by Judith Curry
Big Players of any sort distort the normal systemic activity and render the emergent outcomes unstable and unreliable and create an ideal breeding ground for incentives that motivate ideologically biased people to circumvent normal constraints in the name of pursuing a “greater good”.
I have long been concerned about the role of IPCC in torquing the direction of climate science and promoting groupthink. I spotted a link on twitter to this very interesting paper, that has clarified my thinking on this issue.
Causes and consequences of the climate science boom
William Butos and Thomas McQuade
Abstract. Scientific disciplines, like economies, can and do experience booms and busts. We document a boom in climate science, sustained by massive levels of funding by government entities, whose scientific direction is set by an extra-scientific organization, the IPCC, which has emerged as a “big player” in the scientific arena, championing the hypothesis of anthropogenic global warming. We note the difficulties in obtaining definitive empirical clarity due to the complex nature of climate, the feedback between the effects of the IPCC’s advocacy and the government’s willingness to fund the science, the ideological and political agendas at play, the dangers to the integrity of scientific procedure in the context of ideological bias, and the poor performance of the “crony capitalist” enterprises that have grown on the back of politicized science.
Forthcoming in the Independent Review [link].
Big Players in Markets and Science
“Big Players”, in an economic context, are “privileged actors who disrupt markets” in the sense that, though they are not subject to market constraints and to the discipline of market competition, their discretionary actions have widespread impacts on the expectations and actions of market participants.
Their effects are felt in two ways: in the diversion of entrepreneurial concern away from the assessment of fundamental economic data and toward the attempted prediction of the activities of the Big Player, and in blunting the weeding-out effects of the normal market mechanisms on participants less adept at appraising economic data, especially in the cases of those actively favored by the Big Player. In markets, prototypical Big Players are central banks and government agencies empowered with discretionary policymaking. Markets dominated by Big Players are prone to herding, where market participants, with little reliable information as to the Big Player’s next move, look to what others are thinking and doing.
That the phenomenon of the Big Player has relevance in science as well as in markets can be appreciated if it is understood that markets and science, as systems of social interaction, though differing vastly in the particular transactional forms employed, are similar in their structural form.
Scientists interact with each other in ways that are every bit as complex and structured as the interactions between market participants. As scientists, they don’t produce or buy and sell marketable goods, speculate on future asset prices, or seek financial gain and risk financial loss, but they do participate in interactions that have analogous feedback effects. They publish hypotheses and report experimental findings, use or criticize (and cite) the work of their peers, choose areas of research to pursue often based at least in part on anticipated reputational returns, and face the risk of loss of scientific credibility and funding. In both cases, market and science, the repeated interactions between the participants feed back recursively to generate emergent effects: in markets, a spectrum of goods, prices and brand names, that reflect the realities of resource availabilities, production technologies, and consumer tastes; in science, a body of knowledge and attendant scientific reputations that reflect the realities of the world under observation, experimental techniques, and the dictates of good practice.
So, funding by itself, even if directed to favor one hypothesis over another, is not the problem; the problem arises when the provision of funding allows for or even encourages the continuation of research and publication activities which undermine the operation of the feedback inherent in the standard procedures of science – feedback which performs the scientific analog of profit and loss in assessing the scientific value of publications and in furthering the scientific reputations of their authors.
Science, in rare cases, is also susceptible to another sort of Big Player: one with the ability to portray a favored hypothesis as settled, consensus scientific knowledge even in the absence of a substantial body of confirming evidence. The IPCC has taken on that Big Player role in climate science.
The IPCC and the Emergence of Consensus
The UN took up the issue in 1988, forming the IPCC as an independent body of scientists charged with assessing current climate science research (specifically emphasizing the effects of human activity on climate) and producing summary and detailed reports geared to public consumption – on the face of it, a very useful service. But the way in which the IPCC is organized and its methods for eliciting agreement on the conclusions given in its reports have opened it up to be the conduit of pervasive bias.
The IPCC qualifies as a Big Player in science in that it possesses all of the attributes characteristic of Big Players in markets: bigness in terms of influence, insensitivity to the usual constraints, and discretion in its ability to promote a favored direction of research. Its influence in climate science is pervasive, and the complex nature of the climate system and the lack of understanding of the feedbacks at play enable it to champion the most politically attractive of the several plausible hypotheses and to largely ignore uncertainties and potential disconfirmations which are the usual scientific constraints on the acceptance of hypotheses. Professional success in climate science has become more tied to the acceptance of the IPCC’s pronouncements than with the exploration of contrary possibilities; in fact, scientists who profess competing hypotheses are routinely castigated as “deniers” and some have reported unusual difficulties in negotiating the publishing process.
While a large majority of climate scientists are reported as being in general agreement with the AGW hypothesis and with the IPCC’s pronouncements, the accuracy and extent of this consensus has been questioned. The oft-quoted 97% number may be unrealistic and unsupportable, but the general acceptance by the majority of scientists having any connection to climate science seems real enough. This herding is a predictable result of the IPCC’s Big Player presence.
In science, when the “common wisdom” favors a particular hypothesis, there is an incentive, particularly for younger, not-yet-established researchers, to follow it. If the hypothesis turns out to be correct, you are seen to have the good sense to have espoused it; if it turns out to be incorrect, you have plenty of company. Either way, your scientific reputation will not be materially damaged. But if you flout the common wisdom but it turns out to be correct, your reputation will suffer greatly. This raises the following question: is the contrarian stance, with a potentially large reputational gain if the hypothesis turn out to be false, worth the risk? In the context of markets, that depends on the reliability of the evidence.
Incentives influence the choice between idiosyncrasy and herding. If the penalty for a bad idiosyncratic decision is high compared with the penalty for the same bad decision made along with everyone else, then one has an incentive to concentrate less on evaluating reports about the empirical evidence and more on noting what other scientists believe. If the available reports about the empirical evidence are reliable, then they provide a powerful counterweight to this incentive, namely, the large gains to be expected from taking a correct but idiosyncratic position. When most of the available reports have been rendered unreliable by a Big Player’s discretionary interventions, this counterweight no longer exists. The expected gain from taking a dissenting position then becomes too small to discourage herd behavior. Herd behavior is encouraged by discretionary interventions because of the role of reputation in the furtherance of scientific careers. The process is self-reinforcing – if herding begins, the fact that increasing numbers of scientists seem to espouse the common wisdom serves only to cement the appearance of consensus. The less reliable are the available empirical assessments, the longer such self-reinforcing movements can go unchecked.
A few more irresistible tidbits:
Ideological Bias and the Undermining of Scientific Norms. The scientific process for the generation of reliable knowledge relies heavily on general adherence to the norms of publication and citation, with feedback effects on the reputations and credibility of contributing scientists. Were these norms to be systematically violated within a scientific discipline, knowledge and reputations would still be generated, but the knowledge would be counterfeit and unreliable, and the reputations would be similarly tainted. In climate science, the presence of Big Players – the IPCC in its ability to direct the science and the government in its supporting role of ensuring that IPCC-compatible science is funded – has engendered a growing politicization of the discipline, showing up in attempts to restrict publication of dissenting views and to attack the reputations of scientists who question the IPCC “consensus”.
Scientists are no less likely than anyone else to be swayed by a fear of crisis, and this is especially the case for those who see that they, with their relevant expertise, may be able to contribute to doing something about it. It is surprisingly easy for scientists to overstate the certainty of their results and to ignore or attempt to explain away conflicting data, but this is not unusual and it does no lasting damage provided that the basic procedures of science (review, publication, criticism, citation, and the feedbacks to reputation) are functioning normally. But when a particular set of conclusions is widely held not only to fit one’s preconceptions but also to have serious social implications, the harsh and critical judgments necessary for weeding out shaky science can be significantly toned down. The ideologically primed sense of impending global warming catastrophe, transformed into near-certainty by the influential publications of the IPCC, has endangered the integrity of the processes of science.
JC reflections
Climate Etc. has had several previous posts that danced around some of these issues:
Butos and McQuade are economists that have no apparent engagement with the IPCC or climate science. As outsiders, the provide a fresh perspective on the climate science/government/industrial complex.
I think their analysis of herding/crony science versus idiosyncratic/contrarian behavior is spot on. The role of the IPCC in reinforcing herding behavior seems obvious in the context of their Big Player argument.
How to break out of this situation of herding and crony science? There is growing evidence that supports an important if not dominant role of natural variability, which is making it a ‘safer’ reputational position to discuss natural climate variability. Further, the IPCC’s ponderous reports and diminishing reputation (e.g. AR5 was a bit of a yawn, Pachauri’s travails, etc) are working in the direction of cutting this Big Player down to size somewhat.
There remains a strong social contract between scientists who are funded by the government, and the IPCC that supports the government’s political agenda. The feedbacks supporting this social contract in principle can be reversed; it remains to be seen what, if anything, will trigger this reversal. I suspect that it will be the climate itself, if the hiatus/pause/slow down continues.Filed under: Sociology of science