Repository logo

The Therapist as Scientist: A Philosophical Appraisal of Cognitive Behavioural Therapy



Change log


Ratnayake, Theuni Sahanika 


Cognitive Behavioural Therapy (CBT), one of the most popular schools of contemporary psychotherapy, presents itself as "scientific" and "evidence-based", particularly in contrast with its predecessor, psychoanalysis. This scientific image has shaped CBT on a number of levels. For example, the 'origin story' told by CBT proponents of their own history and the early development of CBT, is one in which the psychoanalytic account of depression, which had gone unquestioned and untested, was proved wrong via scientific experiments, setting the stage for CBT's novel theoretical account of depression. This theoretical account characterises the aetiology of depression — and mental illness more generally — in terms of thinking errors or specific failures of epistemic rationality. Mental illness is described in terms that suggest individuals fail to have an accurate picture of the world, the sort of picture that one might have if one were a scientist. Consequently, CBT's therapeutic approach, described as "collaborative empiricism", involves the therapist and the client acting as scientists, trying to uncover and replace the erroneous or "distorted" thoughts associated with mental illness using quasi-scientific techniques such as "hypothesis testing" and "reality checking".

By scrutinising CBT training manuals and psychological research, as well as drawing on relevant literature from history and philosophy of science, medicine and psychiatry, I dismantle this image of CBT on a number of fronts. Firstly, I show that CBT is not unique in presenting itself as 'scientific' or 'evidence-based'. Psychoanalysts were not unconcerned with evidence and Freud in fact saw himself as bringing psychoanalysis firmly under the rubric of natural science. This is not to say that psychoanalysis succeeded in these goals, but rather that CBT's attempt to distinguish itself from psychoanalysis in this manner is misleading. I suggest instead that the transition away from psychoanalysis to CBT is better explained by the fact that what constituted legitimate evidence shifted during this time — from clinical expertise and observations gathered from therapy sessions, to evidence from randomised control trials — a shift that CBT navigated successfully while psychoanalysis floundered.

Secondly, contrary to the description of CBT as evidence-based, this evidence is partial, relating largely to CBT's efficacy rather than its description of mental illness and the mechanisms that underlie its therapeutic approach. To put this another way, we have evidence that CBT works, but we are not entirely sure why it works. In challenging CBT's theoretical adequacy, I question the characterisation of mental illness in terms of thinking errors and epistemic issues.

Finally, I show that even if mental illness involves epistemic issues, CBT does not have the resources to rectify epistemic issues as it presupposes a simplistic understanding of epistemic and scientific investigation. In concluding, I propose an alternate grounding of CBT's account of mental illness and its therapeutic techniques in terms of failures of practical rather than epistemic rationality.





Holton, Richard
Alexandrova, Anna


Cognitive Behavioural Therapy, Philosophy of Psychiatry, Psychotherapy, Philosophy of Medicine, Philosophy of Science, Philosophy of Psychology, History of Psychiatry


Doctor of Philosophy (PhD)

Awarding Institution

University of Cambridge