Hacker News new | past | comments | ask | show | jobs | submit login
The Global Climate Change Consensus: Peter Norvig's Experiment (norvig.com)
33 points by b-man on April 21, 2010 | hide | past | favorite | 28 comments



This was possibly interesting ... before the leak of the emails that made it crystal clear that the "consensus" group systematically kept the other side out of the peer-reviewed literature.

I'd also note that "consensus" is not a word you find in real science; one inconvenient fact can destroy the position of a consensus, although as Kuhn says (http://en.wikipedia.org/wiki/The_Structure_of_Scientific_Rev...), grossly oversimplifying, you might have to wait until enough of the members of the consensus retire or die.


Your point is grossly biased. You are equating the handful of people whose communications were included in the leak with the authors of (according to the article) at least 900 (papers that support the "consensus".

Given that the IPCC AR4 WG1 report that talks about the scientific basis for climate change is based on of order 1000 citations for each of the 10 chapters, I think you are significantly underestimating the body of science underlying the conclusions of the IPCC report. To think that one research group could significantly bias this, you must think they are powerful indeed.

You are also equating them talking about doing something with it actually happening. People talk about doing stuff all the time, so unless you can show evidence that a) they actually tried to act on their statements and b) were in a position to exclude papers from the body of scientific literature, I find no reason to think that the body of published studies is systematically biased.

Furthermore, the statements can also be interpreted as attempting (perhaps overzealously) to exclude shoddy science, not furthering a personal vendetta.


> The leak of emails that made it crystal clear that a handful of members of the consensus group, on a handful of occasions, talked about keeping the other side out of the process of peer-review.

There, fixed that for you.

The consensus of biologists is that evolution is true. The consensus of astronomers is that the Big Bang happened. The consensus of archaeologists is that the KT extinction event happened and that it was due in significant part to the Chicxulub impact. The consensus of physicists is that quantum physics and relativity are accurate models of the universe at the appropriate scales.

The consensus of climate scientists is that the planet is warming and it is anthropogenic. Why is this fact irrelevant here, but not in other fields?


The talking only confirmed their community's observed actions up to that point. Anyone paying attention knew what they were doing, the emails just made it "crystal clear".

And it's funny you should bring up Chicxulub and the K-T extinction event. When the Alvarez team proposed that hypothesis, it was not ... warmly welcomed. But they were allowed to argue their point and the community eventually agreed they were (likely) correct. If their work had been in the climate science domain I don't think we would have seen the same outcome. I know we wouldn't have seen the same process.

With regards to the "consensus" that the "planet is warming", that "fact" has not been true for the last decade or so, so that "fact" is indeed potentially irrelevant (we need to wait a bit longer to be sure).


Regarding the "planet has not been warming for the last decade" claim, here's why that doesn't disprove global warming, even if true (tl;dr: 1998 was an anomaly):

http://scienceblogs.com/illconsidered/2006/04/warming-stoppe...


> If their work had been in the climate science domain I don't think we would have seen the same outcome. I know we wouldn't have seen the same process.

But of course, that is entirely your own conjecture. You "know" this for a fact.


I "know" this as a "fact" of human behavior. Yes, it's an appeal to authority (myself :-), but I've been watching politicized science for about three and a half decades so I'm pretty confident about my judgment.


Because more money is involved. Because this is actually actionable and requires immediate consideration in long term business planning.

And maybe, with some people, because they don't really like the idea of shouldering that much responsibility.


From wikipedia: With a budget of 9 billion US dollars (approx. €6300M or £5600M as of Jan 2010), the LHC is one of the most expensive scientific instruments ever built in human history.

The Large Hadron Collider is pretty expensive! Why does nobody ever accuse particle physicists of "making up" the Higgs?


$9 billion is nothing compared to the scope of government or environmental spending. In the lifetime of the LHC, US oil alone will process literally tens of trillions of dollars. Making policy decisions about a system that big has nontrivial financial impact. :)

Personally, I like to believe the LHC is too awesome to fail. After the SSC debacle, nobody wants to back out of a second collider. :)


The LHC crew are on safer ground. If they don't find the Higgs, they're still likely to find something interesting. (After all, various people have been worried they'd find something a bit too interesting. :-)


I'd also note that "consensus" is not a word you find in real science;

As someone who has done real science (albeit in quantum mechanics, not climatology), I call bullshit.

We use "consensus" all the time. There's also an understanding that consensus has historically opposed better models, which suggests we all need to keep an open mind. That doesn't change the fact that often, scientists mostly agree on stuff.


Even when I was a pure mathematician, where a layperson might not expect it, we used the word "consensus" all the time.

"The consensus is that Foo's proof, once checked, will be found to be correct".

"The consensus is that the Wibble Conjecture must be true, even though nobody has any idea how to prove it yet".

etc.


In reference to both the comments up thread, real scientists don't use a "consensus" to bludgeon someone bringing up an inconvenient fact.

WRT to the above, if Foo's proof upon checking turns out not to be correct, real scientists/mathematicians will prefect or throw away the proof, not the check that falsified it.

When someone says "The consensus of climate scientists is that the planet is warming and it is anthropogenic" and it's pointed out that contrary to their very specific predictions the planet has been cooling for the last decade, what do you say real scientists should do?


To respond to your example:

Point out that we don't have a high confidence time series that says the planet is cooling over the last decade? For that matter, nobody has a 10-year time series for global average temp which passes any real significance test. It's just not a long enough period to deal with high frequency fluctuations: you need 30 or so years for that.

(Actually, to the best of my knowledge, the last decade's average surface temp has been rising, even if the significance is below two-sigma.)

To respond to the general argument, you're absolutely correct. Proofs, statistically rigorous arguments absolutely call into question the consensus. However, vague counterarguments are not sufficient to overturn carefully gathered evidence. One must weigh the arguments on the strength of their evidence and analysis. There is just not very much substantial peer reviewed research supporting a climate model with zero radiative forcing for CO2.


"There is just not very much substantial peer reviewed research supporting a climate model with zero radiative forcing for CO2"

I wonder why that is true? (See the start of this thread if you really wonder why, although I'll add that you just don't get funding if you disagree with the "consensus" (that based off a friend's interview of MIT professor Lindzen at the end of the '80s).)

I agree and have noted elsewhere in this thread that the observations of the last 10 years are not definitive. However, they don't match the confident predictions of the "consensus" and the leaked emails show them quite concerned about that.

Which suggests keeping an open mind about all this instead of declaring the science is settled and using the "consensus" to bludgeon people who disagree, keep them out of the peer reviewed literature, call them "deniers" and call for a new Nuremberg, etc. etc.


Are you aware regarding to which specific points the other side, as you say, disagrees with the majority view? I have found virtually every single mainstream article about climate change to be spin, I'd like to see for myself.

The wording of the consensus view may also play a role. How much in "Human activities ... are modifying the concentration of atmospheric constituents ... that absorb or scatter radiant energy. Most of the observed warming over the last 50 years is likely to have been due to the increase in greenhouse gas concentrations." is there to disagree on? It's enormously vague, is it not?


"The other side" seems to push a number of arguments from the hypothesis that the greenhouse effect violates fundamental physics (it doesn't, no serious scientist would argue that no matter how much they disagree with the IPCC conclusions) to disagreeing on whether the increase in CO2 is driven by warming climate instead of the other way to arguing about the specific numbers of projected radiative forcing and climate sensitivity.

But you are correct. If the official IPCC estimate is that the increased radiative forcing due to greenhouse gases is [+2.07 - +2.53] W/m^2, are you "disagreeing with the consensus" if you think it's +2.06? What if it's +1.5? The real world is not binary, it's a matter of degree.

Edit: For a list of things that people disagree on, you can look at the "How to talk to a climate sceptic" post, which regardless of whether you believe the responses or not is a pretty comprehensive list: http://scienceblogs.com/illconsidered/2008/07/how_to_talk_to...


Thanks! Was looking for a link like that.


For those of you downmodding hga, the leaked emails contained conversations between some of the researcher that specifically talk about how to keep a scientist who disagree with them out of the peer reviewed press.


I think we have to stop considering Climate Research as a legitimate peer-reviewed journal. Perhaps we should encourage our colleagues in the climate research community to no longer submit to, or cite papers in, this journal.

The other paper by MM is just garbage. [...] I can't see either of these papers being in the next IPCC report. Kevin and I will keep them out somehow — even if we have to redefine what the peer-review literature is!

I don’t really understand what’s incriminating about this. Please explain.


I just want to make the point that "Me talking about how to keep you out of HN" != "you being kept out of HN".


I don't have the time to double check this hypothesis, I'd feel fairly confident that after reviewing the titles of the journals that these articles were published in, that the bulk of the authors are not climatologists, but rather have expertise in other fields such as biology, meteorology, or agronomy. And that their work speculates on the impact of "climate change" on their respective fields, and that they rely heavily on the work of a few select climatologists.

I think it is also very important to remember that we're likely seeing a very large selection bias with respect to what kinds of papers get published. There are very few institutions that will give research money to a scientist who thinks that everything "will be ok". The people who are yelling that we're all gonna die and the earth will be destroyed (unless you give them money), are the types that get funding.


I'm not sure I understand your last point: Usually when you ask for research money, you don't know what the outcome of the study you want to do is.

Besides that, I'm not even convinced that you are right: I would think that a whole bunch of people would be interested in concluding that AGW is not a problem, given the political climate in the US especially. If you could convince everyone that the scientists got it completely wrong, I'm pretty sure even most of them would be relieved. It's just that you can't do that by attacking one argument here and one methodology there. You need to present a comprehensive and consistent framework that shows that AGW is not happening, and if someone did that I'm pretty sure it would be taken seriously. But noone does.

Instead of grasping at straws, why don't you just accept the maximum likelihood solution: most of the papers are correct and global warming is a problem.


Selection bias is not necessarily a bad thing. I bet the articles and papers are also skewed in that they disproportionately aren't written by folks that think that there are large bears on Mars, invisible unicorns in Kansas and that wrestling is the greatest sport ever.

Just a comment on one of your points. On the whole, I think your hypothesis is reasonable.


I found Norwig's little experiment interesting. Even more interesting is the response it raised with our local HN deniers who seem more interested in discrediting global warming and climate change than in building our understanding of the current state and the climate. They are the moral equivalent, in my opinion, of the folks who dispute evolution.

The denier culture is sociologically interesting. Those interested in the process should a look at John Mashey's report

The current(185-page) PDF, has a much higher production quality and a lot more information on funding and activity patterns than earlier versions. The current version is at:

http://www.desmogblog.com/crescendo-climategate-cacophony John suggests that you read the first 4 pages, and, then, if you want to read more, do the following:

a) Print pages 2-4, the navigational aids

b) Download the full PDF, and for on-screen reading, open a second window (Acrobat: Window>new). Use one window to read the mainline narrative, and the second for rummaging the Appendices. Different people are familiar with different subsets, so I know of no way to linearize it that makes sense.

A few highlights:

p.167-168 on plagiarism:

Not just the tree-rings, but a big chunk of the “social network” part of the Wegman Report seems plagiarized from the Wasserman and Faust(1994) textbook. Deep Climate’s 4-page side-by-side is:

http://deepclimate.files.wordpress.com/2010/02/wegman-social...

That one is interesting because it seems unlikely to have originated with McIntyre and/or McKitrick.

DC is doing a further piece on that, but after that, I may look into letting those authors know.

See Figure 2.1, p.10, for the overall flow of anti-science memes and money-laundering “cloud”

See Fig A.2.2, p.46 ExxonMobil & Foundation Chronological Funding for Some Think Tanks

That identifies only visible funding for (Annapolis Center, CEI, CFACT, GMI, and Heartland), which leaves ((84%, 78%, 53%, 36%, 87%) unidentified, but it is certainly enough to be interesting. Someone with subpoena power could find out more, as there are many potential funding routes.

For amusement, see Fig A.3.1 p.50 “What’s in a Name?” to see how often names like “Institute” and “Science” pop up in entities that basically do PR and lobbying, despite mostly being tax-free 501(c)3s. I don’t know the (murky) 501(c)3 law enough to understand whether the amount of lobbying they do is OK or not …

However, here’s an interesting GoogleMap:

http://maps.google.com/maps/ms?hl=en&ie=UTF8&msa=0&#...

Hint: zoom in and see how many of these are located within one block of a Washington’s K Street. Not every organization located there does lobbying, but I doubt that it is a low-rent district.

See Table A.6.2, especially p.96-99 which lists active people versus visible anti-science activities, more or less chronologically from 1990 to now (Climategate), including continuing attacks A.Santer, A.Oreskes (on those folks), A.GATE (the current climategate etc, might be called A.Jones), and then shows organizational connections. I’d guess someone might do some more social network analysis.

At point, I was going to do PeopleXpeople … but in that group, no one was more than 1 hop away

The attacks on the hockey stick (which could also be called A.Mann) are now split:

A.Hockey is the 2002-current visible use of attacks on the hockey stick as a pillar of climate anti-science, basically used by almost everybody in Figure A.6.2(a).

A.HOCKX is my label for the 1998-2006 effort culminating in the Wegman Report, much of which was behind the scenes.

That’s the part that may well be investigated for 18USC1001 (misleading Congress) and 18USC371 (conspiracy) as per A.14, Possible Legal issues, p.184.


It would be interesting if some topic in the field of software got as much public attention and debate by non-practioners as does climate change.

On one side (because there have to be two sides, and presented in a "fair and balanced" way, of course), you'd have the vast majority of working programmers all agree with the position that, say, writing software in a high level language is much more productive and useful than writing everything in machine language. And on the other side you'd have some non-programmers who are "skeptical" and coming up with all kinds of arguments why machine language is best. The skeptics would point out what they say are flaws in the reasoning, gaps in the data, etc.

This line of thought is not meant to argue that man-made global warming or catastrophic climate change are true and urgent, but rather to argue that perhaps that it doesn't make sense for non-scientists to take a stand on this issue -- especially taking a stand against the position of those who have vastly more knowledge and experience with the topic. Might they be wrong? Of course. But they are probably way more likely to be right than the average bloke on the street.


Non-scientists can take a legitimate stand on the methods, e.g. the statisticians from Canada who destroyed the hockey stick. We'd see even more of that if the data, code and methods were public (pity the original raw collected data was lost (much of it might be recoverable if someone were to go back to the original national weather service sources)).

I think you're also misrepresenting the nature of the debate; it would be a lot more like everyone arguing about managed (GCed) code vs. unmanaged (C/C++/Objective C etc. with manual memory management). The other side, and plenty of them are scientists and even specialists in this area, are not arguing the equivalent of "machine code is best", an idea that died very shortly after the advent of symbolic assemblers.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: