Why Dave Snowden Doesn't Know Who Shot JFK

Yesterday I spoke about the presentation on SharePoint implementation at the recent NSW KM Roundtable. Today I want to talk briefly about the second presentation, which was on storytelling.

Well, it was billed as being about storytelling, because that how the audience thought of it, I suspect. It was actually about narrative and how to use it to supplement existing business reporting. So of course, when we're talking about narrative & business in Australia, who better to have come in than Mark Schenk from Anecdote.

Mark kicked off with a new story (or at least, one I haven't heard before), which I enjoyed. I won't tell it - It'll probably turn up on the Anecdote site sooner or later, so keep an eye out for a new "bathroom" story. Then he presented the framework for narrative, the old Cynefin, and used this to build on how to use narrative in business. All good interesting, and well worth listening to.

Anyway, I was looking at the Cynefin framework, and in particular the complex domain. Its been often (and better) explained in other places and forums, but basically, a complex problem is one in which you can only really make sense of what happened after it happened. Mark mentioned the Canberra bushfires as an example of this. Before the fire, we Canbrites (as I was at the time, and possibly still am given my despair over the performance of the Brumbies this year) blithely went on with life, giving little regard to the obvious danger we were in. The danger, however, was only so glaringly obvious in retrospect. After the fires, we could see all the factors that made it clear that 400 hundred homes would be destroyed on that fateful Saturday. This phenomenon of only being able to make sense of something after the fact has its own wonderful term (several, actually, but I like this one): retrospective coherence. Looking at it though, I began to wonder, "Isn't that how you get conspiracy theories?". This wonderful ability of humans to see patterns in the events that occurr around us also gives rise to the issue of seeing patterns where there are none:

The World Trade Centre attacks were funded by the US Government,

there had to be a second shooter behind the grassy knoll,

global warning has already doomed all life on earth and governments are covering it up, etc.

These are all acts of retrospective coherence - albeit flavoured with a little bit of cognitive bias, I would guess - but I wonder how you get around this when you are looking at complex problems.

I didn't have the chance to quiz Mark on this, as I didn't want to derail the line of questioning into something that was probably going to be more advanced than the audience was interested in. I'm curious though - where does understanding a complex problem end, and building a conspiracy theory begin? And how do you prevent one from becoming the other? Do you sometimes want to ensure that one does become the other?