A long list of National Academy of Science members, including several Nobel Prize winners, have published a letter in Science supporting the idea that humans have caused/will cause serious global warming. The letter is striking in several ways — how preachy it is, how it overstates its case, how it fails to provide evidence, and how it ignores the main arguments of skeptics (at least, intelligent skeptics).
It begins:
All citizens should understand some basic scientific facts. There is always some uncertainty associated with scientific conclusions; science never absolutely proves anything. When someone says that society should wait until scientists are absolutely certain before taking any action, it is the same as saying society should never take action.
“Citizens”, huh? This might interest third-graders; if they think that the brighter skeptics or most readers of Science don’t know these “basic scientific facts” they are mistaken.
The letter goes on to claim that the idea that humans are seriously warming the planet is as well established — at least, in the same category of firmly-established theories — as the conclusion that “today’s organisms evolved from ones living in the past”. That is an overstatement.
And the letter ends with hand-waving. In place of evidence that supports what they claim, they simply repeat the claims in detail (e.g., “Natural causes always play a role in changing Earth’s climate, but are now being overwhelmed by human-induced changes”).
The letter is unintentionally revealing. Here’s what I would consider reasonable evidence for serious human-generated global warming:
- Temperature higher now than in the past.
- Temperature increasing at a higher rate now than in the past.
- Good (= verified) model shows serious human-generated warming.
No. 1 isn’t clearly true; the Medieval Warm Period appears to be as warm as now. (Mann et al. understood this point; they tried to diminish the Medieval Warm Period.) No. 2 isn’t clearly true. For example, the 1930s may have been as warm as recent decades. No. 3 isn’t true. Models such as Hansen’s haven’t been shown to predict correctly. There’s no reason to take them seriously.
So No. 3 is off the table (current models are untrustworthy). That leaves Nos. 1 and 2, the failure of which to be clearly true points in the direction of no serious human-generated warming. If a theory makes two predictions, both of which appear wrong, it would be wise to start doubting the theory rather than lecture the rest of us on “basic scientific facts”.
This line of reasoning (ask whether the humans-have-caused-serious-warming idea makes correct predictions) isn’t complicated or obscure but does require you think for yourself rather than accept what you’re told. Apparently no one in this long list of distinguished scientists has done so.
If a letter from 100 United States Senators was full of spelling and grammar errors, would you trust it? Well, no . . . and you might wonder about a world with such a poorly-educated ruling class.
So…were the prize winners actually responsible for the writing, or did they just sign off on it? It doesn’t seem to scan like the stuff these people write to each other.
I would be interested in knowing which one of the signed names is actually responsible for the initiative.
It certainly looks bad, but there are a lot of reasons to suspect it has little or nothing to do with what the signatories actually care about. To bring it near a blog theme, an unmotivated gatekeeper is just about the last person to go to if you really need frank advice on a fraught subject.
Andrew, my expertise is more relevant than you seem to realize. Hal Pashler and I wrote a paper about modeling that said that a 50-year tradition (within psychology) was absurd. I submit that how much to trust models is relevant in the climate-change debate. In the post you link to (by Phil), Phil appears unaware of a long literature in psychology about overconfidence in confidence intervals. On a wide range of topics, people’s estimated confidence intervals are too narrow. Here again my expertise (in psychology) is relevant.
Feel free to punch holes in my easy-to-understand argument. I don’t think it’s “laughable” to disagree with expert opinion because the experts can be glaringly wrong, as the books I call the Emperor’s New Clothes Trilogy illustrated. Don’t you have a copy of The Experts Speak?
Bravo, Andrew. Well said.
What bugs me about posts in this vein, which seem to be appearing with increasing frequency, is that what started out as a sentiment that expressed something like: “just because someone’s an expert doesn’t mean you should cede to them all responsibility to think and reason” has become something much closer to: “he’s an expert, so he must be some combination of a buffoon and a criminal.”
Like that string of posts about how what motivates academics is to be purposefully useless and obscure. Probably there is, somewhere, an academic with such a bizarre and twisted motivation, but in any substantial number? Really? This is the atmosphere at Berkeley, then, is it? Well, not among the people that I know. And not at my institution, either — not a single person that I’ve met believes this, or acts like it.
Every so often there are real gems on this blog. But mostly I find myself reading it because it infuriates me; and I have this idea that it’s good to be infuriated by stuff to keep you honest.
Shane, thanks for reading and commenting. You write: “Probably there is, somewhere, an academic with such a bizarre and twisted motivation, but in any substantial number?”
And outsiders call academia an “ivory tower” because . . . ? Surely academics don’t do research that is useless by mistake. It’s Veblen’s idea, not mine, that academics try to be useless and obscure. In The Theory of the Leisure Class, he gave plenty of evidence for such motivations inside and outside of academia.
Alrenous, Peter Gleick is responsible for the letter. See
https://dotearth.blogs.nytimes.com/2010/05/09/sweating-the-details-in-climate-discourse/
for more details.
@Shane:
I actually have heard a scientist say in response to the question: “How will your research in neuroscience help people make better decisions?”:
“My goal isn’t to produce theories that help people to make better decisions. If the do that’s fine but I’m rather interested in doing basic research.”
Sure no academic would admit that they want to be useless but they want to do pure basic research instead of solving real problems.
Just like artists paint for the sake of art there are scientists who want to do science for the sake of science.
Just like art loses it’s purity when it’s done for money science can lose it’s purity when it’s done to solve practical problems.
Thanks Seth.
So the actual author of the letter is the co-founder of the Pacific Institute, with the mission, “We envision a world where the basic needs of all people are met, where resources are managed sustainably…” and I need not read more. If global warming gets dethroned, such an institute can expect to lose substantial amounts of both leverage and funding.
Conflict of interest? What’s that?
Hmm, I clearly should have checked the hosting url of the letter…but on the other hand I got you to do the work for me. Never underestimate a human’s ability to ensnare others to their goals.
Ironically, I have to agree with Gleick’s water policy suggestions. (E.g. he thinks you should charge for use where water is scarce.) But then, that’s where his actual expertise is. What he actually cares about.
Seth what do you think about the post by “Phil” linked to above, specifically in terms of climate sensitivity? The question is, how many degrees C will the temperature increase, given a doubling of CO2? Would you be willing to share your ideas in what this value should be?
As far as your 3 points, ISTM that we could still face serious human caused warming even if 1 and 2 were false. Even if there were warm and warming periods in the past worse than today, eventually those past warming periods were followed by cooling. But today with our high CO2 load, we have grounds to be much more concerned that current warming trends will not be reversed, due to the aforementioned climate sensitivity.
thanks Seth,
The so called “experts on global warming” need a grade school refresher course in the basics of life of the majority of plants and animals and combustion on this panet.
Most plants require CO2 ( carbon dioxide) to live … and give off O2 (oxygen).
Most animals require O2 to live … and give off CO2.
combustion requires O2 to occur … and gives off CO2.
More CO2 = more plants = more O2 for the planet.
This is our plant’s major natural balancing mechanism between plants and animals and combustion, which keeps CO2 and O2 in dynamic equilibrium.
What do I think of Phil’s analysis? I think it ignores two facts that argue against big sensitivity of global temperature to CO2. One is the way carbon dioxide concentration lagged — not preceded — global temperature changes for a long time. Presumably as the temperature changed, the oceans changed how much CO2 they dissolved. If CO2 sensitivity was great, as the CO2 concentration increased, the CO2/temperature relationship should have changed somehow. It didn’t. The other is the fact that global temp is not now increasing at an unprecedented rate, in spite of the fact that CO2 is now increasing at unprecedented rate.
yes, Kim, I agree: absence of evidence is evidence of absence — and it is sort of amazing that people claim otherwise. Likewise, correlation really is evidence for causation, even if there are other explanations. There are always other explanations.
Kim, your web page about evidence of absence is very clear and convincing. If you do a proof for correlation/causation I will link to it.
@seth: the ‘ivory tower’ thing is real, for sure, in the sense that people are suspicious and contemptful of academics. My tendency, based on extensive personal experience with humans, is to think that says a lot more about the people than about the academics.
@ChristianKl, there’s a huge, HUGE difference between 1) not worrying too much about how the things you do will be used by other people, and 2) trying, on purpose, to do things that are purposefully useless to other people. You have given anecdotal evidence for the former, which I would never contest. As to the latter, though, like I said, I’ve never seen hide nor hair of it, except when uttered by number theorists in a fit of pique.
I think I’ll read Veblen this summer, though. I know Seth is smart, and at this point I’d like to see what he finds so compelling.
Andrew, you mention “the physicists who signed that letter.” The NAS has about 2000 members. The letter was signed by an eighth of them. Making it likely that most NAS physicists didn’t sign that letter. If physicists are divided on this issue, your reliance on Phil because he knows more physics than me doesn’t make sense. If Phil is in the minority in his profession, it makes even less sense.
Shane, I don’t know if you are a professor or not but really among many professors there is a disdain for useful work. “Basic research” (sometimes called “pure research”) is higher status than “applied research”, for example. Even in departments that you might think have to be useful. In engineering, an assistant professor complained to me that all her colleagues valued was theory, actually-useful stuff was nothing. The attitude goes beyond “not worrying too much about how the things you do will be used by other people.” Again, professors are just part of a much larger pattern, which Veblen was the first to describe.
Of course research scientists aren’t interested in solving problems. Solving problems is called engineering. People who are interested in solving problems, and can and care to develop the skills to do it, become engineers. The rest do other things. Research scientists do research. That’s good, because not many people have the peculiar sort of patience needed to do research. I wouldn’t want research scientists wasting their time being bad engineers, when there’s so many questions of basic resarch going begging.
People who want to understand things become scientists. They don’t always remain scientists, though, because real science is frustrating, rarely yielding up the answer you wanted. Many become, instead, believers in whatever they were taught, or in whatever feels more comfortable. The less they know about something, the less scientific feel they need to be about it.
“Of course research scientists aren’t interested in solving problems.” Of course? Physics began with problem solving — aiming artillery. Chemistry began with alchemy. Statistics began with practical problems, such as improving agricultural yields. Because being useful is low-status and difficult, scientists moved away from practical applications as soon as possible.
Nathan Myers — he was talking about engineering.
I agree that if there were substantial disagreement among the members of the NAS, that would be evidence against global warming. But if that were the case, I’d expect the dissenters to vocally object to this much publicized letter which was released in their name. What do you think are the chances that we will see a counter-letter from the NAS?
Of course engineering is a low-status activity among scientists. When you’re a scientist, being useful is somebody else’s job. Your job as a scientist is to understand and explain what is mysterious. If you want applause for engineering, look to other engineers, or to people with problems they need a solution invented for. That’s what engineers do.
Similarly, making stuff work right is low-status among engineers, for the same reason: that’s what technicians do. An engineer doing technical work is probably procrastinating. Technicians you’re likely to encounter include auto mechanics, airline pilots, cooks, and clinical physicians. Other technicians, and people who need things done right, appreciate technicians.
Physics (really, dynamics) started with artillery and developed into a science. That is always how it happens: practice, followed by systematic refinement, followed by theory. Scientists pretend it happens the other way, to pump up their status. People are often fooled.
I should mention that professors of engineering are far from the same as engineers. It’s rare to find engineering being done by the faculty at an engineering school. (Likewise, it’s rare to find the faculty at a hairdressing school cutting hair.) The students do it. Real engineering is done out of sight by professional engineers at Boeing, at Intel, at Bechtel, and at a million other companies.
Hal, this letter wasn’t supposed to represent the official NAS view or anything like that — just the views of those who signed it.
Seth, OK, well it will be interesting nevertheless if a contrarian letter appears from others in the NAS.
My real interest in this topic is as an example of applied rationality and the methodologies we use to get at the truth. Along these lines I thought I might mention a perhaps little appreciated requirement of rationality and honesty. Basically, you have to expect to be surprised by the future.
That is, whatever percentage estimate you give to the likelihood (in this case) that anthropogenic global warming theory is true, you have to expect that future data is equally likely to make you either more or less confident in your belief. Your percentage estimate is equally likely to move in either direction. The reason is simply that if you think it is much more likely to move in a certain direction, then you are really misstating your belief today. As a simple example, a person can’t say he is 50-50 on an issue but believes that he is very likely to be 60-40 in the future. That belief means that he is more than 50% believing in it today.
The upshot is that you have to expect to be equally likely to have to say “maybe I was wrong” as “see, I told you so”, as future data comes in. It is a simple requirement of rational and honest, evidence-based belief.
Unfortunately our human psychology doesn’t give nearly as much credit and prestige to people who say “maybe I was wrong” as those who say “see, I told you so”. This causes people to overstate their beliefs, and leads to polarized and hostile discourse, as we so often see.
Seth, as you might imagine I disagree with almost everything you say here. But let me give just one example. You say:
“Here’s what I would consider reasonable evidence for serious human-generated global warming:
1. Temperature higher now than in the past.
2. Temperature increasing at a higher rate now than in the past.
3. Good (= verified) model shows serious human-generated warming.”
But in fact, neither of the first two would be reasonable evidence of serious human-generated global warming if they are true, and the absence of them does not indicate absence of serious human-generated global warming if they are in fact absent.
Here’s the key fact that you seem to be unaware of (but that scientists who study this know very well): MORE THAN ONE PARAMETER AFFECTS THE TEMPERATURE OF THE EARTH. It’s not all about carbon dioxide concentrations.
If high temperatures in the Medieval Warm Period were due in part to higher solar activity, then those temperatures don’t tell us much about climate sensitivity to CO2. So your point 1 doesn’t really make sense. The issue with point 2 is pretty much the same, except with regard to the derivative of temperature rather than temperature.
And for the models…you assert that the models aren’t good enough to estimate temperature sensitivity, but you’re wrong.
Yes, I know that scientists, like everybody else, tend to be overcertain. But I don’t think that means that nobody knows nuthin’.
I asked you before, more than once, to give YOUR estimate of climate sensitivity (defined as the steady-state change in global average over preindustrial levels in response to a doubling of CO2 over preindustrial levels). You still haven’t answered that question. But it seems that for some reason you put very little of your probability in the range that almost all climate researchers think is most likely. So who is being overcertain?
Phil, you write “you assert that the models aren’t good enough to estimate temperature sensitivity, but you’re wrong.” What’s an example of a good-enough model? I predict that the model you put forward as good enough will turn out to not have been verified, “verified” meaning “found to make accurate predictions about global temperature”.
i’m the only one here who can sign a letter. all the rest of you can’t, because your signatures are longer than a letter.
Seth, the Intergovernmental Panel on Climate Change has an uncertainty window about 6 C wide, for their estimate of “climate sensitivity.” That’s very wide.
You refuse to say what your interval is, but it seems that it is narrower than 6C. It also seems that you put far more probability near very low values of climate sensitivity.
So, the implicit “Seth Model,” or perhaps it is a family of models, has a narrower confidence interval and a much lower central estimate than the experts do.
Since you are the one making the extreme claim here, I think YOU are the one who should justify what you are saying. What is YOUR “good-enough” model of global temperatures, that predicts the temperature rise of the past century and especially the past decades, is consistent with solar intensity measurements, is consistent with the physical laws governing energy transfer, but does NOT have a climate sensitivity of at least 2C per doubling of CO2?
Phil, you wrote “but you’re wrong” without any explanation of why I was wrong. I requested an example to try to understand what you wrote. I’ll provide you my confidence interval in a few days.
Check the IPCC report, chapter 8: https://www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-chapter8.pdf there is an extensive discussion of climate modeling and why it is good enough to be sure that anthropogenic climate change is real and why the key parameters are likely to be not extremely far from the (admittedly wide) range of estimates.
But what I am taking issue is with much larger than your three assertions, though I think they are all incorrect.
You are criticizing people for being “overcertain” because their estimate of climate sensitivity has an uncertainty of “only” 6C, but your own estimate has less uncertainty than that, in addition to being centered much lower than the experts think is reasonable. I do not think you have any basis for having a narrower confidence interval than the experts do, and I do not think you have any basis for having a much lower central estimate than the experts do.
To your first two points, there is nothing that says that warming has to be simultaneous with greenhouse forcing. Under present circumstances we in fact expect the contrary: the thermal inertia of the oceans and the masking of industrial dust mask the committed warming. So your proposed tests simply aren’t valid.
Also, climate models make effective predictions of many things, not limited to global mean surface temperature. As a particular point, they show and have shown since the 1980s that greenhouse forced warming is a near-surface phenomenon, accompanied by stratospheric (upper atmosphere) cooling. They also show a pattern of warming that concentrates on land areas in continental interiors. These long standing predictions did in fact emerge. So your claim that climate models are without skill are without foundation.
All of this is silly; it treats “global warming” as a falsifiable theory, a claim of a causation that is either true or false. This is foolishness. The underlying phenomena, (thermal radiation, absorbtion and reradiation in gases) is two hundred year old physics that is as well established as anything in science.
What’s at issue is not “whether” but “how much”. As Phil correctly points out, those who call themselves “skeptics” are not making an argument from ignorance, they are making an argument from certainty. If we really had no idea what the sensitivity was, we should be more concerned than ever. The attitude suggested by people who claim to be “skeptics” is utterly inconsistent with a lack of confidence in the underlying science. It is a claim that the the sensitivity is certain to be much less than all prevailing evidence indicates.
(Phil’s a bit pessimistic; I’d say the consensus range is 2 C to 4.5 C per CO2 doubling. https://www.nature.com/ngeo/journal/v1/n11/abs/ngeo337.html )
And what if it were zero? Would we be out of the woods? No, even with a global sensitivity of zero we could have huge forced climate change, say, warming the poles and cooling the tropics. There is no doubt that the amounts of CO2 and other gases we are adding to the atmosphere change the way energy flows through the system, a system which at heart is fluid and easy to change. Even in the absence of knowledge and a low global sensitivity there are huge risks.
And then there is ocean acidification.
https://en.wikipedia.org/wiki/Ocean_acidification
Sorry. We have to deal with this.
@Phil: If you claim as the scientists did that global warming is a fact that has the same certainty as evolution than your confidence interval is very small.
The members of the academy probably all agree that p(Evolution is wrong)
To add to what Phil said (that the physics of thermal radiation, absorbtion and reradiation in gases is well established), we have in the rock record repeated examples of global climate having been warmed and cooled by alterations in CO2 sinks and flows. We even have record of a methane induced warming event. Is it so hard to believe that injecting a significant percentage of the world’s fossil carbon into the atmosphere at an — and this truly is novel, in the history of the planet — unprecedented rate will have profound, global effects?
” the Medieval Warm Period appears to be as warm as now.”
Are you sure of that?
Here’s a good graph of the various reconstructions:
https://www.ncdc.noaa.gov/paleo/pubs/ipcc2007/ipcc2007.html
Yes – there was a Medieval Warm Period in the reconstructions – but even if ignore most reconstructions and cherry pick the ones that give the warmest results they are still lower than the average current temperatures.
Obviously these temperatures are averaged – there of course can be a day in that period which is higher than a day now !
I think the broad question is, “are humans ruining the planet”. From there, we get the compound, “humans are causing an increase in CO2 that is causing an unusual increase in global temperature that is ruining the planet.” Four parts there:
1. Humans are causing an increase in CO2.
3. Global temperature is increasing.
2. CO2 causes global temperature increase.
4. Increased global temperatures of that scale are bad.
I think we all can agree on #1. Perhaps not, but I’ll not bother addressing it.
I find that #2 and #3 are often presented side-by-side … I have to rely on my intuition and faith in what constitutes bias, and what I think is an “unbiased source”. I have seen models and data that I largely don’t understand, but if I rely on the conclusions of the presenters, the consensus appears to be that CO2 affects global temperature and increases it in a measurable (but not huge manner compared to other forces like solar output). Nobody is saying that CO2 decreases global temperature, so those who disagree claim that CO2 either has no effect, too little an effect, or that the effect of CO2 cannot be measured. I believe it’s possible to isolate factors in muddled data like global temperature just as I believe it’s possible to divine a sum of sine waves from a complex waveform without understanding how Fourier transforms work, so I favor those who can come to a conclusion.
But is #4 true? Is it really that bad? I don’t think it is possible to answer that question without running the experiment. My gut tells me that expending effort to reduce CO2 to avoid further affecting global temperature seems to be the wisest course of action. I strongly believe that #2 is true, but even if #3 is false, I feel the effort gambled is relatively small.
What it comes down to is a gambling problem. The potentially devastating outcome (i.e. massive bad global changes) has some probability P(bad). You are to select an action that may change P(bad). If you find an action that will significantly reduce P(bad), then you’d clearly take it. The trouble is, will acting to suddenly reverse CO2 output actually increase P(bad)? I am under the belief that taking action will not increase P(bad) and may reduce it, so I lean toward taking action. I am frustrated that there is not a better way to isolate P(bad) and how it is affected by changes in behavior.
Jason, the answer to your broad question is surely yes — and the narrow focus on CO2 is a big mistake that wastes a lot of effort. No one has to wonder if pollution is a problem — it is right now. No one has to wonder if we’ve had profound stagnation in development of new forms of transportation. We have, no doubt about it. No one has to wonder whether we should reduce energy costs and improve long-term supply by developing new sources of energy — of course we should. Let’s start by solving the obvious problems.
We have in the geologic record many changes to global climate that had a variety of causes, change in atmospheric chemistry having been one such cause. Local and mass extinctions were sometimes caused by these changes. The change we’re witnessing now in global climate is on the order of the larger of these events — for example, the amount of CO2 we’ve injected and are likely to inject is will probably put us in a climate state similar to the Miocene/Pliocene warm period. The current rate of warming far exceeds anything seen in the Pleistocene, let alone anything in the current interglacial period, including the medieval warm period.
The current extinction rate matches that of the 5 global mass extinctions that have occurred in earth’s history, and the unique imprint of humans is the likely cause: humans have fragmented gene pools, and humans are causing a change in climate that is occurring at an unprecedented rate.
To reiterate: the rate of change is as significant as the change itself. Remember, within Pleistocene glacial periods there were changes in sea levels of up to 50 meters (there were five such sea level maximums during the Wisconsinan, and they mark interstadials). Despite these radical fluctuations, mass extinction did not result. Our period is unique in that we’ve already exceeded some of the variations that were typical of the last few million years.
We have in the geologic record many changes to global climate that had a variety of causes, change in atmospheric chemistry having been one such cause. Local and mass extinctions were sometimes caused by these changes. The change we’re witnessing now in global climate is on the order of the larger of these events — for example, the amount of CO2 we’ve injected and are likely to inject is will probably put us in a climate state similar to the Miocene/Pliocene warm period. The current rate of warming far exceeds anything seen in the Pleistocene, let alone anything in the current interglacial period, including the medieval warm period.
The current extinction rate matches that of the 5 global mass extinctions that have occurred in earth’s history, and the unique imprint of humans is the likely cause: humans have fragmented gene pools, and humans are causing a change in climate that is occurring at an unprecedented rate.
To reiterate: the rate of change is as significant as the change itself. Remember, within Pleistocene glacial periods there were changes in sea levels of up to 50 meters (there were five such sea level maximums during the Wisconsinan, and they mark interstadials). Despite these radical fluctuations, mass extinction did not result. Our period is unique in that we’ve already exceeded some of the variations that were typical of the last few million years.
oops,this phrase: “we’ve injected and are likely to inject is will…” should just read “we’ve injected and are likely to inject will…”