Monday, July 07, 2014

On Pseudo science

I read a post today by Dave Snowden on the followers of Elliott Jacques and the Requisite Organisation. The post talked about pseudoscience, acolyte syndrome and cultism and the language seemed pretty strong to me.

One of the comments to this post was 'I get accused of this (Acolyte Syndrome) by my wife, all the time, in reference to Cynefin' which made me think, as useful as Cynefin might be - is it falsifiable? Does it predict? Is it not pseudo science itself?

The demarcation between science and pseudo science is a long fraught affair with numerous luminaries from Karl Popper to Paul Thagard. Now this boundary is permeable. Activities often move from science to pseudo science and vice versa over time. It's not a question of what is right or what is more useful but instead what is science.

Falsifiability is an important part of that question both for Popper and Thagard. It's fair enough to say it is central to Popper but even Thagard asks whether the supporters of a concept 'actively attempt confirmation or disconfirmation?' 

Now, I'm not aware of any predictive capabilities of Cynefin. It's a sense making framework, a classification system to understand an environment. Without any predictive capabilities, how can you test it? How can you measure it? For me, it falls on the pseudo science line but then so do many useful management concepts and theories.

Hence, I thought the original post was a bit rich and called it out.
What followed was an interesting conversation which covered some statements on predictability
and some personal views
Now, this is interesting for two reasons. First, pseudoscience is simply a classification and many useful things are pseudoscience (e.g. a lot of economic debate is in this camp). In my experience, whether something is viewed as derogatory depends mainly upon the perspective of the viewer.

The second interesting point is that though Snowden claims Cynefin has no predictive capability (which means it is not falsifiable) other supporters of Cynefin claim it is scientific.
So, I asked Dave

To which, I received the response
To be honest, I found it difficult to get a simple 'yes' or 'no' answer and Dave has kindly agreed to write a post to explain this and why I'm wrong.

So what is Cynefin? Well, I happen to view it as useful way of examining an environment but given it's not predictive then it's not falsifiable nor scientific. Now whilst Snowden makes no claim to it being scientific, his supporters do.  Hence, I'm going to put this into my pseudo science classification.  I don't see this as a negative but instead what something is. There's a lot of useful and often subjective concepts in that classification e.g. Gartner Hype Cycle.

--- Update 7th July

I just came across this excellent post by Tom Graves on whether 'Cynefin is a cult'. It describes how Cyenfin fails the 'science' test which is a paradox because 'most of us find that Cynefin is a very useful tool'. It provides a pretty hefty set of questions to be asked from isolation to non-falsifiability.

Overall the post is a fascinating read which concludes with 'Is Cynefin a pseudoscience, a cult? Short answer, as we’ve seen above, is “probably not” – but you’ll probably need a little bit of magic to help you prove it!'

For me, this is personally interesting because I have that same paradox. I find Cynefin useful but not scientific despite any claims made by supporters. In the same way I find the Hype Cycle useful despite knowing it's not based upon any physical measurement but instead aggregated opinion. The reason why I find them both useful is that they help encourage discussion.

I do understand Dave Snowden's view that 'pseudoscience' is uniformly derogatory, though I don't share that opinion and I certainly don't express this for that reason. His original post raised some questions in my mind over what is and isn't science. There maybe a missing 'useful but not quite scientific' category out there e.g. a proto-science.