Nurse's Dimbleby lecture

Paul Nurse gave the BBC's prestigious Dimbleby lecture last night, addressing science and its place in society.

I sensed that Nurse was desperately trying to keep his political side under wraps, as is only appropriate for such an occasion, but I think it's fair to say that hints of his activism crept to the surface occasionally. The presumption that the most likely solutions to global warming would come through some kind of world action and regulation of the nation state was one such, although it is fair to say that he also noted the problems of scaremongering by those with a predisposition towards world government.

I was struck also by this quote.

It is the ability to prove that something is not true which is at the centre of science. This distinguishes it from beliefs based on religion and ideology, which place much more emphasis on faith, tradition and opinion. As a scientist I have to come up with ideas that can be tested. Then I think of experiments to test the idea further. If the experiment does not support the idea then I reject it or modify it and test it again.

The contrast with the CAGW hypothesis is striking. Being able to test a hypothesis in principle seem to me to be very different to actually having tested it in fact. Global warming to me looks more like a "higher speculation" than a working hypothesis.

The full lecture is here.

Update on Feb 29, 2012 by Bishop Hill

The text of the lecture is here.

Comments

[confused]

Check out Judith Curry and Roy Spencers sites...we do indeed live in interesting times.

[Peter Walsh]

(St) Pauls conversion on the road to his Damascus...?

I doubt it.

[Aynsley Kellow]

'It is the ability to prove that something is not true which is at the centre of science.' What a silly statement from an FRS, let alone the Pres! Does he not know the impossibility of proving a negative? Is the philosophy of science taught nowhere? Prove you are not a witch, Sir Paul!

[Gixxerboy]

Bish

Small point of order: I think you would agree that global warming is a well documented fact of life, at least since the Little Ice Age.

Anthropogenic Global Warming, on the hand, is moot, depending on competing hypotheses of feedback.

Catastrophic AGW is extremely tendentious - perhaps a "higher speculation", as you say - dependent on validation of the worst possible of the above hypotheses. And all manner of other tendentiousness.

Just have to be careful, your Grace, or some nutjob will start painting you as some kind of 'denier' by deliberately misrepresenting a casual shorthand.

[MIke Haseler]

"It is the ability to prove that something is not true which is at the centre of science."

WRONG!!! Yet again we see a totally incompetent understanding of science. In science everything is considered "wrong" until it is proven true (to a certain level of confidence).

Instead Nurse is trying to reverse the null hypothesis ... he is trying to say that everything any scientists (like him) pontificates is true .... until (he) decides to give someone the money to prove it wrong.

In other words, let's rephase Nurse's inane comments to law:

<b>It is the ability to prove that someone is not guilty which is at the centre of our criminal system."</b>

AKA "Guilty until proven innocent!"

[Nicholas Hallam]

Mike Haseler

I take it that Nurse is making the Popperian point that a single observation at variance with a theory is sufficient to refute it, whereas any number of confirmations do not prove the theory or make it any more likely to be true.

[Brent Hargreaves]

Aynsley, I think you miss Nurse's point. When a theory is subjected to fierce attempts to disprove it, and it comes through intact, it becomes a solid brick in the solid edifice of science. Furthermore, 'falsifiability' is a key attribute of science: The person who can disprove that F=ma or that E=mc^2 makes a major breakthrough and science takes a leap forward.

I constantly challenge warmists to state what real-world data would cause them to abandon their CAGW theory. Nary a one will bite; this reveals them to be adherents of a religion rather than scientists.

[simpleseekeraftertruth]

He hung the lecture on his concerns for;

Food security.
Climate change.
Human health.
Sustainable economies.

Asked that scientists be listened to on these and indeed anything that they would like to be listened on while drawing on Lysenko as a warning.

As nurse is a cell biologist the take-away message for me was that if you encounter a scientist proselytising on either of those four subjects: remember Lysenko.

[Jeremy Harvey]

Aynsley, Mike, give the guy a break. Proving that something is not true does not require that you believed it to be true in the first place. The paragraph quoted above is pretty impeccably Popperian. As the Bishop points out, it is just not very consistent with consensus-upholding scientists' approach to the high-sensitivity AGW hypothesis.

[Barry Woods]

I was pleasantly suprised by it..

[Aynsley Kellow]

Generating predictions that are falsified by relevant evidence does not amount to proof that it is wrong. It establishes that it does not provide an adequate explanation. Another theory that generates more robust predictions (subjected to repeated attempts at falsification) is to be preferred. But you cannot prove a negative.

[Ian E]

'Generating predictions that are falsified by relevant evidence does not amount to proof that it is wrong. It establishes that it does not provide an adequate explanation. Another theory that generates more robust predictions (subjected to repeated attempts at falsification) is to be preferred. But you cannot prove a negative. '

Semantics really : you CAN disprove a positive (subject, of course, to repeatability, good experimental design and careful experimentation)!

[Thinkcarefully]

"Prove that you are not a witch"
"But you cannot prove a negative."

Less of this careless talk please. You can prove the square root of two is not a rational number in a few lines.

[Aynsley Kellow]

Not just semantics, Ian E. The Logic of Scientific Discovery.

[James P]

"I have to come up with ideas that can be tested"

Shame he didn't explore that in his Horizon programme, then!

[geronimo]

"I was pleasantly suprised by it.."

Leopards, spots Barry. Remember he had his fingers so burnt by the execrable Horizon programme that he is bound to be more wary now. I could be wrong, but I came to the conclusion that when he did the Horizon programme he massively underestimated the knowledge and intelligence of the people at large. Frankly it was patronising, embarrassing and irritating in equal measure. He may be thick, but he's not stupid and is taking more care than he did having got the guage of the the opposition. But remember, he's already struck the Jolly Roger, we are the opposition, and will remain so to these zealots.

[Aynsley Kellow]

Think Carefully:
' You can prove the square root of two is not a rational number in a few lines.'

That is because it is what Sir Alfred Ayer called an analytic proposition. We are talking here of synthetic propositions, those that make statements about the world. They both differ from ethical propositions or metaphysical propositions.

One problem with climate science is that too many ethical and metaphysical propositions seem to find their was into the synthetic ones.

[Aynsley Kellow]

Sorry. Should read 'find their way'

[Huroner]

Aynsley, I didn't expect to be sticking up for Sir Paul, but you have got this wrong. Your claim that you can't prove a negative misses the point. As one or two other posters have explained, Sir Paul is stating that for a theory to be legitimately scientific it must be falsifiable. He is quite correct and no one with any understanding of the scientific method would disagree, regardless of where they stand on AGW.

[steveta_uk]

Confused suggested that we should "Check out Judith Curry and Roy Spencers sites...we do indeed live in interesting times."

However, Confused did not provide any reason for suggesting this, nor any specific item of interest that we should "check out".

Methinks "Confused" is a good moniker.

[Aynsley Kellow]

Huroner:
No synthetic proposition can ever be considered 'true' - just a better explanation for the time being. the laws of Newtonian physics, despite their canonical status, were never true - they just provided better explanations.

[Bishop Hill]

I must say, I too read Nurse's remarks to be referring only to falsifiability and therefore quite uncontroversial.

[timheyes]

Andrew
"Being able to test a hypothesis in principle seem to me to be very different to actually having tested it in fact."

I agree. Compare your words here to Richard Lindzen's comments regarding model validation rather than testing in his recent presentation in Westminster. He made the point that much effort is put into proving that models give both the "right" answer compared to other models and historical temp data, rather testing them by requiring falsifiable predictions.

I want a name for this kind of "science". Pseudoscience is too perjorative as is parascience. It seems to be a quasi-science where all of the machinery to do the science is in place but the output doesn't make a prediction which can be tested. It's almost like a "cargo-cult" enterprise - but this would probably be too perjorative a term as well.

It is very worrying that this approach to science seems to be becoming accepted as normal by so many learned societies. If engineers operated like this we'd still be living under bushes scratching around for grubs.

I haven't seen the lecture yet though.

[Thinkcarefully]

Aynsley Kellow: Kant would have called "The square root of two is a rational number" a synthetic proposition, would he not? Is there not still debate about the analytic/synthetic divide?

[Selgovae]

"you cannot prove a negative"

I see this notion presented a lot these days, but fail to understand its origins. I remember being taught the difficulty of proving the non-existence of something, which is surely a different thing altogether. Proving someone is "not a witch" should be relatively easy. Witches have three defining characteristics: they can fly broomsticks, they wear pointy hats, and they have an extra nipple. The latter two are easily discoverable. On the other hand, proving the "positive" in this case will be more difficult. Although all witches possess those three characteristics, it doesn't follow that all people having those characteristics are witches.

[Pluck]

The art of science is making the truth obvious. To make the truth obvious, one approach is, using a testable assertion, to make a compelling demonstration upon which the truth hinges. Unfortunately, not every testable assertion is demonstrably true or false and the demonstration fails: the assertion may be true or false but beyond our abilities to distinguish because even with the best instrumentation and careful work, we may still fail to exclude a null-hypothesis as our result. As an approach to making the truth obvious, demonstrations based on such testable assertions is a poor approach. Lord Rutherford is noted for saying, “If your experiment needs statistics, you ought to have done a better experiment.” This is not to say that statistics are of no use to a scientist. Not so, whatsoever. Statistics are a very useful tool: they warn scientists and engineers when they attempt to over-interpret a set of observations and to make a quantitative estimate as to how much mere chance could be influencing their measurements and the calculated results. Statistics have a great utility. One important utility is allowing scientists to design their experiments so as to minimize the contributions of chance to their results. Unfortunately, not all scientists find a proper use for statistics. Some, upon their introduction to statistics early in their training, misapprehend the utility of statistics and wrongly conclude that statistics are a great way to prove things because you can treat any process as a black-box (you don’t need to understand how it works), and you can still prove things about the process. I don’t have an estimate as to how many scientists learn to think this way, but it is a fair proportion. I feel that there is a better approach to science and the art of making the truth obvious. It is an ancient approach, one mastered by Hipparchus and John Dalton (among many others): it is to make careful observations and, over time to refine those observations until, their weight leads reluctantly, then inescapably, to conclusions. These men never started out with conclusions: they tried to understand their observations and ended up discovering something that had never before been thought of.

[John Shade]

We need to tease out of Sir Paul (a socialist knight of the realm no less), a few testable theories based around CO2 as an important driver of climate. For example, if one wishes to replace theories with the operation of computer models programmed in specified ways, then one might point to the predicted hotspot in the troposphere. Or, if one was to think about heat being somehow building up in the climate system, one might devise a test involving ocean heat content. And so on. Or, one might build greenhouses of different types of glass and observe temperatures within them (oh, wait, that was done about 100 years ago!). It is, however, not up to us, the bemused observers of these scientists, to come up with such tests. It is up to them. We might well, though, ask for expert statistical advice on the rigour of their proposed data collection and analysis for each test, since it is well known that scientists in general, and climate scientists in particular, have no great skill in that area. A great many testable hypotheses come to mind for areas such as as ice extent in the Himalayas, on on Kilimanjaro, or around the poles, or sea level, or snow in the UK, or hurricanes in the North Atlantic, and so on and on. Surely, in there somewhere, are testable hypotheses? If not, and if they have not already been tested to the satisfaction of such as Sir Paul, why is he, and others like him, so profoundly convinced that we are facing a CO2-driven crisis?

[Aynsley Kellow]

Thinkcarefully, Mathematics says nothing about the world. Things are true by definition, and proof simply show the chain that connects the two things. The statement 'There is a tiger in the room' is a synthetic proposition. How does one prove it is not true? Repeated observations might produce no evidence. The proposition might look to be inadequate. But there might have been one in that box in the corner that we didn't notice. Since we simplify our observations of the world they are always partial and there were lots of 'boxes' that Newton didn't notice.

[Roger Carr]

Bishop: A side note on formatting.
Twice today I have seen posts with the tags showing, but the formatting not.
The tags have been correctly entered, but do not show when I test the text on an editor.
Mike Haseler has this: <b>It is the ability to prove that someone is not guilty which is at the centre of our criminal system."</b> above.
This leaves the tags showing, but the text in not bold (strong).
I'll keep a screen cap in case you or your webmaster wish to see.

[Thinkcarefully]

Aynsley Kellow: OK well I disagree with you on mathematics having nothing to say about the world, I would have thought the incompleteness theorem if nothing else is relevant here, but I sense that won't go anywhere...

But how about this. Is the proposition "Propositions are either analytic or synthetic" a synthetic or an analytic proposition?

[Aynsley Kellow]

Thinkcarefully, There be Wittgensteinian word games and probably ultimately postmodernism! I'd rather not go there and abandon any ability to make sense of the world!!! Call me old fashioned! :-)

[Gareth]

Nurse has provided a 3 step plan - How To Be A Scientist For Dummies:

1. Come up with ideas that can be tested.
2. Think of experiments to test the idea further.
3. If the experiment does not support the idea then reject it or modify it and test it again.

Has AGW as a whole achieved step one or not? I don't think it has.

[Roger Carr]

Pluck: "...This is not to say that statistics are of no use to a scientist. Not so, whatsoever. Statistics are a very useful tool..."

In checking the design of, say... hockey sticks?

[Nicholas Hallam]

Gareth

"3. If the experiment does not support the idea then reject it or modify it and test it again."

This methodological principle seems to be a charter for what Popper derided as "ad hoc immunizing hypotheses" which we see quite frequently in climate science.

Amazing how scientists feel qualified to pontificate on the philosophy of science with so slight an acquaintance with the subject.

[oldtimer]

I watched and listened to the lecture broadcast last night curious to hear what he would say about global warming and whether there would be any discernible response to our hosts recent critique of the past three presidents of the Royal Society.

I was not persuaded by his comparing the concensus opinion of the doctors dealing with his recent heart condition to the concensus opinion of climate scientists on global warming. I suspect there is today a much better understanding of the working of the human heart than there is on the working of the climate. Further I recall that the medical profession itself does not have an impeccable record - lobotomies, phrenology and eugenics come to mind. But later on in his lecture he did uphold the need for evidence to support hypotheses, and the need to get the science done before the politics takes over. I concluded that if/when the RS changes its position on CAGW, then he has his get out clause on the record.

[Pluck]

Roger Carr:

If we are to go into what can happen when statistics, or any calculation, is ineptly applied, there is no end to it. Statistics can indeed be correctly applied, and when correctly applied have a great use. But that use is often overlooked, and statistics and "models" are used to fashion conclusions that should not be relied on, at best the conclusions should be regarded as speculative. The hocky-stick is an instance of first inventing a "new" statistical method (one that is infact completely unsound) and then relying on the results of the unsound method to reach improper conclusions. It is not a case of disagreeing with the conclusion per se, it is a case of saying that the conclusion cannot be drawn in this way, and even if properly drawn, probably should not be relied on.

[timheyes]

Well I watched it. I didn't hear anything profound and certainly not "provocative" as David Dimbleby introduced it. Just appeals for more scientific engagement by all and more money from goverment and industry. Much of it seemed to be self promotion for his new Francis Crick centre.

Very dull.

[Harry]

Pretty straightforward stuff. many scientists are rather poor at looking at science, hence paul nurse's rather ordinary lecture, which like Rees before him probably thinks is profound.

At least it wasn't given by Brywan Cox who would have overdosed on the words 'amazing' and 'fantastic' and underdosed on content.

[stephen richards]

Richard Dimbleby would be tossing in his grave if he could see how far the BBC has descended into the mire of merde.

These current managers and journalists have totally degraded the reputation and impartiality of a once revered organisation.

[stephen richards]

Feb 29, 2012 at 10:37 AM | Harry

You forgot majestic

[John R T]

BBC World Service [USA edition] presented this morning hours of discussions, from within the Beeb: ee.g., 'Where will BBC Radio be in two years,... in twenty years?' 'What stories are we missing?' 'How do we gain/hold the attention, and how do we tell the story? ' We are even promised a big surprise, @ 2200 hrs.
Granted I was listening with half-an-ear, I heard nothing re atmosphere, weather, climate; 'global' did occur, soley in the context of BBC coverage. I particularly was hoping to hear editorial discussion of the economic disasters related to, for example, Spanish and other European pro-'green' taxes and subsidies. Maybe that will be the big surprise: the Beeb will demonstrate journalistic excellence by examining CAGW: its theories, and its effects.
I might miss the news; where should North Americans look for coverage?

[timheyes]

"At least it wasn't given by Brywan Cox who would have overdosed on the words 'amazing' and 'fantastic' and underdosed on content."

Ha ha. Have a listen to the Now Show on Radio 4 iPlayer. John Culshaw does an "amazing" Brian Cox impersonation.

[dearieme]

It's a bit bloody late for the public Nurse to affect to be a scientist, whether or not he was one earlier in his career. He and his two predecessors have done awful damage to the Royal Society, and all for thirty pieces of silver.

[Nicholas Hallam]

timheyes, re Brian Cox

The Brian Cox consequence of Pauli's Exclusion Principle (that "Everything is connected to Everything else") was so "amazing" it was wrong.

http://motls.blogspot.com/2012/02/brian-cox-misunderstands-locality-pauli.html

You might enjoy

[Mac]

I would argue that supporting science is totally different from being an activist for a certain world view based on "consensus", "settlement", "incontrovertability" and "authority". None of these have any place in real science but reflect a growing conservatism and politcally correct stance being established with climate science.

What we are seeing is that supersitition and dogma are being allowed back into science via the climate back door after they were banished many centuries ago. Any appeal by Paul Nurse to a New Enlightenment should renew its commitment to a rationality and demand openess and transparency is the scientific effort.

If academic freedom is to mean anything then it is time to stand up to those environmental activists that compromise that right to research and teach science without fear or favour.

[Pharos]

There was nothing outrageous in the speech, but again no attempt to rectify the basic cunning deception. As Lindzen has observed, there is scientific substance in the CO2 effect, but the triviality of this effect, the inherent stability of the system and the benign nature of likely outcomes is cleverly concealed, worse by adopting ominous and foreboding innuendo, alarm is fostered and promoted, and all valid debate on the magnitude of the CO2 effect suppressed.

This is neatly summarised in one of Lindzen's slides, which is adapted below.

'Carbon Dioxide has been increasing
There is a greenhouse effect
There has very probably been about 0.8 C warming in the past 150 years
Increasing CO2 alone should cause some warming (about 1C for each doubling)
There has been a doubling of
equivalent CO2 over the past 150 years
Nothing (of the above) is controversial among serious climate scientists.
Nothing (of the above) implies alarm. Indeed the actual warming is consistent with less than 1C warming for a doubling.

Unfortunately, denial of the (above) facts ... has made the public presentation of the science by those promoting alarm much easier. They merely have to defend the trivially true points (above); declare that it is only a matter of well-known physics; and relegate the real basis for alarm to a peripheral footnote – even as they slyly acknowledge that this basis is subject to great uncertainty. We will soon see examples of this by the American Physical Society and by Martin Rees and Ralph Cicerone.'

Nurse cunningly perpetuates the deceit.

[HuhneToTheSlammer]

stephen richards

"Richard Dimbleby would be tossing in his grave ....."

Exactly what I thought watching just the lead-in to the video. O tempora ......

[SteveW]

@Selgovae - best to burn them in any case, hust to be on the safe side.

@Aynsleyu Kellow

Impossible to prove a negative? Really?

I hypothesise that "All objects always fall directly away from the centre of the most massive local body."

Easily proven to be false by dropping any object, anywhere, ever (excepting very clever positioning of said object directly betwen two massive bodies, slightly further from the more massive of the two than the Lagrange point).

[Josh]

The text is here

http://royalsociety.org/uploadedFiles/Royal_Society_Content/people/fellows/2012-02-29-Dimbleby.pdf

[Perry]

I'm so impressed that you were able to perserve in listening or watching. It's not for me. Reading the comments here, I have gained the impression Nurse is recanting just enough to keep his job.

"And yet it moves!" would seem to be his underlying tone, although Galileo was right and Nurse is wrong!

Best wishes,

Perry