Gary Farber, of Amygdala fame, is back. He can't leave the game, the game need him.

In the letter I posted, Robert Lyman makes at least five points that I find completely persuasive:

1. Climatologists' models have very poor predictive value.
2. You can't test a model on the same data that you used to construct the model. (Of course, you can wall off some of the data, build your model, and test it on the reserved stuff. That’s what I’d do, anyway.)
3. There's only one, spotty set of climatological data to use.
4. Climatology is wicked hard.
5. Grants are often easier to get for following in the footsteps of a respected member of your field. (Although, I still maintain that revising a published model is great for the career of an ambitious academic, much better than just measuring something more precisely. I saw that during my academic career, but I don’t want to push it too hard.)

Here's the biggest thing that makes me balk. These statements are not only true about climatology; how many fields have successfully modelled complex systems? I don’t think that anyone can do it (I’m sure I’ll be corrected if I’m wrong). That doesn't make fields that study complex systems useless and invalid. Even if you can’t predict how a variable will move, you may be able to say intelligent things about its relationship with other variables. Let’s go back to economics:

Ask an economically literate person, "Does demand for a product increase when prices decline?" and that person will say yes.

Ask that person, “Can the government promote inflationary trends by decreasing the prime rate?” The person will say yes, perhaps with some caveats.

Now, ask the greatest economists of our generation, “What will the US GDP be in 2015?” If they have honest bone in their body, they'll just laugh in your face.

I made this argument in my comments: Economics, like climatology, is another field built on modelling infinitely complex systems with finite equations. Experiments can’t be done, and the long-term predictive validity of economics models are somewhere between “bupkus” and “ehhh.” But we constantly shape government policy and political arguments around the wisdom of economists. This is not a bad thing. In my mind, (a) economics can teach us a lot of useful things that should shape our decisions, but (b) economists suck at predictions. (B) doesn’t invalidate (A).

I have some experience building economic models. Enron spent mind-boggling amounts of money building them and buying them. I, personally, never seen one that had significant predictive value; our team certainly never came close.

But, but, but. We were able to make arguments that we had found patterns and relationships. "When variable X increases by 100, Y increases by about 10" and so forth. We couldn’t tell you when X was going to increase, but when it did, we’d bet that Y would increase as well.

Robert says that climatologists’ models can’t be disproven, because they don’t fit the data in the first place. But statistical models don't get thrown out after a binary decision about whether they do or do not fit data. They just get endlessly tweaked, as people try to improve their predictive validity with the fewest variables. That's what those hungry PhDs should be doing- inserting or removing variables into the models, seeing if they improve the validity of the results, and making the argument about why the new variable is a logical addition.

I don’t know jack about climatology, but it seems to me that all these things could be true.

1. Climatologists’ models of the atmosphere are woefully incomplete and can’t predict the future.
2. But, they have been able to establish strong evidence that there’s a global warming trend, and they’ve established a strong relationship between global warming and man-made release of greenhouse gases.
3. Given a reasonable estimate of greenhouse gas emissions, climatologists can make an educated guess about the effects of global warming, and we should probably pay attention.

As a footnote, I only recently realized how ironic this argument is:

1. The models that the climatologists use are incomplete and have poor predictive validity.
2. And, proposed efforts to fight greenhouse gases will badly hurt our economy.
3. Just look at these economic models that prove it!

*********************************

Heather* McArdle has a long, reasonable post about GW here. It's getting late, so let me just say two things:

- After saying several nice things, Heather says this about my post:

Et tu, Ted? People who disbelieve global warming or other environmental disaster predictions belong to that majority of Americans who do not want polluted water, ravaged landscapes, or the earth's mean temperature increased above the boiling point of water. No matter how much it may feel like it, neither side has staked out the "against" position in the "total destruction of life as we know it" debate. Everyone is for the environment, in abstract, just as everyone is for mother love and puppies. It's not a question of whether or not most Americans are against those things, but of what they are willing to pay to avoid them.

With all due respect to Heather, duh. If I sounded like I would disagree with that, all I can do is apologize.

- She also says this:

GW non-believer groups have considerable data to back up the conclusion that halting anthropogenic global warming would be disastrously costly. They hurt their scientific credibility, however, by resisting comparison with the costs of global warming.

I don’t want to assign her homework, but I’ve long been curious about that. I've often wondered about a good estimate of the economic costs of various GW predictions would be. I don't think I've ever seen them compared to the estimated costs of Kyoto. Obviously, the economic effect of rising water and increased droughts wouldn't be negligible. This kind of open-ended modelling is ripe for mischief, but I'm still curious. Anyone have any links?

* Megan. Jeezus, Ted...

I got an email from physics grad student Robert Lyman the other day with a serious challenge to man-made global warming. I have some significant points of difference with him, but he's got a lot of good things to say, and I thought that I should share it. I'll post my reply as soon as I get time.

I can remember the exact moment I stopped believing in anthropogenic global warming as a serious scientific crisis.

I was at a conference about the interaction between the ocean and the atmosphere. There was a guy giving a presentation on his model and what would happen in the next 50 years in terms of temperature rise.

I wasn't even supposed to be there; my wife was on a Fulbright doing work on oceanographic chemistry; I was just tagging along, helping set up and take down equipment, hearing a few talks.

So this guy compared his computer prediction from 1800-2050 with the known data. About 1940 the data have a little hump which his prediction doesn't have--he writes it off as "sunspots" or some such excuse. I'm in the audience thinking "so you have a known warming mechanism which you DON'T INCLUDE in your model????"

At the end of the talk, he expressed frustration that his model wasn't stable--it wouldn't run at a constant temperature, it always insisted on warming or cooling, depending on the assumptions. I raised my hand and got to ask the first question.

"What evidence is there that the actual climate will run at a fixed temperature, given what we know?" Given that Europe's "Little Ice Age" didn't end until the mid-to-late 19th century, constant temperatures there in the period 1800-2050 seems unlikely.

Well, he blew me off. He has no such evidence. He had no idea what warming or cooling was occurring naturally. He ignored the bump in actual temperatures instead of trying to make his model fit it.

The room let him get away with it. The asked all sorts of fascinating technical questions about how his model worked, and what sorts of ocean-atmosphere interactions he was exploring. The fact that his model clearly DOES NOT WORK and is USELESS as a predictive tool was not addressed. Actually, they reminded me of a bunch of leftists discussing the nuances of the latest welfare bill--lots of minutiae, lots of vague theory, no actual relationship to, you know, the REAL world.

"Climatologists who use seriously flawed models are at the mercy of peer review before publication, and every hungry PhD. after publication."

My evidence is undeniably anectodal, but I have good reason to doubt your claim here. "Peer" review can easily mean "one of the guys who agrees with you."

"Scientists are in desperate competition for fellowships, grants, postdoc positions and actual jobs. "

And the best way to get such things is to STICK TO CONVENTIONAL WISDOM! Another conference: the "Nobel Prize conference" in Lindau, later that year. Scientist after scientist decried the tendency of funding agencies and journals to take the "safe" route--fund projects which are likely to succeed, do NOT fund projects without clearly defined goals. "I intend to measure X with the same technique I used last year to measure Y" will usually get funded. "I intend to try something no one has ever tried, just to see what happens" rarely gets funded. Also, "Last years I measured X to 2 decimal places better than anyone else" is considered a success, whereas "Last year I did something speculative, and got no measurable results" is considered a failure.

When I pointed out to Bill Phillips of NIST that HE, as a NOBEL PRIZE WINNER, was probably the first guy anyone asks when the want an opinion on what to fund, he admitted that yes, he had been part of the very problem he was decrying.

It's worse in climate science because anyone who doubts the global-warming shibboleth is a "conservative," which is worse than being a terrorist (ahem, freedom fighter) in some circles. At the atmosphere-ocean conference, I had several people tell me that they distrusted the modelers, but since the modelers make predictions, they get attention and money from politicians--and publically rebuking them means the threat of lost funding.

"They'd love to earn their bones by taking down the top models and theories. "

HOW? How can you possibly falsify a climate model? Not fitting the data doesn't disqualify you. What on earth would? In any case, it is an axiom of science that one should not test hypotheses with same data with which they were developed, yet only 1 very incomplete data set is available for anyone to use. I have occasionally thought that I could fit some polynomial to the temp data for the last couple hundred years and then extrapolate, and probably get published.

BTW, my wife's work was to determine if a particular bacterium produced a particular class of chemicals. The answer apparently has an impact in global-warming studies--and the climate-model freaks had ALREADY included these bacteria in their calculations on the ASSUMPTION that they did, in fact, produce the chemicals in question. The had no evidence, but somehow it "felt right" to them.

Climate science is the toughest thing we have ever tackled as a species (that is, in the academic realm). Nobody knows anything particularly meaningful about the coming century; let's all hope for the best.

J Bowen from No Watermelons has a detailed post about the costs of nuclear power.

This just in: Greenpeace Will Now Oppose Everything

"It's all bad, it all needs to stop," said a Greenpeace spokesperson, who added the group will no longer send out action alerts calling for opposition to specific issues, but will instead issue daily alerts to all members that read, "No" in 37 different languages.

The new directive took effect immediately after midnight, as the famous Greenpeace ship Rainbow Warrior was ordered to oppose the first thing available, which turned out to be Barbados.

"Stupid Barbados. You must be stopped!" yelled wild-eyed Rainbow Warrior captain Niels Sturngen as he drove the bow directly into what turned out to be a beach. "Surrender," Sturngen added.

It just gets better.

This is pretty funny, too: Tom Ridge Promoted to Cabinet-Level Whipping Boy

And this: Conspiracy-A-Go-Go

Ah, the healing power of laughter.

Mark Poyser writes:

There is extensive coverage of newly released FBI records about California politicians and politics of the 1950's through to the 1970's. Well worth reading.

Main link: Reagan, Hoover, and the Red Scare

Of interest is this item about Reagan
Excerpts from The governor's race:

-------------------------
As governor, Reagan would have access to UC's atomic research data. The Atomic Energy Act required the FBI to conduct a comprehensive background investigation of him. The process started on Dec. 18, 1966, when Reagan filled out a Personnel Security Questionnaire that asked, among other questions:

"Are you now, or have you ever been, a member of any organization which has been designated by the United States Attorney General as required under the provisions of Executive Order 10450?

"Are you now, or have you ever been, a member of any foreign or domestic organization, association, movement, group, or combination of persons which is totalitarian, fascist, communist, or subversive . . . ?" Applicants were required to list any such groups and the dates they were involved with them.

Reagan answered "no" to both questions on the form, which contained a warning that "any false statement herein may be punished as a felony."

--------------------------

But files of the Los Angeles FBI office showed that in 1946 Reagan had been a sponsor and director of the Committee for a Democratic Far East Policy, which had been designated as subversive by the U.S. Attorney General under Executive Order 10450.

The records also showed that also in 1946 Reagan had been a member of the American Veterans Committee, the California section of which had been cited in a report by the predecessor of the Burns committee as "communist dominated and (as) a vociferous, decadent minority in national AVC affairs."

But [the] head of the L.A. office, approved a report that conformed to Reagan's Personnel Security Questionnaire -- omitting Reagan's association with the two groups officially deemed subversive.
--------------------------------------------------------------------------------

The failure by Reagan to answer the question truthfully puts him in the same company as Bill Clinton (who denied having sexual relations with Lewinsky). One gets an airport named after him, the other gets impeached.

Josh Marshall rudely points out that Bush had no plan when he made his speech about the re-organization of the intelligence agencies. He hadn't even told Tom Ridge. My prediction that the Rove White House would schedule a war against Iraq for maximum political benefit looks better all the time.

As noted by Instapundit, the Ontario Clean Air Alliance (which is apparently not an industry front) is promoting nuclear power. Excellent. From what I learned at Enron, the hydropower resources in this continent are pretty much tapped out, and wind and solar don't pay for themselves in a lot of the country. (Having said that, I have absolutely no sympathy for pricks who block wind farm development because they consider them an eyesore. They should try having their waterfront filled with oil rigs, a la Houston.) In most of the country, we're going to meet our needs with nuclear power or with coal. Nuclear power is so much easier on the earth than coal. It seems open and shut to me.

Go, nuclear! Rah rah rah!

Brad DeLong has a good question.

Alex Whitlock wrote to me about this post:

The writer of the original post points out that people get PhDs so we should listen to them. Implied is that they've studied the facts and know what's right. What that overlooks is that some people enter their science fields BECAUSE they have an ax to grind. I doubt any of them said "I don't have strong opinions on the environment, so I will get a PhD!"... more likely a number of them said "I want to protect the environment, so I will get a PhD!" Such people are likely to be hyperprotective over the environment and treat every possible problem as an emergency that will lead to the collapse of the world as we know it if it is not immediately addressed. "The Chilling", The "Population Bomb", "AIDS will destroy the mankind within 20 years", etc.

Now that is an interesting point. It's especially interesting to me because it's congruent with my experience. I met a number of people in grad school who went into their field with a definite agenda. I'm thinking right now of the woman who was bisexual and wrote her MA thesis on the idea that bisexuality was swell. I exaggerate, but not by much; I thought her thesis was a pseudointellectual exercise in the promotion of her own sexuality, which substituted passion for rigor. I'm also thinking of the Muslim guy who argued that Westerners were wrong to think of Muslim women who wear the veil as lacking freedom, because in fact they were actively embracing the freedom that was most important to them: the freedom from unwanted male attention. I thought this was just a trick of the language. It took advantage of the dual nature of the English word "freedom", which can express two entirely different concepts: "freedom to" and "freedom from". The thesis literally couldnt be written in another language, where "ability to do X" and "absence of X" are expressed with different words.

On the other hand, I saw the woman present her thesis in a casual discussion format, and the liberal professors in a generally liberal university (the University of Chicago) ripped her a new asshole. They asked her very tough questions about her methodology, vigorously questioned her basic premises, and hammered away at her inconsistencies. They made her admit that her premise was non-falsifiable and therefore not scientific. And this was just a casual discussion. Even though they were probably all sympathetic to her, they took her apart with a scalpel. That's what academics do.

This process would continue throughout her academic career, if she had one. Journal publications would be reviewed by experts in her field. Every conference would expose her to sharp questions from skeptics who get points for pointing out flaws in arguments. Every book would be a big fat target for rivals to review it. Academia gets more abuse than it deserves, IMHO. I'm not going to be the guy to defend women's studies or queer theory, where groupthink can get pretty freakin' dumb. And of course, there are some bad apples. But in general, academics have to defend their ideas against a genuinely rigorous assault. No political writer has to face the same standards as an academic; when David Horowitz wants to write a piece blaming Clinton for sunspots, he doesn't have to show his data to a skeptical peer, or hold up publication for months while someone re-runs his equations. (Curiously, there's a blind spot when it comes to book publication; the peer-review process doesn't apply. But even there, when an academic is revealed in a big lie, a la Michael Bellesiles, the academic's career is ruined. When the Wall Street Journal editorial page publishes a big lie, it's just another Wall Street Journal editorial page.)

So I have some faith. Climatologists who use seriously flawed models are at the mercy of peer review before publication, and every hungry PhD. after publication. Scientists are in desperate competition for fellowships, grants, postdoc positions and actual jobs. They'd love to earn their bones by taking down the top models and theories. While false consensus can and will rule scientific disciplines from time to time, I think that ambition will win over ideology in the medium run. All the right incentives are in place. I think that this was true in experimental psychology, and I suspect that it would be even more true in climatology, where everyone shares the same data.

On the other hand, the bisexual woman was at the University of Chicago in the first place. And the Muslim guy got into the PhD. program with that thesis. I'm not totally convinced that "the truth will out", because what actually seems to happen is that you can just find pockets of different consensus (consenses?) co-existing about almost any question. They fire broadsides at each other constantly, but if you want to shop for academics who believe your position, you can find some.

This is a difficult question, and I'm not sure how I feel about it. I'd be very interested to hear reactions.

Who says you can't make fun of Larry King?

I've been out of touch since Wednesday night. If you sent an email since then, I'm not ignoring you; I just haven't had a chance to read it. I was in Ohio visiting my family and celebrating my birthday. Happy birthday to me. I got to see number one rock stars the Hives on Friday night, which was a great treat.

I may or not be back today, but I want to draw attention to two great new links, both in the comedy section: Laughing Boy, and Jim Treacher. They're working overtime for your amusement. Also, Mil Millington, of "Mil's Apology/ Things My Girlfriend and I Have Argued About" fame, has a book coming out. I hope it makes him a barrelfull of money.