Loading summary
Transcript73 lines
- [00:00]
A
Very good evening. Welcome to this evening's lecture. It's, of course, a great pleasure to introduce to you Evelyn Fox Keller. She is well known for her work in philosophy of science, in the history and philosophy of modern biology, and the study of gender and science, to mention just a few domains. She is the author of several books, including Reflections on Gender and Science, the Century of the Gene, Making Sense of Life, and her most recent book has just come out about a year ago. She's now professor Emeritus at the Massachusetts Institute of Technology. Before joining mit, she taught at the University of California at Berkeley, Northeastern University and New York University. Once again, just to mention a few of the places she has been at, she's been awarded numerous prizes and awards, and she has been elected a member of the American Academy of Arts and Science. It's a great pleasure to have her here tonight. Thank you so much for coming. Evelyn, thank you very much. Over to you.
- [01:23]
B
It's a pleasure to be here always. Now, the topic I'm speaking about is a change of temper and tone for me, but I was driven to it by preoccupations that I'm sure many people here share. So the question of how people perceive and respond to risk has received a great deal of attention in recent years. Is that okay? Can you hear me? Okay? Research over the last three decades has shown that because of the cognitive limitations or defects of the brains, we have evolved. We are generally prone to misestimating and misperceiving risk, tending especially to overestimate the magnitude of those risks that are highly evocative and ignore less evocative ones. As the author of a recent Time magazine article explains, we pride ourselves on being the only species that understands the concept of risk. Yet we have a confounding habit of worrying about mere possibilities while ignoring probabilities, building barricades against perceived dangers while leaving ourselves exposed to real ones. But how do we know which risks are the real ones? Generally, we turn to experts who assess the risk of some hazard objectively, that is, by computing first, the magnitude of the hazard and second, the probability that it will occur, and finally, multiplying the two numbers together. Ordinarily, however, people do not even tacitly perform such calculations. Instead, they use various kinds of shortcuts, heuristics, or rules of thumb that permit them to make rapid and intuitive, if somewhat biased, assessments. One such shortcut, for example, is known as the availability heuristic, the rule of thumb by which people evaluate the probability of an event according to the ease with which relevant instances come to mind. Cass Sunstein is a legal scholar currently serving as the administrator of the White House Office of Information and Regulatory affairs, oira, and he has been strongly influenced by the work of Daniel Kahneman and Amos Tversky. Sunstein is especially concerned with the difficulty of fashioning rational public policy in the face of the threat of disastrous events. Such events engage intense emotions, and these emotions, together with the availability heuristic, lead us to focus on the worst case, even if it is highly improbable. Close quote Indeed, Sunstein writes quote when intense emotions are engaged, people tend to focus on the adverse outcome and not on its likelihood. This disposition to misestimate or to altogether overlook probabilities has predictable costs. Especially, Sunstein argues, it inclines the public to overspend on regulation. His conclusion? Because ordinary people cannot be relied upon to make rational assessments of the risk they face, the protection of public welfare requires that regulatory policy be based on expert judgment and not on popular sentiment. A common explanation for the disparity between lay and expert judgments is that biological evolution has endowed humans with two different systems for apprehending reality. One most commonly used in everyday life is nonverbal, experiential, and rapid. The other, relied upon by experts, is analytic, deliberative, rational, and slow but reliable. But it is reliable, or at least that is the presumption. Paul Slovik studies the role of affect in decision making, and he argues that affect is a central characteristic of the experiential system. To most readers this might simply imply serve to underscore traditional views of emotion as antithetical to reason and IPSE dixit take into account for the relative unreliability of the affect base system. Sunstein himself often seems to accept such an account, but Slovik regards that view as overly simplistic and credits at least the possibility of a constructive role for emotion in decision making, especially in an uncertain and hazardous world. Slovik writes, although analysis is certainly important in some decision making circumstances, reliance on affect and emotion is a quicker, easier, and more efficient way to navigate in a complex, uncertain, and sometimes dangerous world. Slovik is not alone. The relation between emotion and rationality has come in for extensive re examination in recent years, and a number of authors have joined Slovik in this effort, often looking to findings of contemporary neuroscience for support, where it has been argued that emotional responses, because they have been honed by evolution, might at times be useful supplements or even alternatives to conventional understandings of rational decision making. Among all the emotions, the role of fear is surely the most controversial. Fear is also the emotion most obviously relevant to the threat of catastrophic events, and its relation to the misperception of risk has been a particularly prominent issue in these discussions. Slovik has emphasized the importance of what he calls the dread factor, arguing that differences between lay and expert estimates of the risk of serious hazard is due largely to lay tendencies to focus on catastrophic potential. The higher hazard score on this factor, the higher its perceived risk, the more people want to see its current risks reduced, and the more they want to see strict regulation employed to achieve the desired reduction in risk, which is exactly what Sunstein complains about. Other studies have also cited fear as a factor that magnifies the perceived risk. But fear has also been a topic of concern as a consequence, rather than a cause of heightened risk perception. As Franklin Roosevelt. Franklin Delano Roosevelt long ago reminded us, fear is something to be feared in and of itself. Sunstein agrees. He writes, fear is a real social cost, and it is likely to lead to other social costs. And elsewhere, he writes, if information greatly increases people's fear, it will to that extent reduce welfare. Close quote. Reviewers of Sunstein's 2005 book on the laws of fear note that Sunstein clearly shares Roosevelt's concern. In Sunstein's view, the authors write, the major proponents of democratically grounded risk regulation have to fear, in essence, is fear itself. Moreover, Sunstein seems to represent what has become a common view among economists, cognitive psychologists, and lay readers of this literature, especially since 9 11. That shock gave rise to an enormous literature on the political abuses of fear, virtually all of which takes as its starting assumption the prima facie counterproductivity of arousing public fear. The 1% doctrine of the Bush administration has come in for particular criticism, especially by those critical of the use of a discourse of fear to promote the war on terror and to justify the war in Iraq. In November 2001, concerned about the possibility of a second attack, Vice President Cheney argued, quote, if there's a 1% chance that Pakistani scientists are helping Al Qaeda build or develop a nuclear weapon, we have to treat it as a certainty. In terms of our response, it's not about our analysis, it's about our response. Close quote. From the standpoint of decision theory, such a policy would be impossible to endorse. But as a political strategy, it was unarguably successful, promoting political decisions that would otherwise have been extremely difficult to defend. Later, as these decisions came to be regretted, the doctrine came under mounting criticism for inducing a culture of fear in the American public. As one commentator put it, the War on terror has been about scaring people, not protecting them. And for this, the 1% doctrine was especially effective. Sunstein has noted that Cheyne appeared to be endorsing a version of the precautionary principle, according to which it is appropriate to respond aggressively to low probability high impact events, while also noting that the more familiar context in which the precautionary principle is generally discussed lies elsewhere, namely in climate change. Indeed, he writes, another Vice President Al Gore, can be understood to be arguing for a precautionary principle for climate change. Although Gore believes that the chance of disaster is well over 1%. Close quote. In point of fact, climate change poses a multitude of risks of quite high probability. But for the present, I want to focus on risks so devastating that they threaten the survival of civilization as we know it. In short, on catastrophic risks that that most experts judge to be of low, even if non negligible probability. Here, terrorism and climate change clearly share the same dilemma. How to respond rationally to the threat of such events, how much to spend and how aggressively to act in the effort to avoid catastrophic events. But the forms of action envisioned are quite different. Cheney envisioned a war on terror and on Iraq. Gore thought about regulation also. To the extent that the precautionary principle is understood as a mandate to do no harm, invoking that principle in the two contexts brings to the fore how very different can be the kinds of harm anticipated or ignored. Clearly, for climate change, the political shoe is on the other foot than it is for terrorism. But Sunstein's criticism of this principle is more general. He seeks to base his critique on logic, not on politics. Sunstein worries that reliance on the precautionary principle reinforces heuristics like the availability heuristic that lead both to excessive fear and neglect of more important risks. He argues that codifying that principle embodies distortions of risk perception that themselves result in serious harm. And he suggests even that it might be better to be called the paralyzing principle. The real problem, he writes, is that it the precautionary principle offers no real guidance. Not that it is wrong, but that it forbids all courses of action, including regulation, because every possible action, including regulation, risks doing harm to someone. The the bottom line, human judgment is flawed and because of the fear they arouse, worst case scenarios have an especially distorting effect on that judgment. For Sunstein, the solution is in the advice of professional analysts and not of popular will. But this presupposes that expert analysis delivers rational assessments of what the risk really is and the obvious question is what notion of rationality is being invoked in these discussions and what is meant by real risk. So rationality, broadly construed rationality refers simply to the exercise of reason, to the deliberative process by which humans draw conclusions. But recently, especially in economics and political science, the term has come to be used more narrowly, referring not to reasoning in general, but to the particular kinds of reasoning required for analytic assessment of risk, risk followed by maximization of the net benefits associated with any suggested policy. A rational choice is an optimal choice based on an objective and quantitative assessment of costs and benefits. Other choices may be based on reason, but they are suboptimal and by definition, less rational. Thus, where studies of human judgment focus on how people do behave, irrational choice, rational choice theory shows us how they should behave. And for decades now it is the latter, the narrower definition of rationality, that has prevailed as the gold standard for American public policy. For example, oira, the office of which Sunstein is currently head, is responsible for vetting and approving all regulation proposed by any federal agency or department before submitting it for White House approval. OIRA is mandated to base its recommendations on the regulatory principles as laid down by executive order, and the most recent formulation of that order reaffirms the basic principles that had first been laid down in 1993. Quote as stated in that executive order and to the extent permitted by law, each agency must, among other things, one propose or adopt a regulation only upon a reasoned determination that its benefits justify its costs, recognizing that some benefits and costs are difficult to quantify 2 tailor its regulations to impose the least burden on society, consistent with obtaining regulatory objectives, taking into account, among other things and to the extent practicable, the cost of cumulative regulations and three to select, in choosing among alternative regulatory approaches, those approaches that maximize net benefits. Close quote the premises underlying both rational choice theory and cost benefit analyses have been extensively critiqued by economists, by philosophers, the social scientists, and on a variety of different grounds. One line of argument began more than 50 years ago with Herbert Simon. Simon argued that because of the finitude of both computational capacity and informational access, all rationality, including expert reasoning, is bounded. His goal was, quote, to replace the global rationality of economic man with a kind of rational behavior that is compatible with the access to information and the computational capacities that are actually possessed by organisms, including man, in the kinds of environment in in which such organisms exist. Close quote Bounded rationality is the best we can achieve, and much of Simon's subsequent effort was devoted to developing models of how such bounded rationality would operate. Central to this focus, to this effort was his focus on how, in practice, behavioral choices depend not only on an actor's computational ability, but also on prior experience with the structure of the environment in which action is required. As he put it, human rational behavior is shaped by a scissors whose two blades are the structure of the task environment and the computational capabilities of the actor. More recently, Gerd Gigerenzer and his colleagues at the Max Planck Institute have taken up Simon's challenge and extended his efforts through through years of careful observation of how humans actually go about making decisions. And they bemoan what they see as rampant abuse of the term bounded rationality. Bounded rationality, they write, is not simply a discrepancy between human reasoning and the laws of probability or some form of optimization. It dispenses. Bounded rationality dispenses with the very notion of optimization and usually with probabilities and utilities as well. It provides an alternative to current norms, not an account that accepts current norms and studies when humans deviate from those norms. Bounded rationality means rethinking the norms as well as studying the actual behavior of minds and institutions. Gigerensa is especially critical of discussions that ignore the environment in which behavior occurs and he suggests, suggests ecological rationality as a better term. And Gigerenzer's view, ecological or bounded rationality does not make use of computations at all. Rather, it employs a set of conscious or unconscious heuristics, an adaptive toolbox that have been honed by biological both by biological evolution and by individual and cultural experience experience. It is these heuristics that give rise to what we call refer to as our gut feelings, our intuitions and hunches, where the terms gut feeling, intuition, or hunch interchangeably are used to refer to a judgment first, that appears quickly in consciousness, second, whose underlying reasons we are not fully aware of, and third, that is strong enough to act upon. Gigerens claim is that gut feelings provide a basis for action that not only need not be less rational than computation, but that in the appropriate environment can sometimes even be superior. His best example is catching a ball, reflecting a view that is widespread in cognitive psychology. Richard Dawkins and the Selfish Gene offered a description of the procedure by which a ball player catches that a ballplayer uses to catch a ball. Quote When a man throws a ball high in the air and catches it again, he behaves as if he had solved a set of differential equations in predicting the trajectory of the ball. He may neither know nor care what a differential equation is. But this does not affect his skill with the ball. At some subconscious level, something functionally equivalent to the mathematical calculation is going on. Now, because Dawkins description is in such striking contrast with how ball players actually proceed, it provides Gigerenza with a good point of departure. Real ball players do nothing like calculating the ball's trajectory. Instead, they employ various heuristics that are simultaneously both easier and more effective. One especially effective procedure, apparently also used by dogs, is what Gigarenza calls the gaze heuristic. The gaze heuristic is a stunningly simple rule of thumb that enables the player to be at that precise spot just when the ball lands, and hence to catch the ball. But it does not enable him or her to predict where it will land. It requires nothing more than fixing one's eye on the ball when it is high and running in a direction that maintains a constant angle between the line of sight and the ground. As the ball comes down, it does not require knowing its original position. And velocity does not require knowing Newton's laws or even the fact of gravity. But because humans have evolved in a world governed by gravity, the fact of gravity is, as it were, already built into their adaptive capacities. The Gay's heuristic is only one of many of Gigerens examples of adaptive thinking, but it well illustrates a central moral that he wishes to draw. Quote, Rationality is said to be a means toward an end, but the end is dependent on the interpretation of bounded rationality that is being taken for optimization under constraints. The end is to estimate the point at which the ball will land. Knowing the cognitive process can inform us, however, that the end might be a different one. In the case of the gaze heuristic, the player's goal is not to predict the landing point, but just to be there where the ball lands. The rationality of heuristics is not simply a means to a given end. The heuristic itself can define what that end is. And for that goal, the Gay's heuristic is not only more practical, but it's also at least equally and often a good deal more reliable. So how rational is the guidance of rational choice theory in the case of the gaze extreme events? Assessing the costs and benefits of human health or lives is an obvious challenge for rational choice theory, and it is one with which economists have long struggled. Low probability and high impact events occurring in the distant future pose two especially difficult challenges for cost benefit analysis. First is the problem, especially acute in the case of climate change, of assessing future costs, that is costs that are not to be shouldered by us but by future generations. And second is the difficulty, often impossibility, of computing either the magnitude of such events or the probability of their occurrence. Economic analyses of future costs depend critically on the choice of discount rate, that is, the rate at which the present value of a future cost decreases with time. For optimization models, the convention is to look to the market rate of interest for a determination of the appropriate discount rate and to the compensation people are willing to accept for a marginal increase of safety or longevity. But the greatest difficulty arises in estimating discount rates for a time when present actors will no longer be alive. Controversy about this question erupted sharply with the publication of the Stern Review on the Economics of climate change in 2006. Described in the news release as the most comprehensive review ever carried out on the economics of climate change, it was commissioned by the Chancellor of the UK and carried out by Sir Nicholas Stern. The main conclusion of the review that the benefits of strong action now at the cost of roughly 1% of the gross demand domestic product, greatly outweigh the expected costs of climate change. That conclusion came in for heavy criticism from mainstream American economists, and the criticism centered mainly on Stern's choice of discount rate. Convinced that discounting at a heavy rate would be viewed by most people as unethical because it involves discrimination between individuals by date of birth, Stern adopted a discount rate of 0.1%. American economists, by contrast, at least initially favored a rate closer to 3%. In fact, even a discount rate as low as 0.1% discriminates against those born in the future. But obviously the discrimination implied by a discount rate of 3% is far greater. In fact, it equates the value of one life today with that of 3.27 million lives in 500 years. The explicit dependence of Stern's choice of discount rate on ethics provided an easy target for a number of prominent environmental economists in the US William D. Nordhaus, for example, argued that such choices must rest not on an inevitably personal ethical judgment viewpoint, but on actual market data. He wrote, the review's unambiguous conclusions about the need for extreme immediate action will not survive the substitution of discounting assumptions that are consistent with today's marketplace. Close quote. Similarly, Martin Weitzman, although elsewhere arguing that the far distant future should be discounted at the lowest possible rate, criticized the Stern Review for relying on a priori philosopher king ethical judgments. Close quote. Yet, given the conspicuous moral implications of such calculations, I find it difficult to see how anyone might claim that any choice of discount rate, and hence any estimate of cost could be free of ethical or value judgment. But I'm not going to talk about discount rates today. The second. I'm going to talk about the second problem that's raised by the prospect of extreme events in the far future. And that prospect may pose an even more fundamental challenge to the rationality of expert judgments of risk. It derives first from the inherent, that is Structural uncertainty of climate science, second, from the uncertainty of behavioral responses to such changes, and finally, from the implications of all this uncertainty for cost benefit analysis. Climate science can tell us that catastrophic climate changes are possible. They can even tell us that the probability of such change is not negligible, but it cannot provide reliable estimates of just what that probability is. Indeed, the more extreme the event, the more uncertain the probability of its incurrence. And the obvious question is how, in the absence of quantitative estimates of the probabilities of extreme events on the one hand, and of the magnitude of their outcomes on the other, can one estimate the expected costs they would incur? One answer is simply to limit the outcomes to be considered to those whose likelihood is judged to exceed some minimal value. But even that requires some quantitative estimate. A number of authors have recently argued that conventional decisions theory systematically undervalues the effect of hard to predict but high impact events, popularly referred to as black swans. Indeed, the response of many cost benefit analysts is to bypass this problem altogether by assuming as the size of an event increases, assuming that its probability decreases so rapidly that the right hand tail of the distribution can simply be ignored, as for example in normal distributions of events. In other words, extreme events need not even be taken into consideration in estimating costs because of their very low probability. Unfortunately, all available indications argue that high impact events are not distributed normally, that is the tail does not drop off so sharply, not even sharply at all in some cases. Furthermore, the very uncertainties inherent in the dynamics of climate change, as well of course as those due to the incompleteness of our knowledge of these dynamics, adds to the fatness of the right hand tail. Right hand tail being this magnitude of the events right hand tail of the distribution, thereby raising estimates based on the data we already have of the probability of a large catastrophe. That probability may still be small, but it can nonetheless make a huge contribution to the costs involved. Ditto for estimating the economic impact of extreme events. The absence of prior experience and hence of prior knowledge, requires extensive speculation about how to extrapolate the impact beyond what is known. And small modifications in these speculations can have an enormous impact on the final computation, indeed larger even than changes in discount rate. For this reason, Martin Weitzman concludes that, quote, the answers to the big policy question of what to do about climate change stand or fall to a large extent on the issue of how the high temperature damages and tail probabilities are conceptualized and modeled by implication. He goes on, the policy advice coming from coming out of conventional thin tail that is Normally distributed cost benefit analyses of climate change must be treated with extreme skepticism. Weizmann goes on to suggest various ways in which conventional analyses might be modified to take proper or better account of the risk of extreme events. But the point I want to emphasize is that what is normally taken as providing the basis for rational decision making, namely cost benefit analysis, the standard against which human behavior is judged lacking in rationality, is itself deeply problematic. For the sorts of problems we face in this area, the tools needed to connect rational decision theory with our predicament are simply not available. And an obvious question arises While environmental economists search for better, that is more rational ways to account for the impact of extreme events, might it not be possible to identify heuristics that, however imperfect, provide a more reliable basis for future action than do the dominant modes of analysis now in use? Indeed, might even ordinary people have evolved or developed heuristics that can outperform standard cost benefit analyses? Gigerenzer and his colleague Klaus Fiedler seem to think that they have. For example, they suggest that Slovik's data indicating a central role of dread in skewing perceptions of risk in the face of extreme hazards might be reinterpreted as evidence of ecological rationality. Catastrophe avoidance, they write, need not be seen as a socially expensive subjective whim, but instead as attention to the third moment of the frequency distribution. The dread risk dimension corresponds to the skewness of the distribution. This is still their writing. Still within the quote, attention to skewness corresponds to dread risk, and the degree of skewness measures the degree of dread. Moreover, they close quote. Moreover, they suggest that in assessing the risk of low probability high impact events, people's attention to skewness may be perfectly reasonable. Although the authors do not explicitly say so, a reader might conclude that in environments in which the frequency distribution of hazards is substantially skewed, fat tailed distributions dread risk might be seen as an appropriate, even effective heuristic. Indeed is an example of what Slovik himself calls affective rationality. Okay, so far my discussion has focused on the reliability risk assessment and the closely related question of what people believe. But there's another obvious problem. The elephant in the room belief is only a precursor to action and certainly not in itself sufficient to guide behavior. Indeed, the gap between belief and action is huge and a subject of much commentary. For example, when public confidence in the reports of climate scientists was at its peak in the US and belief in the imminent dangers of global warming seemed to be shared by a majority of American citizens, people nonetheless expressed a widespread reluctance to make any sacrifices that might help in lessening the dangers. Indeed, there seemed to be a growing gap between intellectual awareness of the problems and a willingness to enact effective precautions. Why should this be? Psychologists have generally attributed this gap to a lack of emotional engagement with either the urgency or the magnitude of the threat. People don't seem to feel at risk. A task force meeting in 2008-2009 found at least a partial explanation in the different ways in which affect driven and analytic processes function. They wrote that the two types of processes typically operate in parallel and interact with each other. Analytic reasoning cannot be effective unless it is guided and assisted by emotion and affect, quoting Dimasio here, global climate change appears to be an example where a dissociation between the output of the analytic and the affective system results in less concern than is advisable with analytic consideration, suggesting to most people that global warming is a serious concern but the affective system failing to send an early warning signal. Here, too, the relevance of fear or dread has particular salience. Emotions, by this argument, are rational because they enable us to act, especially under conditions where rational analysis either fails or or is inconclusive. This claim bears especially on fear, an emotion sometimes responsible for the difference between life and death. According to Joseph LeDoux. For example, if you were a small animal threatened by a predator and had to make a deliberate decision about what to do, you would have to consider the likelihood of each possible choice succeeding or failing, and could get so bogged down in decision making that you would be eaten before you made the choice. Under such circumstances, fear would clearly seem to be a useful heuristic. Most of us, however, remain wary of fear, and for good reason. Our experiences with the political uses of this emotion after 911 clearly underscore just how powerful an emotion fear is and how unwise the choices it can lead us to make make. Furthermore, as a motivating force, its effect is notoriously double edged. While fear can spur people to action, it can also impede action, even if a necessary ingredient for translating belief into action, as some neuroscientists have argued. It can also lead to avoidance, denial, and inaction. And the issue apparently is one of context Climate scientists, even those who are themselves alarmed, may be especially wary of evoking fear in their readers. Indeed, there are powerful constraints inhibiting all scientists from directly seeking such engagement. Theirs, after all, is the domain of the rational, not of the emotional. Their aim is to inform, to evoke in the reader the rational response to what they, as scientists, have learned, and of all emotions. Fear is generally regarded as especially counterproductive to the forming of rational responses. Furthermore, because of our current sensitivity to the ease with which it can be politically manipulated, the fear of fear has now itself become a political weapon in the debates about climate change. As climate scientists know better than most, this is a weapon that climate skeptics do not hesitate to deploy. Those like Jim Hansen, who elaborate on scenarios that cannot but be frightening are called alarmists, fear mongerers. They're accused of creating a climate of fear, of spreading climate porn, of narratives of fear. No one wants to be guilty, appear guilty of such charges, especially not climate scientists. But if conventional decision theory has routinely undervalued the risk of catastrophic events, if conventional uses of cost benefit analysis cannot be taken as a standard against which to judge non expert estimates of such risks, and if fear is identified as the central factor leading us to overestimate such risks, might not fear also be viewed, at least in this particular case, as compensating for the underestimates common to expert reasoning? Might not, in the face of risks before which conventional theorizing about risk so conspicuously fails, might not it be the case that fear has the effect of counterbalancing a bias that seems to pervade that theorizing? Might not it be the case that, rather than something to be avoided, fear can, at least under some circumstances, or at least under these circumstances, serve as a useful heuristic for a more rational response? One reaction to the difficulty of assigning probabilities to inherently unpredictable events, such as, for example, the tipping point of runaway climate change, is to give up on computations that depend on them and instead attempt to avoid such events in whatever ways are possible. The precautionary principle is an obvious form this response takes. Cass Sunstein has done an excellent job in enumerating many of the problems with this principle, at least in its crude form of to do no harm. And he focuses primarily on the great variety of ways in which harm can be done, including by the very exercise of preparation, precaution. And he's right. These costs too need to be included in the calculation. But a key flaw in Sunstein's efforts to amend the precautionary principle is that he fails to address the fundamental problem that invites its formulation, namely the difficulty or impossibility of performing such calculations. And for this, the writings of the continental philosopher Hans Jonas may be more to the point. Writing over 30 years ago, Jonas was already at that early point in time, worried about the future of the environment, and he sought to articulate an ethics for the future, especially an ethics of responsibility for distant contingencies, where that which is to be feared has never yet happened and has perhaps no analogies in past or present experience. Indeed, he argued that it is precisely when scientific knowledge is insufficient for predicting the future that an ethically required extrapolation must take over. For Jonas, the mere knowledge of possibilities suffices for such extrapolation, that is, for the identification of appropriate ethical principles. As he wrote, it is the content, not the certainty of the then thus offered to the imagination as possible, which can bring to light principles of morality heretofore unknown. Central to Jonas methods is what he calls the heuristic of fear. His argument is often likened to, or even conflated with, the precautionary principle, but I think this is a misreading. For Jonas, the heuristics of fear is more in the nature of a requisite, free for the moral considerations that need not just the moral, but also the technical considerations that need to underlie a precautionary principle or any other such principle. By his reasoning, we learn what it is that we value, what we are committed to preserving only when that something is under threat. Accordingly, moral philosophy must consult our fears prior to our wishes in order to learn what it is we truly cherish. The particular challenge raised by distant and future threats is that appropriate fear may not be in evidence. Despite the pervasiveness of fear as a natural autonomic response to present or imminent danger, future threats require an effort of reason and imagination in order to evoke the appropriate fear, that is, the fear required to guide our responses, our response to dangers that are imagined and distant. The response to dangers that are imagined and distant are in that sense less natural. They require not only reason and imagination, but also education. Jonas writes, quote, we must educate our soul to a willingness to let itself be affected by the mere thought of positive possible calamities to future generations. Bringing ourselves to this emotional readiness, developing an attitude open to the stirrings of fear in the face of merely conjectural and distant forecasts concerning man's destiny, that requires a new kind of sentimental education. Close quote. Such an education must proceed, in fact, not just a moral analysis, but it must precede any assessment of costs in order to know what value to place on the risks. Thirty years ago, the forecast to which Jonas refers may have been merely conjectural, but they are no longer so today. Yet an appropriate response to the measurements and predictions of contemporary climate science has not been forthcoming and indeed seems to be ever receding. The summer before last, the hottest on record, the U.S. senate declined to even consider legislation to regulate greenhouse gas emissions. Those whose hopes had been raised by Obama's earlier promises were devastated. As the Canadian political scientist Homer Dixon wrote, climate policy is gridlocked, and there's virtually no chance of a breakthrough. Three months later, the outcome of the 2010 elections lent his prediction stark affirmation. Many factors contributed to this denouement, but surely one critical contributor was the advice our legislators had received from the country's most respected economic analysts. Regulations might be desirable, but the calculations most commonly employed purport to show that it cannot be justified on strictly economic grounds. Quite simply, it doesn't pass the acid test of positive net benefit. In other words, economic analysis have an enormous influence on public policy. And if the assumptions on which such judgments are based are faulty, we all bear the consequences. Indeed, the confidence with which policy analysts have accepted the application of mainstream standards of economic rationality to the particular problems posed by climate change seem something of a puzzle to me, especially given the mounting criticism of these standards that we have begun to see among economists themselves, and especially in comparison with the scrutiny under which the claims of climate scientists have recently been put. Perhaps I'm suggesting it is time to put economists to the same kind of scrutiny. To be sure, the task of estimating the cost of climate change taxes them with the difficulty of dealing with problems extending well beyond their traditional domain. And it is of little wonder that questions about the applicability of traditional analytic or rational criteria in this new domain should arise. As Jonas so presciently observed, the specter of such distant threats requires new ways of thinking, ways of thinking that are unfamiliar to contemporary human sciences, but may prove to be more reliable and hence more rational than those with which we are most familiar. More rational for the simple reason that they better prepare us to act appropriately. Thank you.
- [47:19]
A
Thank you very much. So we have time for discussion now. The microphones are circulating. Please say briefly who you are and then ask your questions. Yeah, I'm Richard Bradley.
- [47:35]
C
Thank you.
- [47:35]
D
So, I'm Richard Bradley. I'm in the philosophy department.
- [47:38]
B
What is your name?
- [47:39]
D
Richard Bradley. So I think you're surely right to draw our Attention to the tremendous difficulties facing application of the expected benefit principle in these cases where there are catastrophic risks well and long time in the future, outcomes that are very difficult to assess, and so forth. But I really fail to see the, the basis for the claim that we can expect heuristics or even hope that heuristics can do better for us than decision making based on this kind of procedure. The important thing is we're in a very different kind of environment to one in which things like the gaze heuristic have their success. I mean, in these cases we know what outcome we seek. What the gaze heuristic leads us to do is precisely what computation tells us to do. So we have a clear measure of what it performing reliably amounts to. But in the case of responses to catastrophic risk, we have no such basis. I mean, we don't have instances where we can't observe instances where decisions have been based on some kind of heuristic and have successfully led to success, because we don't really know have a clear idea of what success would have amounted to in these cases, and we don't know what the humor heuristics are doing. So it seems to me that at this point there's really no sort of basis for pointing to the possibility of a heuristics doing better than the expected benefit principle. Indeed, if we could point to something like that, then of course the expected benefit principle would tell us at some kind of higher order level just to use the heuristics in order to sort ourselves out of this mess. So it's not clear to me at all that it can replace the, the expected benefit principle when that principle is what's going to be able to tell us when the heuristics are successful or not?
- [49:26]
B
Well, no, I mean, it's a very important point that you're raising, but it's not clear that the heuristics or that the computation can tell us can measure the success of the heuristics. What we can say is that by any reasonable notion of rational analysis, the practice of estimating the likelihood of extreme risks is very faulty and systematically under calculates the magnitude of the risks. So by that logic, and it is that standard against which ordinary judgments of risk are said to be inflated. So the question is, you know, my suggestion is, rather than an argument, is that it may be the case that our ordinary so called irrational heuristics that people use to may be compensating, may serve to compensate to some degree for the undervaluing of the risk in conventional analyses. And I invoke Guillarenza and his colleagues analysis of the skewedness of the distribution as possible support for this suggestion. But it is just a suggestion. It's certainly the case. The main argument is that traditional methods of assessing the cost of extreme events cannot be used as a gold standard of rationality.
- [51:04]
D
Can I respond?
- [51:05]
B
Yeah.
- [51:06]
D
So it seems to me there's two different things. One is the situation which I believe we're in, which we find it very difficult to calculate the magnitude of the risks that we face because we're unable, I mean, as you pointed out, very clearly unable to arrive at clear probability estimates. We don't really know what sort of utility functions to slap on the possible outcomes and so on. So that's a situation which we don't know how to quantify. And there's another kind of situation in which we do know and we find that we often underestimate them. So in order to be able to say something like we routinely underestimate the risks, we need to be able to calculate what those risks are and compare it to what our normal calculation of these are. And that's not the situation in which we're. In the case of Ketoshrot, we don't know what the concern correct evaluation of the nas, we don't know whether we're undervaluing or not.
- [51:57]
B
We don't know what the correct calculation is. But we do can say that if you cut off the tail of the distribution in your computation, you are underestimating. And that is a quite common practice. So.
- [52:16]
A
Christian, maybe you can just pass to microphone on. Thank you.
- [52:23]
E
Yeah, Christian List. I'm also in the philosophy department, so my question is related to Richard's previous question. So these fear based heuristics, which we may have a psychological tendency to use, may very well have been ecologically rational for our ancestors living in the savannah or in some completely different environment where under evolutionary pressures over many generations, those psychological dispositions ended up being shaped. But in the environments that we now operate in, where a lot of the decisions that we have to make are a lot more abstract than the decisions faced by our ancestors in the savanna, it's much less clear that there is this nice calibration between the environment and the heuristic that would still make the very same kind of heuristic ecologically rational. And so then a lot depends on which scenarios are sufficiently psychologically salient for us to activate the use of the fear based heuristic and which scenarios are insufficiently salient. And likewise, if we were to use the Hansiona style approach, we would still have to somehow come up with a criterion for deciding which scenarios to make sufficiently salient to fear and then to try to avoid, and which scenarios we can set aside. And selecting scenarios as relevant or irrelevant for these kinds of considerations seems to me to be precisely the task where we still need some form of conventional decision theory or statistical analysis, because otherwise things will just depend on what happens to be salient to us for completely contingent reasons. I mean, as people like Sunstein point out, the types of problems for which a precautionary approach, let's say, tends to be employed in the US can be quite different from the types of problems to which a precautionary approach is employed in Europe, simply because the nature of the culture, public opinion and media coverage and so on is somewhat different. So that different scenarios are salient in people's minds.
- [55:07]
B
Well, it's not obvious that that to me that that accounts for the discrepancy between European attitudes and American attitudes to the precautionary principle. But I think it's a mistake, it's a really serious mistake to think of rational decision theory as the only alternative to intuitive heuristic estimates of risk. I think that it's also a mistake to assume that heuristics, that useful or relevant heuristics need to be products of biological evolution. They could also be the products of individual experience or cultural experience. And I think it's also a mistake to sever. And I didn't really, I just hinted at this at the paper because it's such a huge subject. But I think that the relation between the moral implications, the moral evaluation and the assessment and the response are absolutely central to this whole problem. And that's why I think Jonas approach is interesting because he is talking about a moral education, an education for him the heuristic fear is useful in, in informing us as to what it is we truly cherish, what we value, what the moral value of an action is. And I think he's right to insist that it precedes the implications, that it precedes any attempt to put any kind of objective value on, on a particular consequence. But let me just, you know, I don't pretend to have a solution to this problem. I am saying, just let me reiterate what I think the central points are, that the conventional understanding of rational decision making or analysis is inadequate in the face of extreme, extreme risks. It's inadequate, it underestimates by just the criteria that I mentioned, just A few minutes ago. It underestimates the cost of extreme consequences, extreme risk. And it's not an adequate appropriate standard by which to judge other ways of estimating risk. Irrational. That's the central point. And another central point is that fear has been emotion in general has been excluded from decision making, from rational decision making. I think inappropriately or to a degree that is inappropriate. And fear especially has gotten a very bad rap and especially since 2009, especially since 9 11, that fear does. Fear is a very complicated emotion and has very complicated consequences. But I think that we cannot think about public responses or we cannot think about proper estimations of consequences without thinking about the uses or the appropriate uses of fear. Now let me give you an anecdote which was part of my. My motivation in developing this argument. In World War II, in the Manhattan Project, when the first atomic bomb test was being run, Enrico Fermi, or at least scuttlebutt is that he was sitting under a tree computing the probability of H the test would ignite the atmosphere. And he concluded that the probability was very, very low. And therefore he stopped worrying about it. And in fact the calculation was published some months later in the Physical Review. It struck me that the probability of igniting the atmosphere was like 10 to the minus 20. Very, very low. Suppose the computation had yielded 10% risk. Just a thought experiment. It strikes me that neither Fermorie nor any of the other scientists in the project. Would have wanted to proceed with the test. That there comes a point somewhere, I don't know where it is, but it comes a point where people can be relied on for a moral response to a risk that they cannot calculate but is higher than acceptable. That is somehow they can't. They may not have to be able to give it an exact as quantity, but they can say it's order of magnitude. They will put. They will make a decision and it will be a decision based on the value on valuing the future of life, on valuing the future of civilization. I'm valuing the lives of children born 500 years from now, et cetera, et cetera. So I was trying to get at what kind of calculus would ignite. Acknowledge that piece of the consideration. And I think fear is crucial.
- [61:05]
A
For quick comeback. So the list is growing. Let's speed up.
- [61:09]
E
Just very quick remark on this last example. I mean, it seems to me that the very last example that you gave could also be captured by standard decision figures theory simply if we acknowledge that the utility or disutility of the atmosphere igniting is obviously Enormous. And so even if we discount this by 10% risk or 5% risk, we would still want to take the action of not playing with a fire there. So this all depends on what we assume about the utilities in this particular case.
- [61:51]
B
I think maybe we can continue this conversation later, Alex.
- [61:57]
C
Yes.
- [61:57]
F
So regarding, you made an interesting point, Alex Voover. Sorry, philosophy department as well. You made an interesting point about the disconnect between. At some point American public believed that that global warming was happening, but then they were unwilling to do something about it. And you suggested that this might have something to do with the right emotions not being aroused.
- [62:20]
B
That wasn't my suggestion. It was the conclusion of the report of the American Psychological Association.
- [62:26]
F
I take it that the reason that the emotion of fear isn't going to be aroused is because many people, even if they do believe climate change is going to happen, will happen, they believe it will. Will happen only severely to generations quite far after them and worse be worse for people that they hardly care about in their everyday lives. So fear and other emotions actually are going to be extremely hard to arouse when really we're concerned with distant, very distant strangers.
- [62:58]
B
Precisely. Jonas's argument.
- [63:02]
D
Right.
- [63:02]
F
But then his argument seems to get it backwards. He says, what would you fear? And then that reveals your values to you. Well, actually, I think here we need a moral education. That's right, but it has to work the other way around because we're not programmed, neither biologically nor socially, nor even morally, to have strong emotions about harms that will come to people very distant from.
- [63:28]
C
From us.
- [63:28]
F
So to consult our emotions to decide whether we care about it seems to get it the wrong way around. I mean, we do need moral education. So I agree with that conclusion, but. And we need to activate certain emotions, but that make that knowledge, that realization may come to us in a cool hour rather than from our gut, so to speak.
- [63:54]
B
Again, a very important point that you're making. But the argument against the attack on, for example, the attack on Jim Hansen or other climate scientists that are generating this anxiety, and Sunstein's argument generally, is that even though we don't, by Jonas's argument and by your argument, have the appropriate fear, appropriate to the consequences in the distant future, our fear is already so great that it distorts our risk perception that it makes us ask for regulations that are based, you know, that are inappropriate. So there are two levels of the argument. One is what to do about the fear that already is aroused, and the other is how to provide the moral education that would sensitize People to future consequences.
- [64:53]
F
If I may briefly just on this. So my issue is I don't fear climate change. So fear is a response to something, usually either that concerns me or my direct loved ones. And I might fear that there's a massacre going on in homes, for example, now. But that's. If I'm perfectly honest, that's not a particularly hot emotional state. Right. I'm appalled by it, et cetera. So I think it's morally wrong, but it's not a strong emotional response. And I think that's typical of human beings.
- [65:25]
A
So.
- [65:27]
F
I think that the people accusing climate scientists of spreading fear have it completely the wrong way around. We're incapable of feeling strong fear on behalf of people with whom we have very little ties.
- [65:44]
B
How strong does the fear have to be? I mean, again, this is.
- [65:48]
F
Well, we're looking for something else than fear, I think, concern.
- [65:54]
B
But the argument that I was addressing, Sunstein's argument, was that fear is provoking us to. To the wrong response, evoking the wrong response. It is that the fear that is the call of concern that is aroused is inclining us to spend too much on regulation. And we have to trust. We have to look not to popular sentiment, but we have to look to expert reasoning. Expert reasoning, I claim, is not a reliable guide here. That again, it's the levels of fear that are being considered. So maybe concern is another word. Maybe you need to have grandchildren to be afraid.
- [66:36]
A
Okay, there is a response right on this. The lady in the green scarf. Oh, it was a separate question. Sorry, I thought you were. Okay, then we go back to the queue. That's the gentleman who wrote to jump ahead.
- [66:54]
G
Hi, I'm Brian McGillivray from Psychology in Cardiff. I guess I've got three points to do specifically with heuristics, and the first one carries on from the last question, which is the idea that rather than rely on expert analysis, we should defer to the dread heuristic or the effect heuristic in dealing with climate change. And to me, that just seems like a recipe for inaction because as the guy said, people don't dread or fear climate change. The psychological research shows that people perceive it as a distant phenomenon that will affect other people in other places. So that just seems to me to be a non starter. And the second one is. I thought you were a little bit unfair on decision theory. It's not monolithic. It's not just rational choice. There's also minimax, for example, there's quite a large literature in that and that was developed specifically to deal with situations where we can attach probabilities to events. So it's essentially a formal version of the precautionary principle, and it's a heuristic of sorts. And I wonder whether that wasn't a more appropriate way forward. And the third one is the dichotomy you're drawing and other people are drawing between heuristic judgments and formal analysis, I think, is a false one. I think as soon as you look at formal analysis, you find expert heuristics in defining what counts as valid data and how to extrapolate from. From beyond data points and in choice of structural models to apply, and in Cotton, guiding causal inference as well. So don't really buy the whole. It's formal analysis versus heuristics. I think the two pervade each other.
- [68:25]
B
That is a lot to take on. Okay, can I come back to it later? I mean, let me give that some thought and come back to it.
- [68:39]
A
Okay, Luke's next, then. I'm sorry, Rodin. You're getting a workout here.
- [68:53]
C
Yeah. So Luke Bovenslaus from the philosophy department. So I can see that, you know, there is a way in which physical fear sort of has epistemic value. It teaches us about where the risks are. And then I think there is a way in which we have to include fear in our utility function. That is that the fact that we will experience fear when we do certain things needs to be taken into account, and that's part of the utility. So I can sort of see that there is a role for fear. But then I think there are these many cases where taking into account fear just gets it wrong. Like, for instance, when people started driving to their vacation destinations in the States because they didn't dare to take any planes anymore, and the number of casualties just went up. And that's a case, I think, where the population needs to be informed that this is a stupid thing to do, that their risks now go up, and that this is because of their fear of flying and overcome it. Because you should be much more fearful with all the driving that you do right now. Or there are these cases where people are very fearful. If there's a polluting plant and the risk of leukemia goes up from 0% to 1%, then they're really very afraid. But if it goes up from 5% to 10%, then they're not bothered very much. These are cases, I think, where. Where you just have to say, you know, where you have to educate the population not to fear. And now I agree it's extremely difficult to determine where you have to educate the population not to fear and where you have to take the fear into account for its epistemic value and for its entering into the utility function. So it's just very difficult.
- [70:42]
B
It is extremely difficult. And what I would like to turn the tables and ask. The problem that I'm trying to address is the problem of the standard, the ways in which our customary modes of analysis in economics underestimate the risk of extreme events. So let me throw it back at you, and if fear seems too dangerous to you as a tactic, a strategy to invoke, what would you suggest?
- [71:29]
A
Okay, and next, Luke has been deprived of his microphone.
- [71:35]
C
So.
- [71:39]
B
Do you want to shout an answer? Do you have a suggestion for me?
- [71:46]
C
I mean, that's what I was doing in the case of the plane accidents.
- [71:51]
F
And the car accidents.
- [71:53]
B
Well, there you can do it, but you can't do it for extreme, really extreme events. That's the problem. You can't measure those frequencies. Frequencies. We make all sorts of estimates about what the effects of climate change are, but the ways in which we, we estimate the extreme consequences are really extremely faulty because they're systematically ignoring a huge part of the data. So I don't know anybody who has. I mean, Martin Weissman is trying to develop techniques for, including, for doing, for calculating those. The cost of those detail consequences. And I think that's. I think that's very important. But until we have better ways, the question is how to proceed. And I'm particularly struck by the force of, you know, it's a common perception in the US that at the economic costs of regulation are not justified by the risk.
- [73:25]
A
Okay, now to you.
- [73:28]
H
Thank you. So my name is Millie and I'm in philosophy at Roehampton. I wanted to link what you were saying to Susan Sontag's work on illness, illness and its metaphors. So she talks about how the perception of different illnesses has changed as we understood or failed to understand their causes and their cures. So in the 19th century, tuberculosis was. Was greatly feared and was seen as a kind of metaphysical evil. And then cancer took that role in the early 20th century, and now it's AIDS. And as we're beginning to learn how to manage AIDS, it's becoming things like swine flu. And the principle here is that something that isn't understood is feared in a different way than other parallel phenomena that can be understood and managed. Because sometimes you're using the word fear and sometimes dread. And it seems to me that the apocalyptic imagination has been part of humanity. It's Part of human history, we always kind of fear these catastrophic things. The only thing that changes is what we consider to be catastrophic. So I guess my question is, is this a problem that can be solved or is it simply relocated as one potential catastrophe? We learn how to deal with it, we find a new thing to find, catastrophic. And that's not necessarily a bad thing.
- [74:38]
B
It's a problem that has to be solved or at least addressed. And whether, whether we, whether fear has that circulating dynamic or not seems to me it's an interesting question, but it's not necessary. It's not germane to the need to respond to what the predictions are or what the we need. That assessment of risk is very faulty, but we have no better tool for managing our way into the future. I'm not worried about whether we're a more or less fearful society. I'm worried about climate change. I mean, that is my starting point, principle, starting point. I think that we are in a situation that requires action and we are seem to be incapable, at least in the US of any appropriate action. So I'm not worried about the mental health of our population. I'm worried about the physical health of future generations.
- [75:59]
D
Yeah.
- [76:09]
I
Hello, I'm Hee Sun Hwang from anthropology department. And thank you for your inspiring lecture. I was just wondering if we consider fear in a different way in terms with regards with our decision making and moral responsibility, then some other sort of other understanding of moral responsibility is required or not. Because I've been thinking why fear was more or less excluded in rational decision making processes. And then it seemed to me that it has something to do with that, with the fact that fear is a way of dealing, emotional way of dealing with the situation, which from the beginning we can't deal with properly. So it seems like there is either flight or to act against ourselves, something like that. And in the sense, especially in case of catastrophic events, it seems like these are kind of unpredictable events in itself and it makes us very vulnerable to the situation. It's more like a very radical opening to the situation itself. And I think it's more or less opposed to the, the very concept of rationality, in fact, because it frames the external events as something calculatable or understandable. Or we can.
- [77:50]
B
What frames it is something. What's the it there? What frames it is calculable.
- [77:57]
I
The catastrophe or decision making. I'm sorry, I lost my own context, sorry. But it seems like, like fear is a kind of emotional response that is really usually understood as reactive response, which means that it rather takes some subjective or autonomy, autonomous subjectivity from the supposedly rational actors. So it has to be dealt with something different, but in the other way. It has something to do with what we cannot properly deal with because catastrophic events are something that cannot be predicted, its full result or the most rational choices of our responses to the predicted problems as such. But it seemed to me. Oh yeah, well, I'm sorry, can you wrap up? Yeah, yeah, yeah, yeah. But it seemed to me that your focus on our perception or sensitivity to these kinds of catastrophic events cannot be the other way of framing the moral responses to the situations in the situations of catastrophic events because it's a. It rather. Emphasize more about the perceptibility to the external events rather than reclaiming the autonomous subjectivity into the rational calculations.
- [79:44]
H
Okay.
- [79:44]
I
Oh, sorry.
- [79:45]
H
Thank you.
- [79:45]
I
I was too long.
- [79:48]
B
I think. Well, just say, I think I'll mainly take that as a comment. But it's a comment I want to counter comment, make a counter comment. And that is that that suggests that in a situation where we cannot do computations that are required for rational decision making that we just not think about it. I mean that the only alternative is a so called subjective response. And I want to resist that dichotomy. I think that this is precisely what we. That the need to, the need to anticipate the future I think is very, I'm assuming is fundamental and I'm assuming that our scientific modes of analysis are the best, are the best, we have the best tools that we have available. But that it may be, I'm suggesting that doesn't mean that they cannot be improved upon and it doesn't mean that. And then the only alternative is to. I'm not invoking fear as a subjective response. I'm invoking fear as an informing response that contributes to this attempt, this effort to anticipate the future.
- [81:25]
A
Last question here in the back.
- [81:27]
J
Yes. My name is Elsa Lanking. I'm in the philosophy department at King's College London. So I was thinking about your suggestion that there might be other heuristics around that and simply fear, because we've been focusing a lot of fear, because I'm worried that the kind of structure you're setting up is very highly determined by the kind of problem you're wanting to focus on, which is climate change. So I've recently hung around far too much time around people who deal with catastrophic risk and rational responses in different ways. Context, which is pregnancy, where the exact thing happens that Sunstein highlights, where fear of a catastrophic risk, which is something, anything happening to the fetus sort of obscures any kind of thinking about the likelihood of such an event happening. And since almost anything you can do can result in something bad happening to the fetus, there is basically nothing at all somebody can do. Now that seems to be a clear case in which what we might call rational or risk specific or utilitarian or rational decision thinking could be really helpful in negotiating the sphere. But that's also a section where we have really good evidence on frequencies and.
- [82:40]
B
It'S exactly comparable to the risk of flying.
- [82:44]
J
Exactly right. What that suggests is that what you're really talking about about is not some kind of trade off between rational decision making versus heuristics, but rather that there should be a prior question of identifying those areas where we kind of have good evidence and those areas where we don't have good evidence. And that in those areas where we don't have good evidence, we should radically discount the predictions made by rational choice theory or any kind of evidence estimations like that. Not only in those kind of cases, we should also look at various other methods for coming to conclusions. Even if we recognize that in areas where we had good evidence, the kind of good evidence question superimposes the characteristics.
- [83:33]
B
That's very well put and I agree completely. And it addresses at least part of the, I don't know one of the questions that you raised about the dichotomy between heuristics, and I don't want to suggest a dichotomy between heuristics and rational analysis. I think that the particular notion of rationality invoked in, well, in Nordhaus's work, for example, is, is rather, it does pose, evoke a dichotomy that I think is inappropriate because especially since the conditions required for his mode of analyses aren't available. So I'm not suggesting fear as an alternative, but as a corrective and under those circumstances. But I realize it's extremely tricky because precisely because of the issues that both of you raised that the fear does have, often has, even if it sometimes can be a useful and maybe even necessary heuristic, it's a very dangerous one because it has all sorts of negative consequences. And I fully acknowledge that. And I hope that the focus, my focus on fear, maybe that's just. I'm struck by how much of your responses, all of which have been very interesting, but I'm struck by how much of your responses have been focused just on fear. I would like to turn the question, well, what are the alternatives? What else can we do to improve to what is in our toolkit for assessing the future or anticipating the future.
- [85:37]
A
Very, very briefly.
- [85:39]
J
So one very brief suggestion that struck me is a different kind of heuristic is the kind of don't mess with nature heuristic. And of course people have often argued it as an argument. It's is really bad because we mess with nature all the time. So, you know, as an argument against our arguments, that's not a particularly good argument. But maybe it's one of the kind of heuristics that in the. We really don't know what we're doing and we have kind of very little evidence. It's actually one of those heuristics that give an intuitive kind of precautionary principle where there seems to be some kind of balance, be reluctant to do something. So that was just a suggestion of a different, maybe less fearful heuristic.
- [86:19]
B
I think I'll.
- [86:22]
J
I know there's much to say about that not endorsing this.
- [86:25]
B
It's a suggestion.
- [86:26]
A
Well, just by way of wrapping up, I wanted to ask you whether you would venture a prediction here. So following your methodology, we would get us of the expert panel with suitably informed layper persons and let them think about climate change and anti terrorism. What would they come up with and how would it be different from what we're seeing in current politics? Omenamo, I'm asking you to read the crystal ball, of course, but just what is your gut feeling?
- [87:00]
B
I think that the answer might be quite different for those two cases. I think that because the questions cannot be separated from their political and social context, in fact. And I think that Sunstein's grouping of them together was a little bit mischievous. So let's suppose that we have. Okay, let's introduce into the economic analysis of the cost of extreme events. Let's introduce critics of economic rationality. Let's introduce psychologists who are interested in the role of, the role of emotion in guiding events. And let's see if we can. And let's introduce, let's resurrect Hans Jonas and bring him onto this committee. I think that the.
- [87:58]
C
I think that.
- [87:59]
B
The acceptance of the Stern Report would go up dramatically.
- [88:05]
A
I'm sure Nick would like to hear that. Maybe that's a good note to end the discussion. Well, thank you so much for an exciting talk and a very nice discussion.