Thursday, August 5, 2010

Absolute and Provisional Certainty in Science

There is a story about the Physicist Wolfgang Pauli that illustrates something about how science is done. After a seminar at which the speaker described some recent work, Pauli was asked his opinion of the work described. He responded by saying "it wasn't even wrong." Pauli was a tough audience, but his comment illustrates what sort of things occur as scientists study a problem. His point was that the problem the fellow had studied was so simple that the results probably were correct, but were not interesting. If he had been studying a challenging problem, the results would have been wrong, but in determining why they were wrong, others would have learned more about the subtleties of the subject. Progress is not made by studying easy problems.

I do not recall seeing reports in the media that nuclear physicists did not know what they were doing because many published studies about the structure of an atom's nucleus turned out to be wrong. Perhaps this was because few people understood what was being done, and perhaps it was because the studies did not seem to have an impact on people's beliefs or on how they lived their lives . On the other hand, current controversies over scientific studies, such as Evolution and Global Warming exist, perhaps because one affects traditional religious teachings, and the other has the potential to affect the standard of living in the industrialized nations.

Global warming provides a good example of the kind of scientific study that attracts public attention and criticism. It is also an example of how modern scientific studies are done because the subject is sufficiently complex that many teams of scientists with different specialties are studying it, using a number of different approaches. One interesting result of these studies is that the different teams, using different approaches seem to be arriving at largely similar conclusions. The studies conducted by these teams, from several different countries, examining different aspects of the subject conclude that the likelihood is very high that human activities associated with the emission of carbon to the atmosphere have been causing the earth's average temperature to rise. A report from a U.N. committee based on these studies recommends reducing CO2 emissions drastically and quickly to prevent dire consequences to human populations from climate change.

It should be emphasized that recommendation that CO2 emissions to the atmosphere be reduced is based on the evidence that exists and on analyses of that evidence. The evidence consists in part, of temperature measurements taken since the early 19th Century that coincide with the development of the Industrial Revolution, which show that world-wide temperatures have been increasing for the last 200 years. The evidence includes about 50 years of C02 measurements taken at a station on the top of the volcano Mauna Kea, in Hawaii, far from any industrial activities, which demonstrate that values of C02 in the atmosphere have been increasing for the last 50 years. And the analyses include predictions of future average temperatures based on computer models of the atmosphere-ocean system, several different versions of which, developed by different teams of climate scientists in different countries, predict future warming of the planet if CO2 emissions continue to increase.

Unlike the example of nuclear physics, the work on global warming has been criticized strongly. As is the case in all scientific studies, we have to recognize that the results of these studies are provisional. The predictions of future conditions on the planet based on the analyses of the data sets and on the modeling studies may turn out to be wrong, but no one has as yet shown that they are wrong. That is important.

No one has shown that the correlation between temperature measurements taken since the early 19th Century and the emission of C02 is spurious and for that reason we should not infer that human activities are causing global warming. No one has developed computer models that predict that CO2 emissions will not cause global warming in the future. And no one has developed a theory of atmospheric chemistry that shows that CO2 emissions could not cause warming. Yet critics of the results of climate research persist.

Those who express doubts about global warming have no evidence to substantiate their claims, so they are reduced to complaining that the studies indicating warming is occurring are not conclusive. For example, they claim that the correlation between rising temperatures and CO2 emissions does not prove a causal relationship between the two sets of data. This is certainly true, but Biologists working at the Tobacco Research Institute once made the same argument about relationships between smoking and lung cancer. Or, the critics of global warming complain that the computer models that predict serious warming in the future unless CO2 emissions are reduced do not take into account everything that could affect the climate, so the models' predictions cannot be trusted. They do not mention that those who build and use the models are aware of their limitations, and have developed ways to determine by how much the limitations affect the reliability of the models' predictions.

Because implementation of the recommendations of the U.N. report dealing with global warming will require serious societal changes, those who refuse to accept the conclusions of the report usually base their reluctance on the need for absolute certainty that the changes are warranted. Due to the costs of the changes, they require that the scientists be absolutely correct before they will consider implementing the recommendations. Of course, that assurance will never come. It is interesting to note that editorial positions in Conservative media outlets, which deplore any attempts to deal with the threat of global warming, also contend that if there is one chance in a thousand that terrorists will attack the U.S. again, we should spend billions of dollars to ferret them out and prevent the attack. This inconsistency in the use of the "precautionary principle" suggests that a great deal of "compartmentalizing" occurs in the minds of some people.

How do we get the general public to understand that the assurance they require is not possible and that definitive statements about complex phenomena are usually the result of oversimplifying the topics? How do we get the public to become comfortable with the idea that even if scientific knowledge is provisional, it is not necessarily wrong? How do we get people to accept the idea that policy decisions made on the basis of a provisional understanding of a process is not foolhardy because the provisional understanding is the best that can exist with the current state of knowledge?

Consider what happens if global warming is happening and we do nothing about it because absolute certainty is not available. The results to society will be devastating: inundation of coastal cities and island nations, changes in agricultural patterns, severe droughts, etc. On the other hand, if we take steps to reduce carbon emissions, and global warming is not occurring, the only cost to society will be monetary. Comparing the two sets of consequences, does it make sense to wait for certainty?

No comments:

Post a Comment