Thursday, December 17, 2009 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

Climate: James Randi vs mindless consensus pseudoscientists

Error! Chances are that you came to this blog from "Pharyngula". If it is the case, I urge you to instantly go away because your presence violates the very basic rules of hygiene. After two showers and five prayers following your recent "Pharyngula" visit, you may slowly start tofa consider walking among decent people once again.

(And believe me, this condition means something because I am an atheist, too.)

BTW the self-described godless ejaculating scumbag P.Z. Myers who calls me names only has 137 citations for his 4th article, less than my figure (one missing) - that's despite his being a very old zombie who is still a parasite upon the establishment we call Academia, unlike your humble correspondent. Aggressive left-wing activists of Pharyngula's caliber have thoroughly contaminated the Academia - so that no fine person can breath there any longer.



Jorge P. has brought my attention to an essay about climate change written by James Randi:
James Randi Educational Foundation: AGW, Revisited
Randi who may be the world's #1 symbol of skepticism towards pseudoscientific charlatans (and magicians claiming to have special abilities: he reproduced lots of their tricks without any paranormal abilities) turns out to be consistent in his skepticism: he is skeptical towards the climate judgement day pseudoscience, too.

Randi's arguments are kind of obviously valid. He enumerates many solar, galactic, geomagnetic, lunar, and other influences that change the temperature by quantities comparable to 1 °C per century and that are not under theoretical control. It follows that the climate "equation" that would reliably predict a century of temperature changes with such an accuracy or a better one cannot be written down at present which is a reason why sensible people shouldn't make far-reaching claims about the future temperature.




Randi also mentions the large number of scientists (signed under various petitions etc.) who have reached similar conclusions.

His newly discovered skepticism may explain why Phil Plait who is not a skeptic but rather an uncritical irrational believer when it comes to te atmospheric Armageddon theories is no longer the president of the James Randi Educational Foundation. Well, he may have been simply yet diplomatically fired by Randi for having brutally violated the main principle that underlies the work of JREF - scientific skepticism.

Of course, the far-left blogosphere is devastated in the same way as an Islamic fundamentalist who has just learned that Allah means nothing in Central Europe. A self-described ejaculating godless liberal P.Z. Myers called his article "Say it ain't so, Randi". There are no arguments in his text, except for the usual frequently repeated lies and prayers worshiping the scientific consensus so I won't discuss Myers here.

His slick far-left comrade Sean Carroll wrote some more idiosyncratic comments about the "incredibly depressing news". First, Carroll seems to have understood that the hockey stick papers by Mann and similar authors were wrong. But he promotes some sources that suggest that the hockey stick shape returned in newer, more correct papers, anyway. Well, it didn't.



More up-to-date reconstructions independent of the "hockey team" of Mann and Jones such as "alarmist" Moberg et al. (up) and "skeptic" Loehle (down) don't look like hockey sticks and they don't show any statistically significant "unprecedented changes" in the industrial era.

They may differ on the question whether the Middle Ages were warmer than the present but this information neither implies nor rules out the man-made character of recent changes, anyway. As far as I can say, the question "which time was warmer" is just an irrelevant pissing contest and we may never be certain about the correct answer.

Carroll hysterically repeats some insults about the "denialists" - a group that now includes "their" James Randi. But I want to dedicate some time to the question in the following paragraph he wrote:
Still, there remains a somewhat intractable problem: when people are arguing about issues that necessarily require expert knowledge that not everyone can possibly take the time to acquire for themselves, how do we make judgments about who to believe?
This is a good question but the very way how he formulates it shows that his answer to the question is profoundly incorrect. Obviously, sensible people should try to develop some sense for who is more likely to be right, who is more intelligent, honest, unbiased, uncorrupt, experienced and achieved in similar questions, educated, and who has verified his statements in more detail. They need it because they are often forced to make decisions about matters in which they're not top experts.

But what is fundamentally wrong about Carroll's answer pretending to be a question is that "we" should agree on an algorithm how to make judgments whom to believe. It's actually very important that different people use different algorithms to decide about such matters so that their biases don't become systematic errors. Some algorithms may be expected to be more efficient, and may be actually more successful statistically, but it's still important that people use different algorithms.

If six billion people were using exactly the same algorithm to decide whom they should believe about a particular topic, they would probably end up with the same answer, XY. It would seem like a "consensus" of six billion people. But that wouldn't mean that XY's opinions about the scientific questions are actually correct! In fact, 5,999,999,999 of the people's personal opinions would be redundant. There would only be one sociological argument, or one algorithm, that favors a particular answer. It would be a complete mistake - an incorrectly added multiplicative factor of 6 billion - to think that there are "six billion arguments" to believe XY.

This is a not-so-subtle point that all actual skeptics and scientists understand very well (because repetition doesn't make statements "actually" more true) but people like Sean Carroll and other ideologically driven pseudoscientists, consensus scientists, and left-wing hacks in general don't. In fact, Carroll displays his complete misunderstanding of these matters quite explicitly:
This gets to the heart of why I’ve always been skeptical of the valorization of “skepticism.” I don’t want to be skeptical for the sake of being skeptical — I want to be right. To maximize my chances of being right, I will try to collect what information I can and evaluate it rationally. But part of that information has to include the nature of the people making arguments on either side of a debate. If one side consists of scientists who have spent years trying to understand a complicated system, and the other is a ragtag collection of individuals with perfectly obvious vested interests in the outcome, it makes sense to evaluate their claims accordingly.
Well, this may be a vaguely sensible strategy except that the correct outcome of such reasoning in the climatological context is exactly the opposite one than Carroll suggests.

I don't think that there are any climate scientists who are alarmists and who have spent years trying to impartially understand the climate system. On the other hand, there are hundreds of alarmists employed as climate scientists who have spent a decade or two trying to prove one particular predetermined answer, using legitimate as well as illegitimate tools, and who have indisputable vested interests such as those GBP 13.7 million for Phil Jones, an average fraudster who would be irrelevant and poor if the climate hysteria didn't exist. The ClimateGate files have certainly been enough to prove this modest proposition.

But even if there were real scientists behind the climate alarm, the argument above would not imply that they are right. The case would get much weaker if the scientists were actually not independent but connected via a tight mailing list and pushed by the same ideological and funding pressures. As all of us know, this is the case of the climate science, especially the kind of climate science that generates the hockey sticks and perhaps other bogus arguments supporting man-made climate change.

Once again, Carroll's final paragraph says:
This gets to the heart of why I’ve always been skeptical of the valorization of “skepticism.” I don’t want to be skeptical for the sake of being skeptical — I want to be right. To maximize my chances of being right, I will try to collect what information I can and evaluate it rationally. But part of that information has to include the nature of the people making arguments on either side of a debate. If one side consists of scientists who have spent years trying to understand a complicated system, and the other is a ragtag collection of individuals with perfectly obvious vested interests in the outcome, it makes sense to evaluate their claims accordingly.
I don't want to skeptical for the sake of being skeptical, either. But in the same way, I don't want to be a "slick mainstream believer" for the sake of being a "slick mainstream believer". By being a slick mainstream believer of Carroll's type, you may think that you maximize your chances of being right, but you can still be wrong. Climate change seems to be an obvious example. And the history of science shows hundreds of additional big examples when this strategy to "maximize the odds of being right" led to wrong results. Consensus just doesn't work in science, especially not if it becomes a frequent part of the methodology. As Feynman has defined it, science is the belief in the ignorance of experts.

This quote due to Feynman is no typo. It is not irrelevant, either.

The actual scientists who are doing real science always have to doubt "experts". The outsiders and laymen sometimes have to adopt the opinions of others, preferably of the people who are better experts than they are. But the latter step should never be confused with science itself. Adopting other people's opinions according to predetermined fixed algorithms is not a part of the scientific method. Science is never about the copying of opinions, and the number of copies of an opinion can never serve as an enhanced scientific argument in favor of a hypothesis.

The original hockey stick graph paper, which is now agreed by Carroll to have been wrong, shows very clearly why and how the consensus science reasoning doesn't work. The trick was invented by one person - Michael Mann - who convinced two co-authors to add their names to a paper even though they didn't help to invent the basic flawed methodology and they have probably never quite believed it.

Because three authors were enough, the paper was viewed as a source of some great new "momentum" and its wrong methodology was accepted by a dozen of "experts" in this subdiscipline who were either ignorant, stupid, lazy, corrupt, dishonest, biased, cowardly, or submissive enough to miss the point that the methodology was completely wrong (or to see it but be completely silent about it). From this dozen of people, the wrong result was almost directly adopted by 2,500 who wanted to hear and "believe" a similar conclusion, anyway, for different reasons, and then billions of careless people who "wanted to maximize their odds of being right" so they believed the "scientific consensus".

Obviously, they didn't maximize anything in this case. They would have done much better if they actually tried to work harder and to create their own reconstruction.

Clearly, the pathological propagation of wrong ideas in the hockey stick example above - from one lousy pseudoscientist called Michael Mann literally to billions of people - was no anomaly. It is virtually guaranteed to occur when people give up science and replace it by the "consensus science" or "Carroll science", a method to "maximize your odds" by parroting the chosen people that optimize how much slick you become. The more Carroll science, groupthink, and sloppy thinking you find in a discipline, the less likely it is that the opinions of the majority are right.

If you eliminated the scientific method - independent empirical and logical testing and refinements of hypotheses - entirely, the probability of the majority's being right would drop to 50% as long as there would be no systematic biases. Of course, in climate science, there are lots of biases and there actually exists solid science that shows that the climate alarm is a myth, so the probability that the "IPCC-like majority" is correct about the "danger of climate change" is close to 0%.

Add to del.icio.us Digg this Add to reddit

snail feedback (7) :


reader Brad Tittle said...

According to a post yesterday by Randi, Phil is still the president of JREF. Although I disagree vehemently with Phil's stand on AGW, I agree with him on most other subjects scientific.

The amazing thing about this discussion is that Phil, PZ and others fail to recognize the N-Rays nature of the climate debate.

"Oh, that adjustment there, damn I got it in the wrong direction, BUT ALL THE REST OF THE DATA IS GOOD".

"Oh that other piece is wrong too. Ok, but the rest of it is perfect."

"Oh, I put a function in to make the temperatures appear hotter -- My bad -- but the rest of the data is good I swear!"

Every time the data is scrutinized, errors in the favor of the Global Warming crowd are discovered.

Phil is a good guy. I suspect that he doesn't have religion though. This leaves a gaping hole for other religions to infiltrate. We are human. Even the best scientist is challenged to be completely unbiased.


reader Randi said...

As I anticipated, my readers who jumped to the unfounded conclusion that my comments indicated I'd joined the "denialists" encouraged all sorts of cheering and jeering from those who will now try to take support from those very false assumptions. No, Dr. Plait was NOT fired, for any reason. He left in order to pursue a very important project that took his whole attention. I do NOT support the denialists, especially after seeing how they blatantly misrepresented themselves in their "Oregon Project" scheme. I was poorly informed, and I have so admitted. Being wrong and then correcting the error, is part of the scientific process. I have met that requirement.


reader Randi said...

As I anticipated, my readers who jumped to the unfounded conclusion that my comments indicated I'd joined the "denialists" encouraged all sorts of cheering and jeering from those who will now try to take support from those very false assumptions. No, Dr. Plait was NOT fired, for any reason. He left in order to pursue a very important project that took his whole attention. I do NOT support the denialists, especially after seeing how they blatantly misrepresented themselves in their "Oregon Project" scheme. I was poorly informed, and I have so admitted. Being wrong and then correcting the error, is part of the scientific process. I have met that requirement.


reader Lumo said...

Well, Dr Randi, let me assume that it's you within this reply: if you think that Phil Plait's uncritical, religious defense of the scientific misconduct in the wake of ClimateGate is *not* a reason to fire him from your "skeptical" society, the only thing I can say is "Shame on you."

If you have been truly violently, physically blackmailed and harassed by the AGW fanatics, I could understand what you just wrote. If you were not, let me just state that in that case, you became a morally worthless human in my eyes.

The "denialist" dictionary you adopted and the attacks against the Oregon Petition are pretty disgusting.


reader al fin said...

It is important to understand that Mr. Randi has a very large following of persons who have clung to him on the basis of the causes that Randi debunked: supernaturalism, magic, etc.

These one-cause supporters of Randi proved to be "fair-weather friends" when Randi expressed the slightest skepticism over the "A" component in "AGW."

Randi is undergoing chemotherapy for colon cancer with a 50% chance of survival. He is physically weak, and his acute mental powers are conceivably strained when interacting with the public -- particularly the fickle public of AGW true believers.

The fanatics of climate hysteria jumped on Randi like rabid hyenas in their eagerness to squelch all heretical expression.

I don't blame Randi for attempting a strategic withdrawal -- like James Watson and Larry Summers were likewise forced to do in other circumstances. He is not currently strong enough for an all out battle with the mind-locked devotees of carbon death.


reader Lumo said...

It's sad that such things are happening even in such situations but they're happening and we know who started them.

Well, I think his achievements are fun even though I am unlikely to claim that I need to see a Mr Randi to think that magicians don't have supernatural powers. ;-)


reader Rathnakumar said...

James Delingpole's take on the episode
http://blogs.telegraph.co.uk/news/jamesdelingpole/100020309/climategate-james-randi-forced-to-recant-by-warmist-thugs-for-showing-wrong-kind-of-scepticism/