Posted by: Barry Bickmore | April 30, 2010

Consensus? What Consensus?

The principle of “underdetermination” is a standard tenet in the philosophy of science.  The idea is that humans are only capable of making and processing so many observations, so even if we develop explanations (i.e., hypotheses and theories) that account for all the observations, it’s still possible that our explanations aren’t exactly true.  In other words, we can NEVER collect enough data to make sure our theories are exactly correct.  There could always be other, better explanations for the data that we haven’t thought of.

The practical consequence of this fact of life for scientists is that we can’t expect absolute proof of any scientific theory.  Instead, we are stuck weighing the evidence for and against.  But if there is always at least a little room for doubt, it also follows that there will always be doubters.  At least when it comes to theories that are complex to any degree, you can always find scientists who doubt those theories, no matter how well supported they are.  There are still a few geologists (with real PhDs) who don’t accept the theory of Plate Tectonics.  Some tobacco researchers still argue that second-hand smoke doesn’t have significant adverse health effects.  If there’s no such thing as absolute proof, who is to say how much evidence should be “enough” to satisfy everyone?  (For a great discussion of issues like this in the context of climate change, see this book chapter by science historian Naomi Oreskes.)

Even if we can’t obtain absolute proof, however, we still have to make decisions based on the limited information we have.  We use fallible economic forecasts to do state and federal budget projections.  We use fallible diagnoses from doctors to make decisions about healthcare.  We use fallible weather forecasts to decide whether to bring a coat when we go out of the house.  Similarly, state and federal governments need to make decisions about what to do about climate change based on the best information we have.

There’s no use whining that we don’t have absolute proof or 100% agreement among scientists, since that will never happen about any complex scientific issue, but that’s exactly the tactic used by Mike Noel and his extremist pals in the Utah Legislature to encourage complete inaction with respect to climate change.  After he was publicly criticized for inviting a parade of non-experts and fake experts to testify about climate change to his legislative committee, he buckled to the pressure and invited Jim Steenburgh, chair of the Atmospheric Sciences department at the University of Utah, to testify about possible effects of climate change on the state of Utah.  But to provide “balance,” Noel also invited Roy Spencer, a climatologist from the University of Alabama at Huntsville.

Roy Spencer is one of the few climatologists who disagree with the consensus view that humans are significantly affecting the global climate.  During the legislative committee hearing, Steenburgh pointed out a recent study published in the journal Eos (put out by the American Geophysical Union).  This study showed that 97.4% of active climate scientists surveyed agreed that “human activity is a significant contributing factor to changing mean global temperatures”.  In my experience, it is just not that easy to get such a large fraction of a scientific community to agree on any question of similar complexity.  Scientists get brownie points for showing that others’ work is flawed in some way, so if there isn’t a great deal of evidence you just don’t get that kind of broad agreement.

But for extremists who already have their minds made up, any little excuse for skepticism will do.  For someone like Roy Spencer, who is an actual climatologist, it’s easy to point to an area of the science that admittedly is poorly understood, like cloud physics.  He harps on the idea that large uncertainties in the way clouds behave could lead to large uncertainties in climate model projections.  But what he doesn’t tell you is that the climate models currently used have been tested extensively on data regarding past climate.   For instance, standard climate models do a pretty good job of mimicking the large temperature swings during the glacial-interglacial periods of the last million years or so.  We know what was ultimately driving these temperature swings (changes in the distribution of solar input due to changes in the Earth’s orbit,) and we know how the greenhouse gas concentrations changed in response to the warming and cooling.  When you put all that as input into a standard climate model, you get pretty good mimicking of the temperature data.  But has Roy Spencer tried to reproduce any such past climate data by using a climate model with significantly different cloud behavior?  No, he has not.  Why?  Because it wouldn’t work.  Certainly the models might be getting the right answers for the wrong reasons.  That’s always a possibility with complex models.  Spencer just hasn’t given any compelling reason to throw out models that seem to work pretty well and replace them with…nothing.

Let’s face facts, though.  Mike Noel and Co. don’t have the background to tell whether they should believe someone like Jim Steenburgh or someone like Roy Spencer.  I’m not saying they are stupid–just that it takes people YEARS of schooling and experience to get up to speed on a technical issue like this, to the point that they are in a position to make informed judgements about disputes between experts.  It seems to me that the rational thing to do in this situation would have been to go with the vast majority of the experts.  If 4 out of 5 dentists say sugar-free gum is ok for my teeth, well then I’ll chew my Trident without worry.  But since Noel and his pals are extremists whose only object was to cast doubt on mainstream climate science, they just ended up wringing their hands about how the scientists can’t make their minds up.

Rep. Lori Fowlke (R-Orem) moaned, “I hope the scientific community in some way will be able to reach a broader consensus….And I appreciate those who are willing to look at other options before we invest our entire future on science that may not be conclusive.”

Rep. Mike Noel (R-Kanab) added, “But if there is still a question out there, like Dr. Spencer, I’m not sure I want to subject my constituents to that in this state.”

You get the idea.  A whopping 2.6% of the experts disagree with the rest of the community, so they throw up their hands and complain, “How can we decide what to do when the scientists can’t even agree among themselves?”

Normal adults can accept that the world isn’t black and white, and that we have to make decisions based on incomplete information.  But extremists like Noel and Co. just don’t care about trifling things like the weight of the evidence and broad consensus in the scientific community.  What’s important to them is obtaining the desired result–not having to spend money on alternative energy and limiting government regulation of energy consumption.

More on the political angle later.



  1. Excellent post. Keep fighting the good fight.

  2. […] your colleagues.” Brigham Young University geochemistry professor Barry Bickmore made a similar observation at his personal blog, saying “scientists get brownie points for showing that others’ work is flawed in some […]

  3. […] really a problem, then why impose new regulations?  The problem with this approach is that there is always some uncertainty involved in any complex scientific conclusions, so the mere fact that uncertainty exists is beside the […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: