Weakness of the Will

November 10, 2009

George “WTF” Will has a new piece out in Newsweek, and it’s a real howler. Our skeptical castaway with implausibly many Ys in his last name has a few corrections for Will. More interesting to me, however, is the extent to which public pundits can get away with repeating the same tired lines, over-and-over again.

How many times have we heard that the climate is cooling now? How many times have we heard reference to the “global cooling” article in Newsweek from 1975? How many times have we heard that climate change is little more than a media phenomenon?

Most of the scientific blog community has set to demonstrating the falsity of Will’s claims by appealing to trend lines, offering five million objection-specific responses, or pointing out Will’s supposed similarity to Alfalfa. (Alfalfa? Really? Sherman is a far better match.) Journalists have done as much as well.

What’s crazy about Will’s claims is not what he asserts, but how self-defeating his argument is. It is asinine how dependent he is on climate science to support his criticism of climate science.

For instance…

If you say with one breath that climate models are imperfect and cannot sufficiently characterize fluctuations in temperature to make claims about the warming of the climate, and then you say with the next breath that there is some evidence to suggest that the globe is cooling (which implies therefore that there are strong enough climate models to support this view), then one must ask how you have arrived at the position that the globe is cooling. That’s supposed to be impossible, or problematic, or something over which one should be deeply skeptical. Either all we have are a bunch of raw data about temperature, or we have loose-fitting models that help us string this data together in a coherent fashion. Seems to me that Will can’t have it both ways.

Moreover, simply observing that the media can sometimes create the impression that things like shark attacks are occurring with greater frequency than they actually are, doesn’t at all impugn the science behind shark attacks. Matter of fact, it calls into question the media, not the science. Again, Will depends on the science of shark attacks to support his point about the alleged weakness of science. Then, somehow, we are supposed to bend our feeble minds around his idea that the climate science community is more like the media on the shark attack story, and the media is playing the role of science community by debunking the hype.

It’s a contortionism only Pretzelman McSwizzlestick could love.

To be fair, Will is careful not to take on the scientific establishment directly. He first associates the scientific position with the political star-power of Al Gore and his politicized oompa-loompas on the U.N. IPCC (thereby calling into question the authority of any report issuing therefrom). He then takes an indisputable claim — that models are imperfect — and backs it up with an appeal to a false authority (Jeffrey Ball, of the Wall Street Journal), peppered with a few appeals to anonymous authorities (some scientists). He tops it all off by introducing the Summer of the Shark. And that’s where it gets crazy.

What Will has told us is that the media can hype a string of events in the world, the public can fall for it, and the guardians of data can help us sort out fact from fiction.

Fortunately, it’s a good damn thing he points this out. That’s one reason we have scientists: to help us debunk silly folk theories. (Hell, why stop at scientists? That’s one reason we have academics! Go philosophy, rah rah!)

This all makes it mighty odd, then, that Will, a geeky desk-jockey at the Washington Post, snorts out this fluff about media hype and global warming. Is he, a member of the media, sending us a veiled warning about his own work, about his own institution? Maybe we should take his message to heart and appeal to scientists for clarity.


  1. It strikes me that with conservatives in clear ascendancy in the Republicans and with a Democratic national government, that many columnists who were more wiling to diverge from the orthodoxy in recent years are now towing the line on key issues where Republicans want uniformity, and this is one example. I’ve seen this with other commentators as well. It’s a “you’re with us or you’re with them” thing.

    There are of course plenty of places where one can find room to debate and disagree on climate change and be within the bounds of scientific discourse. But Will, who in the past I would have expected to search out that space, no longer seems to care.

    I think that this is common when ideologies are out of power. Better to unify against the threat you agree on (liberals in govt in this case) than the other aspects that divide you. Those become problematic when you are in the position of making policy and need to actually make choices.

  2. I have not read George Will’s piece.

    However, I was stuck by this statement in your post:

    If you say with one breath that climate models are imperfect and cannot sufficiently characterize fluctuations in temperature to make claims about the warming of the climate, and then you say with the next breath that there is some evidence to suggest that the globe is cooling (which implies therefore that there are strong enough climate models to support this view), then one must ask how you have arrived at the position that the globe is cooling.

    Now there is a big difference between climate models and observations. Climate models attempt to project what will happen in the future, while observations are merely measurements of what actually happened.

    So, without having read the piece, I have to disagree with your point, just based on the difference between climate models and measuring the climate. One can measure temperature over a period of years and draw the conclusion that it is cooling without using a climate model at all. Those are merely observations, averaged over some period of time, allowing you to compute a trend and draw a conclusion.

    Climate models on not the same as evidence.

    A simple climate model might mathematically related CO2 with temperature. You could then plug in the CO2 level and the model cranks out the temperature. You could graph the projected rise in CO2 and also graph the projected rise in temperature.

    If the observed temperatures are different than the projected model temperatures – and a graph of actual temperature diverges from the projected temperature – even while the CO2 is rising as projected, then one can question the climate model based on nothing more than observations.

    Now I don’t know if that is what George Will is doing – but I do think it is important to distinguish between climate models and scientific observations.

    After all, it is scientific observations which are used to validate or reject climate models.

    I think using actual measurements and observations to reject (or validate) climate models is actually scientific.

    It is my general impression that most if not all climate models are on their way to being rejected. Another four or five years of downward temperature trend will cause their rejection with 95% certainty.

    By the way – you should also keep in mind that none of the climate models have been scientifically validated and the IPCC AR4 report makes that point.

    Right now – each climate model is just someone’s opinion about what they think is going to happen and none have been scientifically validated against actual observations.

  3. Well that’s not true, is it? Some models are not predictive, but instead explanatory. The question is whether the world is warming or cooling. Many models say that, based on the data, based on the trend lines, it is warming. That’s not predictive. That’s explanatory. Plus, there are many other questions regarding models that might fly in the face of your claim.

    Here, enjoy:


  4. I would quibble with your question a bit. I don’t think anybody is arguing that the world has not warmed over the last 100-200 years. I think the question is whether the observed warming is natural or caused by humans.

    During the little ice age the world cooled – but why did it cool? Was it a drop in solar radiation? Was it an increase in volcanism? Or did humans cause it by buring to much wood?

    Since the little ice age the world has warmed – but what is causing it to warm? Is it a natural process, or human caused, or a combination?

    If it is natural, then changing our behavior (i.e. putting out less CO2) will not have an effect on future temperature increases. If humans are causing the warming then changing our behavior should change future temperatures.

    When we look at the temperatures and say the world is warming – I agree with you – that is not a model and it is not predictive – that is observation.

    When a climate model projects that the world will warm between 1.5 and 4.5 degrees C by 2100 that is predictive – it is based on a mathematical model.

    I guess when I think of climate models I am thinking of the 22 different models which the IPCC used in AR4 to project the future climate – and they are predictive – they are mathematical.

    Until the climate models are scientifically validated, we don’t know if there predications will be correct.

    We have certainly observed many new factors which are being added to models to “fix” them (i.e. make them more accurately match observations). Cloud cover, land-use, carbon black, aerosols, ocean currents, carbon cycle, solar magnetic fields, etc. have all been found to correlate with temperature and many of these factors are being worked into the climate models.

    A lot of science has been done to show that the models of the past 20 years are not accurate.

    Drew Shindell just published a paper showing that CO2 is only 43% of the cause of global warming – and that methane plays a much greater role than previously modeled.

    Dr. Pielke Sr. published a paper that showed that land use changes may be an important factor on temperature changes and they are not being modeled at all.

    My guess is that the models will keep adding more an more new factors – which is good and to be encourged. However, the current models are not good enough, in my opinion, to make decisions to spend trillions of dollars, just because some equation says the world may be 1.5 to 4.5 C warmer in 90 years.

  5. It’s true that it matters whether we’re doing it, but the scientific angle isn’t the only question; and it’s definitely not the case that just because it’s not all CO2, that therefore we’re off the hook. If there are more anthropogenic forcings, as Pielke suggests, then those implicate other human activities as well…

  6. RickA, by your fourth paragraph it became entitrely clear that you have little idea as to what the science actually says about this stuff. Why not find out?

    BTW, the land use impact stuff isn’t original with Pielke Sr., although you’d think it was to hear him go on about it. The case he tries to make for CO2 not being “dominate” is very thin indeed. A key difference to bear in mind is that it’s CO2, not land use change, that threatens the ice sheets.

    Re Shindell, guess who he works for and who originated the idea that methane is so important. Same guy, as it happens. It’s a small world.

  7. Oh, re land use chnages not being modeled at all: What an ignorant claim. Actually I doubt that RP Sr. said any such thing. Time for you to apply to start writing a WaPo column, RickA.

    • That’s an interesting assertion. I had no idea that they had this level of sophistication. Who is doing this modeling and what form of data are they using for the inputs ? Just collecting the inputs must have been a herculean task. Thanks, this look like a very interesting topic to look into further !

  8. Steve Bloom, the point is that new science keeps coming out showing that the old science (the climate models) are not modeling things correctly or are overlooking important things which influence temperature.

    Now I have no doubt that the models will keep getting better and better. Maybe in 50 years we will actually have the science correct enough to actually use the models to make political decisions.

    I just don’t think we are there yet.

  9. Steve Bloom – I read this statement as indicating that land use is not properly modeled:

    Because global climate models do not accurately simulate (or even include) several of these other first- order human climate forcings, policy makers must be made aware of the inability of the current generation of models to accurately forecast regional climate risks to resources on multidecadal time scales.

  10. RickA, where do you get these misunderstandings? We don’t need the models to inform public policy. Present observations plus paleoclimate are more than enough. The models inform us about the rate of change and tipping points.

    • The models are already informing public policy. It is the models which project a warming of 1.5 to 4.5C over by 2100. It is the models which give rise to the call to hold warming to 2C by capping carbon output. Models are being pointed to as evidence that political action should be taken.

      The models are either right or wrong.

      If we take action assuming the models are correct and they are correct than that seems like a good thing to me.

      If we take action assuming the models are correct and they are wrong than that seems like a bad thing to me.

      What are the ethics if we fully implementing cap and trade, get the world to fully cooperate and then find out over the next 90 years that reducing our emissions to 1990 levels is having no effect on global warming. In other words we have spent trillions to no effect.

      What are the ethics if the models are wrong and the cooling which has occured from 2002 to present continues and becomes a 30 year trend which cannot be ignored. What happens if implementing cap and trade actually amplifies the cooling and makes it worse than it would have been without cap and trade. I would think this would be an ethically suspect outcome. I believe studies have shown that the same temperature decrease kills more people than an increase.

      So how reliable are the models. If you look at the evolution of the climate models over the last 20 years I think you will find they have changed tremendously. How will they change over the next 20 years. I believe they will change tremendously – because I don’t think we fully understand the climate yet – at least not enough to correctly model it over 100 year periods.

      • Rick, why not stop the hand-waving and make a serious study of the subject? You might start, e.g., by looking into the paleoclimate studies confirming that the range of climate sensitivity produced by the models is approximately correct. Then you might review the recent studies in which the glacial cycles have been successfully modeled. There’s a lot more you’ve never heard of.

      • “It is the models which give rise to the call to hold warming to 2C by capping carbon output.”

        Wrong, actually. The basis for this is paleoclimate studies.

  11. RickA, look at the complete list. It includes a number of minor terms that aren’t modeled. Land use change isn’t one of them. The statement itself begs the question of whether the missing factors are enough to affect results significantly.

Leave a Reply

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

WordPress.com Logo

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

Google photo

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

Twitter picture

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

Facebook photo

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

Connecting to %s

%d bloggers like this: