The past eight months have been revolutionary, and the scale of change has taken most of us by surprise. A few months ago it was all quiet on the policy front.

Now I am lost in a crescendo of political and social change, unclear about priorities and connections - but all too clear about some of the internal inconsistencies. And there may be further announcements any time, any place.

Changes at every level have affected many areas of society. In the NHS we have health action zones, health improvement programmes, primary care groups, the National Institute for Clinical Excellence, the Commission for Health Improvement, national service frameworks, clinical governance and professional self-regulation.

Expect a human resources strategy, a social services white paper and a mental health strategy by the summer, a mental health service framework by the autumn, together with a paper on flexibility across health and social care and a review of the Mental Health Act.

How many of us also have a good grasp of all the changes in local government, such as best value, or in education and transport? And I have omitted public health.

Many people cannot keep up with the announcements and all their implications. One perceives a reinforcement of rapid action, as if change is always better than the failed status quo.

So inspiring local leaders invent their own solutions based on subjective interpretations of policy documents, and forge ahead with radical change. A superficial reading of The New NHS white paper suggests single mental health trusts, and several major reconfigurations have already gone ahead, creating large single providers.

This may be good news if the idea behind it is to centralise care around powerful monopoly trusts led by the experts and directly accountable to health authorities or regional offices. But it's bad news if it is intended that PCGs should shape care locally, or if eventually we would like primary care trusts to include mental health. It does not seem to fit with the integrated care model.

An alternative tactic is to play safe, keeping your head down. Why take risks if the quality police are everywhere, out to get you? I doubt whether a quality strategy as formulated in A First Class Service encourages creativity. One wrong move and CHI moves in.

This seems to be the fundamental tension in government policy. On the one hand is a genuine wish for local initiative, with incentives to create partnerships accountable to local users and carers. And simultaneously, there is a deep distrust of decentralisation and an absolute intolerance of failure. Therefore, we see the imposition of tight national quality standards and control mechanisms.

But we know that intermittent failure is characteristic of both command and control type systems and locally led entrepreneurial ones, though for different reasons.

The challenge is to produce an accountability framework within which all stakeholders feel valued, rather than constantly on the edge of prosecution. Nowhere is this more sensitive than in mental health, with its history of scandals, inquiries and negative practice.

This is a crucial moment, and I foresee two scenarios.

The first option is an NHS management framework built around punishment, on the apparently justifiable premise that bad practice has to be smoked out.

National service frameworks will set a rigid example of care with precise standards to be monitored constantly. The consequence of an approach based on tough standards and constant inspection is that clinicians will become demoralised, resenting the scrutiny of all aspects of their practice locally by audit and clinical governance, and nationally by CHI.

There will continue to be tragedies, and a further drift towards inspection, defensive practice, heavy record-keeping and low risk-taking. Mental health could become a no-go area for doctors, nurses and managers. Anyone accusing me of exaggeration should study the latest vacancy figures.

But psychology and MBAs teach us that rewards are effective and punishment fails, and many therapies are derived from these principles. Our inability to generalise knowledge across policy areas is amazing. There is one law for our patients and another for us, it seems.

The second scenario is based on incentives and rewards. The mental health policy paper could promise resources to produce a balanced range of services, with an emphasis on the development of good local human resources strategies. The national service framework could inspire planners and clinicians by outlining good practice, with an emphasis on diversity and the scope for local invention. NICE could inspire change by offering information about effective services here and internationally, and best practice could be rewarded.

In return, audit and more rigorous self-regulation will be acceptable to practitioners as good practice, co-ordinated by CHI. We could learn to trust each other in working towards the same objectives.

We are on a knife-edge, and policy could move us in either direction over the next few months. I hope politicians are aware that every development is viewed with trepidation by a confused field, keen to make it work but unclear what is expected. The threat is that there is too much pressure and too little time to deliver, with too strong a sense of punishment as the motivator.

My plea would be for politicians to ease off and give the mental health field a bit of time and lots of faith. It took 50 years to get where we are: it will take more time to arrive where we would like to be. Let's first be very clear about our destination and mode of transport.