Is there such a thing as best practice? The Government Accountability Office in the US thinks so. So does the UK's National Audit Office, which frequently recommends it to government in fields such as contract management. And the Australian federal government has established best practices for functions such as defining new regulations.

The idea of best practice is alluring. It suggests that there are ways of working that can be identified, codified, transferred and embedded - and that they are not just good but the best. It is part of a view that the world can be comprehensively ordered and measured, and that science can be applied to management.

This view and the way the idea of best practice has been used lie behind a series of disasters.

A first problem: what do we really know?

Even in the hard sciences, the frontiers of knowledge shift as we find new ways of modelling the world. A proposition might turn out to be wrong, e.g. that the sun and the planets revolve around the earth — but only in certain circumstances. Newtonian physics on the whole works well in everyday life, but quantum mechanics and relativity work better under a wider set of conditions.

A scientific proposition, therefore, represents our best current understanding, rather than an eternal truth. This might seem more relevant to physics than management, but science is facing a crisis of replication: the results of studies cannot be reproduced. One investigation found that nearly 90% of peer-reviewed findings in the top journals failed to stand up to scrutiny.

The idea of best practice is alluring. It suggests that there are ways of working that can be identified, codified, transferred and embedded - and that they are not just good but the best.

The problem is particularly serious in studies of human behaviour and, by extension, management studies. No doubt this is partly caused by academics' incentives to churn out papers to meet publication targets, but a more fundamental reason is that many of the questions we're interested in relate to a complex reality. This means that many things are inherently unpredictable, and also that a proposition established in one context might not apply to another.

A second problem: are we free to speak our minds?

Best practice is of a piece with an environment in which it's hard to tell the boss the truth. In a culture where best practice has already been established, and the staff's role is simply to implement it, there is little scope for discussion.  The exhaustive Chilcot inquiry describes how there was insufficient challenge to “mitigate any tendency towards groupthink” in the UK's policy on Iraq. Staying with the Middle East - in the approach to the Yom Kippur war in 1973, the head of Israeli military intelligence and his Egypt specialist disregarded evidence that Egypt and Syria were about to attack. The philosopher Quassim Cassam uses this failure to illustrate closed-mindedness and an unwillingness to be challenged by subordinates.

In the UK more recently a “good news culture” in the Department of Work and Pensions under Iain Duncan-Smith contributed to problems with the development of Universal Credit. The House of Commons' Public Accounts Committee was critical of the department's “flawed culture of reporting good news and denying that problems had emerged.”

Similar issues can be seen in aircraft crashes, which have been linked to the flight crew's failure to challenge a pilot's authority. If best practice has already been defined, the flight crew can feel inhibited from reacting to unexpected situations.

However, it's clear we can be more certain in some circumstances than others. Compare a patient having heart surgery with a patient making lifestyle changes to reduce the risk of heart disease. A heart operation can be conducted anywhere in the world using the same procedures with a similar chance of success. These procedures could be called best practice. By contrast, while a doctor might provide lifestyle advice - eat healthily, stop smoking, exercise more - the patient's response to this advice is unpredictable, and the patient and doctor will interact in unpredictable ways.

But it would be a mistake to conclude that we can draw a clear bright line between certain and uncertain situations. As with the flight crew, one of the influences on the chances of success in a heart operation is how well the clinical staff are working as a team. Human interactions, with all their unpredictability, form part of even the most technical of projects. The misuse of best practice, and other scientific approaches to management, downplay these interactions.

The three varieties of knowledge

What then should our approach be? The philosopher Donald Davidson argued that there are three varieties of knowledge: knowledge of oneself (subjective), knowledge of others' minds (intersubjective), and knowledge of the world (objective). These three varieties depend upon each other, and we need to triangulate between them.

So far, so good — but what has this got to do with public sector management? For me, the three varieties highlight the importance of each individual's perspective, of developing a shared perspective and language with others, and of establishing as much as possible objectively. The tendency in management is to exaggerate the scope for objective knowledge - because it offers a comforting certainty, while disregarding the others, even though each of the three varieties deserves attention.

Individuals need the chance to be heard. Everyone's perspective is unique, and this can be forgotten when people are put into categories or managed through abstractions.

The limits to our knowledge of others' minds mean that -  as Davidson put it - “communication is always incomplete.” Although we can be reasonably confident that we have shared understandings of everyday objects, the same cannot be said of things like “change management.”

If best practices are proposed, continuous work is needed to develop a shared understanding of those practices. This work cannot be assumed away.

Finally, while there is no situation in which certainty can be achieved, we can have more objective knowledge in some domains than others. Practices cannot be reliably established as “best” regardless of context, but they can be informed by objective knowledge. Organisational life is a continuous process, and we can see triangulation as a metaphor which highlights the need to constantly iterate between our own perspectives, others' perspectives, and what we can reliably establish about the world.

How should government deal with uncertainty?

Triangulation does not offer the comforting certainties of best practice. Rather it recognises that uncertainty needs to be struggled with.

This approach is aligned with the idea of enablement. It points towards government's need to devolve power, so that citizens can respond to their own environment, rather than applying someone else's idea of best practice. We know that when this happens, people engage with the problems at hand and come up with their own, more effective approaches. Creating space for this to happen means embracing humility, particularly at the top of organisations, and making it safe to be uncertain.

As the American pragmatist philosopher C.S. Peirce wrote:

“Enquiry is not standing upon the bedrock of fact. It is walking upon a bog, and can only say, this ground seems to hold for the present. Here I will stay until it begins to give way.”

A longer version of this article with references can be found here.