(The following post is based on my talk, Epistemic Injustice in Psychotherapy, which was part of an APA Eastern symposium on Agency and Epistemic Injustice in Psychiatry, organized by Şerife Tekin.)
If you’ve ever gone to psychotherapy (I’ll just refer to “therapy” for the rest of this post) you’re probably familiar with finishing a session and feeling pretty awful. This can happen for a host of reasons: maybe recollections of hard experiences came up, maybe your therapist pushes you to examine some things about yourself that you don’t really like, maybe it feels like you and your therapist are talking past each other the whole time. A lot of the time, these experiences are acceptable, because they are expected. After all, therapy isn’t supposed to be fun, and sometimes things get worse before they get better. However, sometimes these experiences might signal problems. There are, unfortunately, bad therapists. Or there are therapists who are good but aren’t a good fit for what you might need. Alternatively, there are types of therapy that aren’t helpful for everyone. In short, lots of things can go wrong in therapeutic settings and therapy may generate significant harms. It’s surprising, therefore, that even with a lot more attention being paid to studying the effectiveness of various kinds of therapy, very little attention has been paid to the kinds of harms that therapy can produce or how frequently they occur.
Let’s go back to one of the things I listed as having the potential to make you leave a therapy session feeling worse than when you started: the sense that you and your therapist kept talking past each other. This can happen no matter what the content of the discussion is, but I want to focus on cases where you and your therapist seem to have very different understandings of who you are. Your understanding of yourself appears to be very different from who your therapist thinks you are in some way that really matters. Maybe they insist that you procrastinate on writing because of anxiety, while you feel sure that you’re just lazy. Often it is likely that your therapist is the one who’s right, but not always. Perhaps in this case, you do suffer from an anxiety disorder that affects many areas of your life, but your ability to write isn’t affected by anxiety. Yet, your therapist insists. This kind of talking pass each other, with respect to your self-knowledge about the relationship between laziness and writing may lead to a significant kind of harm, which I call epistemic injustice with respect to self-knowledge in psychotherapy.
As a consequence of your therapist’s insistence, you may become convinced that you don’t write due to anxiety, you now hold a false belief about yourself where previously you held a true one. Even if you remain steadfast in believing that laziness explains your failure to write, you may now have a lower degree of confidence in that belief. Of course, just because you now understand yourself less well doesn’t by itself mean that any epistemic injustice has occurred. It could just be a case of epistemic bad luck, so I will need to say more about this. For now, though, consider the significance of going to therapy and losing self-knowledge. While the role of epistemic injustice in restricting self-knowledge is not unique to this context, here it is particularly devastating since: 1) it’s more direct, and 2) one of the primary aims of therapy is in fact increasing self-knowledge. It’s important both for thinking about therapy and for thinking about epistemic injustice, that we identify and characterize this harm. In the remainder of this post, I’ll give a sketch of how I think it goes.
People go to therapy for a lot of different reasons, and there are a wide range of contexts in which people get therapy. Some of these differences matter, so I’m going to narrow my focus here to cases where someone is getting therapy as part of treatment for a diagnosed mental illness. (It won’t matter for the argument if the reader rejects the concept of mental illness, though if the therapist conceives of the patient as having a mental illness and the patient identifies instead as mad, this would require changing some of the things I’ll say.) A problem familiar to many who engage in therapy for a mental illness is the question: what is me, and what is the illness? You might ask yourself whether a particular thing you thought, said, or did was you illness talking. John Sadler has given the difficulty of distinguishing yourself from your mental illness a name: self-illness ambiguity. He and others have argued that reducing or resolving self-illness ambiguity is necessary (though not sufficient) for recovery. Part of the task of therapy, therefore, involves the patient working to understand who they are as a person with or recovering from a particular illness. Dings and Glas have offered an analysis the epistemological obstacles a person can encounter in trying to reduce self-illness ambiguity. Among the obstacles they identify is the fact that the loss of trust in your own understanding of yourself (which frequently accompanies mental illness) requires trusting others and taking them to have authority and expertise when it comes to disambiguating you and your illness. This is where I think the potential for epistemic injustice emerges.
The therapist’s approach to the project of disambiguation is shaped in large part by their understanding of the illness. In analyzing epistemic injustice in psychiatry, Crichton, Carel, and Kidd distinguish between global and specific contributory conditions. Global contributory conditions are those that can affect any patient with mental illness (for example, negative stereotypes about people with mental illnesses in general). Specific contributory conditions reflect features of particular mental illnesses. For instance, if a person who has an eating disorder claims to hate the texture of certain calorie-dense foods and refuses to eat them, a therapist might insist that they don’t dislike the texture and that their refusal instead reflects the rules of their eating disorder and that their insistence that this is not the case indicates a lack of insight into their illness. It seems likely that in the context of therapy, specific contributory conditions will be the ones that actually lead to epistemic injustice.
The need to rely in part on other people to understand ourselves is certainly not specific to therapy; our understanding of ourselves is always shaped by others. This creates what McConnell identifies as a vulnerability to co-authoring of our self-narrative. According to narrative views of the self, we constitute ourselves in significant part via the narratives or stories we construct about our selves, our past, and our future. We and our co-authors rely in part upon narrative archetypes to develop our self-understanding. While these archetypes generally play a role earlier in our lives, before we develop our own authorial skills, finding ourselves in new and unanticipated circumstances can increase our need to draw on archetypes to develop a new self-understanding. Significant illness or distress are certainly among the kinds of circumstances that may have this result. Psychotherapy creates what McConnell calls a context-dependent vulnerability to co-authoring since the therapist has a strong influence on the patient’s self-narrative and the potential to undermine their authorial power. In considering what consequences this vulnerability has for healthcare professionals in general, McConnell emphasizes that listening carefully to the patient’s self-narration is essential, and a failure to do so undermines their authorial power. He also suggests that the healthcare professional should provide the patient with narrative resources that are adequate to construct a recovery or illness-management narrative. As an example, he notes that the integration of a type 2 diabetes diagnosis into a patient’s self-narrative can help them better manage their own care. In the case of psychotherapy, while patients may certainly be dealing with illnesses such as diabetes or cancer, the relevant narrative archetypes will also include psychiatric diagnoses based on the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). While it certainly isn’t the case that other kinds of illnesses are easily and unproblematically incorporated into self-narratives, mental illnesses present a special challenge. Part of this is due to the ways in which aspects of the self and of self-understanding are constitutive of mental disorders, but another part is due to the way in which DSM diagnoses function as archetypes. Tekin has argued that the DSM is a double-edged sword when it comes to patients’ self-understanding. While it does offer narrative resources for understanding oneself in the context of specific illness, treatment, and recover, the fact that those resources are built on the biomedical model of disease and don’t incorporate many aspects of patients’ lives that may contribute to their distress means that understanding oneself and one’s illness using DSM diagnoses as archetypes may lead to flawed self-understanding.
Returning to Sadler, he argues that the third-person view of the clinician and the first-person view of the patient constitutes a “perspective gap” that is to be closed in the course of psychotherapy. The job of the therapist is to translate the patient’s experience into a scientific or theoretical construct. To the extent that the therapist uses the DSM as a basis to guide the patient in reducing self-illness ambiguity and insofar as the patient takes this as authoritative (continuing to mistrust at least some beliefs about themselves that conflict with this third person knowledge), the potential for epistemic injustice to contribute to impoverished or incorrect self-knowledge increases. Group therapy contexts can heighten this risk due to increased chance of taking the DSM diagnosis as unifying the experiences of the group, and because the shared resources of the group contribute to patients whose self-understanding violates the shared understanding of illness and recovery having another source of trusted authority challenging the reliability of their own authorial capacity.
Let me now review the fundamental argument I make in the paper. One of the tasks of therapy is to reduce or resolve self-illness ambiguity and offer the patient better self-understanding. Self-illness ambiguity can lead to doubt or distrust of one’s own experience, meaning that the patient will often need to rely more heavily on the knowledge and experience of others, including their therapist. This puts the patient in a position to be vulnerable to co-authoring. Because the DSM offers a flawed framework for self-understanding, because it can form the basis of specific contributory conditions to epistemic injustice, and because it can serve an important resource (to both therapist and patient) in resolving self-illness ambiguity, there is a real opportunity for epistemic injustice with respect to self-knowledge to arise in the course of psychotherapy. Since an important task for therapy is to increase self-knowledge, this seems to me to be an especially perverse kind of epistemic injustice.
Megan Delehanty
Megan Delehanty is a philosopher of science at the University of Calgary. Her recent work is on philosophy of medicine and psychiatry. Her cat Pyxis is also a philosopher.
