Accountability in an Uncertain World

By Greg Fisher

What does accountability look like in a world that is uncertain and in which innovation is not only prevalent but also essential?  In this blog I would like to argue that how we currently “do” accountability stifles innovation.  This is an important issue given how rapidly the world is now changing with the on-going diffusion of information and communications technology.  I conclude by calling for a serious re-think about what accountability looks like in organisational life, notably for those organisations operating in highly dynamic environments.
The genesis of this blog came while reading an excellent draft paper written by NESTA and MindLab, which looked at “Decision making in complexity and uncertainty”.  The draft was provocative and focused on resilience through innovation in public policy.  However, much of the content was about human decision-making processes in general, so it was relevant in all areas of our lives, including in the private sector.

I should emphasise up front (an obvious point) that accountability and innovation are both valuable.  If, for instance, someone in the public sector were being paid a wage by the taxpayer to do some job, it is important they are held accountable for what they do.  I doubt many would disagree.  In addition, it is eminently sensible that we, as a species, innovate.  We have been doing this for a very, very long time and it would not be unreasonable to say that innovation is substantially responsible for the standards of living we enjoy in the West today.  But have we organised our governance and management systems such that there is a conflict between accountability and innovation, both of which we value?  More specifically, does our current approach to accountability stifle innovation?  I suspect this is generally the case, both in the public and private sector.  Let me articulate what I mean.

In a nutshell, I believe that how we currently practice accountability in organisations is based on linear thinking in which the future is treated as predictable; but in reality the world is constantly evolving in unpredictable ways i.e. it is inherently uncertain.  These two points together are, I believe, the source of tension between accountability and innovation.

I should emphasise, however, that this tension should be viewed as a sub-set of a broader tension between two different “teleological” frameworks, or “mental models”, of how we understand and frame human systems.  For example, do we think of organisations as machines, as organisms, or something else?  In the language of Ralph Stacey, we need to make a distinction between formative teleology and transformative teleology.  Formative teleology is where human systems are framed as if they were like predictable machines in which people are automatons, and where the future unfolds in a predictable way.  Transformative teleology is the recognition that human systems are changing continuously: organisations are in a constant state of flux because of the adaptive nature of human beings.  (For those more technically minded, there is an equivalent debate concerning the difference between ergodic and non-ergodic systems.)  Stacey has emphasised that orthodox management science is orientated around prediction and control (within a formative teleological framework) whereas reality is transformative.  Our methods for holding people to account are based on the formative teleology i.e. it is designed as if the future were controllable and predictable and as if human behaviour and relationships were constant over time.
Our conventional approach to accountability typically requires the setting of targets to achieve some aim or set of aims.  When set, these targets will be indicative of success in achieving the aim(s): they form a benchmark for performance evaluation.  Ideally the targets will be quantified so that success is measured objectively.  The distinction between aims and targets is important because if the environment in which the person being held to account changes, the original targets may be decoupled from the original aims i.e. they become inappropriate.  In addition, an environmental context may change so much that even the original aims might to be called in to question.  But in orthodox management science, the inherent changing nature of the world tends not to be recognised.  Or, at the very least, it is under-emphasised, which means the re-consideration of targets may not be institutionalised in organisational practices.

In a world that is in a constant (and unpredictable) state of flux, there is a fundamental problem with this approach to accountability.  The crux of the problem is this: targets set up front can only ever be about “knowns” and “unknown knowns”.  It is impossible to set targets about “unknown unknowns”.  That would require us to anticipate the inherently unpredictable, which is a paradox.  Moreover, if the environment in which someone is working changes, such that the original targets become inappropriate, time might be spent pursuing those targets regardless, leaving less time for any innovation that might be required to adapt to the new environment.  What this tends to mean is that the prescriptive approach to accountability described above will leave too little time and attention for innovation.

As an aside, it is worth highlighting here the nature of co-evolution in innovation, which is to say that one person’s innovation is another person’s environmental change.  So, for example, when an employee of a company notices that the “environmental context” in which they are working has changed, they might need to innovate to adapt as best they can (such as a small change in their working practices or a change in a supply chain).  Their innovation will change the environmental context of the people with whom they are interacting, who in turn may have to adapt, and so on.  This process of continuous of iterative, mutual adaptation is called co-evolution in Complexity theory.  Of course, innovation is not only triggered by a need to adapt.  It can happen spontaneously simply by someone reflecting on some problem, or simply by two people interacting.

Now, back to accountability.  I have over-generalised in my comments above about accountability and innovation in at least two ways, to help emphasise my point.  First, in practice people can and do recognise that the world evolves and they will probably allow time for innovation and for targets to change too.  Often people will do this through basic common sense.  But it will be against the framing of what Stacey calls the “dominant discourse” in management science.  In that framing, any change ought to have been anticipated ahead of time, and built in to the articulation of aims and targets.

Second, the rate and nature of change will vary between different organisations and industries, and the value of the approach to accountability described above will be contingent on this point.  Compare for example a mature market such as milk production and the hi-tech industry of Silicon Valley.  Rapid-change environments can make the type of accountability described above both tricky and even harmful if it leads to the pursuit of inappropriate targets.  Such environments require near-continuous innovative thinking, and this can render future target setting rather pointless.  But in mature industries, where production, distribution, and consumption patterns are expected to be stable, the type of accountability measures described above might be reasonable at both the organisation and employee level (though I am not saying these industries would never need to innovate).  The point I am making here is that most organisations operate somewhere between these two extremes; and the more dynamic the environment, the more the traditional approach to accountability will stifle innovation.

There was a very good example of some of the points raised in this blog outlined in the NESTA and MindLab draft paper regarding the experience with “Public-Private Initiative” (PPI) in Denmark.  After some difficulties with PPI, there was an understandable demand by the public for accountability measures to be taken, including “authoritative guidelines” and “a blueprint for guiding practice” for PPI.  This can be interpreted as a desire by the general public for those involved with PPI to nail down certainty about the challenges faced.  The problem with this is that PPI was (i) operating in multiple, idiosyncratic domains; (ii) it involved the interaction of the public and private sector (which is likely to be a source of creativity itself); and (iii) it is a relatively new initiative (relative to mature industries that is).  All of these points mean that guidelines and accountability measures would be difficult to draw up because the environment is one of change, idiosyncrasies, innovation, and uncertainty.

Curiously, this subject relates to something that Stuart Kauffman, one of the leading figures in Complexity theory, has emphasised: the need to reduce our emphasis on quantification and the equivalent need to raise our emphasis on the quantitative aspects of life.

To conclude, innovation is necessary in dynamic environments but accountability is important too.  So what should accountability look like in environments in which innovation is essential?  For what it’s worth, I think there are at least two important and complementary ways of starting to address this question.  One would be to look at the most successful innovative and creative organisations to see how they do it, to see if we can learn any general principles.  Most (if not all) successful organisations operating in highly dynamic environments will probably have scrapped the approach to accountability described above long ago (else they wouldn’t be successful).  Second, these empirical experiences should be complemented with a conceptual approach that encompasses transformative teleology and the inherent nature and importance of innovation in social systems.


Follow us on Twitter – @synthesisips

6 Responses to “Accountability in an Uncertain World”

  1. Hi Greg

    As someone who works a lot with the idea of ‘change happening all the time’, I have come to a conclusion (preliminary, of course!) that the difficulty with target-setting is that the target is only useful within a small range of futures. Targets often become an impediment where things change and they get too hard (in which case everyone loses enthusiasm and it’s all hopeless), or too easy (things go better than planned and people then throttle back or focus their energies elsewhere).

    One type of solution to this is to establish a direction, not a target. (A target always implies a direction, but the direction is vector only, without a scalar quantity). Then progress in the direction is built and monitored, without the did-we didn’t-we aspect of meeting a target. Another option is embrace the ‘agile’ mindset and have much shorter-term targets – for a matter of days or weeks, rather than months or years. Then progress is monitored more continuously and matters adjusted as the future unfolds.

    • gregfisher says:

      Nicely written, Mark, and a constructive addition to the blog. I would be curious to know how prevalent the accountability approach I described is, in your experience. And the same goes for Arthur…

  2. Arthur Battram says:

    Hi Mark,

    And it was exactly thinking like that that led me to develop my ‘Stepping stones in focus’ model, featuring a river, fog and stepping stones and Kaufman’s ‘adjacent possibles’.

  3. Eileen Conn says:

    Greg thanks for an excellent analysis of the prevailing target driven cultures. In my field which is community dynamics, which I call the horizontal peer system in interaction with that target driven world, the normal notions of management are very inappropriately applied. We need different forms of structures and processes for good governance in the horizontal peer. What you, Mark & Arthur say fits very well in that arena.

    You know of my paper on this – for other readers of yr blog see these two links:;

    And hello to both Mark and Arthur from different stages in my life to this point! Glad to see digital media help similar paths to connect again.

  4. Ed Mayo says:

    Notice the axis of the linear model of accountability. Public sector models tends always to vertical accountability – even when public policy bends over backwards to stress the opposite (the agency models, localism, public service mutuals, free schools – the weight of the vertical over the horizontal wins out, whether in subtle forms of reporting or a good old fashioned lurch back to centralism). That is one reason why the co-operative model, which enfranchises stakeholders through a democratic member model, is a better bet if you want self governance because it invests in the capability of those involved to resist the default pull of central or municipal government – or in a market context, big business and the allure of speculative investment.

    The wider move away from any linear model, vertical or horizontal, is important in what you write here – reminiscent perhaps of the critique of audit over recent years. Simon Zadek has done a lot of excellent work over time around collaborative governance and accountability – see

  5. Arthur Battram says:

    I like the thrust of Ed’s point about vertical, linear accountability. I have suggested that accountability should be a judgment of outcomes. This harsh-sounding perspective comes from natural selection in the biological sciences: it is not intended as a solution, more a ‘provocation’ for rethinking accountability.

    The obvious counter-argument is the ‘what about Chernobyl?’ type. In practice, few problems are of the high-risk huge engineering investment variety (and note that both the Chernobyl disaster and the Mir incident [remote piloted fuel rocket crashed through solar panels taking out 60% power] are examples of a peculiar Russian, post-Glasnost human risk-taking, not an inherent flaw in engineering. Chernobyl happened because dozens of safety circuits had to be conciously disabled for an experiment. The consequences were exacerbated by human group think- a kind of anti-dialogue.

    What they both tell us is that the Guardian role of government is crucial, and should be kept ‘clean’ untainted by the Trader role.

    In the context of the public sector there are many more contexts where small, fast, learning-focussed, stakeholder-engaged experiments could be carried out for little investment and low risk (MLTQ – many little things quickly, as I call it).

    My case-story ‘Self-Organising For Success'(SO4S) is a case in point. Of course this was easier to do in the days before contracted-out services, but as SO4S demonstrates, it is in everyone’s interest to think differently. As ever the managers – the most committed to a control and order perspective, the most trapped in targets and tick box assessments – are the most resistant.

    Humans are uniquely able to carry out gedanken without betting the farm. A mandatory requirement for a group process of judgement for accountability could be prescribed in advance of a project’s implementation, without specifying its particular form.

    I throw this out as a provocation to explore rather than as a notion to be disagreed with: what if…

Leave a Reply