In my last post I started to open up the question of how we use natural science to reduce the energy cost of change. I argued for testing an idea for coherence and excluding ideas which are incoherent. I know what to look at and the types of ideas we can have and in doing so I’m going to go back to some of the basics of scientific discovery.
There are many theories in science, that is to say, a structured body of knowledge which is generally acknowledged as the best current way we have of understanding an area. We know it may be wrong in part, possibly even as a whole, but it’s the best guess for the moment. Generally, if you look at this over time there is a class of theories that have their limitations exposed but still have utility; Newton’s Laws of motion for example. They still work, but not universally, only within constraints. We also have whole theories that are thrown out by new discoveries; creationism is the obvious example but to that, we can add Locke’s theory that we are born with a Tabula rasa and Phlogiston, which I always felt should be true!
Then we get scientific laws, in effect an aspect of a theory. These should have a high degree of proof and experimental support but can be challenged or limited by subsequent discovery. We probably need to make a distinction between a scientific law and some of the social laws that have been laid down over the year to enforce power relationships and so on. Laws give us a baseline to extrapolate and in human systems, we can choose to impose a moral law which is a special case.
A hypothesis is some idea which can be tested experimentally and proved or disproved from which theory or laws may be generated. We may, as we do in Cynefin, not attempt to prove or disprove a hypothesis but instead, construct a safe-to-fail experiment based on that hypothesis which runs in parallel with others. Yes, this may prove or disprove the hypothesis, but the multiple parallel processes of hypothesis-based probes may change the dynamics of the situation to open up new possibilities for action. The value of the hypotheses can lie in this, it doesn’t matter so much if they are right or wrong.
But that use of hypothesis is closer to a conjecture which is an idea which is not demonstrably wrong but is not currently susceptible to experimental verification. The key thing here is that it is coherent which is a more succinct way of saying it’s not demonstrably wrong. Encouraging different conjectures about what a situation is or what we should do about it, is more generative of ideas than hypotheses as there is no burden of proof, only a requirement to make sure they are not incoherent to what we already know
Finally, we have unsupported beliefs which still have interest if they are widely shared. But we need to move quickly to determine if they are coherent, in which case we can classify them as conjectures or hypotheses. If they aren’t then we have to determine if they are incoherent in which case they should be rejected. There is probably some ambiguity in the boundary between those states and depending on our risk profile and time for action we may draw the line at a different point.
So that gives me some core language in this quest on the legitimate role of natural science. Next up are ways to test for coherence.
And as a postscript
Part of the trigger for writing this series was the intelligent discussion with Mike that I referenced in my last post. But another trigger was dealing with people, who from a position of social safety in a privileged society, make claims that you should never exclude something as incoherent, all beliefs were equally valid etc. That sort of self-indulgence always concerns me and if you want an example of its consequences read this story from the Guardian about the way in which beliefs can corrupt necessary scientific enquiry.
The banner photo is cut from an original by Eugene Zhyvchik on Unsplash
Cognitive Edge Ltd. & Cognitive Edge Pte. trading as The Cynefin Company and The Cynefin Centre.
© COPYRIGHT 2023
This past week saw me head up to the University of Hull to give the ...
In my first post in this emerging series on the naturalising in Naturalising Sense-making, I talked about ...
Leave a Reply