Systemic change typology – some doubts

In my last post I shared my thinking around a possible typology of systemic change that could help us come to grips with the different concepts and ideas that are connected to systemic change. I have received some feedback on the post and I have done some more thinking that I would like to share in this follow up post.

Concretely, I want to share my thoughts about the usefulness of systemic change as a concept on a more fundamental level, based on the obvious fact that systems change anyway, also without purposeful interventions.

Aren’t systems changing anyway?

I was prompted by one email I received as a reaction to my last post to again question the need for this concept of ‘systemic change’. The feedback included the phrase “systemic change is an omnipresent process.” So, I asked myself again, what relevance does it have for development activities? If it is always there, it cannot be a measurable outcome of an intervention. This reminded me of what Shawn and I wrote back in 2016 in our paper for the BEAM Exchange on ‘rethinking systemic change’ in the context of market systems development [1:3, emphasis in the original]:

… the concept of systemic change needs to be rethought. But first and foremost, it is not systemic change per se that market systems approaches are looking for. Systems continuously change also without external interventions. Rather than seeking to ‘make’ change happen, the aim of market systems approaches ought to be to enhance the evolutionary process in an economy and create access for all levels of the society to contribute to and shape this process.

So we need to ask a different to whether systemic change has happened or not as a consequence of a project intervention. It is whether the project intervention has been able to enhance or influence the evolutionary process in a desirable way. Or to put it slightly differently: systemic change is not about achieving change per se, but about influencing the change that is occurring anyway so it is shifting in a more favourable direction.

And to go a step further, we also made the point in the paper the it is not about development agents to change the evolutionary direction of a system in a country we work in. Rather, we should enable and support the people who live in these countries to make their own choices with regards to the change process and engage in such a process. We can work with and on the institutional landscape so such a process can take place in an inclusive way.

Can we change systems?

A more fundamental question is whether this is actually possible. Margaret Wheatley makes a – in this regard provocative – claim by saying: “emergent systems can’t be changed”. In her most recent book, she writes [2:226]:

Life doesn’t change the way we want it to. Our costly attempts at organizational and societal change have mostly failed because we made two mistakes. We tried to change individual behaviours, and we used linear approaches of goal setting, measurement and accountability. Logically it all made good sense. Individual behaviours caused problems. Complex problems needed to be broken into chunks and then designated as tasks to specific individuals or teams. If everyone knew they would be held accountable for results, the’d be motivated to do the work. And change would happen.

Obviously, there is more than a bit cynicism in this quote. Wheatley goes on [2:227-228]:

Culture in any form, at any scale, is an emergent phenomenon. Once formed, you can’t change it using reductionism. […] We spend so much time as individuals and leaders trying to undo things in order to fix them and create better alternatives. we were well schooled in reductionist thinking, have done it for years, so by now we’re experts. We change the players, we focus on specific behaviours, we create new incentive systems for the same people. All for naught. The hard-to-accept news about emergence is that once a culture or pattern of response has emerged, you can’t work backwards. There is nothing to do but start over.

Wheatley advises that one needs to make use emergence itself to turn the system into something different, something new, fundamentally restructuring the current system.

But wait, this is still change, no?

What Wheatley means when she says “emergent systems can’t be changed” is that systems cannot be changed incrementally. Rather, they need to be transformed into something new, something different. Looking at the typology from last week, this would represent the second type of systemic change – the transformation of the dominant regime.

I’m not sure I’m buying Wheatley’s argument that there is no incremental change in complex dynamic systems. Indeed, evolution itself is an incremental process, improving on what is already there. Hence, from a development perspective, the first type of systemic change, the one without a transformation of the regime, might still be valuable. In that sense, systemic change is not about changing a system, but rather about a change that plays out on a system level – i.e. a change that is felt beyond a few individual that were directly touched by an intervention. A change that incrementally improves the lives of the people targeted by development efforts.

But back comes my doubt: is incremental systemic change enough to change the fate of the large number of poor people. And beyond poverty, is incremental change enough to allow our species to survive on this planet given the social and environmental challenges we are currently facing? Not according to the field of resilience thinking. But more about this in the next post.


[1] CUNNINGHAM, S. & JENAL, M. 2016. Rethinking systemic change: economic evolution and institution. Discussion paper. The BEAM Exchange.

[2] WHEATLEY, M.J. 2017. Who Do We Choose To Be?: Facing Reality, Claiming Leadership, Restoring Sanity. Berrett-Koehler Publishers: Oakland, CA.

3 thoughts on “Systemic change typology – some doubts

  1. Bhav

    I guess there is a difference between making and measuring incremental change.

    If I am making incremental changes, it means I know how the system works, where I am now, where I want to be, and what I need to do to get there.

    If I am measuring incremental change, it means I am influencing the system to see if it changes in the hoped for direction, and then I can measure whether that is happening or not.

    So I wonder if Meg is saying you can’t make changes in a clear causal sense. However you can still check whether incremental change is happening.

    Does that make any sense?

    1. Marcus Post author

      I’m not sure if you can ‘make’ incremental changes. All changes in complex systems are essentially trial and error that require you to measure well to see what happens as a result of your intervention(s). What you describe as ‘make changes’ sounds more like working in a complicated system.


Leave a Reply