Bayesian Maths

26 Apr 2017

The Conversation is a news journal well known for publishing gobbledegook articles. It also publishes incisive articles too. I'm not sure what bracket this one goes under - go to https://phys.org/print412240382.html. It seems to be an argument in favour of modelling over common sense. It also seems to criticise science prior to the modern era, even perhaps prior to the 21st century AD. We can see why Bayesian methodology has been used in dating archaeological event as it averages out a raft of C14 dates and settles on particular date. It is then the accepted date of the event - or the object. For example, the arrival of Neolithic people in Britain was dated to 4150BC by the Bayesian method, 100 years after a prominent tree ring anomaly (a cold snap with evidence of abandoned and burnt settlements on the continent). Did the method provide the true date? One has to wonder.  Bayesian methodology is now being applied to biology, in order to produce one Tree of Life that everybody will accept - it is hoped. Why this is an improvement on critical thinking is unclear but one supposes the idea is to produce a consensus opinion - and exclude maverick scientists with a different point of view. Does this mean that consensus science is the way forward - in the West. What is fake science - anything that disagrees with the Bayesian methodology? ... or has The Conversation got the hold of the wrong end of the stick?