Involuntary detention and treatment is one of the most controversial aspects of psychiatric care and has been subject to endless legislation, campaigns, criticism and ethical debate across the globe. The current situation in many countries is that persons who are suspected or known to have a severe mental disorder can be detained and treated against their will under national or regional legislation. Differences exist between the letter of the law, its interpretation, developments of the law through case law, regional custom and practice, and codes of practice where they exist. This can cause varying local practice within similar legislative frameworks.^{Reference Steinert and Lepping1} Ethically, most healthcare legislation tries to find a compromise between the protection of a patient’s autonomy and the protection of the patient and others from harm. Autonomy, as part of a rights-based legislative framework, has increasingly dominated our discussions about coercion. However, there are some serious problems with this approach.

1. (a) Even proponents of autonomy such as the British-Latvian philosopher Isaiah Berlin have emphasised that autonomy can never exist in a vacuum, but always relies on relationships. Thus any autonomous choice will have the potential to interfere with someone else’s autonomous choice. Berlin separates two types of liberty - the freedom from something (negative liberty) and the freedom to do something (positive liberty) - to which, he argued, often only the elite of any given society has access.^{Reference Berlin2}

2. (b) All serious ethical frameworks consider a number of competing ethical principles, of which autonomy is only one. Most famously, Beauchamp & Childress postulated beneficence (‘do good’), non-maleficence (‘do no harm’), autonomy and justice.^{Reference Beauchamp and Childress3} These principles were always meant to be taken into account together when solving ethical problems. No preference has been given to any particular one. Indeed, it does not make sense to give a priori preference to any ethical principle. They are not rules, and we all know that there will be times when one is more important than another in any individual situation.

3. (c) The ethics of care movement has criticised rights-based ethics as predominantly male, and limiting ethics to detached abstract rights rather than empathic contextualised care. Bloch & Green point out that ‘care ethics views utilitarian impartiality and the deontological stress on respect for autonomy as derivatives of perceptions that misrepresent actual relationships between people who are wrestling with day-to-day moral questions’.^{Reference Bloch and Green4} In contrast, care ethics puts empathy and relationships at the heart of decision-making. Not to help a patient in need is viewed akin to neglect.

4. (d) Carson & Lepping have argued that psychiatry is about helping people with mental illness. By looking at the patient’s, carers’ and medical narratives and treating them as parallel truths without a priori judgement, the ethic of helping people comes first. This ethic is the foundation of psychiatric practice. ‘Although this is a banal truism, it is often overlooked in everyday practice. This is because practising psychiatrists often see their everyday practice as “problem solving”. However, solving problems is not the same thing as helping people’.^{Reference Carson and Lepping5} Problem-solving seems to place the issue firmly within the individual concerned, whereas helping people is about the social context in which the issue occurs. This emphasises the patient’s role in the wider context of social relationships, including treating teams.

5. (e) Patients feel coerced to various degrees regardless of their formal legal status in hospital or the community. In a UK sample the lifetime experience of some form of coercive leverage was reported by 35% of voluntary psychiatric patients.^{Reference Burns, Yeeles, Molodynski, Nightingale, Vazquez-Montes and Sheehan6} Equally, a smaller but sizeable minority of involuntary patients do not feel coerced. Studies have shown that patients who regain insight after mental illness do not, by and large, object to the treating teams having taken benevolent decisions on their behalf.^{Reference Owen, David, Hayward, Richardson, Szmukler and Hotopf7} This shows that the boundaries between voluntary and involuntary are much more complex than an exclusive focus on autonomy might suggest. Autonomy, as asserted by Berlin, always needs to be contextualised and cannot exist in a vacuum without considering relationships.

We suggest that recent developments in psychiatry have favoured the principle of autonomy disproportionately over other ethical principles. By putting most of our ethical eggs in the autonomy basket we have neglected other important principles. This has led to clinical situations in which basic needs of patients such as safety, food and social contacts can get neglected for the higher good of autonomy. In most high-income countries, current, rights-dominated legislation is based on capacity. If patients have capacity to make decisions, their autonomous decision-making trumps all other considerations. However, there is reasonable evidence to suggest that clinicians overestimate patients’ capacity.^{Reference Lepping8} Furthermore, capacity as a concept lacks utility in many acute mental illnesses, with fluctuating capacity and differences of judgement with regard to illness severity and need for treatment. In contrast, it may be a more valid concept in situations of chronic and more cognitive impairment such as dementia or severe intellectual disabilities.