posted
I thought this was an interesting read. My own opinion on what he says here has yet to be fully formulated, but here's what I think so far:
I think his criticism is valid, but that he overreaches. Consensus is important, it just isn't all-important. It is one of many factors to consider when you want to formulate policy about something that cannot be directly proven. His point about predicting the weather 12 hours from now is a total non-starter -- what would be the point of collecting seasonal temperature data, then? The bigger question is really: are the model predictions about global warming being borne out so far? But he doesn't address that, since it wouldn't fit in with the criticism he's trying to make. He also glosses over data on secondhand smoke with a lot of hand-waving.
The bit about the behaviour of the scientific community, however, is to me the most compelling part of the whole thing. The scientific community is supposed to be a place that fosters dissent, not a place of suppression and attack of dissenting opinions. That's what America is for.
Essentially, I think that before making such sweeping criticisms you ought to have precisely the sort of hard data on hand that you're complaining everyone else does not have. And he often doesn't have that.
quote:The bigger question is really: are the model predictions about global warming being borne out so far?
If his contention that the models will become increasingly inaccurate over time, then under both his and the modelers' view the models could be accurate early on. I liked the idea of the double blind methodology.
quote: Although Richard Feynman was characteristically blunt, saying, "I really don't think these guys know what they're talking about," other prominent scientists were noticeably reticent.
This fascination with computer models is something I understand very well. Richard Feynmann called it a disease.
Have I mentioned I love Feynman? As for the rest, I'm pretty sure I've made some of these arguments over in the global warming thread. And twink, you missed the point. Not only is concensus not important, it can be dangerous. Science is about agreement with reality, not agreement with the scientists.
Posts: 2283 | Registered: Dec 2003
| IP: Logged |
posted
That was an excellent essay! Thanks for posting it. Crichton identified for me the growing skepticism I have recently developed of scientific claims. So often these days I hear claims from each side of an issue made by supposedly reputable scientists. When I can I dig into the details a bit to try to develop a sense for the material, but lets face it, I am not going to become a climatologist to understand global warming or an astronomer to understand cosmology.
I think that the idea of double blind science, especially for matters of public policy, is an excellent one.
Posts: 4548 | Registered: May 2001
| IP: Logged |
posted
If we followed Crichton's recipie, predictions about future events would be worthless. What's the point of trying to predict the future when we can only observe the present? I mean, tomorrow Newtonian mechanics might suddenly cease to work for predicting the behaviour of moving objects at non-reletavistic speeds. Quick, let's stop launching space shuttles.
My opinion is unchanged from my initial post. His point has validity, but he's going way too far with it, and that takes quite a lot away from his argument.
Posts: 10886 | Registered: Feb 2000
| IP: Logged |
posted
twink, I think one of Crichton's major points is the fact that climate may be a chaotic system, and is definitely an incredibly unpredictable one. His point was that BECAUSE weather is so volatile that we cannot reliably predict trends twelve hours in advance, THEREFORE it is absurd to place such implicit trust in predictions that span thousands of such twelve-hour periods.
He then applies the same logic to predictions about human society. "If current trends continue" is a very difficult premise to assume when you are talking about human behavior. Humans do new things all the time which throw out all previous assumptions. In other words, we are as inherently unpredictable as the weather, and long-term predictions about our future are inherently shaky and suspect.
However, the laws of physics have no such inherent limitation. They can actually be reduced to simple, testable scenarios, and their behavior has been shown to be remarkably predictable. So your example is a little flawed.
quote:His point was that BECAUSE weather is so volatile that we cannot reliably predict trends twelve hours in advance, THEREFORE it is absurd to place such implicit trust in predictions that span thousands of such twelve-hour periods.
But this is a complete non-starter. Otherwise how would we know what seasonal temperatures are for a given area? And what about continental drift?
posted
Crichton is laying the blame on scientific consensus itself, when the real problems are more complex (inflexibility of the prevailing paradigm, intolerance of new viewpoints, using bad science as a political bludgeoning device, etc.). He lays out these problems in a pretty compelling way (though I'm not sure if it's all accurate), but then he goes back and blames his demon of consensus. Consensus is good. It's how we get things done. It only becomes a bad thing when it's wrong.
I think the double-blind experimentation idea is an interesting one, but I'm not holding my breath.
Posts: 1810 | Registered: Jan 1999
| IP: Logged |
posted
Mike, I think the problem wasn't concensus so much as when it's invoked. As Chrichton says, no one invokes concensus to validate experimentally proven theories. The results are allowed to speak for themselves.
Concensus on g=-9.8 m/s^2 is unnecessary. Concensus on F=ma is never mentioned. Everyone believes it because it's been tested. They can see the data.
NASA won't come down one way or the other on global warming.
quote: So far scientists have not been able to pinpoint with certainty any changes in weather due to global warming over the last century.
If NASA can't come to a concensus on something they study a lot, why can everyone else? Why did the global warming thread boil down to "Everyone smart knows that." "Nuh-uh."?
It's not that concensus is bad. Having to invoke concensus to win your argument is bad. What's the quote in Children of the Mind? If all you can say in your defense is "I am so smart" it's not worth saying.
Posts: 2283 | Registered: Dec 2003
| IP: Logged |
posted
And while I can't prove it because I don't have their entire site archived (hmmm, note to self, check archive.org), I bet their site under Clinton didn't have that statement about it being indeterminate.
Posts: 15770 | Registered: Dec 2001
| IP: Logged |
posted
From what I can find on NASA's archived site from 1999 (the furthest back the earthobservatory archive on archive.org goes) they hadn't really looked much at global warming The few articles available are on other earth-related subjects. Its not exactly in NASA's typical purview, y'know?
Posts: 15770 | Registered: Dec 2001
| IP: Logged |
posted
I think the comparison of the global-warming-consensus with the other examples, ie Pasteur, nuclear winter, etc, is a bit forced. Most of his examples are from periods before the real start of science in that field, ie Semmelweis, so they are not actually a consensus of scientists. Nuclear winter pretty clearly was a drummed-up 'consensus' that didn't actually exist. But in global warming, the consensus is being invoked because there is an opposing consensus among businessmen and politicians, saying 'global warming would be bad for business, so it can't exist'. And they look for steadily more far-out ways of disproving it. Well, there comes a point when you just have to say, "Ok, you want to do politics, two can play at that game."
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
posted
If you flip a coin, you can't predict what it will land on. However, if you flip a coin 10,000 times, you can make a pretty good estimate of what the result is going to be.
For the same reason, not being able to predict what goes on in the next 12 hours does not mean you can't get good predictions of the trend across decades.
Posts: 2432 | Registered: Feb 2001
| IP: Logged |
Printer-friendly view of this topic
![[Wink]](wink.gif)
![[Smile]](smile.gif)
![[Roll Eyes]](rolleyes.gif)
