Statistics from Altmetric.com
We are grateful for the thoughtful attention the commentators and editors have given our paper. They raise many substantive points that warrant a response, but for reasons of journal space our reply must be brief. In our paper, we argue for an amended hybrid account of ‘disease’ in human medicine that takes normative ethics seriously, guards against pernicious classifications of disease and reconnects the concept with the goals of healthcare institutions in which disease diagnosis is embedded. Carel and Tekin, in their respective pieces, raise issues about the objectivity and effective operationalisation of our account, Agar makes an ‘evolutionary’ case for disease concept eliminativism in mental health, and Wakefield and Conrad contend that our proposed modification to the harmful dysfunction account is unnecessary and undesirable. We will respond to each in turn.
The rational moral justification component of our account raises difficult moral epistemological questions regarding which reasons are the weightiest ones and who should be allowed to participate in the rational justification process (though so too, we would note, does much bioethical inquiry). Carel suggests that inevitable disagreement on this point undermines the alleged objectivity of our proposal. We agree with Tekin, however, that this is a feature of our account rather than a bug. What constitutes a good reason in moral deliberation is a question that goes to the very philosophical foundation of ethics (see1 for a recent discussion); but disagreement over which reasons are the weightiest ones does not prevent ethicists and policy makers from objectively assessing the strength of moral arguments by evaluating their empirical adequacy and logical structure, and this can take us a long way toward adjudication. The fact that rational justification has been used to support immoral practices and institutions does not warrant abandoning a reason-based approach to ethics any more than the fact that the legal excuse of self-defence has been misused for nefarious purposes warrants abandoning that concept, even if its historical misuse gives us cause for scepticism. Indeed, only a reason-based approach can identify these moral risks and errors in the first place. So, if any method of moral adjudication is to succeed, it will be a reason-based one; if none can succeed, then ethics is a lost cause. And if, as we argue, disease is a properly moral term, then we have no recourse other than to engage in a practice of reason-giving and argument assessment. Carel also worries that our focus on rational justification makes no provision for the input of first-person perspectives of patients and other stakeholders who are routinely overlooked in the process of disease diagnosis. Yet far from excluding such perspectives from the process of rational justification, our account necessitates their inclusion insofar as considerations of patient well-being and autonomy are central to any evaluative analysis of dysfunction.
Tekin shows how our proposal can be used to re-evaluate the criteria used for diagnosing mental illness, which she nicely illustrates with a discussion of depression in the context of bereavement. In contrast, Agar argues that while the thickly normative hybrid account of disease is well-suited for human medical physiology, it is poorly suited to mental health. Agar’s case for a purely welfarist approach to neuro-psychological diversity relies on the decoupling of biological fitness and well-being, which he contends may be especially pronounced in the psychological domain. Although we are sympathetic to welfarist approaches to disease in general, we are of the view that retaining a unified concept of disease across human medicine is ethically preferable, for several reasons. First, many mental illnesses implicate plausible physiological and psychological dysfunctions even if the structure of their underlying mechanisms remains poorly understood. Second, although human cognitive and social environments are now very different from those of the late Pleistocene in which many human-specific psychological functions are thought to have congealed, it is not obvious that mental or cognitive dysfunctions will generally fail to track well-being in modern human environments, even if we can identify some cases in which this is true. Third, acknowledging underlying dysfunction can help reduce the stigma of mental illness by disabusing people of the belief that mental health illnesses are less significant, less real, or less unchosen than their physiological counterparts. We believe that the divergence between dysfunction and well-being in mental health is better accommodated by prioritising certain non-disease states in healthcare allocation, rather than by eliminating the disease concept altogether in that domain.
Wakefield and Conrad argue that there is less daylight between our proposal and the traditional harmful dysfunction view than we suggest. In particular, they maintain we have erroneously interpreted the ‘social value’ component of the harmful dysfunction model as being indexed to prevailing moral attitudes, whereas that term was intended to capture cultural standards of value that exist irrespective of sociological patterns of evaluation. Because these cultural standards are subject to critical interpretation, revision and reasonable disagreement over how to resolve value conflicts, Wakefield and Conrad clarify, claims about the social value of dysfunction are subject to ‘objective’ analysis within particular cultural systems. We think even this thicker normative glossing on the moral value component is not sufficiently normative and hence that there is a bigger conceptual gap between our views than the commentators suppose.
First, the harmful dysfunction model, even on this clarified interpretation, remains an explicit attempt to describe patterns of diagnostic practice—not a normative account of how the disease concept ought to be specified given the moral goals of healthcare institutions. That said, their clarified account appears to slide back and forth between descriptive and normative desiderata, and this equivocation is likely responsible for some of the interpretative confusion. If it is intended to be descriptive, then the harmful dysfunction model is engaged in a completely different project from our own. On the other hand, if it is intended to be fundamentally normative and not simply a conceptual analysis of diagnostic practices, then it appears committed to a version of cultural relativism (notwithstanding the authors’ express disavowal of any metaethical commitments), although one that provides room for critical scrutiny within cultural traditions.
We will bypass the well-trodden problematic territory of cultural relativism, such as how cultural traditions are delineated, how prevailing standards are determined, and how on such views we can make sense of human rights and moral progress. Instead, we shall underscore the fact that, contrary to the authors’ claims, the harmful dysfunction model is incapable of handling pernicious classifications of disease (such as homosexuality) when immoral moral standards come to predominate in a putative cultural tradition. If one wants to claim (ad hoc) that all cultural traditions would reject certain pernicious moral norms if their moral standards were appropriately scrutinised from some idealised standpoint within their respective cultural domains, then why not go the extra mile and accept the universalisability of certain evaluations? This is not to say that the factors that figure into individual flourishing are not culturally contextual, nor does the ethical pluralism that might result from such cultural context-sensitivity entail ethical relativism. Neither, however, does it preclude universalisable moral norms.2 Indeed, Wakefield and Conrad’s assertion that only intrinsic harms should be considered in assessments of the ‘harmful’ component (more on this below), and their understandable worries about moral imperialism, are themselves premised on thick normative commitments that apply across cultures—a logical implication which relativists famously deny at the pain of self-contradiction.3
Second, Wakefield and Conrad contend that by limiting the value analysis to intrinsic harm, the harmful dysfunction model can elide the pernicious norm problem mentioned above. Setting aside the fact that the motivation for this limitation looks like a globally normative one (and not a mere model of diagnostic practices or local moral evaluation), we see two problems with it. First, few neuropsychological disorders are intrinsically harmful (like migraines), or harmful when abstracted away from their social context. If even uncontroversial mental disorders are only disvaluable in a given social context—such as how they affect participation in social enterprises, access to cooperative goods, or the ability to engage in personal relationships—then there will be few mental disorders to speak of because few will meet the intrinsic harm criterion. Second, some mental disorders (such as psychopathy) may not be intrinsically (or subjectively) harmful and yet be properly disvalued because they pose a threat of harm to others and valuable institutions. Some cultural traditions may view homosexuality as precisely such a threat. Of course, one can respond that in such cases the threat perception is incorrect, irrational, or outweighed by interests of justice or autonomy. But these would be global, not culturally localised, assessments which rely on conceptual resources that our thickly normative account provides and the harmful dysfunction model lacks.
This response merely scratches the surface of a complex metaethical and moral epistemological discussion about the kinds of reasons and diversity of perspectives that should be given weight in bioethical policy and deliberation. We are pleased that our paper has generated such challenging and constructive commentaries and we look forward to continuing the discussion.
Correction notice This article has been amended since it was first published online. The authors' affiliation was published with a typo. This has now been corrected.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; internally peer reviewed.
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.