Debunking Cool It Critique
My short answer to the critique on Cool It is that it is mostly right about the details, but completely wrong about the big picture. The most comprehensive resource of critique on the book can be found on http://www.lomborg-errors.dk/. Kåre Fog and others have spent many years creating an inventory of every mistake – big and small – that Lomborg has made over the last decade.
The idea behind the site is the following:
For every piece of information in the books, you have to check if it is true and if the presentation is balanced. If the concrete information given by Lomborg is correct and balanced, then it follows that his main conclusions are also correct. But if the information is flawed, then the main conclusions are biased or wrong.
This is in clear disagreement with my statement that one can actually be mostly right on the details, but completely wrong on the big picture. One could consider this a pointless philosophical debate, but since it is the main premise of his entire effort, there is no harm in analyzing the statement.
It sounds very intuitive that if the details are wrong, than the whole must also be wrong. On the other hand we have this conventional wisdom that the whole is more than the sum of its parts. What does this tell us, apart from that it is generally a bad idea to listen to your intuition without proper training?
There are actually plenty of examples where all the details are “wrong”, but the big picture is right. Many individual measurements of the length of an object may all be wrong, but their average is pretty accurate. The algorithm that Google uses to rank pages may very well be wrong on each individual page, but the results are impressive nonetheless. Most climate models divide the world into arbitrary squares that have nothing to do with reality, yet their results are useful. The same goes for many other assumptions and simplifications in computer models. Newton’s theory about the universe is completely wrong, but the predictions are nonetheless useful. The vast majority of authors on Wikipedia are not the worlds best experts at what they write, but collectively their produce a similar or – according to some – even better result than those same experts would have achieved.
Fogs hypothesis here is false. Even if all Lomborg’s facts are completely wrong, his main conclusions can still be right. Vice verse even more so.
Recently the site also tracks mistakes made by Al Gore and Fog has come to the following conclusion:
However, when Al Gore and Lomborg are judged by the same standards, there is a wide difference in credibility. In those texts that deal with the climate issue, Lomborg has on average about one flaw or error per page. By comparison, Al Gore´s book has 325 pages. Even if we consider that, because of photos and large letters, this would compare to only 100 pages of Lomborg´s type, that would amount to only 0,13 flaw or error per page. In the film, there is on average one flaw or error every 9th minute. You have to watch the whole film in order to meet as many distortions as there are in 10 pages of one of Lomborg´s books.
Although I appreciate Fogs gesture to correct for the number of images in Gores book, this is complete nonsense. Perhaps he should also compensate for the fact that a picture is worth a thousand words?
If comparing the “credibility” of two works based on the number of errors makes sense, then what would you say in this case?:
Person A : “On pllus oone iz too.” (5 errors)
Person B : “Murder is legal.” (1 error)
Another interesting point of view from Fog is the following:
Especially during the first years of his public appearance, this was Lomborg´s style again and again, and it has – justifiably – made a lot of his opponents become his enemies.
Very professional indeed…
But perhaps Fog’s actual arguments do make a case against the main messages of Lomborg’s book?
Before we deal with them – and rest assured I will not be very kind about Lomborg’s examples either – let me summarize what I think are the core take home messages of Cool It. That is, what I take home from it, not necessarily what Lomborg intents to convey; I can’t read his mind.
My personal take home messages from Cool It
- Climate change should be analyzed from the perspective of the individual issues it affects. The effectiveness of climate change policy should be judged on their merits for these issues and contrasted with other ways to tackle these issues. Not even the IPCC does this currently.
- For most, if not all, known issues, controlling climate is the least effective way to tackle them.
- The Kyoto protocol in its current form is not realistic, which leads to countries failing their responsibilities. This in term uses up precious international willingness to cooperate on other issues.
- The debate around climate change has lost its scientific objectivity and resembles a witch hunt.
- The Precautionary Principe is abused to rally unreasonable effort to deal with speculative scenarios.
How bad are Lomborg’s Errors?
In his book Lomborg deals with a number of worldly issues that climate change has an effect on. On lomborg-errors.dk Fog analyzes most of these issues. I do not claim that these are the only issues that climate change has an effect on and I can’t recall Lomborg claiming that either.
Any other issues should be (re)investigated from the same perspective, with the same question: how does adjusting climate relate to other ways of tackling the issue.
But there are also issues, you might argue, that can’t be analyzed that way, because we don’t understand them well enough? I will discuss that in the next section about the Precautionary Principle.
The issues for which I will assess Fogs critique are polar bears, heat versus cold deaths, melting glaciers, sea level rise, hurricanes and malaria. For each of these issues I will quickly summarize Lomborg’s analysis. In particular, I will repeat Lomborg’s estimate for the relative effectiveness of climate control versus other measures. I will cite Fogs most important critique on the issue. By most important I mean the things that would mostly impact the Lomborg’s estimate of the relative effectiveness of climate control versus other measures.
I can save you some reading by telling you that the answer is usually: not much (except perhaps for heat deaths). A recurring theme is that even if Lomborg’s analysis are off by a huge margin, climate control is so unimaginably ineffective that it dwarfs even the most blatant errors. It is like trying to shoot at a skyscraper from two meters away: very hard to miss. Fog shows no sign of realizing this and in stead tends to ignore all factors but climate change.
Short story: some environmentalists suggest that polar bears are being driven to extinction and that reductions in carbon are necessary to prevent this. Lomborg claims the extinction trend is dubious, as well as the direct link to climate change. Most importantly he concludes that in order to save the polar bear it is far more effective to adjust hunting policy than to change carbon emissions.
Before I go into this discussion, I would like to take you on a detour to show you some common forms of fuzzy thinking.
First of all, let me remind you that there are millions of species on this planet and many of them are going extinct. Obsessing with a single species just because it is big and hugable is unwise.
Second, the fact that a species exists is not enough reason to save it. Malaria exists, yet we love to get rid of it. Is it not unreasonable to ask why we need the polar bear in the first place.
A common response to this argument is that we do not know the value of a polar bear or any other species for that matter. First of all this in not true for a lot of species that we regularly interact with. We use cows for food and milk and malaria tries to kill us. We have gotten pretty good at putting a price on these species, whether it is for getting more of them or wiping them out. Second, if we do not know the value of a certain species, it is not a valid reason to just put a random (usually high) price on it. I will explain this in part 3 where I discuss the common misinterpretation of the precautionary principle.
Despite this, it is important to note that in Lomborg’s paradigm, one does not have to put a price on a species. This is one of the reasons why it is so useful. I will get back to this point when I discuss Ackermans paper in part 3.
Another issue that is often left unspoken, is our obsession with increasing or decreasing populations. When it comes to plagues of otherwise benevolent species, most people are well aware that increase is not always a good thing. But when it comes to more regular numbers, we suddenly associate increase with good and decrease with bad, for no particular reason.
We should ask ourselves how many polar bears we actually want. We don’t want them to roam the streets of Vancouver or be at risk of extinction. But with no further information, there is no point in saying more is good and less is bad.
Lomborg avoids this question by taking an arbitrary year as a reference point and Fog avoids it by arguing about that particular choice. Both are implicitly assuming that whatever is ‘normal’ for nature is good for us and they only disagree on what exactly is normal.
The reason behind the assumption that whatever is normal for nature must be good for us, is that we see nature as the constant factor and humans as the agent of change. This is pretty arrogant. While humans are certainly responsible for a lot of changes in the world, nature is not a passive bystander. It is constantly trying new things and it won’t politely stop if one of those things wipes us off the planet. The other problem is that even if nature is a passive bystander, our own changes make us sensitive to new factors. What may have been a harmless polar bear population 20.000 years ago, may now be a threat. It is just one more example of something that sounds reasonable but isn’t. Think about this next time you here somebody say “but it’s natural!”.
But all of this is irrelevant to the question at hand: what are the most effective ways to control the polar bear population?
Here is Fogs critique.
It seems Fog agrees with Lomborg that adjusting hunting pressure is the easiest way to adjust bear population, but he doesn’t put it that way. In stead, he claims that Lomborg does not understand the concept of sustainable yield. Turns out that Fog doesn’t understand it either when he says:
This is nonsense due to Lomborg´s lack of understanding of the concept of sustainable yield (as explained above). If we want a stable population, we can shoot 49 bears annually without compromising the stability. If the 49 bears were not shot, a similar number of bears would die from other causes.
In fact, sustainable yield is a little more tricky than that. It is about keeping hunting pressure low enough so it does not damage vital ecosystem functions that would lead to the extinction of that species. If you take away the hunting pressure and you end up with a population explosion, not with bears conveniently dying from other causes. The explosion in turn might be undone if there is not enough food to support this growth, but this is not essential.
Compare it to your bank account: with $100 and an interest rate of 5%, using $5 per year is a sustainable yield. But the amount on your account (i.e. the population size) is arbitrary. You could reduce it to $50 dollars and end up with a sustainable yield of $2.50 per year. Likewise, you could leave it alone for a while and watch it go up to $200; the extra money doesn’t disappear ‘from other causes’.
For ecosystems there are some lower and higher limits to the population size, but there is more than one possible sustainable yield. So no, if the 49 bears were not shot, a similar number of bears would not necessarily die from other causes.
Next, Fog basically admits that Lomborg is right:
On the other hand, when the environment becomes less favorable, then the sustainable yield decreases, and the number of bears shot annually will have to be reduced to avoid an acceleration of the ongoing decline.
Fog does not even mention reducing carbon emissions in his critique. He does not present a new analysis that shows how reducing carbon emissions is even remotely as effective as reducing hunting pressure.
There is no need to reassess Lomborg’s conclusion on this matter, based on this information.
Heat and Cold deaths
The short story: Lomborg claims that increasing temperatures actually save lives. This is because many more people die due to cold than due to heat and because most of the temperature increase will go towards warmer nights and warmer winters. Heat deaths will increase a bit, but cold deaths will more than compensate for it, according to his analysis.
This seems to be Lomborg’s favorite example, which is a pity because it also turns out it is the easiest to criticize.
But first of all, let’s get some perspective here. While our two friends are arguing about a plus or minus sign for 150,000 deaths per year, ask yourself, How many people actually die every year on this planet?
About 62 million according to a possibly unreliable source. Here’s the top ten according to WHO:
- 12.6% Ischaemic heart disease
- 9.7% Cerebrovascular disease
- 6.8% Lower respiratory infections
- 4.9% HIV/AIDS
- 4.8% Chronic obstructive pulmonary disease
- 3.2% Diarrhoeal diseases
- 2.7% Tuberculosis
- 2.2% Malaria
- 2.2% Trachea/bronchus/lung cancers
- 2.1% Road traffic accidents
Do you see heat and cold deaths on this list? No. So why are we even talking about this?
Once again, for each of these diseases, ask yourself how we can get rid of them. How effective is reducing CO2 emissions relative to others measures such as medicine?
Fog makes it pretty clear that Lomborg’s analysis on heat and cold deaths is dodgy at best. But he leaves the most important question unanswered:
As explained in the introduction to this page, the study referred to is based on calculations that deal with seasonal variation and moderate warmth, but not with heat waves. Therefore, what Lomborg concludes on the basis of this study does not refer to heat waves.
Great, but how does reducing carbon emissions affect the number of deaths from heatwaves? What other ways are there to reduce the number of deaths from heat waves? How do they compare? Perhaps this quote sheds some light?
In more northerly cities there is still a scope for adaptation, such as more widespread air conditioning. However, a recent study by Kalkstein (link here) finds that a warmer climate will cause more heat-related deaths here, even when adaptation is included in the models.
As an aside, I always find it fascinating when scientists claim to completely understand the next hundred years of human ingenuity by stating that “adaptation is included in the models”.
Fog seems to conclude here that apart from reducing carbon there is absolutely nothing we can do for these people. To translate that to Lomborg’s paradigm: there is only one way to tackle heat death, which is therefore the best way.
I doubt it, but will leave it as an open question.
Lomborg asks the critical question: how do people use glaciers? How does climate change effect such usage and in what ways can we compensate for these changes. Fog seems to interpret these questions as a mere distraction:
Glaciers worldwide have been receding at an accelerating pace during recent decades, concomitant with the general rise in air temperatures worldwide. Lomborg manages to shift focus completely away from this fact, by various distracting excursions.
More concerning is this – incorrect – remark:
Modern recession of glaciers is presented mainly as a consequence of emergence from the little ice age
Lomborg actually says that because of the little ice age, humans have already had a recent taste of receding glaciers even before humans accelerated the process. That initial experience hasn’t killed us, so it is not immediately obvious that continuation of this process will kill us. Several critical remarks go without any explanation as to their relevance, such as:
but the recent acceleration in glacier melt is left unmentioned
The only tropical glacier mentioned by Lomborg is that on Mount Kilimanjaro
These statements lack a conclusion, like: “because the recent acceleration in glacier melt is X, the cost of Lomborg’s proposed measures increase by Y and thus enter a similar price range as cutting carbon.”. Presenting them without further argument, turns them into rhetorical arguments. It is then left to the reader to assume Fog knows why this is important but did not have time to specify.
Here is Fogs final conclusion:
The overall balance is projected to be as follows: In the upper Indus, there will be initial increases of water flow ranging from +14 to +90 % over the first few decades, followed by decreases of -30 to -90 % after 100 years
That seems to be roughly along the lines of Lomborg’s conclusion. Perhaps a few decades more or less, but no orders of magnitude difference. So I see no reason to assume that Lomborg’s proposed measures will turn out orders of magnitude more expensive, based on this new information.
This remark puzzles me, so perhaps I misunderstood:
Lomborg does not explain how it should be possible that water flowing in the rivers during winter should be stored and kept until summer.
This is what dams do, isn’t it?
Once again, Fog does not show that CO2 reduction is the cheapest way to solve this particular problem.
Sea level rise
Lomborg reminds us that the twenty meter flooding pictures shown by Al Gore are completely ridiculous and that the latest consensus is closer to 30 cm. More importantly he shows that if left unchecked it will cost us only a tiny fraction of the total available landmass, albeit a pretty important fraction. However, he argues that building (better) defenses against the sea is dramatically cheaper than trying to slow sea level rise.
Fogs introduction to this topic is complete speculation. He conveniently throws the IPCC reports in the bin due to recent discoveries and talks about all sorts of possible sea level rises.
I understand his arguments: the last ice sheet model that I saw was dodgy at best. Not to mention the fact that many climate models are written in Fortran, which is enough reason not to trust them 🙂
Fogs assumption that CO2 reduction will prevent such catastrophes is just as much speculation. One can not give a sensible estimate on the odds that we are not already beyond tipping point or that Kyoto’s reduction will keep us below such tipping point. It is safe to assume that any measures will decrease the chance, but such decrease might be anywhere between 0.00001% and 100%, which means it’s just a guess.
By sidestepping peer reviewed works and simply guessing there is no longer a fair comparison. We can have the discussion again at the next IPCC report, or if an alternative scenario makes it through peer review. At that point, Lomborg should also revise his analysis. See also my discussion on the precautionary principle later on.
Eventually Fog seems to settle, with some reservations, for 50 cm. But he does not show what the effect of this change is on Lomborg’s calculation and conclusion.
Furthermore I find not even a single comment on Lomborg’s proposed solutions to flooding mitigation.
Given the extreme difference in effectiveness between these proposals and changing carbon levels, I doubt that another 30 cm makes a difference.
Hurricanes and extremes
Lomborg argues that most of the increase in hurricane damages is not due to more severe storms, but due to increased value at risk. Had some of the storms from many decades ago hit today, they would have caused similar or even more damage than Katrina. Fog counters this argument as follows:
as if we had to choose whether damages were due to denser infrastructure or to increasing wind speeds, instead of admitting that both factors contribute in tandem
If this argument sounds reasonable, you are still not getting the point: go back to my crash course on dodos.
When we hear ‘in tandem’ we are inclined to think 50-50. But there is no reason for this to be the case. It could just as well be 99.999:0.001. Even more importantly, even if temperature increase and denser infrastructure increase damage on a 50-50 basis, it doesn’t mean their solutions deserve equal attention and resources.
Lomborg argues that changing CO2 concentrations will reduce future damage far less than other methods, per dollar invested. To visualize how big this difference is, let me remind you of the legendary Hans Brinker who put his finger in a dike to prevent a disastrous flooding. Why didn’t he just buy carbon credits?
The following is so typical about Fogs methodology that I will just show it, even though it is irrelevant:
‘Page 100 bottom : “Had it hit today, it would have caused damage of about $100 billion . . “. Flaw: Lomborg´s source, Pielke et al. 2007, gives a figure of $78 billion. “‘
Sorry, but $78 billion is about $100 billion. Lomborg is trying to argue that storms really haven’t changed by a factor 10 like others suggest. To make his point, that 10% discrepancy is irrelevant on such a huge difference and simply makes it easier to read. Not to mention the fact that converting historic dollar values to present day values is black magic to a certain extend anyway.
Fog marked this as a flaw, which is a relatively severe rating (his system starts with comments, followed by remarks, flaws and the worst rating is error).
Fog also makes exactly the kind mistake Lomborg was warning against, when he compares the number of dead people per hurricane:
Page 100 bottom, note 509: In the note: “Notice that both hurricanes . . . caused many deaths . . “.
Flaw: The important omission is that hurricane Katrine caused the death of 1,200 – 1,300 people, making it the second most fatal hurricane after the Galveston hurricane. Estimates for the Great Miami hurricane range form 373 to 800 dead
The population at risk has grown so the number of deaths per million at risk has probably decreased.
Fog seems to completely disagree with Lomborg’s suggestions for alleviating hurricane damage:
That better warning systems today would have evacuated much of the city is at odds with what actually happened when Katrina struck in 2005.
As far as I know the way the US goverment dealt with Katrina was one of the worst blunders in their history. It is not just about warning systems, it is also about what you do when the storm actually hits. So when you look at it with that in mind it turns out that even a poorly managed flooding disaster in 2005 results in relatively less casualties than a long time ago.
But most importantly, once again, Fog is not comparing the effectiveness of CO2 reduction to that of other measures. He simply dismisses such other methods.
Let’s hear what the experts have to say on Lomborg’s story on storms. This is where things get interesting, because here we find Chris Mooney explicitly disagreeing with Lomborg, but implicitly agreeing. His main critique seems to be this:
Lomborg seems to ignore worst-case scenarios and precautionary thinking.
Interesting, but I will deal with “precautionary thinking” in part 3. Suffice it to say Lomborg is correct when he ignores worst-case scenario’s that have no explicit odds associated.
Although he spends much time discussing how societal changes–the moving of persons and property into harm’s way–make us increasingly vulnerable to hurricanes, he fails to seriously consider the idea that when you add global warming to said societal changes, the result could be a double whammy.
Alas, it’s not an “either-or,” it’s a “both-and.”
So if you ignore them and only focused on addressing that vulnerability, you’re only dealing with a part of the problem
This is exactly the same flawed reasoning as with the Hans Brinker example above. It is not a double whammy, more like a 1.03 whammy. Further more, every dollar spent on lowering that 1.03 is better spent on other measures than carbon reduction. At least show a reasonable alternative distribution. The current argument is completely qualitative.
And here he basically admits Lomborg is right, or at least that Fog is completely wrong:
the most immediate policy prescription ought instead to be investing in better hurricane preparedness
The fact that I actually found this article through Fogs links is rather confusing.
Lomborg is not ignoring the effect of a carbon reduction measure, he is putting a number on them and comparing that number to other measures. It’s fair to debate his number, but neither critics are doing that. In fact nobody is doing that, and that is the problem.
Lomborg’s argument is pretty straight forward here: Malaria used to be one of the biggest problems for western societies, even in temperate climates. Not anymore. Yes, Malaria likes it when the planet heats up a bit, but our medical systems overwhelmingly compensate for that. Malaria is solved as soon as suffering countries get help or get their act together and do something about it.
Fogs arguments on this issue are the most fuzzy of all yet. In stead of properly countering Lomborg’s arguments, he speculates that poverty in Africa won’t be solved. Once again he conveniently ignores scientific consensus here. Perhaps it’s time for a new phrase, to join Climate Deniers. How about Africa Deniers?
The second part of that comment is also worth reading:
‘Comment: Nobody can know for sure what will happen. We do not know for certain if Africa will manage to get out of its poverty trap, and we do not know if the malaria parasite will develop resistance faster than we can develop new anti-malaria drugs.’
The argument about drug immunity has nothing to do with climate change. How is this supposed to prove Lomborg is wrong?
This is by far the weakest attack of all. In no way is he convincing me that the effect of climate change on malaria is even nearly as relevant as the effect of policy. Using carbon emissions to get rid of malaria is completely absurd.
Final score Lomborg versus Fog
For none of the above issues does Fog convince me that climate control is a relatively effective way to tackle them, with a possible exception for heat deaths. On some issues he gains a few orders of magnitude at best, but turning a turtle in a race car does not get you to the next star.
I encourage other scientists to try harder: my main point of this post is to convince people to change the way they think about climate change. Secondary objective is to show that tackling climate change directly is a waste of resources.
Critics will have to come up with significantly bigger – and well founded – numbers to win this race.