While catching up on things this morning, a link at Bishop Hill took me to one of the most penetrating and concise commentaries on the Hockey Stick controversy that I have seen, and it comes from a rather surprising source.
I’m not going to attempt to summarise what it says, mainly because if I did so it would probably give the impression that the author – Sam Norton, a philosopher and country parson – is reiterating arguments that most of us have often heard before, and to some extent this is the case. The power of his post comes not from covering new ground, but from the clarity and rigour with which it brings together issues that are often discussed in isolation: the political influences that contaminate climate science, reliance on arguments from authority, and the insights that applying dispassionate philosophical analysis to a scientific controversy can provide.
If you are commenting here on what Sam has to say then please, please, lets not have yet another discussion of what Michael Mann’s work may or may not tell us about climate over the last millennium. That is not what the article is about. The Hockey Stick saga has far more interesting things to tell us about the relationship between politics, science and belief at the beginning of the 21st century than whether the 1990’s were the warmest decade for a thousand years – if that matters – and that 1998 was the warmest year.
If you consider commenting at Sam Norton’s blog, then I advise you to get all your ducks in a row first. He seems to be a very pleasant and courteous chap, but note his reply to ‘Tess’, third comment down.
Kudos to Andrew Montford (aka Bishop Hill) whose book The Hockey Stick Illusion is helping to bring what appears to be a rather grubby scandal to the attention of a far wider circle of people whose views are valuable.
PeterM
Some advice.
DO NOT get into arguments about semantics with Robin.
I did so some years ago on another thread, and learned (to my dismay) that his grasp of the English language is vastly superior to that of most English speakers (most likely due to a “déformation professionelle”).
You (or I) may be better at plotting an Excel graph with trend lines, etc. than he is, but he has us both beat hands down when it comes to the fine points of the English language.
Max
PeterM
To go back to your exchange with Robin and the apparent “disconnects” in comprehension and semantics:
Robin writes:
IOW Robin is saying that “attribution certainty” may be at 90%, but that this does not provide any level of scientific “validation”, as this requires totally different standards of evidence than simple “attribution”.
You miss this point entirely when you respond with:
Scientific validation of a premise requires empirical data based on actual physical observations as support, while attribution requires, at best, an educated guess based on theoretical deliberations as determined (for example) by computer model simulations.
A “90% certain educated guess” is a long way from a “100% validation”.
Do you get the difference here, Peter?
Max
PS Robin, I may have “oversimplified” here (at least I am not guilty of “undersimplifying”).
Robin,
To answer your 3 questions in 110 which I hope you won’t use as an opportunity to avoid the ‘validation’issue.
1) I must admit I don’t know who the 43 are. I could be wrong in my prediction that they won’t turn out be climate scientists themselves. We’ll see.
2) Are their views of “no account”? I wouldn’t say no account. Each will, no doubt, have their own speciality. I’d just ask the question of how much account they themselves would be willing to allow, if publicly challenged by others outside that speciality?
3) How do I define a climate scientist? Everyone on this blog has used the term so many times, including yourself, and so it seems a bit late to start quibbling about definitions. I seem to remember that you once said that “Few if any climate scientists” support the view that the early 20th century warming was largely anthropogenic. Who did you have in mind yourself?
Robin
Back to the RS survey.
43 RS members responded that they do not support the RS official policy stand on AGW.
There are over 1000 members.
How many members were asked for their opinion? Were these members specifically asked to give an opinion or was this simply an anonymous opinion poll?
If only 43 were asked, then one could say that the respondents unanimously rejected the official policy stand on GW.
If it was 172, then one could conclude that a significant minority (25%) of those questioned rejected the official policy stand on AGW.
Does anyone really know how many were asked?
On the other hand how many respondents specifically stated that they endorse the official policy stand on AGW?
A meaningful survey would list:
a) the number of respondents who stated that they specifically reject the RS policy position
b) the number of respondents who stated that they specifically endorse the RS policy position
c) the number of respondents who did not state a position or did not return the survey form
If a) is greater than b), one could conclude that a greater number of members reject rather than endorse the official policy position on AGW.
Does anyone know how this poll was conducted and what it really means?
Max
Peter M, re your #121, I think I know what you mean about the PCBs; no personal experience in my case, but widely reported PCB failure is one of the reasons why I’m still hesitating to buy an energy-saving gas boiler – the technology isn’t exactly at Star Trek levels yet!
Back on the subject of physics (as in “ye cannae change the laws of”) I’d be curious to know your reactions (and those of anyone else here, come to that) to a recent article (link here) regarding the Stefan-Boltzmann formula and the greenhouse effect. As far as I know, the authors have just published this on the internet, rather than via Nature or a similar journal, so whether it passes muster, scientifically speaking, I’m not sure at all. I’ve forgotten much of my secondary-school physics, so don’t have much to say about it that would be useful. (TonyN, please shift this comment to the main thread, if you consider it belongs there instead.)
Max,
Yes maybe Robin is good at playing with words but he doesn’t clearly understand the scientific, or even the artistic, meaning of ‘attribution’.
An art historian, or scientist might say that it is “just possible” that object, or event X, was caused by person or influence Y.
Level of certainty – Less than about 25%
“possible” – Between 25% and 50%
“probable” – Between 50% and 75%
“very probable” – Between 75% and 90%
“very likely” -90% and 95%
“almost certain” above 95%
The word “attribution” would also start to be used withe the word “probable”. At first tentatively but then with increasing confidence.
Robin is implying that “validation” is also necessary and requires nearly, if not actually, 100% certainty. That’s a requirement for absolute proof! And you can’t have that with any historical art-work or over the science of historical climate, which of course is crucially important for the understanding of future climate, either.
PeterM (#125):
It’s strange that, after all this time and despite all your professed focus on the science, you still seem not to understand how the Scientific Method works. Here’s a simple summary:
So you see, “proof” is never possible – as you acknowledge. But what you fail to understand is that, because proof is never possible, the concept of 90% certainty is meaningless in the context of the Scientific Method: evidence either survives the process outlined above or it doesn’t. And, even if it survives, new evidence may subsequently show the hypothesis to be false.
(In #127, Max explains clearly how “attribution certainty” differs from “scientific validation”: it’s the difference between an educated guess and actual physical observation. He neither oversimplified nor, wait for it … “undersimplified”.)
So “sufficient evidence” is, quite simply, evidence that has successfully survived the process outlined above, thereby verifying the hypothesis. “Insufficient evidence” is evidence that has failed to survive that process. And obviously not producing any evidence doesn’t even get to the starting block. Both are failures.
PeterM:
My #132 applies as much to your #131 as your #125: for the reasons explained, percentages are meaningless in the context of the Scientific Method.
PeterM
You are missing the point.
It is not about “playing with words”.
It is about “understanding” them.
I think I explained the difference between “attribution” and “validation” (127).
Your discussion of “levels of certainty” is interesting, but these are totally subjective. Richard Lindzen might assign a totally different “level of certainty” to a projected future climate event than, for example James E. Hansen. Whose “level of certainty” is correct, and why?
IPCC uses this type of categorization, calling it the “assumed likelihood, using expert judgment, of an outcome or a result”, and assigning numerical “probability” estimates (in percentages) to different past or projected events.
In its assessment of “human influence on past weather phenomena and trends and on projections for future extreme weather events” (Table SPM.2, SPM 2007), IPCC lists the assumed “likelihoods” for these, with the footnote that the “magnitude of anthropogenic contributions not assessed” and “attribution for these phenomena based on expert judgment rather than formal attribution studies”.
This boils down to qualitative (as opposed to quantitative) “attribution” by “educated guess”.
In science, “validation” of a hypothesis is something totally different. It is not based on “expert judgment”, “educated guess”, “theoretical deliberations supported by model simulations”, etc., but on “empirical data, derived from actual physical observations”.
This is what is lacking for the premise that AGW has caused most of the past warming and represents a serious potential threat.
And no matter how much you beat around the bush, you have so far been unable to refute this by citing such empirical evidence.
Max
Max #129
The RS survey wasn’t a “poll”. It was just a number of members asking around among their acquaintances. About a third of those asked refused to sign.
Information from a commentator at Bishop Hill (I think)
Robin and PeterM
Looks like Robin’s posts and mine crossed, so may be a bit repetitive.
Are we thus “undersimplifying”?
(I hope not.)
Max
PS I personally like the word “undercomplicating” (with opposite meaning, of course, better.
geoffchambers (135)
Thanks for clearing up the RS survey.
Looks like 43 members (who actually went out of their way to state opposition to the official RS “party line”) is a pretty strong indicator.
Had there been significantly more that 43, who specifically endorsed in writing the RS policy statement, this would have indicated that the 43 were just a (possibly disgruntled) minority.
But since there were no specific “pro” votes, and the abstainers may have had all sorts of reasons for abstaining, it appears that the 43 “nay” votes to the RS “party line” will have to be taken seriously by RS management.
Let’s see how they revise the official RS blurb on AGW (if at all).
Max
Max (re your #129 – BTW arguably this, the “validation” discussion and Alex’s comment should be on the main thread):
I understand (from Harrabin’s article) that there was no survey involved in the RS development. Harrabin quotes one Fellow as saying:
So it would seem to be impossible, at this stage, to answer your questions. It’s interesting however that, according to this report, “a third of those [approached] declined to sign the petition”. But I don’t think it’s possible to draw any useful conclusion from that.
PeterM
In your 128 you questioned whether the 43 RS members who stated that they did not endorse the official RS “party line” on AGW were “climate scientists”.
I do not believe that this makes any difference, Peter. There may actually be a better chance of getting an objective opinion from scientists who are not “climatologists”.
The study entitled “Seductive Simulations”, by Myanna Lahsen,, which I cited on the other thread, “challenges the assumption that knowledge producers always are the best judges of the accuracy of their models”, in other words the climate scientists who produce model simulations may actually be worse judges of how accurate or meaningful their model results actually are than others, who use or analyze the model results.
This phenomenon has been explained earlier (in a more general sense) by Thomas Kuhn, in his treatise on paradigms in science. Data points lying outside the prevailing paradigm are ignored or rejected. In some cases, they are actually physically not seen, according to Kuhn.
Lahsen does not go into a discussion of “agenda driven science”, a second reason why climate scientists might be less objective than others is assessing the accuracy or validity of model-based climate projections. I personally believe that the many recent revelations (Climategate, IPCC exaggerations and fabrications on glacier loss, African droughts, etc.) show that this reason is just as important as the ones discussed by Lahsen.
Max
geoffchambers (135):
Sorry Geoff – you got there first.
And, Max, it seems from Harrabin’s article (see my extract above) that RS management are indeed taking it seriously.
PeterM (#128):
Re the three questions:
1. Yes, your prediction may be wrong. It seems (see #138) that they are “mainly in physical sciences”. But it rather depends on the definition of “climate scientist” – see 3. below.
2. I would expect that, whatever their specialty, they would be open to reasonable and courteous criticism from colleagues outside that specialty. That’s how good scientists operate: as Max points out, it can be possible to get a more objective view from someone outside the field in question.
3. “Quibble” is the wrong word. I’m really interested to know what you mean when you refer to a “climate scientist”. Let me know and I’ll comment – referring no doubt to my view of the matter. I look forward to your answer. Thanks.
Further to my #124 it seems that, for most of its history, the Royal Society may have tried to avoid the characteristics of a public body identified by Hazlitt. Its Philosophical Transactions (the Society’s journal) used to forbid pronouncements by the Society as a whole on any scientific or practical matter:
In my view, it would have been wise to stick to that. As it would have been wise to stick its original motto, Nullius in verba – usually defined as “Take no one’s word for it”. (And not, as once suggested, “never put anything in writing”.)
Er – I didn’t mean “stick its original motto” (that’s what it did). I meant “stick to its original motto”.
Robin,
I seem to remember pointing out to before that there isn’t a single scientific method. For example, experimentation is of course desirable and should be included in the method wherever possible, but sometimes, and not just with climate science, it isn’t.
You brought up the comparison between climate science and art attribution which is a good one. But, are you saying that there is a fundamental difference in methodology between a 90% attribution from a climate scientist and an art historian? Yes/No answer please?
PeterM
Sorry. There is a single scientific method, regardless of what you may have “pointed out” previously.
“Attribution” is fundamentally not equal to “validation”.
Max
Alex, Reur 130, part 2 on S-B law.
I’m about to make a comment on the NS thread
Max,
Of course there isn’t a single scientific method. A major difference would involve experimentation. This is clearly sometimes possible and sometimes not.
Can’t you see that?
If you can’t you might want to read in the sceptics dictionary (where better?)
“There is no single scientific method. Some of the methods of science involve logic, e.g., drawing inferences or deductions from hypotheses, or thinking out the logical implications of causal relationships in terms of necessary…..”
Link to sceptics dictionary
PeterM
Not everything done in science is done following exactly the same methodology or procedures, of course.
But the “scientific method” has a fairly restricted definition. This is one of the key differentiators between “science” and “pseudo-science” (or as Carl Sagan called it: “bamboozle”).
Wiki gives a fairly good summary of the “Scientific Method”:
http://en.wikipedia.org/wiki/Scientific_method
A key point is expressed in the sentences repeated here:
It seems pretty clear to me, Peter.
Max
PeterM (your #144 and #147):
In my #132, I set out a summary of the Scientific Method. Note the word “summary”: the practice of the Method is complex and varies depending on the field of investigation (e.g. Darwin observes the natural world and particle physicists set up the complex experiment at CERN) – perhaps that’s what “the sceptics dictionary” had in mind. Thus Wikipedia refers to “a body of techniques”. It goes on to say:
[My emphasis.]
Hmm … no reference here to “attribution”, I see. It does, however, add something that some “climate scientists” seem to have forgotten – i.e. that there is a
You ask me about art historians. Well, it may be regrettable but, so far as I can see, they do not pursue a code as rigorous as do scientists – indeed the Scientific Method is a factor that differentiates science from other fields of enquiry. Or at least it does when science is properly conducted. Re your question – as already pointed out by Max and myself – for a “climate scientist” to make an attribution is essentially meaningless (see in particular Max’s #134). Therefore, so is your question.