Article Text

Download PDFPDF

Psychiatric taxonomy: at the crossroads of science and ethics
  1. Şerife Tekin
  1. Correspondence to Professor Şerife Tekin, Department of Philosophy and Religious Studies, Daemen College, 4380 Main Street, Buffalo, NY 14226, USA; stekin{at}

Statistics from

The scientific investigation of mental disorders is an invigorating area of inquiry for philosophers of mind and science who are interested in exploring the nature of typical and atypical cognition as well as the overarching scientific project of ‘carving nature at its joints’. It is also important for philosophers of medicine and bioethicists who are concerned with concepts of disease and with the development of effective and ethical treatments of mental disorders and the just distribution of mental health services. Philosophical worries surrounding mental health and its care have recently extended beyond the bounds of academia, becoming a vigorous topic of debate in a variety of public domains in the wake of the publication of the most recent revision of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), the psychiatric classification system used by mental health professionals in the USA, and increasingly, by those around the world.

The DSM-5 lists mental disorders according to the observable symptoms presented by patients. It is designed for pragmatic use across a variety of settings to accomplish several tasks: to facilitate clinical treatment, to provide clear criteria of eligibility for various administrative and policy related purposes (such as the determination of insurance coverage and disability aid (p. xxiii;1 p.xli2)), and primarily, to further scientific research into mental disorder aetiology. Although designed to meet the needs and interests of various stakeholders (including patients and their families, researchers, clinicians and insurance companies), the recently revised manual has not fully satisfied any of them. This failure challenges the assumption that a single manual of mental disorders can adequately fulfil the variety of purposes for which it is intended.

Controversy regarding the changes brought about by the DSM-5 arises in part from its symptom-based approach to classification, which developed as a reaction to earlier aetiological approaches grounded in psychoanalytic theory. These early approaches to mental disorder classification relied on empirically undefended theoretical assumptions rather than outwardly observable correlates of disease. Clusters of symptoms and signs are thought to facilitate objective scientific research and clinical diagnosis, because as Steve Pearce3 puts it, these ‘consensus-based lists’ afford clinicians a sense of certainty in an area of medicine where no physiological tests are plausible, increasing the usefulness of diagnostic categories. However, as Pearce argues, these symptom clusters fail to represent certain complexities of the phenomenology of mental disorder, such as patients’ socio-cultural contexts within their life narratives. The symptom-based approach therefore risks abstracting away the important features of the subjects’ experiences that may be crucial for therapeutic improvement. Pearce addresses here what philosopher and psychiatrist John Sadler has called ‘hyponarrativity’—the abstraction of the illness category from the particular experiences and contingencies (such as gender and race) of the individual patient.4 Because the symptom-based approach lacks sensitivity to such particularities, a clinician may not be able to fully address patients’ needs.

An example of this problem relates to the removal of the bereavement exclusion criterion for diagnosis of major depressive episode. This modification allows grieving persons to be diagnosed with major depression, as the observables for those who have grief-related distress are similar to those exhibited by persons with clinical depression. As JS Blumenthal-Barby rightly points out, the removal of the bereavement exclusion criterion from the DSM-5 leads to the medicalisation of a ubiquitous human experience, as well as the risk of trying to address and manage griever's distress through pharmaceuticals.5 One conceptual concern with this modification to the DSM-5, which was intended to help grief-stricken individuals receive clinical support and insurance coverage, relates to its failure, as Bluementhal-Barby puts it, to distinguish between ‘disorder’ and ‘nondisordered conditions for which we help people’. Because the same manual is used for research, treatment and insurance purposes, the best way to help those who cannot cope with their grief is to reify normal grief as illness, thus initiating a research programme premised on this distinct disease entity. Rachel Bingham and Natalie Banner similarly argue that the DSM-5 fails to adjudicate disorder from non-disorder.6 Using the removal of homosexuality from the DSM in 1973 as a litmus test against which candidate definitions of mental disorder might be evaluated, they convincingly argue that the DSM-5 definition of mental disorder does not rule homosexuality out as a disorder and thus that it fails to mitigate against potential future abuses of psychiatry, especially in societies where same-sex rights are not recognised or enforced.

The complexity of accommodating the interests of various stakeholders in these issues is heightened when considered from the perspective of an individual patient, for whom the quality of life is contingent upon the quality of care and various administrative and financial accommodations. George Szmukler argues that the research, clinical and policy contexts require ‘different notions of diagnosis to tackle the particular problem such a designation is meant to solve’.7 To illustrate this, he distinguishes between a ‘status’ definition of a mental disorder (ie, a diagnostic label or category) and a ‘function’ definition of a mental disorder (ie, how well the patient is able to meet the demands of a test of performance requiring certain capabilities, aptitudes or skills). A status definition can allow a criminal to be not legally accountable for his actions if he or she committed a crime during a psychotic episode or during a severely impaired level of consciousness. Lacking such a diagnosis, the wrongdoer would be held legally accountable for his or her actions. The result of the status definition is that the criminal is committed either to hospital or to prison, neither of which creates meaningful opportunities for their betterment. A functional approach, on the other hand, assesses whether the criminal was capable of making a decision at the time of the crime, which requires a more thorough engagement with them in determining their competency, and addresses their needs in a way that encourages their reintegration back into society.

The DSM-5 has also caused controversy among its research-oriented stakeholders. Martyn Pickersgill argues that a sociology of critique within the psychiatric landscape might lead to a better understanding of the controversy.8 The National Institutes of Mental Health (NIMH), the organisation within the US government that funds most research in psychiatry, has abandoned the DSM for research purposes. The arguments put forward in favour of this decision are that (a) the DSM categories lack validity, and (b) a diagnostic system that aims to scrutinise mental illness should more directly reflect modern brain science, as ‘mental illness will be best understood as disorders of brain structure and function that implicate specific domains of cognition, emotion, and behavior’.9 As an alternative to the DSM, the NIMH announced the Research Domain Criteria (RDoC) project, which is attempting to create a new conceptual framework to describe psychiatric research that brings together the resources provided by various basic sciences, including genetics and neuroscience. Critics of the NIMH's approach have suggested that the primacy afforded to neuroscientific and genetic research in psychopathology continues an unfortunate trend that ignores the crucial role of the phenomenology of mental illness.10 Felicity Callard maintains that it is important not to evaluate these concerns as reflecting a debate between the proponents of scientific objectivity and those who deny the reality of mental illness.11 Rather, diagnosis should be viewed as a temporally distributed process of negotiation, and we should welcome ambivalence ‘vis-à-vis the achievements and problems of psychiatric diagnosis’.

These conflicting views surrounding the nature and purpose of the DSM-5 illustrate that the way we classify mental disorders is not only a theoretical question of accurately individuating life experiences associated with mental distress. It is also an ethical question about how psychiatric taxonomy might directly impoverish or enrich the quality of lives of patients and their caregivers through its influence on clinical treatment and policy decisions.


View Abstract


  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.