The physicist who runs a blog he should call “AndThenIGaveUpScience” has a blog post on:
consensus, the appeal to authority, and how we counter manufactured doubt,
And the video (at end) is a real laugh.
In the talk we hear appeals to ignorance: “we can’t check all scientific claims our self”. That’s rubbish. It’s the same elitist claim that “only someone trained in media studies can tell whether a program is worth watching”. Checking and critiquing something is far easier than doing it our self. Most people are not able to create a TV program, therefore (so goes the appeal from the BBC) only those who make TV programs are able to decide what is “good” TV. Rubbish!
We all know, it is extremely easy to tell what is “good” TV even though we haven’t any skill in making TV programs our self. Anyone can tell a good program: simply look at the viewing figures! People are more than capable of deciding what they think is “good” even if they have no ability to produce something themselves. The same is true for clothes, food, books, computer games, dictionaries, doctors, plumbers, politicians. We don’t need some expert to tell us when an operation goes wrong or when a plumber causes more leaks than they fix.
When we see people reluctant to have their work assessed, trying to avoid going into detail, making up excuses (the ocean ate my heat), and trying to argue that the cracks rapidly appearing in their work are just “normal” – when we know they aren’t. We don’t need a PhD in psychology to know what’s going on.
In all other areas of life, we don’t need experts to judge whether someone has done a good job.
So, why should “science” be unique? Why on earth would the only people who are allowed to judge science be those with an self-interest – the academics?
There is in fact a very simple way to evaluate science. it is the same way any of us evaluate a doctor, a dentist, an engineer – who just like the academics are experts – and that is to see whether what they do works. And the procedure that ensures this in science is called “the scientific method”. So, surely this is how we can evaluate science? No! And this next bit really does come from his blog:
There’s the rather naive view that science is about developing a hypothesis, testing this hypothesis, and either rejecting or accepting the hypothesis depending on whether it passes – or not – the tests.
Like so much of the drivel we get from academia, assessing science is actually very simple: the theories must work in practice. But for obvious reasons, academics don’t like this. Why ruin a good theory by testing whether it works in practice?
But this is how all engineers work. The scientific method isn’t unique to science. Indeed, if anything its the other way around. No engineer needs to be reminded that – if their models, ideas, designs, theories fail in practice then not only will money be lost, but lives – and they could easily find themselves in court.
Engineers are constantly reminded that ideas have to work. It’s an inherent part of engineering. Engineers are constantly checking their theories work. The problem is not that engineers don’t use the scientific method, it is that academics in their ivory towers divorced from the real world – don’t at all like the rigor of testing their work.
And we can see their point! What does testing do? Other academics aren’t pushing for it. The grants keep rolling in without testing – because they’re all doing it that way. What can testing achieve other than stopping the grants rolling in?
And testing is so labour intensive. Why stop the bandwagon dreaming up new “science” by insisting on the time consuming process of testing the theories work in practice? That is why it is oh so easy for academics to go off into dreamland for years even decades at a time and never subject their theories to the rigor of seeing they work in practice.
That is why the scientific method is so important!
It is a polite way to tell academics to to get off their fat arses, stop dreaming and go and check their work in practice. It is also a polite way to tell them to ditch ideas and models that do not work in practice. It is a polite way to tell them to pull up their socks and achieve the much higher standards of checking and verification which is the bread and butter of engineering.
But like so many academics AndThenIGaveUpScience doesn’t want to be tied down to only working on things that have to work. He doesn’t want the hassle of the scientific method:
most science today doesn’t really work like this. Most scientists are not expecting to falsify a fundamental law or do research that leads to a new fundamental law. Most science today is about observing and collecting data and then trying to understand what this data tells us about what is being observed. It often involves modelling, but most of this is based on well-established laws and so even if a model doesn’t correctly represent what’s being observed, one wouldn’t immediately conclude that something had been falsified. It probably just means that the model is missing something.
So basically AndThenIGaveUpScience is saying that even if the model doesn’t represent what’s being observed then that’s no reason to say it’s wrong. That’s no reason to stop the grant money rolling in. That’s no reason to keep taking public money. It might be wrong – but why stop the gravy train!
In other words, he is saying is that if we have a dataset:
(a0, a1, a2, … an)
he would be quite happy to model like so:
y= b0 +b1.x + b2.x2 + … + bn.xn
Then if “a model doesn’t correctly represent what’s being observed, … it probably just means that the model is missing something.”
So when we get a new result an+1 that isn’t predicted by the equation he dreams up, what he will argue is that he can just add a bit onto this equation to get a new one with:
y= b0 +b1.x + b2.x2 + … + bn.xn + b(n+1).x(n+1)
Now I’m sure any skeptic will immediately know that it is a simple law of maths that any dataset of n points can be represented by a polynomial of power n. So there is no set of data that cannot be represented by the model I have selected. It’s an entirely bogus equation! But by adding just one extra terms as AndThenIGaveUpScience wants, this “academic method” is guaranteed to turn total bollocks into “undeniable science” (after the event).
I appreciate this will be difficult for “AndThenIGaveUpScience”, to understand. So now I will try to put it in a simpler way. Let’s say AndThenIGaveUpScience is claiming to be a very good pundit for horses. He puts his money on an accumulator bet for 10 horses:
- Lucy in the first,
- Jess in the 2nd
- Sid in the 3rd
- Meg in the 4th.
The first race is run, but instead of the predicted Lucy winning, in comes another nag we shall call Lin. But because AndThenIGaveUpScience believes:
There’s the rather naive view that betting is about putting a bet, holding to that bet, and either winning or losing the bet depending on whether you win ….
He says he should be allowed to change his bet to “Lin”.
In the next race a horse called Kath wins instead of Jess. He still thinks it “naive” to not allow him to make a small change to his bet after the event so he now says his bet was on “Kath”. Now he’s betting on:
Because he thinks its “naive to be held to your bet” … he just keeps changing the bet each time he is found to be wrong! The next race is won by Sid, etc.
At no time is his accumulator bet that wrong. Indeed after he “fine tunes it”, it’s always perfect!
He only makes a small change each time. (After the adjustment) the bet always accounts 100% for all the available races with a perfect bet aka “undeniable” or “settled science” .
Of course, far from AndThenIGaveUpScience being 100% right with his pefect clairvoyance and “undeniable science” he is in fact 100% wrong.
If like climate, it takes a decade for each result to come in, not only will each academic have quietly forgotten all the changes they had to make to shoehorn the theory to fit – but very soon new academics will take over – and none of them are doing anything more than “minor adjustments”. Which of them are to blame? They all just “fine tuned” the (bogus) theory. And it was always shown to be 100% perfect (after the tune).
Despite doing no better than picking horses at random and failing on every one. They can constantly claim, and even sometimes really believe, that they are a very skilled academic with some special understanding of how to pick the winner!
Yes, the above two scenarios are unrealistic, because in real life each result is not entirely independent from the previous ones. So there’s a high chance that one result be like the next (a warm day tends to be followed by warm days).
However when AndThenIGaveUpScience finds he does not have to constantly change his model it just lead to confirmation bias. The lack of any need to tune his model confirms in his mind that he really does have a god given gift to predict whatever he thinks he can predict.
This is why it is so key that the scientific method is rigorously followed.
Because a dataset where there is dependence between readings will lead into a false belief their models work, it makes it all the more acceptable to make “just the odd change”. Because who wouldn’t allow the academics to make the odd adjustment to “tune their model”?
The result (as in climate) is that a model that has no relationship to reality, can end up being considered to be nigh on perfect, because it is never subject to the rigor of the scientific/engineering method.
The real state of the global warming theories – not much better than pure chance and constantly trying to excuse past failures.
To illustrate just how the so called “undeniable” theory (in reality theories) have performed I have tabulated what has really happened below.
|1970||Camp Century Cycles||Global Cooling||1970s
|1980||Camp Century Cycles +CO2||Global warming||1980s
|1990||Only CO2||Global warming||1990s
|Certain warming||2000s no warming||WRONG|
|Certain warming||2010s no warming||WRONG|
|2014||dito + Ocean ate my heat.||Certain warming||?||?|
The comedy Video which should be called:
“Academic theories never fail”
“Academics are unique! Unlike all the other arrogant professionals whose self-regulation failed – ours is perfect because [we say] outsiders can’t check our work.”
“Why Academic ideas don’t need the boring hassle of being proven in practice”
“Why should I make the data available to you, when your aim is to try and find something wrong with it.” (Phil Jones)