Dave Snowden

The “R” words

RSS Feed

I started to make some real progress on the development of constraint mapping when I realised the typology I was developing split into resilient and robust categories. Inevitably when you use those words the question arises as to Taleb’s anti-fragility and there are other issues. Making words problematic is something I was introduced to by J C Spender at an AoM session in Washington several years ago. He argued that when the meaning of words becomes problematic then advances are made. He suggested this was true for knowledge at the turn of the last century and that my work was making the whole idea of meaning problematic in turn. The meaning word is important but I will leave that for another day. For the moment I want to outline where I am with the R words.

For some years now I have defined robustness as surviving unchanged, while resilience is surviving with continuity of identity over time. I’m not sure that surviving is not too reactive a word but it will do for the moment. Now in using the language in this way I correspond with various dictionary definitions of robustness but resilience is more problematic. Some definitions imply elasticity, i.e. change in the moment but then return to the prior state; in that sense I have generally used elasticity as a word relating to robustness. I’m using it more in the sense of ecological resistance, to quote wikipedia: the capacity of an ecosystem to respond to a perturbation or disturbance by resisting damage and recovering quickly. Such perturbations and disturbances can include stochastic events such as fires, flooding, windstorms, insect population explosions, and human activities such as deforestation. The point I am making is that the various events create change, certain features or aspects survive and I see continuity over time, but change is a part of the capability for survival. In this sense of the world anti-fragility is a type of resilience in which the system improves as a result of the stress. Taleb distinguishes this from systems that return to their prior state. So I think here we have different uses of language rather than a deep seated difference.

I do worry that a gifted thinker and writer like Taleb seems to have been seduced by the desire of many publishers for one simple concept presented as a panacea, but he resists the danger more than most other popular authors. I am more specifically concerned that we get into a pejorative dichotomy in which anti-fragile systems assume some pre-eminence, are assumed to change for the better and are the only form of system which shows a non-linear approach to threat – all of which assumptions can be read into Taleb’s work. I don’t know that that was his intention, hence the qualified language.

Getting the language right is important, as it was with complex v complicated and efficiency v effectiveness. At the moment the key distinction for me is the way I defined it earlier. On the one hand a system which survives by change (hence the header picture) but not necessarily for the better, against a system which survives without change. Now this is a key linking concept to many things. Just to give some examples:

  1. In my reworking of Apex predator theory during ecological stability the main actors are not susceptible to disruption and you adopt a strategy to exploit the stability.  During a period of disruption the stable, robust guys will not survive so you start to see what can be rapidly changed or repurposed to establish a new market.  Continuity but rapid mutation and change.
  2. The distinction gives us a much better and more nuanced and context sensitive use of the so called Forest Cycle beloved of the chaordic systems guys and a few others.   Chaos is not inevitable in human systems – something given the geometric increase in existential threat we need to start taking more seriously.
  3. It allows a language and practice of mutation through experimentation to test sustainability of different approaches.  Change is not always a consequence of threat, it can be induced in the face of difference (another reason to be careful about the anti-fragile idea).
  4. I used the slime mould picture (in text above) to partly make the point.  Slime Moulds shift from animal to vegetable as needed to match conditions of scarcity.   But they remain the same thing.  For me the language of constraints and constraint change allows an emergent approach to strategy.
  5. The libertarian assumptions that are increasingly problematic in some thinkers on complexity assume emergence without continuity.  Panarchy and others have this problem.  Now that is a whole other field and I’m only placing a marker here for that discussion.  But market solutions, alternative tokens to money and so on all to my mind loose the idea of continuity with change in favour of a change and pray strategy that is plain bloody dangerous.

Of course once we have this sorted (and I’m not wedded to the language yet) then there are other key words and concepts to explore including reification and redemption to name but two.

This post like most recent ones are fragmented takes on a wider whole I am developing at the moment. The links between resilience and liminality in Cynefin for example are very important.

  • Sonja Blignaut

    Very useful post, and pertinent as well. I think we tend to use these terms interchangeably without really thinking about what we actualy mean. Another interesting concept from the ecological resilience work is the idea of ‘Transformability’ (https://www.ecologyandsociety.org/vol9/iss2/art5/main.html).

  • Russell Gasser

    Resilience is already a widely used term in humanitarian, disaster preparedness and similar areas, and extends into some other areas like anti-radicalisation work as well. It mostly gets used with a similar meaning to wikipedia’s ecological definition. A community’s resilience to a disaster (e.g. flooding) is how well the community as a whole survives a flood event (not necessarily unchanged). Resistance would be the way that the infrastructure is able to deal with the floods unchanged – so this accords to the same definitions. Some definitions also introduce the ability to recover from a negative event or disturbance in a “timely, efficient and effective” manner. Adding timeliness may be useful to separate resilience from evolution or co-adaptation and efficiency appears relevant in terms of least energy cost of systems.

  • http://kneaver.com/bruno BrunoWinck

    Your definition behind “In this sense of the world anti-fragility” fits to the use of resilience in psychology. Successful adaptation. Resistance is ability to cope with stress. I came across resilience first in metallurgy. The ability to turn stress into energy and release it back afterward. I think it is the root of the concept. It oppose resistance (hardness) which is the ability to take stress without changing. Interestingly French and English research on Psychology went into a different understanding of the phenomena. Resilience is to restore an equivalent state, no need of improvement.

  • Neil Ramshaw

    I’m really interested in point 5 but was wondering if you could develop and clarify some of those libertarian assumptions.

    • Dave Snowden

      Well you have two strands, the Ayn Rand selfishness is ethical and then the more left wing/anarchist strand – both put the individual at the centre of everything. The point here is the lack of desire for restraint

  • Heather Bewers

    I’ve been fascinated by resilience as a concept for a while – and appreciate the distinction between the engineering and the ecological perspectives. What I’ve seen less on (and my apologies if I’ve missed a field which does distinguish this) is the distinction between resilience which survives but changed, and resilience which survives and is enhanced by the experience. I have (and it is only a gut feel) that there is a key distinction here which the debate about resilience of the status quo and resilience post change doesn’t quite capture. Certainly I didn’t get this from Anti-Fragility – it’s these nuances which I’m really interested in – and why I find your blog so fascinating Dave, as frequently it is these nuances you pursue.

    • davesnowden

      Enhancement is a type of change surely? Making distinctions within the field of resilience makes sense; trying to argue that anti-fragile systems are different from resilience systems (largely conflated with robustness by Taleb) is more problematic. Also what type of change is sustainable? Strengthened may not be an option in a catastrophic environmental change. The other big issue for me is the granularity of examples used and their applicability.
      Prof Dave Snowden
      Cynefin Centre & Cognitive Edge
      Sent from my iPad Pro

  • Barrett W Horne

    Confess to having missed the blog posts for a while–in both senses of ‘missed’. Reading this one I am reminded of Heraclitus’s observation about not being able to step into the same river twice. The idea of an identity that survives over time is not without some conundrums. What is the ‘thing’ that ‘survives’ through change and adaptation? In what sense is the slime mold ‘the same thing’?

    • Dave Snowden

      Well its not an ‘essence’ nor is it a thing per se. Continuity and coherence are key words here, you can see true pathway in those terms even through a phase shift. Think of the ship of Thesisus as a sea going ship not a historic artefact