I really apologise for the opening picture, but it was too good not to use. Yesterday I picked up a tweet from Glenda Eoyang in which she said In complex system the trick is to give up control to succeed. Counterintuitive? Depends on your intuition. Now for reasons I will elaborate shortly, I think that sort of statement harms the adoption of complexity and is in any event wrong. So I responded with Disagree profoundly with you there. The issue is what you control, not control to which I should probably have added of itself. Given the time difference the response didn't come in until later and Glenda argued Disciplined intention (needed in complex system IMHO) is not the same as control. Cx implies prediction of outcomes. Now I don't disagree with the idea of disciplined intention per se, in fact I think I emphasised intention in a long email I wrote to Glenda many moons ago when she was researching the subject. However I still disagree with the original statement and I think the definition of control as implied prediction of outcomes is a little retrospectively coherent.
That said, if Glenda originally meant prediction of outcomes when she talked about control then I don't disagree with the thought behind the statement, but its a very odd or at best narrow definition of control and its just going to confuse people so, in part, my original criticism stands. Let me share the definition of control from my Mac Books Dictionary:
[mass noun] the power to influence or direct people's behaviour or the course of events: the whole operation is under the control of a production manager | the situation was slipping out of her control
• the ability to manage a machine, vehicle, or other moving object: he lost control of his car | improve your ball control.
• the restriction of an activity, tendency, or phenomenon: crime control.
• the ability to restrain one's own emotions or actions: she was goaded beyond control.
• [ count noun ] (often controls) a means of limiting or regulating something: growing controls on local spending.
• [ count noun ] a switch or other device by which a device or vehicle is regulated: he had the chance to take the controls and fly the glider | the volume control.
• [ with modifier ] the place from which a system or activity is directed or where a particular item is verified: passport control.
Nowhere is prediction mentioned and all of those applications apply to complex systems. In fact control is one of the most important aspects of managing complexity (and yes you can manage it). As I have long argued complexity is about acting in the situated present, not attempting to direct outcomes towards an ideal future. As a result of that complexity thinking is not a subset of systems theory but something unique and different, if only because it validates Glenda's idea of prediction of the system constraints are strong enough. Complexity deals with systems where can describe and manage in the present, and there are many aspects of a complex system around which the idea of control is not only possible but also desirable.
Now one of my pet bug bears is that people don't pay enough attention to complexity in human systems as opposed to the flocking behaviour of birds, termite nest building and the like. While nature is always complex, humans have learnt how to create order, and also how to order aspects of complex systems. That is one of the reasons that I adopted and extended the constraint based definition of complexity that I first picked up from Alicia Juarrero. A complex system is one in which constraints and agents co-evolve, it is not one without any constraints (thats chaos). We can and do manage constraints, something I have picked up on before in the context of traffic management. In the children's party story we constrain boundaries or, to use more direct language we control them. We can also as it happens make reasonably accurate predictions about some behaviour and also boundary transgressions, or at least we soon learn that as parents. We don't give up control, but we direct our control to enable emergence.
So boundary conditions are one thing we can control. Humans do it all the time with laws, rituals, taboo etc. etc. If you don't attempt it someone else will do so. Its also a really big mistake to assume that just allowing a human system to develop without controls is a good idea. The most basic knowledge of history, or a simply examination of the Banking sector makes that point.
We also control probes and interventions, then we monitor the resulting emergent behaviour and control the amplification or damping factors (I am simply building on the children's party story here by the way). We regulate behaviour, we restrain our own behaviour and that of others. All of this is control. Ok its a neat soundbite to suggest giving up control, but that is all it is, a soundbite. Wise leaders may relax constraints to the point were control is not obvious, but its still there.
In fact a complex system needs more centralised and disciplined control than an ordered one. To maintain a level of alertness to evolutionary potential requires that. You may not direct micro-activities but you set and control both the boundaries and the subsequent resource allocations. If you are wise you distribute cognitive processes, especially scanning; you delegate within constraints but overall control at various levels is exercised and cannot (let alone should not) be avoided.
In any human system constraints are, as the opening cartoon says, liberating and thats not intuition speaking, not is it counter intuitive; its something every parent knows.