcing Eva’s accusations. Yet Bergman’s Charlotte remains impressive; she manifests experience and intelligence. If the intention was to make a film in which the child gives birth to the parent, casting Ingrid Bergman in the role of the yet-to-be born parent was perhaps a serious blunder.

To sum up, my view is that a fully fledged pro-Janov interpretation cannot be squared with all of the available evidence. As for the choice between the two non-Janov readings of this cinematic work, it might be said with a degree of caution that the available evidence does not support a firm judgement in favour of either of them. This is the case because we lack systematic and detailed evidence into the evolution of Bergman’s beliefs and artistic intentions, and also because some of the available evidence about them is fairly ambiguous. For example, the previously cited statement in Images: My Life in Film would appear at first glance to count in favour of the serendipity reading: Bergman identifies an important way in which the film did not turn out as planned, and we have no reason to deem him insincere in this regard. Yet this same evidence can also be interpreted as indicating that Bergman changed his mind about the Janovian premises as he made the film, in which case the story’s critical distance from a Janovian perspective was, finally, the fruit of Bergman’s better artistic (and other) judgements. Perhaps his final, effective artistic intentions regarding the work’s themes and story were influenced by Ingrid Bergman’s strong resistance to the initial characterization of Charlotte as an emotionally ‘unborn’ mother. In any case we may conclude that although there are certainly traces of a Janovian perspective in the story of Autumn Sonata (and that it is indeed very likely that Bergman at least started out with some Janov-informed story premises), the work does not finally resonate fully with a psychological doctrine according to which many or even most of the woes of the world, including physiological afflictions in the order of cerebral palsy, are the product of selfish parenting. Nor does this fiction evoke a world where people can scream their way to well-being. Bergman’s Autumn Sonata hardly supports the idea that music, or the fine arts more generally, are a neurotic or ‘narcissistic’ product that the post-neurotic subject could live happily without. Bergman was no doubt interested in making artistic use of aspects of Janov’s psychology and therapeutic techniques; but in the age-old rivalry between the doctors and the artists, he came down finally on the side of the artist.

A thought that may come to mind at this point is that the psychological insights of Autumn Sonata, at least as I have described them, are sketchy and relatively modest. Perhaps that is right, but it is also relevant to point out that the film is nonetheless thought-provoking and potentially instructive. The moving and at times very eloquent conversational exchanges, illustrated with flashbacks illustrating how things might have seemed from the character’s perspective, establish a dialogical narrational pattern which suffices to cast serious doubt over the blatant simplifications of primal scream psychology. Bergman’s fictional story provides an imaginary counterexample that can be taken as challenging some of the tenets of a psychological theory, the key idea being that in a case like this, an adequate diagnosis would not amount to saying that a successful and extremely talented professional woman is guilty of having single-handedly caused severe mental and physical disorders in her children. The form of the argument is simple: a particular theory says that such-and-such is what happens, and a case is imagined that does not convincingly work that way; consequently, a possible challenge to the theory has been brought to mind.

Acknowledgements

Thanks to Alexis Luko for an informative discussion of the Bergman–Janov connection based on her research at the Ingmar Bergman Archive and on her email interview with Janov.