Bringing together some recent and old threads on #systemsthinking is #complexity is #cybernetics

Mahoo, @SystemsNinja, asked me (possibly michievously):

Hey @antlerboy tell us why complexity thinking is systems thinking, is cybernetics? Nerd face

Here’s my reply:

You tryna stop me working, or what??

I have some of this prepped, off of facebook, so here goes…

Complexity, cybernetics, and systems thinking are an extended family recognisable by a whole set of similarities (and some controversies) which draw from the same roots and influences, and share the same governing intent – understanding.

My ‘acid test’ is that I believe you cannot make a distinction between systems thinking and complexity which will not ‘sweep in’ to each ‘discipline’ something avowedly part of the ‘other’, and ‘sweep’ out from each something which claims it belongs.

some of the roots are demonstrated here:
some quotes on the theme #complexitythinking is #systemsthinking (is #cybernetics)

Look at the Macy conferences, for a start. Look at the overlaps between the early thinkers, the shared thinking, the shared learning societies.
The field is transdisciplinary (and indeed meta-disciplinary), so naturally it has diverse expression and form.

So, why do people believe there is a difference? There are indeed tribes wearing each of the three badges (and some who wear more than one) – and if you squint, you can see some differences between them. But it relies on squinting – narrowing down to what you want to focus on.

Well, there are many reasons why it suits people to say ‘my work is *this* and not *this*’ (it’s the rule of tables – if someone has a table saying ‘left side old, bad, right side new, good’ – they are trying to sell you something).

We might call it ‘wrecking synergy to stake out territory. A nice piece on that concept is here: (formatting not good as exhumed from the internet graveyard)

A good example of that is Castellani’s ‘complexity map’, which is to me a piece of fundamentally poor scholarship for this reason

There are others who I won’t name either because they’re nice people out to learn, or so argumentative as to not allow me to get to bed. (But if you search the archives for ‘curmudgeons’ and ‘popularisers’ you will find some materiel).

What tends to happen (other than simply eliding or ignoring bits of the history which show the overlap across the family resemblance) is that you pick a somewhat populist, simplistic version of the thing you want to do down, you straw-man it a bit further, and thereby produce a strangulated version of the ‘other’ (and announce This Is Wot Everyone Kno as The Thing). Then you post five or seven or 13 points showing why your brand overcomes and surpasses (usually not encompasses) the weaker, wrong part of the family. And that way we are all a little the poorer. Note that there are, in fact, many members of our extended family we potentially aren’t *that* proud of, bless their hearts… but we tolerate them and recognise they don’t represent any particular chunk of the family tree in full.

The risk of this sort of thing (‘down with this sort of thing!’) is what caused me to create the ‘four quadrants of thinking threats’ – systems / complexity / cybernetics thinkers are prone to move into one of the four corners – it’s imperative we try to full ourselves towards the middle…

(this has a modicum of discussion about the quadrants: )

See also for a magisterial take on the topic, the first comment in this link , Gerald Midgely’s excellent facebook comment at

…The constraints on that topic make a huge difference to the possible outcomes that could be concluded – so much so that diametrically opposite findings would arise from different ways of bounding the understandings of Systems and Complexity. In my view, a great PhD on this would have to start by acknowledging the diversity of paradigms (and perspectives within the paradigms) in both fields, so this is not a simplistic question of “theory A says X and theory B says Y”. So, for example, there are systems methodologies that are strong on exploring multiple perspectives, and others that are weak on this. Likewise, there are complexity approaches that are both strong and weak on perspective-taking. So a really strong analysis would, I think, look at the diversity; the various aims that the diversity of approaches are trying to achieve; the various critiques of the different approaches; and then map each approach onto that territory of aims and critiques. Once that has been done, it should be possible to look for patterns – identify how the two research fields differ in terms of number and diversity of approaches, aims that are unique in one field compared to the other, aims that are common across both fields, aims that are very strongly featured in one field, etc. If you’re serious about doing a PhD on this (or a related topic), we could talk by skype. I should flag straight away though that we don’t have funded scholarships. I have a bunch of PhD students, but most are studying part-time and paying for themselves.

For some practical examples, have a look at these two papers and tell me what you learn about the difference or not:

A good chapter IIRC:

A good series of papers IIRC:

And an enquiry:

So. All three labels are multiply defined and probably ‘essentially contested’. And, at the end of the day, it doesn’t matter – there are a bunch of good ideas, which can also steer you wrong – let’s use them.

Where it hurts (us all) is when people feel a need to define their work by doing ‘systems thinking’ down – explicitly or implicitly, subtly or not – in comparing themselves to the model they hold of some crap form of systems thinking. So in fighting against this nonsense, I’m partly creating the pain which I think we should all avoid by doing our work and not putting down other disciplines. But it’s a double bind – you let the mud stick as if you deserve it, or you get down in the mud and wrastle…

I would that I have nothing ‘against’ any person who chooses to label themselves as complexity; I love to hear about and explore and share their work (and will critique it or not based on what my limited understanding suggests it deserves – lord knows there are some poor, limited, self-limiting attempts at systems thinking too – I try to help nudge them to deeper awareness always). I *believe this is all part of the same learning and exploration*, and it turns out to be much harder to make an argument for overlap across and distinctions within-not-between, than it is to straw-man something and define your thing as different. Every time I get into this argument, I discover that my antagonist has picked one view of one set of practices, and held this up as *being* the whole.

And there *are*, of course, some more or less unsatisfactory ways you could try to make a distinction (subject to the arguments above) – at a SCiO group presentation, the only true distinction people form all three ‘camps’ could divine was a set of emotional biases of practitioners. But any definition of ‘complexity’ will fall short by some standards – as I’m arguing – so I won’t go into that here. (SCiO is the systems practitioner organisation – – formerly Systems and Cybernetics in Organisation, now Systems and Complexity in Organisation cos it is undeniably trendier and why not?)

I’ll end with McCulloch on the Macy conferences:
“Even then, working in our shirt sleeves for days on end, at every meeting …. we were unable to behave in a familiar friendly or even civil manner. The first five meetings were intolerable. Some participants left in tears never to return. Margaret Mead records that in the heat of battle she broke a tooth and did not even notice it until after the meeting.”
There has never been an agreed definition, and there probably never will be.

A thousand years ago, you asked ‘Hey @antlerboy, tell us why complexity thinking is systems thinking, is cybernetics?’. The answer is there is no ‘is’ of identity (I’m borrowing Wittgestein’s ‘family resemblances’ concept), but the overlaps are so many and varied, as are the distinctions within the field, that meaningful distinctions can really only be made of small subsets across the space – or for polemical reasons.

Er, so why did you ask?

I can’t I’m being a public intellectual

some quotes on the theme #complexitythinking is #systemsthinking (is #cybernetics)

Lewes 1875: ‘The emergent is unlike its components insofar as these are incommensurable, and it cannot be reduced to their sum or their difference.’

Smuts 1890 – 1926: ‘the tendency in nature to form wholes, that are greater than the sum of its parts, through creative evolution’

‘One of the two most important ideas for the next millennium’ – Einstein

Bertalanffy developed the concepts of open systems in 1934

Ashby’s Self Organising Principle: ‘Complex systems organise themselves’

Beer: ‘the output of a complex probabilistic system (such as a society) is a function of a self regulating, self organizing organization …in which regulatory power is not vested in a ‘controller’ but in the structure of that organization itself.’

Socio-technical systems is the study of how social groups self-organise

Autopoiesis is self-organisation

The viable systems model works with autopoietic & self-organising systems

Meadows: ‘self-organizing, nonlinear feedback systems are inherently unpredictable. They are not controllable.’

Ashby’s 1st Circular Causality Principle: ‘Given positive feedback, radically different end states are possible from the same initial conditions’ Skyttner, 2001

Darkness Principle: ‘No system can be known completely’ Clemson, 1984 (ie ‘compressability’)

Stafford Beer: ‘It is terribly important to appreciate that some things remain obscure to the bitter end.’

Stafford Beer ‘Instead of trying to specify it in full detail, you specify it only somewhat. You then ride on the dynamics of the system in the direction you want to go.’

Smuts: ‘A whole, which is more than the sum of its parts, has something internal, some inwardness of structure and function…some internality of nature that constitutes that ‘more”

Ashby: ‘the characteristic structural and behavioural patterns in a complex system are primarily a result of the interactions amongst the system parts.’

Beer: ‘Relation is the stuff of system’

Ackoff : ‘Never improve any portion of the system unless is also improves the whole.’

Iberal: ‘System stability is possible only if the system’s relaxation time is shorter than the mean time between disturbances.’

Beer: ‘If we cannot adapt, we cannot evolve. Then the instability threatens to be like the wave’s instability – catastrophic’
4th Principle of organization: ‘The operation of the first three principles must be cyclically maintained through time without hiatus or lags.’

Canon: ‘A system survives only so long as all essential variables are maintained within their physiological limits.’

Ashby: ‘The upper limit on the amount of regulation achievable is given by the variety of the regulatory system divided by the variety of the regulated system’

Varela: what is the meaning of ‘wholeness?’ This relates to two key processes. One is the process of recognizing the stable properties of wholes, by interacting with them. The other is the recognition that the stability we see arises from the self-referential, mutual, reciprocal interactions that constitute the system. Thus, the three notions I mentioned are distinction, stability and closure, and are really one and the same.

#complexitythinking, #cybernetics, #systemsthinking