Understanding Climate Risk

Science, policy and decision-making

Merchants of Doubt: review and reflections

with 3 comments

Merchants of Doubt

How a Handful of Scientists Obscured the Truth on Issues from Tobacco Smoke to Global Warming

Naomi Oreskes and Erik M Conway

June 2010, 368 pp, Bloomsbury Press

This is a very good book. It is well written and a clear exposition of the authors’ case, describing how a few scientists involved in US military programs during the cold war then shifted their expertise into selling doubt to preserve the markets of the free world from interference on health or environmental grounds. The chapters deal with the strategic defense initiative, acid rain, the ozone hole, second-hand smoke, global warming and the recent attacks on Rachel Carson’s legacy from Silent Spring. The main merchants of doubt Oreskes and Conway name are Frederick Seitz and Fred Singer, William Nierenberg and Robert Jastrow. Others joined in later on.

The basic strategy of sowing scientific doubt to preserve profit was developed by the tobacco industry in the 1950s. Frederick Seitz was intimately involved. SDI was the impetus for a larger scale attack on scientific efforts to describe environmental risks such nuclear winter and the population-ecology links being made by Paul Erlich and others. Seitz was the lynch-pin, but when acid rain and ozone issues became policy issues, the other players became involved. Their strategy was to dilute the science in order to delay action. In each of these cases, a concerted and deliberate effort was made to sow the seeds of public doubt, using scientific-sounding think tanks and providing ‘balance’ to controversial scientific research. Assisted by a willing media in some quarters, who willingly provided polemic and more generally by the media at large, who were reporting disagreement as news, doubt was being sown, reaching from the White House to the major news outlets in America and across the western world.  

Industry abetted this effort specifically around second-hand smoke and global warming, funding specific strategies to sow doubt, for which the authors provide documentary evidence. They funded think tanks, conferences and reports to put ‘the other side’ of the story. Op-eds by the scientists themselves and sympathetic columnists in major news outlets promoted ‘the other side’ as viable alternatives to mainstream science. By and large, the strategy has been successful as a delaying tactic. Legislation on second-hand smoke was delayed for some years until the evidence of risk to innocents, particularly children, became overwhelming. The case for policy on global warming is still being delayed. Finally, the lies around Rachel Carson and the banning of DDT continue to gain currency, as being document by Tim Lambert at Deltoid and others.

The book is controversial because it not only documents what occurred but builds a case as to why the merchants of doubt acted in the way they did. By imputing motive to their actions, the book moves onto why a handful of scientists obscured the truth on issues from tobacco smoke to global warming. This is not where science history usually goes and is fuelling outrage and derision from opponents. But the authors prosecute a very strong case.

They also detail the style of argument used very well providing insights into the modus operandi of organised denial. Mirroring as a style of argument, for example accusations about scientific conspiracy; “If we’re doing it, they must be doing it, too” becomes an unconscious admittance to the charges being made about global warming being an organised charade of the scientific establishment in order to influence global governance.

If I have to quibble, it would be around the role of DuPont in the ozone hole issue, where as the largest manufacturer of CFCs, DuPont organised a phase out of chemicals in favour of what turned out to be more efficient replacements. This was pretty much glossed over. However, a technological fix was readily available and the delay was in industrial turnover. This is an example where industry can and does, responds to the scientific evidence and plays a positive role. Perhaps the subject for another book.

The publishers have also ensured that no-one who subscribes to the world view of free-markets unhindered by environmental regulation or governance will be persuaded by the book, carrying endorsements by Al Gore, Elizabeth Kolbert and Bill McKibben.

Australia has similar examples to those in MoD, most notably documented in Guy Pearse’s the book High and Dry and the essay Quarry Vision. We have scientists in Australia who are just as prominently sowing doubt with pseudoscience.

The depressing aspect to all of this is that people do not understand abstract reasoning very well. It is poorly taught or not at all, and like mathematics, isn’t something that people need very often. So when fallacious arguments are presented, most people don’t have the cognitive tools to see through them. For example, following a true statement by a false statement, but inferring the latter is true, makes that false statement much more believable. “Environmentalists are advocates, therefore environmental science is advocatory”, is one very commonly put argument. Rhetorical tricks, such as the rule of threes, presenting points in three bites, preferably with rhythm (written or spoken) are also effective. How often have you heard someone say they would swoon if George Clooney read a laundry list? Sometimes even the words are not that important to communicating a message.

Scientific reasoning is also not taught well, but the situation is improving. Science undergraduates might do something in research methods, however, a Doctor of Philosophy can still be obtained without any grounding in the philosophy of science (I can name names, but won’t).

One of the main tools used in scientific reasoning is approaching a hypothesis as both an optimist and a pessimist. The optimist looks for patterns, for links, for how to bridge other hypotheses and assumptions. If a hypothesis such as the global warming effect is widely accepted and backed up with multiple lines of evidence and explanatory power, then it is ok to be an optimist. The pessimist takes a hypothesis apart, trashes assumptions and doubts that the evidence supports that particular conclusion. When preparing new work, a scholar will do both, because if they don’t, the peer reviewers of that work will take it apart for them. Then reject the paper.

The skeptic will tell you that the proper scientist is always a pessimist. This is not the case.

As Sherrilyn Roush puts it,

the pessimistic inductivist …  tries to undermine our confidence, and show that there is one potentially threatening argument form he could use. However, this argument form requires much more than has been expected if it is to serve the pessimist, and pessimists have not actually ever offered what is needed.

The starting point at which questions arise as to what we have a right to believe about our theories is one where we have theories and evidence for them, and we are involved in the activity of apportioning our belief in each particular theory or hypothesis in accord with the strength of the particular evidence. The devil’s advocate sees our innocence and tries his best to sow seeds of doubt. If our starting point is as I say, though, the innocent believer in particular theories does not have to play offense and propose sweeping views about science in general, but only to respond to the skeptic’s challenges; the burden of initial argument is on the skeptic. The greatest strength of the realist attitude lies at this starting point, as Arthur Fine (1996) realized, and I will argue here that no one has given reason from the history of science to give it up. In particular, no pessimistic induction over the history of science has done what it needs to do in order to undermine our right to apportion our beliefs in particular theories to the evidence we have for them.

Roush, S. (2009) Optimism about the Pessimistic Induction, Centre for the Philosophy of Natural and Social Science, The London School of Economics and Political Science, London.

This statement broadly supports the conclusions of earlier philosophers of science about methodical falsification, and was an area upon which Popper, Kuhn and Lakatos all agreed. It places the burden of proof on the pessimist, asks for evidence, and if that evidence is not supported by the body of scientific knowledge, the argument can be disregarded.

All the arguments of the Merchants of Doubt presented by Oreskes and Conway are arguments by induction and all have this limitation. Good scientific communication will contain the argument of the optimist – this is what we know, and the pessimist – and this is what we are uncertain about. When have you ever heard a merchant of doubt speak of their own doubt?


Written by Roger Jones

March 6, 2011 at 10:26 am

3 Responses

Subscribe to comments with RSS.

  1. Roger, thanks for this post.

    I particularly enjoyed the paper by Roash – there’s lots to digest in that one & it’s made a good Sunday read. The analysis of the space of ‘valid’ pessimistic arguments against our scientific reasoning is an interesting tack, and not one that I’d seen before in such a formal fashion, although I suppose we all employ such reasoning about arguments in our day to day lives.

    I especially like her recourse to probability to explain beliefs in a fuzzy scientific system and ‘the preface problem’ (as an aside every mathematics book on probability I have ever read contains the line “this book contains errors with probability 1” in its preface – probabilists need a new joke). From my (admittedly brief) reading, she argues that a redundant system is the best way to justify a belief in science (reading her statements on cross induction vs naive induction). It seems that this is a Lakatos-esque argument about core vs non-core science, and how the two are delimited.

    Just to play devils advocate for a minute here (and this is completely unrelated to climate science), if past failures in the scientific method are no longer relevant to us, whether through the use of new methods or evidence or some other means, then how can we accrue the successes which form the basis of the optimist’s position without falling into the same inductive trap as the pessimist?

    Also, it seems that despite her claim that we can disregard large classes of arguments against our network of beliefs, this doesn’t seem to solve the problem that there might still be a cross-induction out there which will falsify our _entire_ belief network in science. I suppose here one could fall back on the argument that such an event is incredibly unlikely, but how could we quantify the likelihood of such an event, given a (presumably infinite) class of possible cross inductions even once the disregarded classes are taken out?

    We all have to deny this last point unless we want to end up in some Feyerabendian science-less universe, and maybe I’m just restating the problem of induction here.

    I’d be interested to hear your thoughts.


    March 6, 2011 at 7:11 pm

    • Jess,

      the way I read it, a single argument by induction; “theories have been disproved in the past”, is insufficient to suggest that a current theory is vulnerable. The Galilean or Copernican overturning of earlier theories are endlessly trotted out as evidence that AGW could be jettisoned any second.

      The optimist position is subject to the evidence, as Roush says. So for AGW, one actually has to posit evidence that shows the theory is at risk. Her point is that science now is done differently to how it was carried out in the past. Now we have systems, not just single theorems. Therefore, AGW is not just one hypothesis, it is a system of hypotheses.

      This is where Roush’s example of hypotheses, where she suggests that (it is ony an example, not her argument) H1,H2,H3 … Hn where H are hypotheses cannot all be correct in a complex system of science. This is true and she uses the example to show that while the probability of H1 … Hn may be high not all of H will be true.

      What she doesn’t show with this example is the relationship between core and auxiliary theory. So with something like AGW, you get if H1, then H2, then H3, H4, H5 to Hn, with goodness knows how many layers in between.

      H1 is the basic physics of radiative forcing of greenhouse gases. Well known, p(H1)>99.9%. H2 is the positive atmospheric forcing that leads to short term climate sensitivity (as defined by Hansen). Most researchers would also put that at 99%, but let’s be conservative and put it at 95%. The next layers describe the vertical heating of the atmosphere, the spatial fingerprint of warming and differential cooling of the stratosphere and heating of the troposphere. All confirmed by conventional science with very high P. Energy fluxes – there have been measurement problems due to ocean and upper air monitoring but recent projects to address this are bringing the numbers in. Confirmatory evidence – other planets – check, palaeoclimate, check. So the chance of H1 or H2 being disproven are very low, but problems with auxiliary theory are legion.

      Most of the guff that people argue about, global mean air temperature, the hockey stick actually don’t matter. They are well down the scale. Every piece of positive evidence adds support, sure, but it’s not important. Every piece of negative evidence doesn’t matter unless it affects H1 or H2. The attacks currently trying to negate H1 and H2 qualify as pseudoscience if online, or are quickly overturned if they make it into the scientific literature.

      So, large-scale cross-induction? Let’s say it’s possible, just like the point that no theory can be ever be 100% proven. So, until it emerges it is not a problem but is a possibility, but a remote one.

      We need to remember that arguments of doubt want to take hypotheses where the scientific framing of likelihood is very high, to where it becomes a political likelihood, in the 45-55 zone of the swinging voter, or even better.

      But subject to basic reasoning and the scientific evidence, these arguments don’t even make it to first base.

      Roger Jones

      March 6, 2011 at 11:33 pm

  2. Hi Roger, thanks for your reply.

    I don’t think I phrased my previous statements very well – sorry!

    I was more interested (from a purely epistemological viewpoint) in the relationship between old and new science that Roash posits, and her assertation that this destroys the pessimist’s attempts at induction over the history of science. The crux of her argument seems to be that we can’t relate the failures of previous scientists to our theories since we don’t include their failures in our current scientific understanding. However this doesn’t seem very compelling to me. Couldn’t the pessimist just answer that if we have _any_ part of their theories in our current understanding then the induction holds because the bit of their knowledge in our current knowledge provides a bridge for that induction?

    For the record I can’t see any problem with the arguments about large redundant networks being unlikely to be falsified – I just wish more denialists would think carefully about what falsifying such networks entails. Quite apart from anything else I subscribe to a fairly utilitarian view of scientific knowledge pretty much as described in the paper – it’s rational to base ones predictions on what science does well, rather than some abstract notion of absolute truth.


    March 9, 2011 at 11:54 am

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 )

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: