Graham Thornicroft and George Szmukler elegantly demonstrate, through the use of a single table, that the Draft Mental Health Bill 2004 is not informed by the principles enunciated in either national or international mental health policies. They then proceed to argue that primacy has been given to notions of community safety, potentially subverting traditional concerns in mental health legislation with such matters as ‘least restrictive alternative’ and the health benefits to the patient. They raise the spectre of mental health professionals being suborned as agents of social control. I am in total agreement with the authors that risk to others should not become the central issue, let alone the only issue, in formulating mental health legislation, but would suggest it is nevertheless a legitimate concern.

The UK, like my own home Australia, is blessed with a populist government whose principles are forged to no small degree by pollsters, pundits and the media, reinforced by a touch of old time religion. This creates a polity in which fear rules. The proposed Mental Health Bill may be unprincipled but is all of a piece with much of government policy.

Without wishing to defend the Draft Bill against the attack launched by Thornicroft and Szmukler, it might be worth asking to what extent the proposed legislation could be a response to the perceived failings of current mental health services. Relationships have soured between psychiatry and the Government in England and Wales. The homicide inquiries and the culture of blame they instantiate have bred understandable resentment, and mutual trust has not been improved by the Government’s wavering commitment to community care, which contrasts so dramatically with their indecent enthusiasm for Multi-Agency Public Protection Arrangements (MAPPA) and for the Dangerous People with Severe Personality Disorder Initiative (DSPD). However, the mental health professions have to accept their share of the responsibility for the deteriorating relationship, not least in their responses to government initiatives concerning community safety.

Thornicroft and Szmukler label the concerns of many ordinary people about the risks to their safety presented by the mentally disordered as the result of ‘stigmatising stereotypes that associate mental illness and violence’. Sadly, however, attached as we may be to this notion, the burden of current research points to mental disorder being associated with an increased propensity to violence. Schizophrenia, which accounts for a large part of the public mental health service’s case-load, and for most compulsory admissions, is associated with criminal behaviour, and violent criminality in particular (Reference Brennan, Mednick and HodginsBrennan et al, 2000; Reference Walsh, Buchanan and FahyWalsh et al, 2001; Reference Soyka, Morhart-Klute and SchoechSoyka et al, 2004; Reference Wallace, Mullen and BurgessWallace et al, 2004). Those with schizophrenia constitute less than 1% of the population but account for some 5-10% of homicides, which is far from a trivial contribution (Reference Eronen, Tiihonen and HakolaEronen et al, 1996; Reference Wallace, Mullen and BurgessWallace et al, 1998; Reference Erb, Hodgins and FreeseErb et al, 2001; Reference Schanda, Knecht and SchreinzerSchanda et al, 2004). A reluctance by many mental health professionals to face up to the implications of the association between serious violence and serious mental illness contributes to a situation where many ordinary people, and their political representatives, no longer trust our profession to discharge its responsibilities. These constitute responsibilities not only to the community but to our patients.

When someone with schizophrenia attacks, or behaves in a frightening manner, the victim is usually a carer, be that a relative or a professional (Reference Nordstrom and KullgrenNordström & Kullgren, 2003; Reference Joyal, Putkonen and PaavolaJoyal et al, 2004). The actual or threatened violence often alienates those on whom the patient depends, sometimes putting at risk the very support that sustains them in the community. Violence by the seriously mentally ill is a double tragedy, for the victim and the perpetrator. Our responsibilities to our patients demand we address their propensities to violence. In the schizophrenias this propensity is not simply a product of their symptoms but also reflects the social conditions imposed by disability, the disproportionate levels of substance misuse, the personality changes associated with the disorder, and probably common risk factors from childhood and adolescence (Reference TaylorTaylor, 1998; Reference Arsenault, Moffitt and CaspiArsenault et al, 2000; Reference Hodgins and JansonHodgins & Janson, 2002). Psychiatry should embrace approaches to assessment and treatment that recognise the risks of violence to others and the need for a transdisciplinary approach to address the symptoms, criminogenic factors and substance misuse driving that behaviour. It is not good enough to hide our collective heads in the sand and refuse to recognise the reality of the distress and understandable fear engendered by violent propensities, which are both overrepresented among our patients and potentially preventable.

Despite the drafters’ best intentions, the proposed Mental Health Act is unlikely to advance the agenda of community safety. Our response as a profession should, however, not be a rejection of the understandable, and politically unavoidable, drive for better management of the risks of violence to others. The response should be to use the evidence base concerning the association between mental disorders and violence to inform both public policy and mental health practice. A greater willingness on the part of our profession to accept and respond to the general public’s fears of violent behaviour in the mentally ill may produce a political climate more receptive to our advocacy for better services and a greater sensitivity to the rights of our patients. It is particularly important that it is made clear that the best way of reducing criminal behaviour among the seriously mentally ill is neither institutionalisation nor compulsory community treatment orders, but improved services that focus on symptom control, support, appropriate accommodation, behavioural and cognitive therapies and active social and economic rehabilitation (Reference Hodgins and Müller-IsbernerHodgins & Müller-Isberner, 2004; Reference Wallace, Mullen and BurgessWallace et al, 2004).