Blog RSS Feed
05.03.08
Well, Einstein's *job* was patent clerk, he took it only after failing to find a teaching position. He had a degree in physics.
The statistical sense of "bias" is a tendency to an incorrect decision. Reducing bias is then a mathematical question, and so the more sophisticated the math the smaller the bias. This isn't condescension, though it may be daunting. Mathematics that exacerbates a bias is, on the whole, simply wrong. Overcoming bias is a primary goal of mathematics. So it's puzzling to imagine what my correspondent is thinking here.
I think it's important for those of us in science to explain things as well as we can, at a wide variety of levels. It's also important for everybody else to understand that the ways in which we approach things take long years of training. We should in most cases (not always) be able to explain what we are up to to a general audience in general terms, but we can't possibly get you on board with the details. There is a reason it takes decades to season a scientist.
So, for instance, people claim that the greenhouse effect saturates. The argument is that if CO2 absorbs infrared radiation at a particular frequency, past a certain concentration it will have absorbed pretty much all of it, and adding more CO2 won't matter.
It turns out this is wrong. It is not a matter of opinion. It is not a matter of debate. Lots of people believe the wrong answer but it is still wrong.
There are several ways to answer this. The first way is to set up the problem correctly, and build up a case of how multiple layers of the atmosphere interact. It's possible to dance around calculus in the details, but it is still a pretty sophisticated argument. People who didn't do well in college physics and calculus will tend not to understand it.
The second way is to point to Venus, which has a very thick CO2 atmosphere and a very high surface temperature. This is how Mr Gore addresses it in the movie. Now, one might object that Venus is closer to the Sun than the earth is, so it should be hotter. Fortunately, there is Mercury, still closer to the Sun, lacking an atmosphere and relatively very cool compared to Venus, (by hundreds of degrees) though still hot enough to bake a pizza, to be sure. Is that compelling? I don't know. If I were determined to cast doubt on he argument I would ask how we know the surface temperature of Venus under all those clouds, and then the math would really have to start flying fast and furious.
The third way is to assert that I know more than you do about this, so you should trust me. This only works when there is little controversy. When there is enough controversy it is always possible to find somebody who has a credential and disbelieves or at least is willing to pretend to disbelieve the argument. Then we are really in hot water.
It turns out that with climate change, we are in very hot water indeed. Not only is there a prospect of serious problems, but some people don't take it seriously. That's bad enough, but it's much worse that there are ideological predictors of who believes one side vs another on purely scientific questions.
The only way democracy can function is if the general public has a good grasp on important information. If there's a huge culture gap between practicing scientists and the rest of the society, science suffers (its funding declines) but society suffers worse. Decisions get made on the basis of less than the best information.
The real fix is to stop having science happening in cloisters. A few famous institutions can't dominate funding except for the most expensive work. If people don;'t have scientists as friends and neighbors they will be willing to believe strange things about us, and won't be in a position to evaluate bad news fairly when it is delivered.
In the end, though, nobody objects to the idea that a professional basketball player can play basketball better than they can, or that a professional clarinetist can play clarinet better than they can. Most people don't take their barber's medical advice over their MD's. Why should people object to the idea that people who spend all their time competing to do science can do so better than most people?
Is an Elite Elitist?
by
I got this from a persistent if anonymous correspondent in my mailbox and I am puzzled about how to respond:
I went to a lecture recently where I was the only person attending who
wasn't a math major or member of the math club. What I found most interesting
was the overwhelming sense of arrogance and condescension from the math majors
once everyone began talking at the end of the presentation. I was talking with
the professor who is the advisor, and had invited the speaker, and he said that
this attitude makes him profoundly sad, and he believes it's one of the chief
obstacles in recruiting more people in higher math and sciences.
I get that same feeling from you. If you are saying that one must be at a
certain level of qualification to comprehend and debate certain issues, then you
are simply creating a filter that risks bias towards one position. I'm deeply
skeptical of the "you must be this tall to ride the global warming debate"
attitude. Keep in mind that Einstein was a patent clerk. Many groundbreaking
scientific discoveries have come from very non-scientific people.
The statistical sense of "bias" is a tendency to an incorrect decision. Reducing bias is then a mathematical question, and so the more sophisticated the math the smaller the bias. This isn't condescension, though it may be daunting. Mathematics that exacerbates a bias is, on the whole, simply wrong. Overcoming bias is a primary goal of mathematics. So it's puzzling to imagine what my correspondent is thinking here.
I think it's important for those of us in science to explain things as well as we can, at a wide variety of levels. It's also important for everybody else to understand that the ways in which we approach things take long years of training. We should in most cases (not always) be able to explain what we are up to to a general audience in general terms, but we can't possibly get you on board with the details. There is a reason it takes decades to season a scientist.
So, for instance, people claim that the greenhouse effect saturates. The argument is that if CO2 absorbs infrared radiation at a particular frequency, past a certain concentration it will have absorbed pretty much all of it, and adding more CO2 won't matter.
It turns out this is wrong. It is not a matter of opinion. It is not a matter of debate. Lots of people believe the wrong answer but it is still wrong.
There are several ways to answer this. The first way is to set up the problem correctly, and build up a case of how multiple layers of the atmosphere interact. It's possible to dance around calculus in the details, but it is still a pretty sophisticated argument. People who didn't do well in college physics and calculus will tend not to understand it.
The second way is to point to Venus, which has a very thick CO2 atmosphere and a very high surface temperature. This is how Mr Gore addresses it in the movie. Now, one might object that Venus is closer to the Sun than the earth is, so it should be hotter. Fortunately, there is Mercury, still closer to the Sun, lacking an atmosphere and relatively very cool compared to Venus, (by hundreds of degrees) though still hot enough to bake a pizza, to be sure. Is that compelling? I don't know. If I were determined to cast doubt on he argument I would ask how we know the surface temperature of Venus under all those clouds, and then the math would really have to start flying fast and furious.
The third way is to assert that I know more than you do about this, so you should trust me. This only works when there is little controversy. When there is enough controversy it is always possible to find somebody who has a credential and disbelieves or at least is willing to pretend to disbelieve the argument. Then we are really in hot water.
It turns out that with climate change, we are in very hot water indeed. Not only is there a prospect of serious problems, but some people don't take it seriously. That's bad enough, but it's much worse that there are ideological predictors of who believes one side vs another on purely scientific questions.
The only way democracy can function is if the general public has a good grasp on important information. If there's a huge culture gap between practicing scientists and the rest of the society, science suffers (its funding declines) but society suffers worse. Decisions get made on the basis of less than the best information.
The real fix is to stop having science happening in cloisters. A few famous institutions can't dominate funding except for the most expensive work. If people don;'t have scientists as friends and neighbors they will be willing to believe strange things about us, and won't be in a position to evaluate bad news fairly when it is delivered.
In the end, though, nobody objects to the idea that a professional basketball player can play basketball better than they can, or that a professional clarinetist can play clarinet better than they can. Most people don't take their barber's medical advice over their MD's. Why should people object to the idea that people who spend all their time competing to do science can do so better than most people?
Tags: elitism, ideology, mathematics, outreach, science
4 Comments
+ Add Comment
May 24, 2008 6:19 AM
N1XIM
Mike,
Whilst I agree that it would make sense if people would believe those whom are trained to understand something that they haven't been at first blush, there is a good damn reason (evolutionarily and historically) not too. Think about it for a second. Is it really in your best interest to believe my amazingly complex irreducible explanation for something? No, it isn't. (In fact, true scientists would refuse to do so outright.) The reason why is that I cannot connect the parts of your argument together and cement them into my existence. If I cannot do this, then so far as I am concerned you could be lying to me for your own personal gain.
This is the way that "Top Down" complex explanations about things from how ice skates really work (no, they don't melt the ice under them), to how global warming works are approached by common people. If you don't take that into account you're going to get labeled an elitist--or worse.
I've oft used the example of a glass of iced tea (with ice still in the glass) to explain the idea that the energy in a system can be increasing ("warming") without the OBSERVED temperature changing in any notable way. The same argument can be used to disprove a couple of different objections about Global Warming.
You have to consider your audience when you are trying to make a sincere argument about something. Otherwise you are yet another salesman. (We all know what people think of used car salesmen--due to the few disgustingly dishonest ones.)
You buy that?
June 4, 2008 10:28 PM
Alexi Tekhasski
Michael, you wrote:
"Why should people object to the idea that people who spend all their time competing to do science can do so better than most people"
The problem with the above question/statement is in the generalized use of term "science". Not all sciences are equal and have the same rigor. There are exact sciences like physics and mathematics, less exact sciences as chemistry, applied sciences, and observational qualitative sciences. The climatology is a mix of observational and applied sciences, even worse, an intertwined conglomerate of many applied sciences. There is no new physics or chemistry, but all parts of climatology are tightly coupled in mutual inseparable interaction. The issue is that everything in climatology is in a gray area of original sciences, there is no clean limits that allow to make physical reductions based on small factors or large factors, every effect is about of the same order of magnitude, so you cannot neglect one and get correct overall result. This means that each individual subsystem in climatology essentially operates under conditions that were rendered as untreatable in its pure exact original science. For example, all atmosphere is highly turbulent, and constitutes one of the unsolved problem of fluid mechanics. There are people who spent their entire life trying to find useful solution to this problem, and they didn't have much of success. Therefore, if someone who spends all his time "competing to do climatology" implies that the problem of turbulence somehow does not matter for heat, mass, and radiation transfer across the atmosphere, then it is quite reasonable to challenge results of such aggressive climatology.
For example, application of "radiation code" to an averaged model of atmosphere
http://forecast.uchicago.edu/Projects/full_spectrum.html
shows that just 1.4% increase in average low cloud cover would completely negate the whole hypothetical effect of doubling of CO2. What does it mean? It means that climate models have to emulate the entire hydrological cycle to better than 1% accuracy if you want to evaluate effects of CO2 on global climate. For someone who spent his better part of life conducting professional studies of fluid dynamics and turbulence, both at its origin and quite developed, I would say that it is completely impossible to model evaporation, transport, and condensation of water vapor across turbulent and contaminated atmosphere to this accuracy. Period. According to your definition, I know it so better than most people. This is not a matter of belief or disbelief; it is a statement of fact. Unfortunately, since the atmosphere is an inseparable part of climate formation, this cast serious doubt in overall predictions of catastrophic (or even measurable) effects of CO2 on climate.
June 9, 2008 7:51 AM
Michael Tobis
N1X; I don't follow. I argue that the whole thing should be as transparent as possible, and will agree that we have not been successful at that. (In our defense we are neither trained nor rewarded for doing so. One might even argue that the attempt is usually punished rather than rewarded.)
The outlines of our climate problem are not complex. The details are complex and rich. People attempting to confuse the matter bring details into the broad brush picture. We are left in a bind; to address the details requires losing most of the audience, and to fail to do so leaves the impression that our knowledge is shallow.
The best we can do is to address the details people raise, and at the same time emphasize that the outline of the situation is not all that complicated.
June 9, 2008 8:07 AM
Michael Tobis
Alexi:
Yes, we are not as rigorous as pure physics by necessity. The inclusion of various other fields is, to me, the fundamental appeal, though others find it disconcerting. However
"there is no clean limits that allow to make physical reductions based on small factors or large factors, every effect is about of the same order of magnitude, so you cannot neglect one and get correct overall result"
is quite incorrect. The realms of applicability of this or that theory are matters of considerable importance in our field. Your example of turbulence is in fact a case in point. Turbulent motions are inherently unpredictable, but on a long enough time scale the energy and mass fluxes they produce are consistent and may be constrained by combinations of theory and observation. It is not as if we know nothing when we don't know everything. On certain time and space scales, turbulent closures are crucial; on others they can be parameterized adequately to arrive at substantive conclusions about the larger system.
"1.4% increase in average low cloud cover would completely negate the whole hypothetical effect of doubling of CO2. What does it mean? It means that climate models have to emulate the entire hydrological cycle to better than 1% accuracy if you want to evaluate effects of CO2 on global climate. "
Unlike the rest of what you say, this is nonsense. Suppose I meet a person on the internet or on the telephone. I cannot know their height; my skill is limited to the uncertainties implied by the height distribution in the general population. Nevertheless, I can make strong claims about the rate of change of their height (zero for an adult, small and positive for a child). These are independent.
While we may have uncertainties in the individual terms, we can still know that the entire system must be changing. I will be trying to present this argument over the next couple of weeks at a broadly accessible level.
I am sure the rigor will not satisfy you, and I anticipate more of your usual approach to our discussions. Again since we live in the same city I would much prefer to discuss this over coffee and try to convince you that the field as a whole is on a solid footing.
Post your comment