Dialogues concerning Natural Politics: Book Five
Under what conditions do experts make a positive (as opposed to a neutral or detrimental) contribution to the policymaking process?
Book Five Synopsis: Book Five continues the discussion of the role of experts in policymaking. Even on the dubious assumption that policy experts are always infallible, it does not follow that expertise always contributes positively to the quality of policy decisions. The structure of the various sciences does not map onto the structure of the various problems that we might want policy to address. There are more (and different) problems than there are policy-relevant scientific explanations. Some problems are not covered by any scientific explanation. Other problems require input from multiple sciences, each of which operates under ceteris paribus assumptions and which, therefore, cannot be simply combined or integrated into a coherent explanation useful for policy purposes. Relatedly, there are no experts about tradeoffs, about how to balance or choose between multiple priorities that may be in tension. There are no experts about experts that can counsel policymakers when the advice they receive from experts is inconsistent. Under such circumstances, whatever advice experts might be able to offer, even on the dubious assumption that they are infallible within their scientific domains, need not contribute positively to the quality of relevant policy decisions.
Dramatis Personae
Emma – 30 years old; Fifth-year philosophy PhD student, specializing in political philosophy and epistemology; ABD (“all-but-dissertation”); dating Andre
Andre – 35 years old; Fourth-year philosophy PhD student, specializing in philosophy of science; Academic Advisor to undergraduate students in the Philosophy Department; dating Emma
Connor – 24 years old; Second-year philosophy PhD student, specializing in normative ethics; a devout political progressive
Jack – 25 years old; Second-year philosophy PhD student, specializing in logic and epistemology; a devout political libertarian
Book Five
Emma: Boys…can I get back to it?
Jack [in a little boy’s voice]: Aw, shucks, Ma. Don’t ground us. It won’t happen again. We pwomise.
Emma: You “pwomise”? I hate you.
Jack [laughing]: Dammit, will just make your point?
Emma [laughing]: OK, but I do hate you. I want that noted.
Andre [nodding]: It’s on the record.
Emma: As I was saying, political thought from Plato to Glenn Beck, as it were, is a complete muddle on the significance of policymaker knowledge – or, more to the point, of policymaker ignorance. In all political analyses that I know of, if policymakers are not treated as omniscient and omnipotent, they are at least treated as epistemically privileged, again, as if political office somehow bestowed special knowledge and abilities on otherwise normal human beings. This is true, as I noted, even in informal contexts: every voter assumes policymakers know enough to improve upon those states of the world the voter wants improved upon. Indeed, come to think of it, voters tend to make similar epistemic assumptions about members of rival political parties: it is not merely that these rival policymakers want to bring about states of the world that the voter disapproves of, but implicitly, the voter assumes that these policymakers are knowledgeable and capable enough to do so! Everyone reflexively and uncritically attributes unique knowledge and abilities to politicians. That’s nuts! I mean, politicians…they’re not wizards! They’re just – at best – lawyers. Do you know any lawyers, Connie, or, for that matter, any law students?
Jack [groans]: My god, law students are the worst. They think they’re smart! They’re just good at memorizing things and recalling details.
Emma: We agree on that, Zippy.
Connor: Yes, unfortunately, I know some lawyers…and some law students. And you’re right, Zippy.
Jack: “Unfortunately” is right: an almost universal reaction to the mere mentioning of the legal profession! Just hope that you never have to hire one…
Emma: The point is that human beings are not omniscient and omnipotent. People are not gods, neither can they become gods. Indeed, I see no reason to think of any particular class of human beings, especially not a bunch of former law students, as epistemically privileged in the respects relevant to causing effective policies. Civilization, modern society, and world history are products more of evolution than of design. Humans intervene in these evolutionary processes and affect their course, but not necessarily – indeed, I suspect, only rarely – as rationally designed. There is no reason to think that any person or any class of persons possesses special knowledge concerning relevant causal mechanisms sufficient to puppet-master society. I would think, Connor, to reiterate a point that Andre made earlier, that if you’re skeptical of omnipotent and omniscient gods, you should be especially skeptical of omnipotent and omniscient persons. But, I also think you should be skeptical that normal people, simply in virtue of acquiring official governmental duties, suddenly secure epistemic advantages adequate to making effective policies. If you think that the absence of God means that nature is a product of evolutionary processes, you might consider that the non-existence of epistemically-privileged puppet-masters makes society no less a product of variation, mutation, adaptation, development, and selection.
Connor: That’s what policy experts are for! It is not that policymakers become uniquely knowledgeable in virtue of being policymakers. Rather, scientists become experts in virtue of years of intense study of a specialized subject matter and then they offer this wisdom to policymakers. I agree with you that politicians aren’t epistemically special, but improving their knowledge is precisely the role of scientific experts in the political process.
Emma: I agree that experts are involved in the process only because policymakers would otherwise be ignorant, by and large, of much of what they need to know to make effective policies. It is the manifest fact of policymaker ignorance that brings experts into the policymaking process. The question is how well experts play this role or, perhaps better, the question concerns the conditions in which they play this role relatively well and the circumstances in which they add little to policymakers’ rather destitute knowledge reserves. I’m not suggesting that experts never contribute anything to policy effectiveness. Obviously, professional physicists and engineers were crucial to the success of Apollo 11. I’m arguing there is no reason to assume that they always make such a positive contribution to policy effectiveness, which, I think, is how many people conceive of experts and of expertise. If we don’t think of expertise as sufficient, we at least think it necessarily helps to solve a problem. But, there is no reason to think that Eagle landed safely and successfully because the psychologists who advised NASA on the choice of astronauts foresaw the suitability of Armstrong’s individual psychology for the circumstances that he confronted at the crucial moment. There is no reason to think that psychological expertise rather than outrageous fortune put a uniquely capable person in that module. That expertise always makes a positive contribution is a dubious proposition. Indeed, I would argue it is questionable to assume that expertise typicallymakes a positive, rather than a neutral or even a detrimental, contribution to the effectiveness of policymaking.
Connor: You’re suggesting that experts are not really expert?
Emma: No, actually. Well, yes. I do think that expertise is often, perhaps generally, time-sensitive: it often has an expiration date, in the sense that the experts of one generation are frequently the fools of the next. It also tends to be highly contestable. At any given time, on any given topic, there might be competing experts pushing mutually-exclusive accounts of relevant phenomena. One would need an expert about experts to choose between them! Inasmuch as expertise incorporates the theoretical outputs of science, this tentativeness and contestability would seem to be in the nature of expertise, to return once again to Andre and Zippy’s earlier point. But, my worries about expertise as a policy tool do not hinge on the possibility that experts are sometimes wrong. Indeed, even if experts were infallible in their particular domains of expertise, I would still insist that there are few reasons to think they typically contribute positively to policy effectiveness.
Connor: How’s that? How could experts be expert – by which I mean, how could the advice they give to policymakers be true or, in some sense, correct or adequate – yet they fail to make a positive contribution to policy?
Emma: Simply put, because social phenomena are typically far more causally complex than is reflected in the division of expertise into distinct specialized domains. There are more causes involved in most social phenomena than there are distinct domains of expertise. Even granting that experts are always fallible in their particular domains, there may be – indeed, as the relevant phenomena grow increasingly complex, I think, there must be – causal considerations encompassed by no one, by no expert.
Connor: Andre made a similar point earlier. He said there need not be an explanation of every phenomenon that might be pertinent to policymaking. I granted it at the time.
Emma: Yes, but I need to develop his point a bit in order to show the silliness of the assumption that experts always make a positive contribution to effective policymaking.
Connor [sighs]: Oh. Joy.
Emma: Actually, I think I can state my argument relatively straightforwardly.
Connor: I cried “uncle” a while ago, if you recall.
Jack [laughing]: Oh, stop it! You love this. It’s because of discussions like this that you’re a graduate student in philosophy rather than, I don’t know, a corporate accountant. No one has arrested you and forced you to stay. Yet, you stay.
Emma: Indeed, you do.
Connor [smiling]: Proceed, please.
Emma: As you wish…Even if there were infallible scientific experts with regard to each and every individual causal factor that figures in the manifestation of the phenomena that we need to control in order to make some effective policy, even if each expert provided counsel that was true, correct, and adequate with regard to their respective domains of inquiry, there need not – indeed, as a practical matter, would oftentimes not – be experts about any emergent consequences of interactions between and integrations of individual causal factors. The notion that some expertise is always better than no expertise, that experts necessarily make a positive contribution to policy effectiveness implicitly assumes either that there are no relevant emergent phenomena or that such emergent phenomena as are relevant are easily tractable by experts, either because there are experts about the emergent phenomena themselves or because experts about, say, political phenomena, are able to determine, in company with experts about, say, economic and sociological phenomena, the emergent consequences of the interactions and integrations of these diverse phenomena. I see no reason to accept this assumption. I see no reason to assume that experts about distinct phenomena can, individually or collectively, account for what happens, say, when physical and physiological, or psychological, factors interact and amalgamate, or when political considerations interact and commingle with economic and sociological factors. A physicist knows physics, but not – or, at least, not much about – physiology and psychology, and probably less about the consequences of confrontations between specific physical stimuli and particular physiologies and / or psychologies. Psychologists know psychology, but not physics. Go back to Apollo 11: in the last analysis, the success of the mission hinged on the interaction of Neil Armstrong’s physiology and psychology with incoming physical stimuli, an amalgamation of factors the consequences of which are not and may never be explained by any science. There is no science that I’m aware of capable of explaining or predicting exactly what was occurring in Armstrong’s central nervous system such that he successfully responded to incoming stimuli in a way that prevented him and Aldrin from dying like so many flies on the windshield of the moon. We might be able to account for how the human nervous system responds to stimuli in general, but there are no experts who could have known the specific stimuli that would confront Armstrong in the moment and the prevailing condition of his extremely causally-complex nervous system, and then have predicted the consequences of cross-temporal interactions of physical stimuli and nervous system. That Armstrong or, more exactly, that someone with Armstrong’s psychology, rather than someone else was at the controls in that capsule at the operative moment was more a matter of spontaneity – luck or fortune – than of deliberate design.
Andre: Part of what you’re pointing to – I think – is that much of science operates and is learned under ceteris paribus clauses. There aren’t experts about what happens when other things are not equal or constant, when multiple factors operate at the same time.
Emma: Yeah, I think that’s right. So, even if there were infallible experts with regard to all of the individual phenomena relevant to some policy, inasmuch as what is really needed is expertise not about the individual causes, but about their interactions and integrations, the experts need not make a positive contribution to the effectiveness of the policy. As phenomena grow increasingly causally complex, the scope for such ignorance of the policy significance of causal interactions and integrations, and, thus, the scope for the impotence of expertise, expands as well. Indeed, there is no reason to assume that such advice as experts can offer under circumstances like these will not be detrimental to the effectiveness of the policy. The assumption that experts necessarily make a positive contribution to policy effectiveness is without ground.
Jack: Actually, Emma, you’re understating the case. It is not just that the scope for impotent, if not detrimental, expertise expands as complexity grows, it expands exponentially. Under the simplest assumption concerning how causal factors might combine, there is one way to combine two causal factors, four ways to combine three causal factors, uhhh, eleven ways to combine four causal factors, and – um, let me think – twenty-six ways to combine five factors.
Emma: Right, Zippy. Yes, thanks. There’s one more point, related to this last one, that I would like to make.
Connor: Before you go on, can I just add that I hate Zippy, too?
Emma: Of course you can. We’re approaching consensus.
Jack: I have nothing but love for both of you.
Andre [laughs]: You’re such a liar.
Emma: Connie, what about tradeoffs?
Connor: Tradeoffs? What do you mean?
Emma: I mean, sometimes, oftentimes, we cannot have everything we want, we cannot realize all potential goals at the same time or in near succession, but must choose between goals, weigh considerations that are in tension. Who is there to assist the policymaker when competing considerations have to be balanced, when, for example, the negative socioeconomic effects of a public-health policy have to be weighed against the positive health effects of the same policy? Again, I’m assuming here that there are infallible experts about, respectively, socioeconomic and public-health considerations, but where is the expert about tradeoffs between competing factors? Who tells the policymaker what to do in such a situation when, say, the same policy that will save lives will destroy livelihoods?
Connor: That’s not a problem at all. Lives always outweigh economic concerns.
Emma: Oh, really?! So, you would sacrifice the entire world economy – the engine that produces enough food to feed over 7 billion people – to save one human life? No, I’m quite sure, you wouldn’t.
Connor: Well, when you put it that way—
Emma: So, there is some number of lives you would be prepared to sacrifice in order to ensure that the food-producing machine that is the world economy continues to function effectively for the survivors of your culling?
Connor: Yeah. I don’t know how many lives, but you’re right. I am not so much a deontologist that I am willing to sacrifice a reliable supply of food for the rest of the world in order to avert a single death.
Emma: Of course you wouldn’t, but I think you’re missing part of the point. Even a dogmatic deontologist cannot pretend that the issue in this example is simply one of weighing human lives against economic considerations. Economic considerations affect lives, too. There is plenty of evidence concerning the deleterious health effects of economic loss, unemployment, and material poverty. This is particularly obvious when you consider the likely effects of disabling the food-, shelter-, and clothing-producing machine that is the world economy. A lot of people are going to die without this machine in full working order. So, you can’t pretend that what is at stake here is simply lives versus trivial economic concerns of relevance only to the bougie. Indeed, the people likely to suffer most from knee-capping the economy are the poor and those with lower incomes, who can least afford it. The economic concerns at stake are relevant to the lives of everyone.
Connor: I see the point.
Emma: Also, is there not a moral difference between the consequences of a disease and the consequences of human-made policies meant to counter the effects of a disease? Are not one set of consequences more [makes air quotes] “natural,” for lack of a better word, and the other set more “manmade”? Aren’t human policymakers responsible, both causally and morally, for deaths due to their policy measures, but not responsible for deaths due to disease?
Connor: If politicians fail to act to counter the effects of a disease, then they are responsible for its effects.
Emma: Only – at best – on the dubious assumption that they know how to counter these effects. I might agree that, to the extent that policymakers know some way of mitigating the effects of a disease, but fail to act to mitigate them, then they are to a similar extent morally responsible for these consequences, but why assume that they know anything? Why assume a priori, without knowing anything about the relevant circumstances, that they know how to address the problem at hand? As always, like everyone, you’re assuming that the necessary knowledge is well within the grasp of policymakers and their expert helpers. Stop treating this assumption axiomatically. It is an empirical claim and you have no empirical evidence to support it. It is an empirical claim, but you take it on faith. More to the point, in the present example, if policymakers do not know how to effectively counter the effects of the disease, it is either false or just plain meaningless to assert that they are morally responsible for the relevant consequences. Indeed, you yourself would have to agree, on pain of inconsistency, that, if policymakers were so ignorant that their best policies could only exacerbate the problem – by, say, failing to mitigate the effects of the disease while, at the same time, making other circumstances worse, then they should not act to lessen the effects of the disease. You said it yourself, you would not support a policy that you believed policymakers were too ignorant to avoid bungling.
Connor: I did say that.
Emma: So, who are the experts that will advise policymakers in this scenario? Who are the experts about such tradeoffs? Who are the people knowledgeable enough about both the health effects of diseases and the socioeconomic effects of policy efforts to mitigate the health effects of diseases that can accurately predict the consequences of different policy options? Moreover, who are the morality experts qualified to advise policymakers with respect to the thorny ethical questions surrounding acting versus not acting to mitigate the effects of a disease that policymakers and their scientific advisors might not understand very well.
Connor: Econodemiologists? No, I’m kidding, I don’t know.
Emma [laughs]: There are no experts about such tradeoffs. In the end, in such a scenario, policymakers will have to act on such competing expert advice as they receive either from public-health experts or from social scientists – that is, they will have to simply ignore the need for trading off such concerns and choose to favor either mitigating health concerns from the disease while indulging all of the negative socioeconomic and health effects from this policy choice, or favor socioeconomic concerns while indulging all of the negative health effects from the disease – or they will have to decide such tradeoffs themselves, without recourse to an expert on such tradeoffs, who, again, does not exist. Experts cannot advise policymakers with regard to the best policy when the best policy requires that concerns of different kinds be weighed and traded off against each other. All an expert can offer is advice about what would be the best policy with respect to their field of expertise were other considerations irrelevant, but the crux of this example is precisely that other considerations are relevant. What’s more, in cases where what is really needed is a tradeoff between competing goals that cannot be simultaneously realized, there’s no reason to think that knowledge of the policies which would be best, were other considerations irrelevant, is particularly helpful to the policymaker. Such knowledge might just as easily distract the policymaker from discovering a rational tradeoff. In short, there’s no reason to assume a priori that experts can contribute to effective policymaking where either emergent phenomena are relevant or tradeoffs are required.
Connor: I suppose you are going to insist that such cases are typical, that it is only rarely, in relatively causally simple cases, that it might be reasonable to assume that expertise contributes to policy effectiveness.
Emma: You read my mind. Right. It is more reasonable to assume that expertise contributes positively to policy effectiveness the more causally simple the case. Unfortunately, causally simple problems, like the purely physical aspects of the Apollo 11 mission, are comparatively rare in society and politics.
Connor: Yeah, I don’t know. If your premises are true, you’ve given me a lot to think about, but I’m not sure I agree with your premises, though – I admit – it’s not immediately obvious to me which of them to reject, if any.
Book Five: For Further Reading
On Society as a Product of Evolutionary Processes
Bernard Mandeville – The Fable of the Bees or, Private Vices, Publick Benefits ([1714] 1924), Clarendon Press
David Hume – Treatise of Human Nature ([1739] 2000), Oxford
Charles Darwin – The Origin of Species ([1859] 2003), Signet
Carl Menger – On the Origin of Money ([1892] 2016), Ludwig von Mises Institute
F. A. Hayek – The Constitution of Liberty: The Definitive Edition, in The Collected Works of F. A. Hayek, Volume 1: The Market and Other Orders ([1960] 2011), Ronald Hamowy (ed.), University of Chicago Press
Gerald Gaus – “Hayek on the Evolution of Society and Mind,” in The Cambridge Companion to Hayek (2006), Cambridge University Press
On Emergent Phenomena
David J. Chalmers – The Conscious Mind: In Search of a Theory of Conscious Experience (1996), Oxford University Press
Mark Bedau and Paul Humphreys – Emergence: Contemporary Readings in Philosophy and Science (2008), MIT Press
On the Health Effects of Economic Loss, Unemployment, Etc.
P. T. Martikainen and T. Valkonen – “Excess Mortality of Unemployed Men and Women during a Period of Rapidly Increasing Unemployment” (1996), The Lancet, 348 (9032), 909-912
J. Pharr, S. Moonie, and T. Bungum – “The Impact of Unemployment on Mental and Physical Health, Access to Health Care and Health Risk Behaviors,” (2012) ISRN Public Health, 1-7.
A. Milner, A. Page, and A. LaMontagne – “Cause and Effect in Studies on Unemployment, Mental Health and Suicide: A Meta-Analytic and Conceptual Review” (2014), Psychological Medicine, 44 (5): 909-917
J. E. Brand – “The Far-Reaching Impact of Job Loss and Unemployment” (2015), Annual Review of Sociology, 41, 359–375.
On Deontological versus Consequentialist Ethical Theories
Larry Alexander and Michael Moore – “Deontological Ethics” (2020), in The Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/entries/ethics-deontological/
Walter Sinnott-Armstrong – “Consequentialism” (2019), in The Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/entries/consequentialism/
Book Five: Discussion Questions
1. In what sense do people tend to attribute special knowledge and capacities to policymakers of both their preferred party (or political perspective) and the opposing party? Consider and discuss some of the implications of the assumption that policymakers are epistemically privileged.
2. Consider and discuss the significance of the fact that humans cannot be omniscient and omnipotent for the notion that society is more a product of evolution than of rational design.
3. Why is the (presumed) infallibility of experts no guarantee of the positive value of their advice for policymaking?
4. What is a ceteris paribus clause? How do ceteris paribus clauses complicate the application of scientific expertise to the design of government policies?
5. What are emergent phenomena? What is the significance of emergent phenomena for the policymaking value of scientific expertise?
6. Discuss the significance for policymaking and policy experts of the need to make tradeoffs.
7. Under what circumstances is scientific expertise likely to improve (or impair) the quality of policies made on its basis?