Global Warming is in the news again, this time with the reported suppression of a paper written by a highly respected German climatologist because it might contribute to the skepticism expressed in some circles about global warming. What I'll pick up from the news here is a trend which has been called the "democratization" of science. Both sides in this debate decry this trend, but identify it differently.
Advocates for public policy changes to arrest global warming complain about the idea that anyone with a blog (present company certainly included) can contribute to science. Skeptics complain that politicians think science can be subject to public opinion polls and presume to dictate when a debate has been 'settled'.
I come to this debate first and foremost as one who has taught undergraduate courses in critical thinking. Thomas Jefferson noted: "In a republican nation whose citizens are to be led by reason and persuasion and not by force, the art of reasoning becomes of first importance." Founding Fathers like John Adams believed strongly that democracy required the general populace be well-educated in order to be able to study and understand the issues of the day. So my questions surround what appears to be a disconnect between the nature of the inquiry at hand - the changing climate - and the conclusions being presented.
But before I take that up in detail, let me comment on the advocates' complaint about the 'democratization of science'. This problem isn't really new. It happened a few hundred years ago in another form - the 'democratization of theology' - and caused quite a stir. A certain mathematician, who was also a devout Christian, understood well what the Church taught about the cosmos. And then he improved on an earlier invention - the telescope - and began to chart the movement of the planets and stars. And this created quite a problem for him.
By now it should be clear we are talking about Galileo Galilei. But there is something about the history of the 'Galileo affair' which is little understood, and very pertinent to this debate.
The term 'binary logic' is sometimes used by philosophy students to describe a process of thought which tries to reduce everything to propositions which are either true or false (or 1 or 0 if you are a geek like me and make your living in the digital world). For Galileo, what the Church taught about the heavens and what he was observing and documenting clearly presented a binary problem. What the Church was teaching simply could not be true if he was really seeing what he was seeing. Something had to give.
So Galileo presented what he thought was a solution. The passages in the Bible which described the course of the heavens were not to be taken literally, but to be understood as using 'phenomenological' language - that is to say, language which described something as it appeared to the unaided eye rather than how it actually happened. And it was here that the Church really got wrapped around an axle.
The idea that a mathematician would presume to instruct the Church on how she read her Bible did not sit very well. I can imagine the theological academe reacting with scorn: "Um, we have people to discuss and debate how we read the Bible, thank you very much... You just stick to your math, astronomy... or whatever it is that you do..."
And Galileo refused, of course. And the rest is, as they say, history.
This is instructive in light of the debate over global warming. The response to skeptics' questions is often nothing more substantive than an ad hominem attack on the skeptic's lack of a scientific resume. It is as if to say: "Um, we have people to discuss and debate changes in the earth's climate, thank you very much. You just stick to your computers, your blog... or whatever it is that you do."
This is not to claim that anyone can do science. It is to claim, though, that any reasonably well-educated person can question it. And when science is deployed in an effort to influence public policy, in a democracy, that questioning is actually essential. So here I'll try to explain the nature of my questioning - which orbits around a disconnect between the nature of the inquiry and the conclusions being presented.
Science 101: Correlation Does Not Mean Causation
And you do not need a degree in science to understand this. But this does not mean correlations are not important to science. To maybe oversimplify, when science collects and examines data, discovering a statistically significant correlation between two things - say atmospheric CO2 and global mean temperatures - then helps a scientist form hypotheses which can be further tested. But it is important to make the distinction between the correlation on the one hand and empirical evidence on the other which might be discovered as a result of the testing of subsequent hypotheses.
This, then, brings us to the data and the statistical modeling which is done with the data. If the conclusion being presented is the existence of a correlation between concentrations of atmospheric CO2 and global mean temperatures, then the statistics are evidence - evidence of the correlation. But if the conclusion being presented is one of cause-and-effect - that human activity is causing global warming - then the statistics are simply not evidence.
What is left for climate science to do here, then, is to plug the data into statistical models and predict what they think may happen. This is very different from observing and documenting what is actually happening. This is why we hear repeated appeals to consensus. This appeal should immediately prompt us to ask: "Why in the world would a scientist appeal to consensus if they have evidence."
Here we have to go back to the Galileo affair. The consensus was, of course, that the heavens all rotated around the earth. But Galileo was actually observing something entirely different. He had evidence to the contrary. As a scientific matter, the consensus was utterly meaningless in the face of contrary evidence. And this is why we rightly remember Galileo: He thought for himself - even about how his Church read her Bible - and so stood up to the consensus.
The consensus, at least as it is asserted, is that human activity is causing global warming. The IPCC actually attaches a percentage of 'certainty' (95%) to the conclusion. But let's be clear about the nature of the inquiry and the conclusion it is presenting. The appeal is made to consensus because they do not have actual empirical evidence. Global warming science is drawing its conclusions based on statistical models which predict what they think will happen, not what they have actually observed in empirical experiments. The complaint will immediately be made that global warming has been 'observed' in the data. So let's look at the data. But again, we must keep the nature of the conclusions being presented at the forefront.
I like to tell my critical thinking students to always 'start with the conclusion' when examining an argument. The nature of the conclusion governs the manner in which you examine the argument. If we are starting with a conclusion that asserts a percentage 'certainty' of human-caused global warming, it means we are examining an inductive argument.
In a deductive argument, a conclusion follows as a matter of mathematical certainty from its premises. The easiest way to see this is to note that if A=B and B=C, A must, as a mathematical certainty, equal C. In an inductive argument, however, instead of this kind of mathematical certainty we are looking for probability. The conclusion falls somewhere on a spectrum with 'not likely' on the left and 'highly likely' on the right. An inductive argument never really reaches mathematical certainty like a deductive one does. Thus it is unlikely global warming science would ever claim to be '100% certain'. Scientific inquiry simply does not entertain certainties like this.
In an inductive argument, the volume of relevant data is crucial. The more data you collect which can be shown to support your conclusion, the more you can push your conclusion to the 'highly likely' side of the spectrum. So how, then, does climate science go from less data to more certainty? Joseph D’Aleo and Anthony Watts show that around 1990 the reporting stations in the National Oceanographic and Atmospheric Association (NOAA) dataset dropped by no less than 75%. It can also be shown that the stations lost to the dataset were at higher latitudes (further from the equator), higher altitudes, and rural locations, all of which naturally tend to be cooler locales.
But even apart from the statistical bias this would cause, the more fundamental question - again, about the nature of the conclusions and the nature of the inquiry - revolve around the volume of data. You simply do not move an inductive argument more to the 'highly likely' side of the spectrum with less data.
A Product to Sell or a Problem to Solve
Before I continue, let me point out that I will not suggest here that climate science is engaged in a nefarious conspiracy, nor am I suggesting things are 'made up'. But I will suggest that the scientific community is no less subject to budgetary imperatives than any other community. And these imperatives become magnified when you marry the scientific community to quasi-governmental bureaucracies like the United Nations's Intergovernmental Panel on Climate Change (IPCC).
Advocates for global warming are fond of challenging skeptics by noting the source of their funding. So, if we are going to "follow the money," especially in light of how the very existence of a bureaucracy - and its budget - depends on concluding human causation of climate change, then by all means, let's follow the money.
If you trace back the dollars budgeted to a scientific organization, your inquiry will lead to one of two places. Either investors are providing the funds or public and private grant-making institutions are providing the funds. For the investor community to fund the science, it must be done in support of a product to sell. But there is a lot of good science being done which does not relate to a product to be sold. This science is funded by competitive grants. And to compete for the grant funding, you must have a problem to solve.
Climate science clearly does not have a product to sell.
Again, I am not suggesting some nefarious conspiracy. I am suggesting that scientists assert an economic value to their work - in their grant applications - on the basis of being able to solve a problem. If the globe is, in fact, warming (and the culling of the dataset is more than enough to call that assertion into question), but this warming is not being principally caused by human activity, then global warming is not a problem we can solve. Since climate science does not have a product to sell, to admit this possibility would be to call into question the economic value of the work being done.
There are tough questions to be asked, and a democratic society requires an educated populace willing to think for themselves and ask those tough questions. Among the tough questions are these: Is the IPCC now the institutional 'church'? Is the scientific community now the 'priesthood' that we dare not question? Are those who think for themselves in the face of a contrary consensus now to be intellectually imprisoned by the 'magesterium' of that consensus?
[UPDATE: Many thanks to my friend Allan Chhay - who almost always disagrees with me - for providing me this excellent and funny link about the climate change debate. There are a couple places where the language is, shall we say, less that the refined English we would otherwise expect from our British friends (??? - I am not sure, but that's how he sounds to me), but this is actually very revealing.
Watch for the "Statistically Representative Debate." The priests used to wear black. Now, it seems, they wear white lab coats. They gather as the new 'magesterium' to shout down the ones who dare question their consensus. There really is nothing new under the sun.]