Pictured: Reasonable citizens deliberating
In the latest issue of The Critique, Scott Aikin and Robert Talisse have an interesting article on a phenomenon they refer to as the owl of Minerva problem. Roughly, the idea is that the tools we use to evaluate the quality of arguments submitted in public political debate are necessarily retrospective. This is because as soon as we acquire those tools, they get turned loose in first-order disputes. And as soon as that happens, we need new tools to evaluate arguments that deploy those evaluative tools. The cycle repeats indefinitely. To the extent that “deliberative democracy” (in which political outcomes are strongly connected to the outcomes of inclusive processes of reasonable deliberation) requires citizens to have tools with which they can evaluate arguments currently before them, this problem may be fatal to the deliberative ideal.
Aikin and Talisse are two of my very favourite political writers, and I enjoyed the article. But I also disagreed with almost everything in it.
The formulation of the owl of Minerva problem in Aikin and Talisse’s article seems to me to be based on an implausible kind of linguistic determinism — essentially reducing critical thinking skills to knowledge of specific metalanguage — which produces a contradiction in their argument. The identification of critical thinking skills with knowledge of specific metalanguage is what causes the infinite regress they’re worried about, whereby the public’s ability to analyze arguments effectively enough for collective decisions to be based on reason-giving must always lag one step behind the arguments being used. But if critical thinking is so reducible, it should be impossible to identify phenomena requiring new metalanguage. The fact that it does seem to be possible to identify such phenomena has two possible implications. The first is that critical thinking may involve a substantial component of know-how (in addition to declarative knowledge of analytical metalanguage) that is sufficient (at least jointly with some minimum amount of knowledge of specific metalanguage) to generate new metalanguage to deal with new problems. The second is that all problems in argumentation can be analyzed upon mastery of a relatively small and easily acquired analytical vocabulary; new metalanguage may be generated for the sake of convenience, but the phenomena it is concerned with are easily grasped in terms of the basic vocabulary. In either case, the infinite regress would be halted in its tracks.
So perhaps Aikin and Talisse’s theoretical case for the owl of Minerva problem can be set aside. But their theoretical argument is inspired by empirical concerns regarding the quality of public deliberation in actually existing democracies. Does the ease with which metalanguage is appropriated as a tool in first-order disputes pose a challenge to the feasibility of deliberative democracy? The force of Aikin and Talisse’s empirical conclusions depend on the truth of the claim that the deliberative ideal is so widely endorsed (at least as an aspiration) as to be a defining feature of contemporary democracy; if the deliberative ideal is not widely endorsed, then widespread violation of its constitutive norms in political life should come as no surprise. The argument offered in support of this bold claim, however, is not very convincing.
I think many deliberative democrats would bridle at the suggestion that any contemporary democracy truly aspires to — let alone instantiates — the ideals of deliberative democracy. Aikin and Talisse’s claim that our democratic practices only make sense if we really do (currently) aspire to deliberative democracy only gives support to the claim that we aspire to deliberative democracy if our democratic practices do in fact make sense. But why suppose that our democratic practices make sense? If the values of deliberative democracy have not shaped our democratic practices, then a nonsensical, historically and culturally contingent, slapdash hodgepodge of practices is exactly what we should expect. And if actually existing democracies neither instantiate nor even aspire to the deliberative ideal, the shortcomings of actually existing democracy as evaluated from the deliberative perspective have no bearing on the feasibility of that ideal.
Perhaps Aikin and Talisse are actually only claiming that particular democratic practices — not our democratic practices considered collectively as an overall regime — depend on the deliberative ideal for their intelligibility. There are two problems with this. First, particular practices based on the deliberative ideal may be compromised by other practices based on other ideals, or on no ideals at all. If non-deliberative practices compromise deliberative practices, then this only casts condemnation on the non-deliberative practices. Second, Aikin and Talisse’s examples of particular practices that ostensibly depend on the deliberative ideal for their intelligibility strike me as very weak.
For example, are television panel debates to be understood as actual debates in the deliberative sense? Aikin and Talisse seem to take this as self-evidently true. But it seems much more likely to me that pro-and-con panels are normally set up so as to attract a broader viewing audience. The “debate” serves mainly to present contrasting positions, not to deliberate about them. Nor do such faux debates do much to facilitate the overall processes of public and private deliberation; the presentation of each side tends to be cursory at best. On non-commercial television, panel debates tend to be more productive, not because both sides reach an agreement, but because they allow for a more thorough presentation of pro and con positions and the reasons that favour them, and for modelling realistically decent (though not ideal) political argumentation. This contrast between commercial and non-commercial television seems to support the notion that panel debates do not proliferate because of widely shared deliberative values, but because there is a commercial incentive for the bare representation of both sides on television.
What about the claim that “bias, spin, derp, lying, flip-flopping, glad-handing, and all the rest” can only be regarded as vices from the deliberative perspective? This is clearly false. Bias, for example, tends to be regarded as a political vice when it favours an opposing group, threatens the stability of one’s political coalition, leads to rash decision-making, etc., not simply when it compromises a politician’s objectivity or reasonableness. When bias works in one’s favour, we call it something like loyalty or being principles, and it is often regarded as a political virtue. It seems just as easy to give non-deliberative accounts of the rest of these vices. To my knowledge, habitual dishonesty has never been publicly regarded as a political virtue under any system of government. Unless the deliberative ideal has been with us since the dawn of political society, it must be possible to criticize behaviour like lying and flip-flopping from a wide range of perspectives.
Part of the problem with Aikin and Talisse’s empirical argument is that they seem to be working with far too broad a conception of what deliberative democracy entails, eliding what is actually distinctive about it. Deliberative democracy is more than using just arguments in the democratic process; in Joshua Cohen’s words, it “ties the exercise of collective power to reason-giving among those subject to collective decisions.” But in existing democracies, the exercise of collective power more often seems to be linked to the interests of the rich, the organizational success of competing political groupings, and the aggregated unreflective preferences of the electorate. In short, the exercise of collective power is tied to the outcome of power struggle. Political life is power struggle because our societies are unjust; the distribution of wealth, liberties, opportunities and other goods does not conform to reasonable standards. By definition, an unreasonable state of affairs cannot be defended reasonably. But this unjust state of affairs we find ourselves in obviously has many defenders in political life. To the extent that they defend an unjust state of affairs in the political sphere, they are not entering into reasonable debate. Why then should their opponents conduct themselves strictly according to the standards of reasonable debate?
Achieving the deliberative ideal requires eradicating or somehow neutralizing the gross inequalities that pervade existing democratic societies and turn political life into a true battleground. Until this happens, it makes little sense for voters or politicians to always conduct themselves according to the norms of deliberative democracy. As Ken MacLeod put it, we put ourselves at a serious disadvantage when we mistake a fight for an argument. If public debate is and ought to be conducted as a fight, then it is no surprise and no cause for special concern if the tools of argument are appropriated as weapons of war. This phenomenon does not call the feasibility of the deliberative ideal into question. The last word goes to Cohen again:
Deliberative democracy is a normative model of collective decision-making, not a universal political strategy. And commitment to the normative ideal does not require commitment to the belief that collective decision-making through mutual reason-giving is always possible. So it may indeed be the case that some rough background balance of power is required before parties will listen to reason. But observing that does not importantly lessen the attraction of the deliberative ideal; it simply states a condition of its reasonable pursuit.