Comments on: Science and pseudoscience

By: Scott Martens

Scott Martens — Fri, 23 Jan 2004 11:01:14 +0000

Anarch’s got it right. I remember reading somewhere about how New Guinea natives walk funny because they’re afraid a tree will fall on them. By moving around in a particular way, they are able to more easily avoid a tree if it should pick that moment to fall. This, it turns out, is an entirely reasonable fear. As rare as it is for a tree to fall on you, when you spend your entire life living in the woods it turns out that you will on the average have to dodge two or three gigantic trees falling right on top of you in a five decade active lifetime.

By: Dave F

Dave F — Thu, 22 Jan 2004 20:06:01 +0000

Dsquared, I believe you are deliberately ignoring my main point: I don’t see the point of heeding doomsday warnings because only an all-knowing god (if he existed) would be an authority worth heeding, and he or she (if extant) isn’t saying anything. If this time the merchants of doom are right, purely by statistical fluke, we won’t be around to appreciate it. I won’t especially. I will not be moved by this kind of manipulation, and that is what it is. Science has no business in advocacy, which is fatal for objectivity.

By: Anarch

Anarch — Thu, 22 Jan 2004 17:04:20 +0000

Sebastian: You don’t buy such insurance for a 1/10,000 chance of occurence if it represents 1/10 of your yearly income. You have to balance choices. The whole question is how likely is the bad outcome, how costly is the proposed ‘insurance’ and how much of an effect will the proposed ‘insurance’ have. You know better than to forget a part too, Sebastian: not just how likely the bad outcome is, but how bad it is. So we're right back to gambling with the devil and the St Petersburg paradox: how do you strategize when one of your outcomes (global warming exists and will wipe out all human life on the planet) has nonzero probability and infinite disutility?

By: Ken

Ken — Thu, 22 Jan 2004 15:42:06 +0000

The submersion of south Louisiana is already underway, as a result of modifications to our rivers by the Army Corps of Engineers which speeds the flow of soil-bearing water through our fine state and prevents it from depositing replacement soil on its way out as it was previously doing. Some of the land separating New Orleans from the Gulf of Mexico has gone missing, and without nearly universal ownership of skycars, New Orleans is pretty much screwed when the next hurricane makes a direct hit.

By: dsquared

dsquared — Thu, 22 Jan 2004 13:44:18 +0000

Dave: The reason that the record “so far” of doomsaying is bad, is that if the doomsayers had ever been right, we would no longer be around to assess their record. This follows from the definition of “doom”. Sebastian: You write: “The current state of the scientific inquiry suggests that the answers are very unlikely, very costly, and not very much effect if we implement the proposals.” Two points; first, that is not what the current state of scientific inquiry suggests, and second, where do you get your cost estimate from and why should it be regarded as better quality than the climatologists’ estimates?

By: Dave F

Dave F — Thu, 22 Jan 2004 11:54:29 +0000

Dsquared, the context of my remark was that doomsaying in general is self-defeating. No one has any notion of all future events that might affect the outcome of doomsday projections. The reason the record isn’t good so far is that models must exclude unforeseeable events, whic h are likely to be more numerous and widespread than foreseeable ones. Self-defeating because self-dismantling. The past record isn’t good because other effects unforeseen at the time were in play. The people who put forward alarmist scenarios (a million species could die, etc) are asking us to live our lives as if their predictions will inevitably prove accurate and the end of life as we know it. I’m not buying it, and if I am falling off a building, well, it’s been 62 years and we all hit the ground eventually. If global warming is inevitable, lie back and apply the sunscreen. Boo!

By: pkwill

pkwill — Thu, 22 Jan 2004 05:21:10 +0000

I don’t get why people seem to imply that modelers are idiots—they know exactly the limitations of what they’re doing. A lot of effort in science is devoted to understanding and quantifying sources of error, and modelers are no different. It’s harder for modelers to get things right, of course, with a limited set of observations and experiments to check their work with. At first, they were just content to produce realistic weather patterns. Nowadays, a common check is to see how the models check with historical records: ice core samples have given us enough data about the ice ages to be able to run the models in “ice age mode”—and they seem to agree very well with the data. Similarly, after Pinatubo erupted, the models were able to predict the effects well. Finally, the models are clearly not tuned to generate warming in response to CO2 emissions. They’re tuned to predict the weather, more or less. For the past thirty years though, every major model has had a CO2/temperature corellation. (Although the earlier ones aren’t that useful because they missed some important issues in cloud formation.) For way too much information, see http://www.aip.org/history/climate/

By: raj

raj — Thu, 22 Jan 2004 03:41:18 +0000

My boyfriend and I (both males) have no children and don’t intend to have any. Accordingly, we have both agreed not to worry about global warming or any of the other calamaties that have been predicted. If the conservatives—those who are supposedly concerned about fam’ly values—don’t care about what their children and grandchildren might have to face, why should we? I’m being sarcastic, but only slightly so.

By: praktike

praktike — Thu, 22 Jan 2004 02:32:12 +0000

This is all very interesting. I’m of the school that says that a few % points of GPD is a trivial cost for the wealthiest nation on Earth to pay to find out whether anything we can do will prevent potential catastophes like the submersion of Southern Louisiana. I find it odd that many of those who advanced a similar argument WRT the Iraq war eschew such logic here. And finally, I’ll note that the world’s largest reinsurance companies are quite worried about climate change.

By: Sigivald

Sigivald — Thu, 22 Jan 2004 00:09:50 +0000

Matthew: Huh? How does it follow that, since it’s gotten warmer over the past 50-100 years, this cannot be the result of natural events? Remember the Little Ice Age of the middle ages? It got colder for a few centuries. Then the Medieval Warm Period followed. One imagines that at the beginning and end of those there were noticeable changes over 50-100 years. Simply saying “there’s no way the sun’s activity could be related” to that does not make it so; Certainly it’s been observed that the sun’s output changes, and over periods we can easily see. Long-duration shifts over decades are not at all implausible. Remember, we haven’t even been keeping track of temperatures over huge areas, accurately, for more than a century or two. (While someone claimed that “all the models” support global warming caused by humans, they should recall that other people claim that the models were made to provide that conclusion. If one is looking for X, one’s models tend to find it. That’s the eternal problem with model-based solutions – it’s VERY hard to avoid bias in the models, no matter what side you’re on, since if you get a surprising result, it’s very easy and very tempting to assume the model is wrong, not your expectations. And Crichton was absolutely correct that “consensus” is no way to do science. If the data are sound and the theory works, consensus won’t be needed to argue your case; if the data are not sound, your case is wrong even if the fad of the moment produces consensus, or at least a vocal simulacrum of consensus among those making a living studying the field. Odd how profit-motive and greed are never brought up as a motive in this context by people ordinarily ready to admit they’re motives in other contexts, no?)

By: Sebastian Holsclaw

Sebastian Holsclaw — Wed, 21 Jan 2004 23:12:00 +0000

“In general, it is rational to take out insurance against a disaster of unknown but non-zero likelihood and very high severity.” If such insurance is cheap enough. You know better than to forget that part. You don’t buy such insurance for a 1/10,000 chance of occurence if it represents 1/10 of your yearly income. You have to balance choices. The whole question is how likely is the bad outcome, how costly is the proposed ‘insurance’ and how much of an effect will the proposed ‘insurance’ have. The current state of the scientific inquiry suggests that the answers are very unlikely, very costly, and not very much effect if we implement the proposals.

By: Antoni Jaume

Antoni Jaume — Wed, 21 Jan 2004 21:53:47 +0000

“I don’t see any reason to simply assume that our current climate is ideal.” It is not. That is not the point. Rather, from a limited human perspective, that, as we are organized, small changes have the bad habit to be very unsettling. You see, we do not allow people to move freely, and countries which are most directly affected tend to be poor in everything but peoples, so these peoples spill on their neighbours, stressing their societies, and so on. DSW

By: Mark

Mark — Wed, 21 Jan 2004 21:34:04 +0000

Just to add that a lot of “anti-enviro” advocates get away with muddling the debate because of the ignorance of the time-scales. This might be because there is a large intersection between the set of climate change deniers and the set of young-Earth creationists. Neither camp seems to understand how science works, and (as pointed out in the original post) they use the same argument methods. To a large extent that’s because they are the same people.

By: Ken

Ken — Wed, 21 Jan 2004 20:27:40 +0000

“The debate is – or ought to be – whether climate change is affected or exarcebated by human activity. If it is then the question is: what, if anything, should or could we do about it.” And another question which needs to be asked is: would humans be better or worse off given a warmer global climate than we have now. I don’t see any reason to simply assume that our current climate is ideal.

By: loren

loren — Wed, 21 Jan 2004 20:15:44 +0000

dsquared: “In general, it is rational to take out insurance against a disaster of unknown but non-zero likelihood and very high severity.” I keep thinking Rawls rather than Pascal.