NDJ:7 Annette Leavy, LCSW, BCD

Imagination and Atonement

“Tis safer to be that which we destroy, /Than by destruction dwell in doubtful joy,” (Shakespeare, p. 982). So speaks Lady Macbeth as she begins to contend with the crimes she has committed. By the play’s end, Lady Macbeth has gone mad and killed herself, literally becoming that which she destroyed. Thus, Shakespeare’s words and his tragedy’s action suggest that to atone requires a profound identification with those one has harmed.*

Is it possible to achieve the degree of empathic identification needed to atone without destroying oneself? How might this task be accomplished? Another work of literature provides some answers.

At age thirteen, Briony Tallis, protagonist and narrator of Ian McEwan’s novel Atonement (2001), tells a story. As a result, the wrong man, her sister Cecelia’s lover, is convicted of rape and sent to prison. The lovers’ lives are destroyed. Cecelia dies during the bombing of London in World War II and her lover, Robbie at Dunkirk. Briony’s atonement takes the form of the novel she writes to set the record straight, which she is only able to complete fifty-nine years later. While accurate about her own misdeeds, the novel ends with the lovers’ alive, together and demanding Briony’s recantation. The novelist Briony not only breathes life back into Robbie and Cecelia, but also offers herself the chance at reprieve and reconciliation she did not actually have. She tells the  reader: “The problem these fifty-nine years has been this: how can a novelist achieve atonement when, with her absolute power of deciding outcomes, she is also God?”(McEwan, 2001, p. 371).

I wonder if the novelist’s problem with atonement belongs to all those who need to atone. Briony becomes that which she destroys, but up to a point. Perhaps it is not possible to fully experience oneself as doer and done to—the aggression against the self would be unbearable, as it is for Lady Macbeth. The atoner’s identification can only be fictive, a facsimile of the original hurt. Briony concludes: “It was always an impossible task, and that was precisely the point. The attempt was all.”(McEwan, 2001, p. 371).

McEwan’s aim, in part, is to distinguish the brutality of actual events from the potentially brutal power of the imagination. Novelists, it seems to me, are better at making this distinction than psychotherapists. Guilt is our stock in trade, but we are less accustomed to dealing with actual misdeeds. Our patients are often guilty about fantasized crimes, not real ones. But what if a patient needs to decipher if she caused actual harm? What role would therapy have in this process?

A recent consultation highlighted this problem for me. A woman in her early thirties came to see me as her approaching marriage escalated tensions within her family and threatened her often-precarious equilibrium. As a teenager, she had accused her father of sexually molesting her. Teachers, her mother, grandparents, evaluating psychotherapists and a district attorney were convinced of her veracity, and a lawsuit was brought against her father, only for her to recant just before trial.

Some years earlier, her parents had divorced acrimoniously. Following the divorce, her father remarried and fathered new children. Her mother, always mercurial, became even less reliable. My patient’s situation was similar to that of thirteen year-old Briony whose father is in London with his mistress, mother in bed with a headache and older sister consumed with the beloved Robbie. To bind her turmoil and confusion, the real girl, like the fictional one, told a story, and once her story commanded attention, she needed to believe the authenticity of her creation.

In the years that followed, my patient and her family never talked about what she called “the incident. ” It was, to use the Freudian phrase, scotomized, known and disavowed in the same breath. In recounting “the incident,” it was as if she were telling me a story about someone else. All feelings in our discussion seemed to belong to me, and I felt disoriented. I did not know what in my patient’s story was true or false, who was doer, who done to. I did not know if her original claim of incest was true, her retraction or neither.

My patient appeared to share none of my doubt and disorientation. It was as though she had performed a magic trick; by sleight of hand, one truth had replaced another. I, on the other hand, could find meaning only in the confusion I was experiencing. To be confounded was the way I had begun to know my patient. Like Briony Tallis, my patient needed to become a more reliable narrator of her own experience. For our work on this project to be meaningful both of us would inevitably experience ourselves as the aggressor and the aggressed, the doer and the done to. This was my guiding formulation, the story I told myself to make sense of our consultation.

Such an enactment is commonplace in clinical practice, and in this way, the work of therapy could be a precursor to the work of atonement. But can the place of insight also be the place of atonement? McEwan reminds us that all anyone can do is to attempt to atone, and that attempt can take a lifetime. In beginning therapy with my patient, it was essential for me to question if both she and I would have the flexibility, stamina and tolerance for emotional and moral ambiguity such work would require as well as to consider if the attempt, which therapy might offer her, could be her all.

* For the concept of atonement as necessitating an identification with the aggressed, an at-one-ment with the victim, and specifically for the elaboration of this theme in Macbeth, I am indebted to Irwin C. Rosen in his paper, “The Atonement-Forgiveness Dyad: Identification with the Aggressed.”


McEwan, I. (2001). Atonement. London, England: Jonathan Cape.

Rosen, I.C. (2009). The atonement-forgiveness dyad: identification with the aggressed. Psychoanalytic Inquiry 29: 411-425.

Shakespeare, W. (1953). Macbeth. In C.J. Sisson (Ed.) William Shakespeare: the complete works. New York: Harper & Row.