Whenever someone challenges me for not believing in this or that “expert” I have a simple response:
“I’m an expert on experts – I’ve written a book** on the subject and trust me, you can’t believe experts”
Of course, if I am an expert as I say, then my expert opinion means that they should believe me. Which means that they should not trust experts like me. Which means they should not believe me. Which means they should trust experts like me …
However, the deeper question is why am I an expert merely by publishing a book? The answer is that in the past, if you did publish a serious book, it meant that some big publisher had read your content and thought you worthy of publishing. In other words, publishers were deciding who was and was not an “expert”.
Likewise, if a TV program wanted to have an “expert”, a journalist would ring around a few of his buddies from School (Eton) or University (Oxbridge) and whilst it was unlikely that any had actually studied the subject themselves – as long as they had a degree which sounded vaguely relevant, they could pontificate on the subject and fool the audience enough to pass as “an expert”. An again, the people who were acting as “gatekeepers” to this class of “experts” were the universities who accredited them as academics and journalists. So, e.g. an “expert” from Oxbridge was worth more than an “expert” from hull – even if the hull “expert” had actually studied the subject – because the University itself gave a kudos to people.
The High cost of being an expert in the past
Before the internet, in order to be an expert in a subject, you needed access to the very latest research in the subject, that meant buying all the latest books written by others interested in the subject and subscribing to the journals and as such you needed a deep pocket. Almost no individual became an expert using their own money. Instead, they belonged to some organisation – usually a University (paid for by the public) and they used their privileged access to increase their own knowledge immeasurably beyond that of the general public.
Then along came the internet … and we all could become experts
But when the internet came along, anyone with an interest could read and write on a subject at almost no cost. Suddenly, instead of having to have shelf loads of highly expensive books and subscriptions to very expensive journals … almost all the information start to appear on the internet and the general public, with an interest in any subject, could very quickly become as expert. Indeed – due to the knowledge on the internet, and the ability to focus on just one small area – those on the internet could become more expert in any specific subject than the academics.
That was bad enough … but because internet experts do not spend their time just talking to those in the same peer group … because we are not immersed in an “ivory tower” world where we only talk to other ivory towers, we tend engage with people at all levels of knowledge, and so are inherently better at communicating our ideas to other similar people on the internet. And unlike an academic who is forced to teach dumb students and sick to death repeating the subject … experts on the internet are experts through interest and so they write with enthusiasm … not lecture down to people because they are paid to lecture.
It’s not the death of “expertise”, instead it’s the death of a different social class called “experts”
As a journalist – with no expertise – the change is inexplicable. A few years ago, their friend let’s call him Paul (a biologist) could be hauled on the BBC to discuss climate change and no one would complain. But now, as soon as they put this “expert” on the screen, even though his expertise has not changed at all – indeed he’s even read a book on the subject++ – there are howls of protests. Why don’t people believe the “experts” any longer?
The answer, is that there are now thousands, even tens of thousands of experts on the internet with more knowledge than their friend Paul. Paul hasn’t changed, instead the audience is now filled with experts quite capable of taking on their friend Paul in an argument and tearing the opinionated-third-rate-scientist-who-only-gets-on-the-BBC-because-he’s-a-chum-of-the-editor to pieces.
And indeed, even someone who knows less than Paul (which is difficult) knows they can find experts who do know more than Paul discussing the subject on the internet and because they can see a range of experts, they can see Paul’s clear political bias and lack of any real knowledge to back up his views.
Paul is no less of an expert than ever. Paul is no more biased in his views. He’s no more keen to “science-wash” his eco-politics and present his views as science to the public.
The difference is that the public, thanks to the internet, are now far more informed, better at telling who is and who is not an expert, and is sick to death with people like the BBC passing of third rate knowledge and biased opinions – which wouldn’t last a few minutes on the internet – as “experts”.
**The Academic Ape: Instinctive aggression and boundary enforcing behaviour in academia. Available for kindle at a price from Amazon or for free
++Likely “an introduction to climate alarmism for undergraduates”
Therefore if I had responded I would have taken a simple average gradient (LR) and picked those with the highest absolute gradient and depending if +ve or -ve have said they were +1 or -1C respectively. But unfortunately, I was missing one vital piece of information: how many had been adjusted.
Therefore, a bit of stats is required. We know the gradients fall into three buckets: those with no addition, those with +1/century and those with -1/century. Within each bucket the gradients should be a normal distribution (if the time series were random), and we know where each should be centred. So it should be possible to estimate the number that were in each bucket. That is just a question of putting each gradient into the most likely bucket. There’s probably some statistical trick for that, and I could have found it, but as Doug would have known this was the likely approach and have checked to ensure a 90% accuracy was impossible using this approach, so I knew it was pointless.
Instead, in order to win the $100,000 prize, it would have been necessary to “move away” from the optimum which would have been checked to ensure the prize could not be claimed, toward a sub-optimum solution which was sufficiently different that there was a small chance that it might have by accident picked the right ones (but most likely would be wrong).
But what if I did win?
It would then put me in a moral dilemma – since I could very well do with the money, I’m very happy Doug ran the competition – but he clearly never intended it to be won – and certainly not by a sceptic.
However, I would still be interesting to see a graph based on the scale of gradient chosen as the cut-off (above which it is assumed +1 or -1C/century added) and what percentage would be right using this very simple approach for each cut-off. I assume the graph would rise steadily as the threshold was increased because at first each increase would identify more right ones than wrong. But eventually the scale would be too large and it would start identifying more wrong than right.
The Top Graph
It’s just natural variation – it’s just as meaningless as the “global temperature” graph.