#WorkingOutLoud – Black Swans and the Limits of Hierarchy

A very early stage #WorkingOutLoud post today as i start work on a more academic paper entitled ‘Black Swans and the Limits of Hierarchy’. I’m not going to get into much detail here, but the notion of ‘black swan’ events, neatly coined by Taleb, is essentially the idea of the unthinkable: black swans originally being deemed something that you would never find in nature (until they were ‘discovered’, shattering assumptions). The term is used today to mean the unthinkable: the things we are neither trained for nor which fall within our sphere of understanding.

Black Swans and the Limits of Hierarchy

For the paper, i’m coupling this with the notion of the ‘limits of hierarchy’, the idea that formal hierarchies are often poorly suited to respond to these unknown unknowns: whilst hierarchy can be superb at codifying curiosity, diagnosis and response, it often does so within frames of understanding, either of it’s own creation imposed from outside. My premise is that Black Swans challenge that framework.

The illustration is an early stage idea around this, and the narrative runs like this: In everyday life we operate in Known Frames, these are realms of understanding that are known to us. An example i give is that of a sports league: we understand that teams play each other, by set rules, to ascertain a ranking and eventual winner.

Encounter play out against set scripts, and those scripts trigger other scripts in often predictably or understandable ways, all juxtaposed against formal mechanisms of power. In other words, there is an expected script of how encounters play out: resulting in a win, a loss, or sometimes a draw, so whilst we can’t anticipate the result (beyond what stats can predict or intuition tells us) we tend to know that it will fall within a particular script, even if certain outcomes are extremely unlikely. They are still conceivable.

Winning triggers a certain celebratory script, losing a scapegoat script maybe. Scripts are cognitively efficient mechanisms to string together complex social situations: but that efficiency comes at a cost. The formal power mechanisms are codified against these scripts: the outcomes and potential often measured against these.

Outside, two other forces play on this, and relate to Black Swans: unknown scripts and asymmetric power.

Unknown scripts are one that are beyond our experience or understanding. We do not have conditioned responses, or the triggers we experience trigger inappropriate responses. In other words, we may be conditioned into precisely the wrong response (correct in the Known Frame, but entirely at fault in the Unknown Frame that it turns out we are playing within).

Worse: the unknown frame may be governed by asymmetric power: non hierarchical, seemingly chaotic, and utterly unavailable. to us.

I have’t fleshed this idea out fully yet, but essentially the key is this: Scripts can trigger responses, but unknown scripts are not available to us, and asymmetric power drains our momentum or capacity to react in any event. There are at least two dimensions of weakness in the system.

I’m taking this approach, because a weakness i see in hierarchical systems is to try to understand Black Swans from within an existing (Known) frame, codified and reported according to know scripts. I think that this leads to misrepresentation: things identified as Black Swans that are, in fact, simply complex knowns. Unlikely, but conceivable, outcomes. In other words, we are simply observing outliers, not unknowns, and the implication is that we are able to respond and end up in the worse place: with a belief that the system is robust (able to respond) when in fact it’s just highly stressed.

Using the two dimensions of Unknown Scripts and Asymmetric Power i think leads into the theoretical resolution to the problem: a framework to learn the scripts and evolve the power.

My tentative process here is the Dynamic Response Framework: response, because it’s about taking action (something that kinetic systems enjoy and are good at) and Dynamic, because it’s more capable of adapting as the scripts unfold and we learn.

This relates to the core weakness of overly hierarchical systems: they may be very good at problem solving and taking action, but at the cost of not actually understanding HOW they are good and the basis upon which decisions are made and action taken. In other words, such highly hierarchical and kinetic systems may be accidentally great, within a limited but Known Frame.

They would therefore be (and i think are proven to be), at risk from emergent, asymmetric and unpredictable events. In other words, to some extent, they are self delusional about their fitness for an adapted threat.

Early stage thinking, shared in the spirit of #WorkingOutLoud. More to follow…


About julianstodd

A learning and development professional specialising in e-learning and learning technology.
This entry was posted in Authority, Curiosity and tagged , , , , , , , . Bookmark the permalink.

7 Responses to #WorkingOutLoud – Black Swans and the Limits of Hierarchy

  1. Pingback: #WorkingOutLoud – The False Comfort of Category Errors | Julian Stodd's Learning Blog

  2. Pingback: Mugged in Madrid: Black Swans, Scripts and Power | Julian Stodd's Learning Blog

  3. Pingback: #WorkingOutLoud – How Formal Systems Fail | Julian Stodd's Learning Blog

  4. Pingback: Narrative Loops: Disruptive Power | Julian Stodd's Learning Blog

  5. Pingback: The Limits of Hierarchy: Brittle Systems | Julian Stodd's Learning Blog

  6. Pingback: Disrupting Power | Julian Stodd's Learning Blog

  7. Pingback: Risk | Julian Stodd's Learning Blog

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s