It has come to my attention that expert advice and reason has been rejected in recent times. My worry is that today because it is possible to google something that everyone now thinks that they are either an expert or better than an expert on many issues.
We also have the problem that some false experts are giving out advice and opinions on a range of topics, this second problem fuels the first problem.
This blog post is intended as an antidote to both problems. This problem is not merely an academic problem, it is a problem in wider society which has contributed to the dire choice of the UK public to leave the EU.
Now we will start with the false expert, if we consider for example Prof Meadow make a series of statements on “cot death” while working as a expert witness which have resulted in a series of convictions. He stated “one cot death is a tragedy, two is suspicious, three is murder”.
The problem is that we look at L. Tranebjaerg, J. Bathen, J. Tyson and
Now if 1 in 1000 babies were to die at random then if we assume that Mrs Normal has three babies then I would predict that she has a 1 in 333 chance of one of her children suffering this fate. However for her to have two children suffering this fate it is much more unlikely.
Now we assume that one in 1000 women will have lost their first child to SIDS, now these women have a 1 in 500 chance of losing one of the other two babies to SIDS. This works out as 1 in 500000. Now if we go and calculate the chance of a women losing three children to SIDS we can come up with 1 in 1000000000.
The problem was that no one seemed to challenge Prof Meadow (who got into a lot of trouble, he is a flawed expert witness who was struck off and then the striking off was reversed). His striking off was reversed by a judge.
Now back to Meadow’s law, if we assume that 1 in 100 women have a gene which causes their offspring to have a 1 in 10 chance of death as a result of SIDS. Then if we had 1000 women then if they all have one child then 1 in 1000 of the children will die of SIDS.
But if we consider the chances of a woman who is in this group (with the defective gene) from having a second cot death then it is 1 in 50 which is much a higher chance than Meadow predicted if we assume the woman has a total of three babies. Now for the chance that all three die of SIDS would be 1 in 1000. This would suggest that if SIDS can have a genetic cause that it will be clustered in some familes rather than being truely randomly spread through society.
I suspect that some SIDS cases may have the generic cause while others might not, this will make the maths more complex when we consider clustering but we can save that for another day. The big question with Prof Meadow was why did no one use what you would have learnt in A-level (or maybe GCSE) biology. If I had been cross examining Prof Meadow some of the biggest questions I would have had would have been.
- Are you aware of the work of a Roman Catholic Monk named Gregor Mendel ?
- What is the cause of SIDS ?
- Are you able to rule out a genetic cause of SIDS ? If so why and how ?
For those of you who do not know Gregor Johann Mendel was the person who founded the science of genetics. He did his work with plants such as peas, but his work does apply to animals and even humans. I rather like him as a person from history, he even had a sense of humor he said something to the effect that he had had a good life until he was aged by being promoted.
With Prof Meadow the system of experts failed because nobody questioned what the “expert” based his evidence on. I am sure that the sorry tale would have been a lot shorter if he had been subjected to vigorous cross examination in court.
My advice when dealing with an expert is to ask them to justify why they hold view X rather than just accepting view X. What you should do is to ask to see the reasoning and evidence behind their professional judgement. True experts are normally willing, when they have the time, to go through the evidence. I would argue that when acting as an expert witness in court or when offering advance to the public that you should make the time to provide the detailed explanation if it is required.
We will get back to the subject of experts and who to trust soon,