If Complexity was a person, she would be a Socialist. Jean Boulton on the politics of systems thinking.

Jean Boulton (physicist, management consultant and social scientist, right) responds to Owen Barder’s Wednesday post on thinking of Jean-Boulton-pic-199x300development as a property of a complex adaptive system.

I’d like to go a bit further than Owen on the implications of complexity for how we understand power and politics. It is generally the case that the powerful get more powerful and the big get bigger. We know this through bitter experience, captured in complexity language by the notion of ‘positive feedback loops’ which equate to the economists’ ‘increasing returns’. In general there is no reason to expect that economies will self-regulate and find a ‘natural’ balance. Even forests, if left to themselves for long enough, reduce in diversity, increase in efficiency and become ‘locked in’ to ecological patterns that are hard to invade and change and can easily collapse (see below, left). Despite the popularity of the phrase ‘complex adaptive systems’, complex systems do not always adapt.

Instead, complexity suggests that  if we want economic development that equalizes power, reduces inequality and incorporates longer-term environmental goals, there is a need for some sort of regulatory processes to counter the seemingly inevitable coalescing of power and wealth in fewer and fewer hands. Otherwise the rise out of poverty is linked more to growth than to development (development meaning a qualitative change in shape and form of the economy rather than a quantitative change – you can obviously have both). And an economy that is growing can in fact take our attention away from underlying structural exacerbations of inequality. Growth cannot go on forever, as land, water and minerals are consumed – not to mention the impact on climate change – but growth can mask just who captures the bulk of resources and can exert control over governments, markets and societies.


This move away from a balanced self-regulating system is different, I would argue, to the behaviour of ant colonies and flocks of birds. Their self-organising processes are specific to the task. They can merely rearrange the deck chairs, so to speak; they can fly around buildings or build ants’ nests that adapt to local conditions, but they don’t in general change their nature and adopt new behaviours. Nothing new emerges, nothing co-evolves.

Sometimes, something more drastic and disruptive is needed. In situations which have very locked-in political and social factors, the focus needs to be on how to break the deadlock, perhaps with high level political interventions and sanctions rather than more gentle adaptation. No use approaching economic development in Palestine with an adaptive mindset. Equally, if the situation is chaotic, like say in South Sudan, then finding ways to build on (any) emerging shoots of political stability is likely to be a first priority.

So, I think the ways to intervene are not necessarily about improving the ‘capacity to adapt’. Sometimes it is also about opposing the powerful who want to reduce diversity. Sometimes it is about slow systemic change, but sometimes it is about seizing opportunities, building on success or on pockets of best practice. And sometimes it is about pushing all the levers in the same direction; or influencing the one key person or focusing on a key underlying issue without which all else well fail; or doing all of these at different times.

I’m also more keen on analysis than Owen. I think you have to start with the historical background (which sets a sort of complex baseline and identifies the strength and form of current social and political and economic patterns), understand the wider contextual features and, indeed, identify ‘missing ingredients’ and ‘binding constraints’ together with opportunities and future possible ‘critical junctures’. Analysis has to have a time dynamic, and be systemic, but I think it is vital. Then, true, you have to move into action and try things out without expecting to prove exactly what is the best strategy or what it will achieve.

We’ve covered power, but, with last week’s elections in the UK fresh in mind, what about the politics of Complexity? At least if she was interested in social and economic justice, Complexity would never stand for election for a party based on a ‘free market’ ideology because, as discussed above, positive feedback loops lead, almost inexorably, to the big getting bigger and the powerful increasing in power. The reduction in numbers of small banks, the constant pushing of legal boundaries, the size of bankers’ bonuses show what can happen in a deregulated market. Power and money give the means to dominate, to win the advertising campaign, to push governments, to squeeze supply chains. There is no such thing as a free market.

Instead, I would argue that Complexity is more of a socialist than anything else (a very Green one though – she

Red, green and complex
Red, green and complex

understands the need to consider long-term consequences to the system of which we are a part). Complexity understands market failure. She does not take the naïve view that self-organizing processes are shaped by some sort of ‘natural law’ and can be trusted to provide the ‘best’ outcome; she understands the importance of governance and ways of upholding the needs of the less powerful, the poor, the longer-term and the environment. This is not to suggest that she would impose a top-down model of governance dreamed up on a plane by consultants and lawyers and plopped fully formed onto a developing country. Rather she sees the need to facilitate the emergence of socially-owned processes of governance and civic empowerment, and to build on those practices that already exist.

There’s more. Complexity is community-minded (would balance freedom with responsibility), keen to work at the appropriate scale, keen not to impose solutions, but to work with enhancing and protecting what is already there. She is passionate about embracing diversity and brave enough to wander well outside any narrow remit to identify blockages, join things up and say the unsayable. She understands that you have to work from the smallest household to the biggest government or corporation – and back again -to enhance the conditions for economic development in a way that leads to equality and sustainability.

Jean’s book, ‘Embracing Complexity: Strategic Perspectives for an Age of Turbulence’ (written together with Peter Allen and Cliff Bowman) will be out this summer, published by Oxford University Press. 


Subscribe to our Newsletter

You can unsubscribe at any time by clicking the link in the footer of our emails. For information about our privacy practices, please see our Privacy Policy.

We use MailChimp as our marketing platform. By subscribing, you acknowledge that your information will be transferred to MailChimp for processing. Learn more about MailChimp's privacy practices here.


12 Responses to “If Complexity was a person, she would be a Socialist. Jean Boulton on the politics of systems thinking.”
  1. Owen Barder

    Jean – thanks for this. I’m looking forward to the book (even though it sounds as if it might be a bit too Gaia for my taste.)

    Though there are parts about which we disagree, I want to highlight something about which I agree with your very strongly: your point that we should not “take the naïve view that self-organizing processes are shaped by some sort of ‘natural law’ and can be trusted to provide the ‘best’ outcome;”. I wish I had emphasised this more in my earlier post (though I did say this in the comments in reply to Gregg Gonsalves). I do not want to give the impression that I think that any and all results of evolution are desirable; nor do I want to give the impression that we must be fatalistic about the outcome towards which we converge. We can and must shape the evolutionary fitness functions which in turn shape our complex adaptive systems.

    This is of course embodied in the second fundamental theorem of welfare economics. There are infinitely many Pareto-optimal outcomes to which an economic system might converge, and they are not all equally good. In welfare economics it is the initial resource endowments which determine the eventual outcome; in the real world we have, and should use, a rather more multidimensional toolkit.

    The good news is that many of these selective pressures already reflect a considerable amount of human agency rather than any kind of “natural law”. Firms do well when they sell products people want to buy at a price they can afford; political parties do well when they offer policies that people want to vote for. These mechanisms are far from perfect, of course; but they are also far from accepting a “natural law” which somehow trumps human values and desires.

  2. Jean Boulton

    Hi Owen, thanks for this.The book is far from ‘Gaia’ if you mean that in the sense that the planet will survive whatever – and I am of course, like you, arguing with the idea that a laissez-faire free market ideology gets the optimal outcomes, which is of course what is implied in this ideology. I’m merely pointing out that not all complex systems are in a self-regulating balanced mode where they can adapt and where they can be helped to adapt – and we need to address different kinds of interventions with that in mind. So I don’t like the phrase complex adaptive systems as not all complex systems can and do adapt – they collapse or get locked-in. I’m not sure, again, about the idea of an evolutionary fitness function which still feels too mechanical to me, even though more systemic. And, as Brian Arthur has pointed out, there is no reason to assume economic systems converge at all.

  3. Ben Ramalingam

    Much as I respect and admire you Jean, I am afraid I have rather fundamental disagreements with this post. I hope you take these comments in the spirit of debate and discussion.

    In general terms, I am not sure the analogy between scientific approaches and political alignments is a particularly helpful or useful one. Complexity science can help us analyse and understand the complex world we live in better, using methods and tools and approaches that are more suited to these realities. What we do with that understanding depends on our political priorities, our perspectives, and the context in which that analysis is undertaken. Politics of course shapes science, but it is quite another thing to claim an entire field as being aligned with particular political positions.

    I personally believe that the connection between politics and complexity would be rather more productively and usefully framed by thinking through how complexity sciences can be brought to bear on problems of power and politics in different settings. This would not mean taking sides, but seeing how complexity can deepen our understanding of the complex interplay and dynamics of political change and related phenomena. In terms of last weeks election, for example, what are the network effects that explain the ‘shy tory’ phenomenon and the rise of the SNP? Why did the pollsters get it so wrong, and what kinds of biases played out to inhibit the ‘outlier’ messages? When pundits were confidently claiming Ed Miliband had more possible paths to being PM, were they really grasping the whole system and potential pathways? And so on.

    In specific terms, claiming complexity is a socialist is to my mind as partial and narrow a view as those free market ideologues who claim that complexity science justifies minimal states. But to run with your argument and take it to its natural conclusion, surely if complexity was a socialist, then ex-ante planning would work rather better in complex systems? If anything, if complexity ‘sees the need to facilitate the emergence of socially-owned processes of governance and civic empowerment, and to build on those practices that already exist’ then wouldn’t complexity be a liberal, selecting markets and government solutions based on the evidence? After all, in many different ways, in many different contexts, complexity scientists have found the balance between complete control and complete decentralised autonomy to be optimal. See, for example, Gell-Mann on effective complexity, Kauffmann on the edge of chaos, Crutchfield on the balance between order and chaos, Jay Gould on punctuated equilibrium, and so on.

    But I really don’t think it is the job of science to map itself onto political alignments, because doing so risks invalidating the objectivity of the scientific method, by loading it with normative assumptions of what it ‘should’ find. Complexity scientists take great truck in busting others flawed assumptions – by claiming complexity leans in a particular political direction, do we not risk pre-empting any future findings that might lean in other analogous directions – and therefore be prone to flawed assumptions, biases and group-think of our own?

    Overall, I fear this kind of thinking, even in the form of analogy and metaphor, does the field of complexity science little good. I am all for bringing values into science, but I really don’t think that this is the way to do it.


  4. Jean Boulton

    Ooh this is fun. I think it is important to consider the ontology of complexity, its worldview, and then consider what it means for the social world. So the point I am opposing (I think Owen thinks I am saying the opposite) is the systems, left to themselves, as with the philosophy of a free market – leads to inequality and power in the hands of the few. So there is a need for a pushback from this to ensure that the voices of the disadvantaged and considerations of the future sustainability are upheld. We frame most of our thinking in the west either around a mechanical worldview or this idea that leaving things to themselves means they sort of balance up. So I think it is good practice to consider what complexity means for the underlying philosophies and mindsets through which we approach politics, economics, communities, families and everything. As an aside, ene thing we do in the book that is coming out is compare and contrast different approaches to the mathematics of complexity, and the notion of the ‘edge of chaos’ is critiqued as a rather special case. The book also makes the point that non-linear dynamics can give a rather false impression of the nature of the real messy world which is not always subject to mathematical analysis

  5. Jean Boulton

    Oh, Ben,
    and just wondering, it is ok to talk about complexity economics but not complexity politics? What about political economics, would that be ok? 🙂

  6. Ian Thorpe

    I’m probably missing something here but I don’t see how you can ascribe political tendencies to a natural process, and one that creates wildly different results in different contexts (in politics and in life), and most of which are only stable in the short term.
    The interesting human dimension is that we often seek to impose order and logic into these systems and often have some success in success in doing so, but when we do we never quite get the results we (logically) expected and they don’t tend to last.

  7. Ian Toal

    “…they can fly around buildings or build ants’ nests that adapt to local conditions, but they don’t in general change their nature and adopt new behaviours. Nothing new emerges, nothing co-evolves.” Sorry to nit pick, but birds and other living organisms/groups do change their nature, and create new things. Gulls following a tractor have learned that they can get food. This is not a behaviour found in “natural” gulls, but is something they have learned. Complexity can be seen, but not specifically predicted. So a complex system like economics kind of works most of the time, but in truth, is very difficult to use as a means of prediction. As well, I don’t like adaptive complex systems either. There’re seems to be a critical mass to complexity – too much or too little and the system becomes less able to adapt to changes

  8. Jean Boulton

    Yes, Ian (Toal) I agree, birds(and all living creatures) do change their nature, but not in the influential models of, say, bird flight from which much is inferred about self-organising systems. Yes, complexity emphasises the limits to prediction – it is less of a tool and more of a mindset which alerts us to the limitations and dangers of applying tools as if the world were predictable and controllable when in most cases of interest (ie when things are changing or we are trying to engender change) which it is not.

  9. Jean Boulton

    I wanted to go back to Ben’s comments and also to Ian’s (Thorpe) comments. Complex systems, so called, look at the behaviour of interacting agents (from molecules to humans to institutions) where the interactions are not simple and linear, where the boundaries are open to the wider context, and where there is variation and messiness (so the system cannot just be understood analytically through mathematical equations) and (in the general case) there is the potential for learning – ie for the agents themselves to change. Not all complexity mathematical models include all these general aspects of how systems in the real world can change, and so we must be wary of inferring/abstracting too much from particular models when we want to say something about human and social systems.

    The point I am making in this piece is that complex systems are not always in a self-regulating adaptive state; they can become ‘locked-in’ where there is less diversity and the relationships become harder to change. Complex systems can also collapse into chaos (so that aspect of Gaia theory, which tends to assume the planet will necessarily self-regulate, is brought into question too).
    So when we want to work with different situations, we have to take a judgement as to whether in this particular case we have a locked-in situation, a fairly balanced self-regulating one or a chaotic one – or of course, aspects of all three. To assume that all situations can adapt and that we should necessarily approach situations and try to enhance adaptive capacity may not always be helpful.

    The ‘lens’ of complexity thinking is often applied to economics and also to management, policy development, social theory and so on. And of course it is just as legitimate to look at political theory through its lens. I can’t really see how Ben can argue otherwise. The point I am making (along with complexity economist Brian Arthur and social scientist Graham Room, amongst others and of course my co-authors) is that systems, if left to themselves, have a tendency to lose diversity and for the big to get bigger. This is the case in natural systems as much as in social systems and is due to the power of positive feedback loops. So, in economic and political theory, this means that markets, over time, will end up with fewer larger players who constellate power and are hard to contest. This is what we see in many markets today. So complexity theory would suggest that any political ideology which believes the free markets and deregulation will necessarily be the ‘best’ approach would not be supported by complexity thinking. There is a need for some forms of governance and regulation which uphold the rights and needs of the less powerful, the disadvantaged and look to sustain the future. Complexity tells us that ‘market failures’ are the norm, over time.

    So, in political theory terms, then we are looking for ways, ideally, to support ways societies have traditional regulated themselves, to counter this tendency towards dominance. But we also now live in a world where corporations are global and often have more power than governments – how to exercise ‘global governance’ is a really big question. But the need – in whatever way – for processes to counter the general trend towards inequality is clearly made by complexity theory. And that, in political language,is more of a socialist policy. I am not making a party political point here, I am merely arguing that free market ideologies will not engender equality or the needs of those with less power and voice.That is not to argue for controlling everything, for stifling competition and choice and all that; it is arguing that if we take the ‘free market’ approach too far, we value individualism over collective responsibility. There is such a thing as society, in other words.

  10. Eitan Reich

    Thanks Jean, great post and discussion. Your point is well made and taken, that the ‘Adaptive’ in CAS is misleading, and is a kind of trap. However, I also agree with others here that Socialism as much as Free Markets ideologies collapsed and lost diversity in the same loops Capitalism did. So if Complexity was a person she might just be much more flexible in picking his identity…economic, political or gender? I am looking forward to the book too! good luck!

  11. Jean Boulton

    Good to hear from you Eitan! My main point is just that systems, if left to themselves, end up with less diversity and constellated power – you need ways to counter that if you care about the poor and the planet – so less of a comment on socialism more of a comment of the need for creative and embodied ways to ‘regulate’

Leave a Reply

Your e-mail address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.