Bryan Caplan  

Underestimating Climategate

UC Tuition: The Revolt of the ... Hummel Responds...
Neither Tyler nor Robin think that Climategate tells us much about climate science.  Robin says that "it's news about academia, not climate":
[T]this behavior has long been typical when academics form competing groups, whether the public hears about such groups or not.  If you knew how academia worked, this news would not surprise you nor change your opinions on global warming.
I say it's news about both.  If you previously thought that climate science was not very politicized, you should change your mind - and discount mainstream research accordingly.

P.S. Contrary to Steve the comments, "consensus science" is perfectly reasonable as a presumption.  It simply isn't feasible for laymen to personally evaluate scientific research, and it isn't reasonable to ignore evidence that you haven't or can't personally evaluate.

Comments and Sharing

COMMENTS (34 to date)
david writes:

I read Tyler's post as suggesting that we should discount non-mainstream research appropriately, too.

Ryan Vann writes:

"I read Tyler's post as suggesting that we should discount non-mainstream research appropriately, too."

That is part of the joy of reading Mr. Cowen. His wording is generally cryptic enough to interpret it many different ways. He is the “choose your own adventure book” of Econ bloggers.

Snorri Godhi writes:

Having reviewed quite a few scientific papers, I discounted mainstream research a long time ago. I had an email dispute a few years back with an environmentalist blogger who thought that all academics are angels, and anybody who criticizes them must be in the pay of Big Oil; I am afraid I did not manage to persuade him that he is wrong at least on the first count.

Having said that, this scandal is different: it shows that science can be much more corrupt when a lot of political and economic power is at stake. One might have expected it, but here is the evidence.

steve writes:

I think anybody who is familiar with life as a scientist should find none of these revelations surprising or influential in making up their mind about AGW.

However, much of AGW has been sold to the public as, "Those deniers are politicized and we are just honest scientists."

Live by the sword, die by the sword.

Steve writes:

At times like these, it is useful to re-read part of Michael Crichton's lecture to students at Cal Tech

"I want to pause here and talk about this notion of consensus, and the rise of what has been called consensus science. I regard consensus science as an extremely pernicious development that ought to be stopped cold in its tracks. Historically, the claim of consensus has been the first refuge of scoundrels; it is a way to avoid debate by claiming that the matter is already settled. Whenever you hear the consensus of scientists agrees on something or other, reach for your wallet, because you’re being had.

"Let’s be clear: The work of science has nothing whatever to do with consensus. Consensus is the business of politics. Science, on the contrary, requires only one investigator who happens to be right, which means that he or she has results that are verifiable by reference to the real world. In science consensus is irrelevant. What is relevant is reproducible results. The greatest scientists in history are great precisely because they broke with the consensus.

"There is no such thing as consensus science. If it’s consensus, it isn’t science. If it’s science, it isn’t consensus. Period."

Brent Buckner writes:

Given the discussion in the dumped e-mails about adjustments to 1940s temperature record, I think it's information about both....

David writes:

Science is not trustworthy when:
1) It is in an area where making testable predictions and replicating the experiments in question is not feasible and
2) The conclusions involve leverage over a lot of political or economic power

Imagine you are a graduate student doing research for your PhD, or a young professor looking for grant money. There is a strong temptation to cherry pick your results and worse yet, the particular inquiries you pursue in such a fashion as to produce papers that are likely to result in more funding for your research. Why look for results that are just likely to get you blackballed? It is really easy to not find things if you're deliberately not looking for them.

Dan Weber writes:
1) It is in an area where making testable predictions and replicating the experiments in question is not feasible and

So much for economics.

In the real world, there is lots of science that you can do without repeatable experiments. Astronomy, economics, epidemiology, and archeology, just off the top of my head.

Not that this defends the hiding of data. Data should be public when either the public pays for it, or when the data is being used for significant social policy. Both of those apply here. It never should have gotten to the level of needing FOI requests.

TomB writes:

Every scientist in the world could agree that a theory is correct, and in the end, the theory could be totally wrong. Consensus and the number of peer reviewed articles really doesn't tell us anything. Even if reliance on consensus is reasonable for a layman, reliance is not reasonable for someone who is taking a serious look at a topic.

The real test to the validity of any scientific principle is its ability to predict and explain. The reliability of disciplines in which it is impossible run testable, repeatable experiments is diminished, and it requires more evidence to show that the claims are correct. Such as consistently explaining results or parts of results of future events.

Given the extremely long cycles of climate, and the range of climate in the historical record, I personally think there is reason to doubt the claimed certainty of scientists in regards to climate models. Especially when those models have not proven to be a reliable predictor in the short term. It wouldn't matter if every climate scientist in the world thought otherwise, so long as the models failed at predicting future climate changes.

wm13 writes:

"Contrary to Steve [in] the comments, "consensus science" is perfectly reasonable as a presumption. It simply isn't feasible for laymen to personally evaluate scientific research, and it isn't reasonable to ignore evidence that you haven't or can't personally evaluate."

I think the point (my point, anyway, if not Steve's or Michael Crichton's) is that scientists, or partisans for causes endorsed by a large number of scientists, often claim that natural science is an enterprise with more inherent trustworthiness than law, or economics, or philosophy, or whatever. So that you should take any purported consensus of economists about, say, free trade with a grain of salt, but should trust without any personal investigation the scientific consensus about the origin of species.

I would say that knowledge which one has not personally verified should play a very small part in forming one's life plan. So my beliefs about global warming, being uninformed, will not result in any action on my part.

James A. Donald writes:
"consensus science" is perfectly reasonable as a presumption.

Not, however, if consensus science supports the power of the government that pays for it.

If you see a sportsman endorsing Nike running shoes, how much weight should you give his undoubtedly expert opinion?

You are well aware that holding certain opinions is unlikely to advance your career. When you see other economists disagree with your opinions on those matters, how seriously to you take that “consensus”?

When a Nobel prize winning economist endorsed the financial stability and probity of Fannie and Freddy, did that constitute reason to have confidence in Fannie and Freddy?

Les writes:

I think that Steve's quote from Michael Crichton was absolutely correct. So-called "consensus" is seen at lynchings, in Congress, in religious doctrines, and in jury decisions - but never in science. To the contrary, everything in science is provisional and subject to being rendered obsolete.

And whatever laypersons might think has no bearing on the matter.

Finally, any expert statistician can tell whether so-called "global warming" is fact or fiction.

Amaturus writes:


Building consensus is actually a very important aspect of the scientific method. Crichton is wrong to say that consensus cannot be science. A theory certainly isn't valid just because it goes against the grain. I recommend some reading in Thomas J. Hickey's "History of Twenthieth Century Philosophy of Science":

"Research producing scientific change in the normal science phase is controlled by belief in the consensus paradigm, and the resulting scientific change is always a change within the institutional framework defined by the paradigm.
Mature sciences are distinguished by normal science, a type of research that is firmly based in some past scientific achievement, and that the members of the scientific specialty view as supplying the foundations for research. Unlike early science there are normally no competing schools and perpetual quarrels over foundations in a mature science. The achievements that guide normal science research are called paradigms, which consist of accepted examples that provide models from which spring particular traditions of scientific research. A paradigm is an object for further articulation and specification under new and more stringent conditions, and it includes not only articulate rules and theory, but also the tacit knowledge and pre-articulate skills acquired by the scientist. No part of the aim of normal science is to call forth new sorts of phenomena or to invent new theories. This conformism proceeds both from a professional education, which is an indoctrination in the prevailing paradigm set forth in the student's current textbooks and laboratory exercises, and from a consensus belief shared by the members of the scientific specialty that the paradigm seems sufficiently promising as a guide for future research and that acceptance of it is both an obligatory and a justified act of faith. Conformity to the paradigm assumes a recognizable function, which is to focus the group's attention upon a small range of relatively esoteric problems, to investigate these problems in a depth and detail that would not be possible if quarrels over fundamentals were tolerated, and to restrict the research resources of the profession to solvable problems, where the solutions are solvable precisely because they agree with the paradigm and are interpretable in its terms." (Book VI, page 2, available in its entirety online)

Consensus plays a major role in science, and it should, even if Crichton thinks it's pernicious.

Joe Weir writes:

Dan Weber,

Yes, so much for economics and epidemiology. Surely no one who reads this blog is unaware of the limitations of economics, especially macro. I come from a physiology background, so perhaps I am too fond of experimental data, but I have also been covering epidemiology lectures to health science students for a quite a while now. The problems with classical epidemiology are likely less well known to readers here, but epi is probably more of a mess than macro econ. For example, look at the hormone replacement controversy, or all the noise (and very little signal) in nutritional epidemiology. So yes, one can "do" science without real honest to God experiments, but unlike our colleagues in climate research, let us be modest in our conclusions.

nohype writes:

Anyone with a political agenda is tempted to shut down and delegitimize the opposition. The internal data in this case suggests that the temptation was not successfully resisted. However, that raises a question about economics because economics often has political implications. To what extent has economics avoided the corruption of scholarship that seems to be evident in this case? And if we have not had the same level of corruption, why not?

On a related different note, James Taranto seems to have gotten to the bottom line with this comment: "The press's view on global warming rests on an appeal to authority: the consensus among scientists that it is real, dangerous and man-caused. But the authority of scientists rests on the integrity of the scientific process, and a "consensus" based on the suppression of alternative hypotheses is, quite simply, a fraudulent one."

Peter Lentz writes:

Prof. Caplan defends and Dr. Crichton and others attack the concept of scientific consensus. I think the factions are talking about two different things.

Bryan states that “it isn't reasonable to ignore evidence that you haven't or can't personally evaluate" -- not a very controversial statement, but I can’t see what it has to do with the subject under discussion. Nobody proposes that those who are not active scholars in the field should “ignore evidence.” He is on shakier ground when he declares that it isn’t “feasible for laymen to personally evaluate scientific research." But, that subject matter, whether and how members of the public should form opinions or take positions regarding unsettled scientific propositions, is a totally different discussion from the one at hand.

What the rest of the world is talking about is a different matter altogether. The “scientific consensus” that was excoriated by Dr. Crichton is the process by which scientific hypotheses are asserted to be something more than that for the purpose of imposing a point of view about an uncertain matter. Specifically with regard to AGW a cabal of zealots declared a “scientific consensus” for the purpose of forcefully exerting institutional, political, and cultural coercion to drive other points of view from the public forum.

The actions revealed by the East Anglia data illustrates one perversion of scientific enterprise—the conspiracy of researchers active in the field to advance their agenda and to quash alternative views. A related aspect of the consensus phenomenon is illustrated by the embracing of the AGW “consensus” by those vast legions of the scientific community, not necessarily active in the specific field, who are credentialed and arguably competent to “personally evaluate scientific research.” The stampede to sign onto and validate the IPCC report is a classic example of the "information cascade." Global interconnectedness both feeds the desire and enables every academic with even a modest ego to join the glorious in-crowd.

The Wikipedia definition perfectly describes the herd mentality that has been magnified by the web. “An information (or informational) cascade occurs when people observe the actions of others and then make the same choice that the others have made, independently of their own private information signals. Because it is usually sensible to do what other people are doing, the phenomenon is assumed to be the result of rational choice. Nevertheless, information cascades can sometimes lead to arbitrary or even erroneous decisions.” Certain AGW scholars, aided by various activist communities gathered and stampeded the herd.

So when the term describes a manipulative effort by powerful actors to coercively advance an agenda or is used in reference to a malevolent information cascade, the concept of scientific consensus is appropriately criticized.

Matthew C. writes:

Hurrah for Bryan!

Yes, climategate is a big deal. And I suspect it is far from over yet. . .

SydB writes:

1. This theft of private email will soon pass and be forgotten.

2. The information contained in these emails tells us nothing new about human nature nor about the nature of competition. It's a well known fact that competing people look for advantage--even in science. A big "duh" on that one.

3. Science is largely about consensus and Michael Crichton was not really a scientist--and a bit of a loon from what I read, believing in the supernatural amongst other nutty ideas. Consensus can be wrong and science has a method in which consensus is refined and corrected.

4. A sign of weakness is not ad-hominem. To engage in it is human nature. But a true sign of a weak argument is the focus of one side of the argument on the ad-hominem generated by the other side. Global warming skeptics are currently engaged in this with respect to these stolen emails.

5. Libertarians do not believe in theft and therefore should be admonishing those who engaged in this illegal behavior.

Perry writes:


Taking the faux moral high ground here doesn't really prove much of anything.

Some very concerning things happened through the course of reviewing these emails:

1) There is reference made to manipulating a series of data to create the infamous 'hockey stick effect' for showing a trend of recently rising global temperatures.

2) They admit that certain datasets show no warming since 1998, but refuse to admit that this warrants any further discussion. Rather - "the data are surely wrong. Our observing system is inadequate."

3) They discuss ways to avoid releasing their climate model and emails relating to it if forced to through freedom of information laws.

The complete lack of willingness to even consider competing data shows at the very least intellectual laziness. Far more concerning is the huge concern placed on avoiding the release of models and data when every single thing that they are saying to the global community is based on that very data and those models. That really just doesn't pass the smell test. But then again, neither does creating a cherry picked data chart to show a phantom trendline.

Most reasonable people agree that human beings contribute to the degradation of the Earth's natural resources; air, water, atmosphere and etc. Because we are having very real and very serious discussions about these effects and those discussions could lead to policy changes that could change the very way that we exist as people, our climate scientists should hold themselves to the highest standards.

And that is why this disregard of the truth from them is such a big problem. It gives certain people a reason to think that AGW doesn't exist at all and others yet another reason to hate all 'deniers' when most of us REAL people are somewhere in the middle in need of a bit more of that elusive truth. Let the real story sell itself..

Jeremy, Alabama writes:

As in all political crimes, it is the cover-up that gets punished.

This is what the Hockey Stickers could have said:
"Our models that showed humanity-threatening warming in the next few decades have been proven wrong. Global temperatures are following a profile that the models completely failed to predict. While there may still be anthropogenic global warming, we are unable to find that signal in the noise."

But they did NOT. Scientists can be wrong. When they KNOW they are wrong, the cover-up is fraud. Since they were spending public money, their research is covered by FOI and somebody will prosecute them for it. This is not over yet.

Steve writes:

Bryan - my closing thoughts on this thread - "climate-gate" is NOT about "consensus science". No one had ever checked their work. This group of involved scientists refused to release their data sets, and algorithms used to reach their "hockey stick" pronouncement. The more scientists like Steve McIntyre and Roy Spencer pushed for release of the underlying information, the more Mann, Jones, Briffa, and others refused. Addition, this cabal, plus enablers like Gavin Schmidt, undertook a concerted campaign to slander their professional reputations. It is the "true" scientists in the Climate Science community who should hang their collective heads for not throwing the full weight of their support behind the McIntyres and Spencers.

That's my morning vent. Thanks Bryan for asking the hard questions.

Steve Spiller

SydB writes:

Perry: Please provide specific references and a brief analysis for each of the three issues you raise so we can evaluate the nature of the malfeasance. Whenever a crime is committed, it is those making such accusations who must provide the evidence.

Greego writes:

"5. Libertarians do not believe in theft and therefore should be admonishing those who engaged in this illegal behavior."

Trespass, not theft. And considering the importance of this information being made public, it was probably justified.

Perry writes:


1) There is reference made to manipulating a series of data to create the infamous 'hockey stick effect' for showing a trend of recently rising global temperatures

2) They admit that certain datasets show no warming since 1998, but refuse to admit that this warrants any further discussion. Rather - "the data are surely wrong. Our observing system is inadequate."

3) They discuss ways to avoid releasing their climate model and emails if forced to through freedom of information laws.

Links included. I've read everything with a grain of salt where there has been secondhand analysis and otherwise just read the emails themselves. You can read my original post for my analysis.

steve writes:

Ignoring AGW for the moment. This is a pretty interesting case for libertarian theories of justice.

I have read theft and trespass so far. But what about self-defense? Considering the potential degree of force to be used based on the outcome of the AGW debate can this case be made?

Also, would libertarian justice have the equivalent of FOI requests? Of course, the standard government approach when a request for information is defied would be for men with guns to show up and take your computers and other records. This has the drawback of significantly disrupting ongoing operations until the computers, programs, data, etc can be replaced.

Assuming FOI is acceptable under the right circumstances to libertarians, what would a private provider of justice do in the event of unco-operation in the face of such requests? This trespass/theft and open release of the data seems like just such a possibility. i.e. Minimal disruption to the company subject to the FOI combined with independent scrutiny of the data itself.

Yancey Ward writes:

As a scientist, nothing in these e-mails was particualarly surprising to me. I have always believed the AGW camp was guilty of confirmation and assimilation bias- there was already abundant evidence of this. The e-mails simply confirm this from that camp's own mouth.

This is a big deal since the AGW camp's primary political weapon has always been it's claim of moral and intellectual rectitude. This has been demolished utterly by the new revelations. Hopefully, some good can come of it, and both sides will freely release data and methods to all requesters. Sunlight is a great disinfectant in this case.

Snorri Godhi writes:

One concept implicit in many of the above comments is that not all consensus is created equal: the consensus that the Earth is approximately spherical and spins around its axis in 24 hours, is a lot more solid than the former consensus that stomach ulcers are not caused by bacteria.

One thing I have not seen discussed is the point of view of the scientists (if that is the right word to describe them). Suppose that you had published research that had been used to push for the Kyoto Protocol. Suppose that later on, you discover that your data is not as solid as you thought it is. You are now in a position similar to that of Nick Leeson when his speculative trading started to go wrong. What would you do? more to the point, what would you think?

Does anyone else expect the UK and USA to fall below France in next year's Corruption Perception Index?

rpodraza writes:

Piltdown man, the earth is flat, Piltdown man, the sun revolves around the earth. - all scientific consensus.

Science that will not release information, or supresses contrary opinions is not science; it is religion.

There is no "settled" science. If that were so, we would not have discovered plate tectonics (there was a consensus against it prior to Glomar Challenger data), that the rate of expansion of the universe is increasing (the data was supposed to determine a better value for the the Hubble "constant"), and cold fusion would be powering our cities.

The "setted" theory of Global Warming is reasoned somewhat like this:

My dog got out last night.

After he returned, he was sick this morning.

My neighbor does not like dogs.

My neighbor poisened my dog and should be punished.

What's missing is a connection and proof.

SydB writes:

Perry: Since you read the emails, which I just did, I assume you could write a brief summary that describes exactly what malfeasance they have engaged in. For example, he says temperatures in boulder are below average. That one desires higher temperatures to go along with his theories makes sense. That's science, standing by the measurement apparatus looking for a result that doesn't show. "Damn" they say, "what went wrong?"

Did they say they were ignoring the data? No. They said they don't believe it. What did they mean by that? Did they fudge data because of this? Not that I know of.

I would enjoy it if someone could put together a complete set of propositions that indicate the malfeasance. So far it seems, as usual on the internet, people point (meaning link) to this or that and then claim they've proven something.

So far I'm seeing a lot of smoke. I'm not saying there isn't malfeasance. But prove it. I'm not sure anyone has. Why not? Probably because there's no there there.

ps: Is it possible for everyone currently engaged in this discussion here to make available all emails they've written for the past ten years. I'd like to check them out. Thanks.

Matthew C. writes:


Just for starters the conspiracy to violate the FOI act by deleting AR4 emails. . .

By itself it shows that they were acting as clergy for a warmist religion and/or political hacks, not as scientists. Science is a method of inquiry, and it is beyond obvious that CRU did not practice it.

David Kane writes:

See here for the full context of (some of) the FOI requests.

As I said, the issue is not Trenberth or scientists talking smack. It is the illegal evasion of legitmate scientific requests for data needed to replicate a scientific study. Without replication, science cannot move forwards. And when you only give data to friends of yours, and not to people who actually might take a critical look at it, you know what you end up with? A “consensus” …

Read the whole thing. It is quite damning.

Perry writes:


Again as I explicitly stated: I am not a climate scientist and neither are you, i'm assuming. And to meet your standards i'm guessing you would want me to parse exactly what all their science jargon means, which you and I both know is impossible. But taken as a string of connected thoughts its possible to infer as a reasonable person a pattern of at the VERY LEAST ill intent.

1: "I’ve just completed Mike’s Nature trick of adding in the real temps to each series for the last 20 years amd from 1961 for Keith’s to hide the decline."

I've read a LOT of dancing around what this means and justification. To any reasonable layperson, using a "trick" to hide a decline over the last twenty years seems to say exactly what it says - data supports that temperatures are declining but instead of treating the entire dataset with one method, they have chosen to ONLY treat the last 20 years and 1961 with that 'method'. Why would you ONLY treat one specific range of data with some method and not others?

2: Again, please just go ahead and read the quoted part in the email: "the data are surely wrong. Our observing system is inadequate." To any reasonable layperson, this says that they have data that they don't understand but instead of discussing ways that they can figure out and understand this data they presume that the observing system is inadequate and that their own intuition is correct.

3: "As for FOIA Sarah isn't technically employed by UEA and she will likely be paid by Manchester Metropolitan University. I wouldn't worry about the code. If FOIA does ever get used by anyone, there is also IPR to consider as well. Data is covered by all the agreements we sign with people, so I will be hiding behind them."

Again, I don't know Britain's specific FOIA laws but I have worked in jobs where similar laws applied. The only reason that a person would feel comfort in the fact that their programmer/coder is not employed by an agency subject to the FOIA request is because they don't want that model released. You go ahead and tell me why in science you would want to keep a model private instead of opening it up for peer review/confirmation? I can't.

">> Mike,
> Can you delete any emails you may have had with Keith re AR4?
> Keith will do likewise. He's not in at the moment - minor family crisis.
> Can you also email Gene and get him to do the same? I don't
> have his new email address.
> We will be getting Caspar to do likewise."

You tell me why you would be concerned about deleting emails after being concerned with FOIA requests? There certainly are many reasons but a primary one would be to eliminate any record of emails so that the contents of those emails would never become public.

p.s. I'm not the one creating a model or any narrative that may inform public policy discussions that potentially impacts trillions of dollars of value worth of human activities. When I do, i'll make sure to act with a sense of ethics.

Perry writes:

And here an excerpt of comments from one of the programmers for the model that was posted on CBS's blog network, illustrating again that there is absolutely no good reason to keep a complex model and its data private from outside eyes for too long because errors and 'tweaks' keep on building and building until the model breaks.

I am seriously worried that our flagship gridded data product is produced by Delaunay triangulation - apparently linear as well. As far as I can see, this renders the station counts totally meaningless. It also means that we cannot say exactly how the gridded data is arrived at from a statistical perspective - since we're using an off-the-shelf product that isn't documented sufficiently to say that. Why this wasn't coded up in Fortran I don't know - time pressures perhaps? Was too much effort expended on homogenisation, that there wasn't enough time to write a gridding procedure? Of course, it's too late for me to fix it too. Meh.

[Long quote elided. For the remainder of the quoted excerpts, see "Congress May Probe Leaked Global Warming E-Mails," by Declan McCullagh, Nov. 24, 2009.--Econlib Ed.]

Jr writes:

"I've read a LOT of dancing around what this means and justification. To any reasonable layperson, using a "trick" to hide a decline over the last twenty years seems to say exactly what it says - data supports that temperatures are declining but instead of treating the entire dataset with one method, they have chosen to ONLY treat the last 20 years and 1961 with that 'method'. Why would you ONLY treat one specific range of data with some method and not others?"

Quite frankly only someone with complete ignorance of scientific terminology would ascribe any significance to the use of the word "trick". As for the reason for doing the adjustment the reason is that the tree ring data does not correlate as well with temperature measurements after 1960 as before.

Whether the adjustment is appropriate is an issue I have no way of forming an independent judgement about. But since the discrepancy is well known in the peer-review literature I do not see the significance of us finding out that scientists discussed it among themselves as well.

Comments for this entry have been closed
Return to top