Back in 2014, Maria Paula Saffon and Nadia Urbinati published a paper in the journal Political Theory entitled "Procedural Democracy, the Bulwark of Equal Liberty." It's a brilliant article whose contemplation these days gives me some pain, because, while I indicated my indebtedness to Urbinati in a number of places in my 2020 book on democratic theory, I had never seen that particular piece. That stings because the authors not only make a particularly fine case for bringing Hans Kelsen's unmatched justification of democracy to the attention of 21st Century theorists, but also because of the lovely way in which they bash the heads together of (i) epistemic democrats, (ii) advocates of (a certain type of) populism, and (iii) minimalist Schumpeterians as if those alternative theorists were Larry, Moe, and Curly pretending to give a joint talk at an APA convention. [KLONK!]
Saffon and Urbinati defend a position they call "Procedural Democracy" or "Proceduralism," a view that seems to me absolutely right insofar as it stresses the intrinsic value of self-government and pounds home the fact that all the theorists they criticise fail to demonstrate a real grasp of the fact that without a functioning democracy in place it's almost absurd to consider a polity free. Their Proceduralism, which they claim to have largely derived from Kelsen (as well as Norberto Bobbio), never stirs far from the essential point that the freedom and equality of a citizenry cannot be maintained without an adequately functioning democracy.*
Taking these opposing positions in turn, what do Saffon and Urbinati say caused those who defend epistemic democracy (e.g., David Estlund and Helene Landemore) to have entirely missed the main plot of the rise of democracy over the last couple of hundred years? According to the authors, their main mistake was to understand the value of democracy to be entirely extrinsic, a way for a group to produce the most desirable outcomes. This misses majoritarianism's essential contribution to liberty. In addition, they point out that if policy instrumentality really were the main point of having democratic governments, every outcome that is worse than might have been produced by a (either beneficent or confused) despot is a good argument against democratic rule.
I heartily agree with everything in this portion of the Saffon/Urbinati critique. In fact, I made many of the same points in this recent paper (but, you know, like a decade later....)
The authors also do a good job on populists, but only if you restrict "Populism" in such a way that it refers exclusively to particularly disgusting variants of what I believe is actually a fairly wide variety of theories.** Clearly, a Schmittian style fascism is inconsistent with representative democracy, and if that's what populism is or must become, it's silly to call a populist entity democratic.
Now, I'm not suggesting that it's absurd to claim that populism has a fascistic streak. Its forms of "radical democracy" often prefer identification with leaders to more traditional representation. And, of course, if the Leader just IS us, it doesn't make much sense to disagree with him/her. Thus, any desire for pluralism and minority voice in government is commonly cast aside by populist regimes as being detrimental to the real people, the sort of thing that can only be wanted by harmful "elites." Saffon and Urbinati accuse populist sympathizer Ernesto Laclau of that attitude, in spite of conceding his goal of combining "homogeneity (the project of unifying different citizens) with heterogeneity (allowing their plurality of interests to be voiced)." This is because they take several of Laclau's remarks to be antithetical to pluralism. Consider this one: "The classical theory of political representation, reduces the people to a pluralism of interests and values, because it makes of power an empty place occupied by formal rules of the game."
Perhaps they're right. But that attitude certainly does not comport with that of American populists like W.S. U'Ren, who worked tirelessly over a period of many years to give additional voice in government to various minorities. Whatever. I don't want to quibble about the definition of "populism" here. If we take the term to refer exclusively to Schmittian/Trumpian conceptions of government rule as Saffon and Urbinati do, it's hard to deny that Populism will turn out to be anti-democratic to its core.
Interestingly, the authors point out that in its way fascist-leaning populism is akin to epistemic democracy: both are focused on a single feature that is extrinsic to democracy proper, and thus ignore democracy's intrinsic capacity to increase the equal freedom of a populace. The relevant difference between the two theories is that for Populists, it is the unified and unstoppable "power of the people" rather than truth that is the holy grail.
Saffon and Urbinati contrast their rule-centric Proceduralism with one more general democratic theory here: a position they call "Minimalism," (or sometimes "Realism"), a theory that they say originated with Joseph Schumpeter. That position is minimalist in the sense that it calls for only that amount of democracy--and no more--that will prevent a system from being correctly identifiable as either anarchic or despotic. Minimalism is, as Przeworski indicates, basically a way to "get rid of governments without bloodshed," whether or not such governments make any attempt to figure out what the citizenry actually wants. In this way, it could be argued that Minimalism need only fool a populace into abandoning violence in favor of occasional elections that do nothing at all substantive in the area of policy choice. In addition, the Minimalist conception is said to "disregard equality as a key dimension of democracy."
Thus, by Minimalist lights, a little discrimination against this or that group is OK so long as the "rules of the game" are sufficiently kept to ensure that power can be transferred peacefully. In its fear (and arguably also loathing) of the masses, it generally allows no governmental power to be distributed to citizens other than the right to vote for one or more representatives occasionally. Obviously, however, having an election every so often isn't sufficient to produce a democratic polity, particularly if it's OK if these plebiscites are nothing but ruses.***
According to Saffon and Urbinati, the conclusion to draw from the defects of each of these alternatives to Proceduralism is that where authentic citizen preferences for this or that action, policy, and representative are downplayed--or even ignored, so will be the freedom of that populace to choose its own future.
So, in spite of a bit of hand-waving about what seems to me the intractable tension between majoritarian democracy and the problem of persistent minorities, this 2014 paper seems to me to have been a vital contribution to political theory, the sort of article that is sure to have made fans look forward to a sequel. And, within the last month or so, a follow-up paper called "Parties as Agents of Equal Freedom" did appear in the same journal. But, alas, as with so many sequels to excellent creations....
-----
While the place of political parties was not a highly visable feature of the earlier work, the new paper is focused on the claimed indispensibility of parties to any functioning democracy. However, the authors don't try to hide the fact that historically parties have often been less than virtuous. They write, in fact, "Parties have often turned into oligarchic structures disconnected from their bases." "They have also frequently become cartels who [sic] live off the state, protect vested interests, and engage in collusive practices," and note that they are "increasingly immoderate in their opinions and practices, stretching the democratic rules of the game." Saffon and Urbinati nevertheless mostly attend to the positive aspects of political parties, partly because they don't believe it's possible for democracies to amount to much without their crucial contributions to pluralist thought.
Thus, while Saffon and Urbinati maintain that parties are necessary to democracy and at least can be healthy supplements to sound democratic practices, their affection extends only to the good kind of parties, not, e.g., National Socialism or Trumpian Republicanism. That makes sense, of course, but rather than extend this quite reasonable attitude to advocates of non-Procedural democratic theories, they claim that Populists, Minimalists, and "Epistemics"# are antithetical to parties no matter how the latter institutions happen to be constituted or act in actual practice.
[Chart by a Reddit contributor with the moniker Mgmfjesus]
Unfortunately, the conclusions of the critique here are almost entirely predestined by the paradigms they choose for each of their adversaries. We have already seen this with respect to Populism in the first article, when they chose Schmitt as its primary spokesperson. I mean, to take a couple of the most famous Schmittian Populists, both Hitler and Mussolini were quite explicit that there should only be ONE political party. So, if "Populist" is a term for those sorts of positions only, it's pretty clear that there will not be a place for any parties but the single good one. Fascism is simply incompatible with pluralistic democracy of any kind.
In the sequel paper, this same kind of distortive conceit is applied to epistemic democracy, when, instead of Condorcet, they use Rousseau as the original model for that theory. And to make things worse, they imply that epistemic democracy is essentially tied up with deliberationism. Now, it can't be denied that several epistemic democrats (e.g., Elizabeth Anderson and Helene Landemore) have stressed the importance of deliberation in their writings, and that view has also contributed to their hostility to parties. But it also should be clear that the two ideas of deliberative and epistemic democracy are entirely separable. The point of epistemic democracy is that voting is the best way to determine "correct public policies." A primary focus on deliberation is consistent with having no elections at all. While deliberation may be useful in the determination of what proposed policies an epistemic democrat thinks ought to be voted on, jury theorems do not depend on that detail. The two independent theories seem to me to have been conflated here for the sole purpose of insisting that epistemic democracy must be hostile to political parties. Furthermore, Anderson and Landemore notwithstanding, it's not obvious to me that even stand-alone deliberationism must be incompatible with strong parties: it will continue to depend on how the parties operate.##
It is also worth noting that several well-known exponents of epistemic democracy have not taken the position Saffon and Urbinati have set aside for them. Estlund has said almost nothing about this matter, Thomas Christiano, while sometimes antagonistic to parties (particularly when they display features that Saffon and Urbinati also criticize), has pushed for proportional representation, something that is generally considered dependent on a system that includes at least two parties. And for his part, Robert Talisse has mostly been focused of late on expressing the same sorts of concerns about the hyper-polarization of contemporary parties### that Saffon and Urbinati do in their paper.
Again, this misconstrual is not surprising: it's true that Rousseau may not have seen any need for political parties in determining the "general will," but he was not an epistemic democrat at all. And, again, deliberation is neither here nor there in this fight. In fact, although some deliberationists have been epistemic democrats (and vice-versa), it probably would have made more sense to make an additional place for deliberation-focused theorists rather than just lumping them with "Epistemics."
Turning finally to Minimalists, we may take Schumpeter and Riker as our guides. And we will find that, largely because of their dismissive attitude toward general electorates, they both saw parties as indispensible to democracy, even if more along the lines of necessary evils than as anything to get excited about. And it should be remembered that even an "evil" (the party system) that is necessary for something that is itself taken to be a necessary evil (plebiscitary democracy) should be understood to be a "positive influence," rather than, as our authors put it, nothing but "factions that devour the common good." In other words, if democracy is (even reluctantly) considered a good, and parties are necessary to it as Schumpeter and Riker clearly believed, parties are themselves good things. These Minimalists were quite clear about why they thought parties were essential. The sole purpose of democracy on their view is to retain the ability of a population to throw one group out and replace them with another. And for them, it is only parties that can outline the alternative batch of these alternating ne'er do wells in a manner that can be grasped by the plebes.(Przeworski's attitude is along the same lines.)
Now, I don't want to imply that there is nothing worthwhile in this article. There are, in fact, a good many useful recommendations regarding how contemporary parties should be run. And it may be a valuable piece if taken simply as a paeon to the importance of political parties for democratic polities. Indeed the work makes a number of important points along those lines, points which are, no doubt, regularly ignored by political theorists.
But the paper should not have contained the suggestion that either Minimalism, Epistemic Democracy, or Populism (as correctly understood) are necessarily antithetical to parties. There are, certainly, difficult problems to consider when attempting to make parties (and federal structures as well!) consistent with majoritarianism. But that perhaps unpleasant fact is largely an entailment of simple mathematics, and does not have to be connected with any disdain for pluralism. Certainly, those types of thorny problems connected with voting and elections need have nothing whatever to do with the sort of democratic theory one happens to prefer.
------------------------------
*While I also discuss Kelsen's importance to democratic theory in my book, I put more weight on the (likely less-well-known) work of Everett Hall in my provision of a somewhat similar justification for majority rule. Like the Kelsen/Urbinati/Saffon take, the Hall/Horn version also focuses on the intrinsic value of free choices in accordance with principles that I take to be foundational regarding the equal weight assignable to both persons and their wishes.
**For a more comprehensive approach to populism I recommend both of Margaret Canovan's books on the subject. See also this paper, and/or my reviews of Canovan's The People and the Elgar Handbook on Populism here.
***I wrote a bit about "fake elections" here.
#I admit that their frequent use of "the Epistemics" in this paper sometimes caused me to wonder whether they were referring to an 80's boy band or a group of ESP-powered arch rivals to Professor X.
##The soup gets even murkier when the authors mix in sortionism as though it too were intimately connected with either deliberation or epistemic democracy or both.
###See my review of one of his recent books.
No comments:
Post a Comment