Have just been reading Sunstein’s *Worst Case Scenarios.* There is some Douglas/ Wildavsky and a lot of Kahneman/ Tversky in it. Since Sunstein is now our Regulation Czar, it is probably worth analyzing this approach with care. The question he is posing is, basically: why are we (ie. the US) over-responding to the risk of terrorism and under-responding to the risk of climate change? And his answer is cultural and cognitive: “people” do not estimate risks rationally, but rather based on which cultural type they belong to (hierachical, individualist, solidarist, egalitarian) and based on common human cognitive tendencies. They do not engage in cost-benefit analysis. Rather, they are affected by heuristics like availability (it happened recently) probability-neglect (vivid images of a potential event outweigh calculation of its likelihood), and outrage (people like to focus on bad events that can be attributed to a malicious or malfeasant actor).
How would one respond to this from the vantage of a “dispositif”-focused analytics? For one thing: the Sunstein approach seems to rely too much on the centrality of “the public” to the way that catastrophic risks are governed (or not governed). If one takes, for example, the case of smallpox vaccination as Dale has analyzed it: where did the impulse to engage in preparedness for smallpox come from? Not from “the public,” but rather from a fairly small group of security and health experts who were engaged in worst-case thinking. These experts worked to generate the heuristics Sunstein is talking about (availability, probability-neglect) not among the public but rather among officials who could put some kind of preparedness program into effect. Thus, scenario-based exercises such as Dark Winter. Something similar could be said about avian flu preparedness.
An analysis at the level of the forging of different kinds of security apparatuses (which are capable of recognizing certain problems and not others, which engage in certain kinds of calculation and not others) thus seems very much opposed to the cognitivist and culturalist explanations of someone like Sunstein. This seems to be a result of using expertise and technical practices as empirical objects rather than public opinion surveys and psychological tests.
Perhaps the two approaches are not mutually exclusive. If it somehow became clear that a public of some kind was demanding more climate change mitigation measures, then the administration would have more leverage to develop a mitigation-apparatus. Nonetheless, there is something amorphous and unsatisfying about locating the problem of rational risk regulation in the minds of the public. Other thoughts on this?
well, as you know I have some interest in publics that build their own systems of governance as ways of opting out or proposing alternatives to reigning technical systems, i.e. Free Software. So I think your question is very clearly posed, because my argument has always been that the “hacker ethic” and the motivation of the whole project develops out of “expertise and technical practices” and not from an amorphous cognitive or cultural similarity shared by geeks. It turns out, this has been one of the most persistent and repeated questions I have gotten, especially from sociologists and anthropologists. They ask it different ways, but it comes down to a puzzlement over something very similar to what you are describing here… that these geeks must be doing this for some reason related to their gender, upbringing, class, elite social location, (perhaps, ‘heuristics’ in sunstein’s sense) and not simply because of the contingencies of being inducted into, steeped in, and highly skilled at operating certain empirically identifiable and coherent technical practices (writing copyright licenses, using open mailing lists, bug-trackingf systems and source code managment systems, defining standards and protocols, sharing software code in specific ways, etc).
So my argument is that whatever a public is (in this case, a ‘recursive public’), it is not ‘out there’ as Fox Mulder would say, but immanent to the technical practices. The implication, which is also something I think bothers people, is that people not steeped in these practices are not part of the public. To which I answer, yes, not part of *this* public, which is the one that matters when it comes to much of the software and communication infrastructure permeating our world. I think I could probably mount a very similar argument for the emergence of preparedness or risk regulation, but with a far less ‘open’ and ‘egalitarian’ system of inclusion into the technical practices. Same goes for the current discussions of finance (and I note that Wired magazine, predicatbly, suggested that we should “crowdsource” the regulation of financial transactions).
Andy — I think this is really right, and it gets to something fundamental about Sunstein. As we have discussed before, his frustration (and, I guess, Wildavsky’s) seems to be that the public isn’t rational. There is often an unmarked conviction lying behind this sort of argument that the experts know how to be rational. So taking your point one step further, it feels like our instinct has not only been to study technical practices but also the different and often conflicting normative rationalities in which they are articulated.
I wonder what you think of this formulation:
Sunstein wonders about why the public is not rational in its assessment of risk, and explains it in terms of culture.
Jasanoff doubts the legitimacy of expert authority, and reduces it to culture.
We focus on the fact that the experts don’t agree — or that there are different ways that a problem can be problematized. Following Chris, it isn’t a matter of “culture” at all, although there is a different kind of practico-conceptual matrix of knowledge production (which I will NOT call a “style of reasoning”) that is the real object of analysis.
In any case, I agree with your point that these are not necessarily mutually exclusive. But there does seem to be a different critical instinct behind each.
just a further specification on this. It strikes me that the version of ‘cultural’ which Sunstein is employing here (though not the one Jasanoff is) is currently very appealing to legal scholars. I note that one of the most widely read papers on nanotech is the one by Dan Kahan and colleagues at Yale, who are quite sophisticated proponents of this version of culture. He responded to my post about this a couple of years back, and the conversation is interesting.
Here is how I would sort it out:
For Sunstein and the cognitivists, the task is one of enlightening the public. If we (experts) can just sort out the heuristics that are muddling their risk perception, we can move ahead with rational calculation and decision.
For Jasanoff and the critical STS types, it is perhaps more of a task of enlightening the experts: in order for their decisions to have legitimacy, they have to engage the public. And what we (critical analysts) study is the comparative success or failure of various regulatory authorities to constitute – and deliberate with – a public.
For ARC-types, the interest is in, as SJC says (yes, that rolls right off the keypad) the “practico-conceptual matrix of knowledge production”, public or not. And in the shared problem-space within which experts and their critics make claims and counter-claims. So we might initially ask of this debate: how did experts come to think “public risk perception” was the problem? And what devices are they setting up to manage it?
Thank you for this discussion.
Presumably, experts are people too (not simply points of emanation for some kinds of statements (‘claims and counter-claims’)), and so participate in (and reproduce) publics concerned by things like global warming, terrorist attacks, car crashes, pollution, nuclear proliferation, HIV incidence. Wildavsky notwithstanding, Douglas would certainly not pretend that the experts’ reason stands outside of culture, where culture is a term for connecting up morality, social relations, thought, and meaning. In a world where political decisions are being made with respect to threats of all kinds — and from Douglas’s perspective, it could never be otherwise in any setting for this is the nature of politics and power — a method for describing and analyzing connections of various kinds across the expert, the public, expert publics, lay publics, the zeitgeist, Twitter, and so on, is what is badly needed. While the refinement of an approach that locates risk as one object in a ‘practico-conceptual matrix of knowledge production’ helps to describe °that° problem-space, I continue to find the larger repercussions for actual existing and pressing concerns of the public good difficult to discern, murky.
Discourses of risk/danger expose moral visions about what is important when people agree to live together. Chris’s beautiful discussion (Two Bits) of the social imaginary that is reproduced in and as a certain kind of recursive public — that is, his discussion of the ‘cultural significance’ of this public and its means and media — exemplifies an approach to experts as people, and not just “people.”
Obviously, experts are people too – though Sunstein would presumably want to stick with the idea that when regulatory matters are at stake, there is a difference between specialists who speak with a certain authority and what he wants to call “the public.” Certainly in my own work – in which the experts are indeed actual people – I try to show how visions of the collective good are linked up to truth claims in specific contexts. How to analyze these linkages is one of the questions Stephen and I address in our article on “regimes of living”.
Though you may find the repercussions murky, I would maintain that it is of interest to analyze why and how different knowledge claims come into conflict. The characteristics of such conflicts are not always obvious in “public” discussions of issues like risk and security, and that’s one of the reasons to look carefully at the normative rationalities that structure expert practice.