The possibility of self-government

The Possibility of Self-Government Author(s): Colin Bird Source: The American Political Science Review, Vol. 94, No. 3 (Sep., 2000), pp. 563-577 Published by: American Political Science Association Stable URL: http://www.jstor.org/stable/2585831 . Accessed: 05/09/2013 09:51 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. . American Political Science Association is collaborating with JSTOR to digitize, preserve and extend access to The American Political Science Review. http://www.jstor.org This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions American Political Science Review Vol. 94, No. 3 September 2000 ThePossibility of Self-Government COLIN BIRD University of Virginia M z,f any have suggestedthat thefindings of social choice theorydemonstratethat therecan be no "will of the people." This has subversive implications for our intuitive concept of self-government.I explore the relation between the notion of a "social will," that of self-government, and the impossibility theorems of social choice theory. I conclude that although the concept of the social will is essential to that of self-government,thefindings of social choice theorydo not cast doubt upon thepossibility of either. Unlike many attempts to respond to the threatposed by social choice theory, my argument does not requireany appeal to the problematic notion of the common good. T o speak meaningfully of an organization or po- suggestion that a society whose decisions are deter- litical community as self-governing requires that mined for it by a dictator is genuinely following its own it have (or implies that it has) the property of will. Similarly, to impose a blanket prohibition on having, knowing, and consciously pursuing its own society choosing a certain preference ordering, even "will." As an abstract proposition, this claim is not though all its members would otherwise do so, may also especially controversial, but exactly how the standard it seem inconsistent with the notion of society following sets might actually be met remains a matter of deep its own will. Arrow's terminology is telling here: To dispute. I will address a particularly strong form of violate this second proviso is to violate a norm of skepticism about the possibility of satisfying this re- "citizen sovereignty." quirement. This is the claim made by certain social In the years since Arrow's initial contribution, social choice theorists that it is not merely difficult, but choice theorists have both confirmed and deepened the logically incoherent, to believe there are social proce- general point (McKelvey 1979; Plott 1976; Riker 1982; dures under which collectivities can learn their own Schofield 1985; Sen 1970). As a result, they have been will. I hope to show that, whatever other problems increasingly confident in their claims about the impli- theories of the "social will" face, this particularlystrong cations of social choice theory. It is now standard for objection can be answered, and in a way that opens up these theorists to claim that the impossibility theorems, fruitful avenues for further inquiry. I begin by briefly because they establish there can be "no such thing as setting out the objection that I aim to defuse. the will of the people" (Maclean 1991, 509), mandate a certain kind of methodological individualism. Willing, preferring, and choosing are not possible predicates of THE SOCIAL CHOICE OBJECTION societies, and to think otherwise is to engage in a Starting with Kenneth Arrow's ([1951] 1963) seminal primitive kind of anthropomorphism (Riker 1982, 18; monograph, social choice theorists have sought to Schelling 1984, 93). The comments of Charles Plott identify procedures under which a collectivity might (1976, 525) are representative: identify the preferences that could reasonably count as The concept of social preferenceinvolvesan illegitimate its own. Taking for granted relatively uncontroversial transferof the propertiesof an individualto the properties liberal democratic assumptions, the social choice the- of a collection of individuals.... The Arrow theorem orists asked a simple question: How can one devise demonstratesthat the concept of social preference in- procedures to "process" or "aggregate" the impartially volvesthe classicfallacyof composition,and it is shocking weighted preferences of individual members of a col- only because the thoughts of social philosophersfrom lectivity into a preference that can stand for the whichwe have developedour intuitionsabout such mat- collectivity as a whole? The answer Arrow and his ters are subjectto the same fallacy. successors repeatedly gave was that the only proce- I seek to refute this kind of claim and to show how dures capable of identifying a coherent and unambig- having a will might be a possible property of both uous collective preference or "will" involve either societies and individuals. Before I tackle these claims, deferring to the judgment of a dictator or narrowing in I want to set the technical issues raised by social choice advance the range of preferences that citizens are theory within the broader context of discourse about permitted to express. Neither of these ways of gener- self-government. This is essential if we are to under- ating a coherent collective preference-seems to satisfy stand the nature of the difficulty identified by social our intuitive sense of what it means for something to choice theory and its relation to the larger theoretical have its own will. No one would be satisfied by the issues that it often claims to resolve. Colin Bird is Assistant Professor of Government and Foreign Affairs, University of Virginia, Charlottesville, VA 22903. SELF-GOVERNMENT: SOME The author thanks Talbot Brewer, James Cargile, Dan Devereaux, PRELIMINARIES Joshua Dienstag, Paul Freedman, George Klosko, Charles Kromkowski, Debra Morris, Steven Wall, Brantly Womack, and Very roughly, to say that a political community or several anonymous referees for their extremely helpful comments on organization is self-governing is to say that in some earlier versions of this article. sense or senses the actions taken or controls imposed 563 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions The Possibility of Self-Government September 2000 by its governing institutions can be thought of as sectional or partisan groups within the community. For originating from within that community or organiza- instance, juries might seem to be self-governing in a tion. There is no essential or canonical specification of variety of other senses (e.g., juries do act in accordance the senses in which self-government demands that with their will), but they are not self-governing in this authoritative action be internally "sourced" in this way. third sense, for a jury does not derive justification for a To set the stage for my argument, however,. I will verdict on a particular occasion from a set of claims identify three salient ways in which one might claim about its own well-being but, rather, looks to some- that authoritative action originates from within a self- thing beyond itself-the fate of a third party. governing political community or organization. I will Similarly, a society whose ruling class justifies the then use these three senses of self-government to make exploitation of the rest of the population by reference several preliminary points about self-government, de- to its own partisan interests, rather than to those of the mocracy, and social choice theory. whole society impartially considered, is not self-govern- A first and relatively undemanding sense in which a ing in this third sense. In such a case, justifications for group can be self-governing operates at a basic institu- public action are not grounded in claims about the best tional level. A society can be self-governing simply to interests of society as a whole but in claims about the the extent that the apparatus of government and interests of a partisan subset of the community. Here, administration is operated exclusively by officials and it seems more accurate to say that one social group is participants drawn from within, and authorized by, the simply subjugated to another in that the former's citizen body, not by outsiders. In this institutional interests are not given due weight in the (purported) sense, imperial domination is the antithesis of self- justifications for public action offered by the ruling government. class. One can reasonably claim that this sort of A second and more demanding interpretation of the subjugation is inconsistent with at least ideal forms of idea that public action originates from within makes self-government. Certainly, the history of modern con- use of our organizing idea of a social will. In this view cepts of self-government, from Rousseau on, is replete of self-government, public decisions must be plausibly with examples of theorists and commentators who have understood by members of the collectivity as reflecting, understood the concept in this way.1 expressing, or revealing a will that is authentically their To set the stage for my argument, I now make own, and there must at least be social consensus on several connected observations about these three procedures for determining or verifying the content of senses of self-government. this will, such that one can in principle assess the extent to which public action fulfills or deviates from it. Self-Government and Democracy Self-government in this sense requires that the political community itself be the subject of political action at The concept of self-government has an elective affinity least in the following minimal way: Both governors and with the idea of democratic rule. But if democracy governed must recognize accepted "will-revealing" requires simply that all citizens in a given state (have a procedures and view them as communicating instruc- right to) participate in the political process, then tions, which governing agencies are then expected to strictly speaking it is not true (1) that any of these three execute as a matter of course. If this condition were senses of self-government requires democracy or (2) satisfied, we could meaningfully speak of a society or that democracy entails self-government in any of the organization being the-author of its own fate, at least three senses. Hobbes's political theory illustrates the insofar as the actions it takes successfully fulfill a will first point: Citizens can will their collective security and that has been appropriately certified as its own. It is the self-preservation by authorizing highly undemocratic possibility of ever satisfying this condition that social forms of government; even a monarchical Leviathan choice theory calls into question. instantiates Hobbesian self-government. A third and still more stringent sense in which public The second point is illustrated by those empirical action might have an internal source results in a highly theories of democracy inspired by the work of Joseph moralized (and I believe marginal) conception of self- Schumpeter (1942) and his antecedents (Michels government. In this account, society is self-governing to 1916). These modern theorists of democracy regarded the extent that its government acts only in the light of the notion of self-government as obsolete and presci- a plausible conception of citizens' common good. The entific, in part because in their view it tends to obscure idea is that officials take the whole community's welfare the systematically oligarchical reality of political orga- (the "public interest") to be the main object of public nizations. Despite this, they believed that a theoreti- action, and they appeal to a plausible account of this cally useful theory of democracy could be retrieved common good as the grounds for public action. In this from the conceptual swamp left behind by the philos- sense, a community is self-governing insofar as public ophers of self-government.2 The findings of social action is oriented only to those goals and ends identi- fiable as proper, rather than alien, to the whole com- 1 This sort of claim is normally brought in under the aegis of claims munity. This rules out justifications for public action about the virtues of citizenship. Such virtues require that citizens be that either appeal to goals and values that go beyond motivated by a concern for the public good rather than by their self-interest (their "particular wills," in Rousseau's terminology). the welfare of the relevant community (the progress of Self-government, in this view, requires a virtuous citizenry. For a mankind in general, fulfilling God's providential de- recent statement of the idea, see Barber 1993, 66. sign, and so on) or privilege the interests of narrower 2 For a superb illustration of this tendency to dissociate democracy 564 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions American Political Science Review Vol. 94, No. 3 choice theory tend to reinforce the intuitions of those good and its imperatives, the relevant society is not in the Schumpeterian tradition. In this view, of which fully self-governing. But this seems far too strong. It William Riker (1982) was the most vociferous propo- might make sense to say that such a society falls short nent, the social choice objection demonstrates that the of the moral ideal of a perfectly self-governing com- only viable conceptions of democracy available to us munity, but it is implausible to say that it is violating are ones that drop the idea that democratic votes conditions that are either strictly necessary or sufficient express a popular will. We are left with a much less for us to be able to describe it as genuinely self- ambitious-in Riker's terms, "liberal"-conception of governing. democracy. It demands merely that all citizens (have the right to) participate in a set of political processes that provide some kind of external check on the The Central Sense of Self-Government activities of oligarchical political elites. The ability of If the third sense seems too strong, the first seems too democratic political systems to institutionalize this sort weak by itself to count as a sufficient condition for of check does not, however, imply that when the check self-government. Yet, the first sense does appear to is invoked, "society" as a whole is concertedly acting in embody a necessary condition for regarding a commu- accordance with something called "its own will" (Riker nity or organization as authentically self-governing. 1982, 243-5). Rather, the democratic system can be Although the absence of (say) imperial domination thought of as a rather crude, although at least inclusive, does not automatically guarantee that a society is arbitration device. According to Riker and his follow- self-governing (there have been plenty of nonimperial ers, social choice theory provides a cast-iron justifica- autocracies), it is clear enough that any form of impe- tion for severing the connection between democracy rial domination is strictly incompatible with self-gov- and the problematic idea of self-government. This ernment. By contrast, the second sense looks very conclusion is apt to seem shocking: If authentic democ- much like a sufficient condition for self-government: If racy requires anything, it is surely that public action we can show that a society is acting upon a will that it track the popular will. But it is just this conception of knows to be its own, it is unlikely that we need to democratic self-government that social choice theory establish anything more in order to convince ourselves calls into question. In order to retrieve that conception, that this society is self-governing. There might be a we must defuse the social choice objection and show further question about whether this society is good, but how there can in principle be such a thing as the we have already seen that a self-governing society need popular will. not be good: The mere fact that a certain society is pursuing policies inconsistent with the common good or social justice, or that are in some other way ethically Self-Government and the Common Good problematic, does not automatically establish that it is Self-government in the third and most demanding not self-governing. As long as public policy fulfills the sense shades into, and becomes difficult to distinguish social will (for good or ill), that seems sufficient to show from, some more general account of political justice that this society is self-governing. and the social good. For this reason, we should regard this third sense as lying at the very periphery, rather than near the core, of the concept of self-government. The Independence of the Second and Third This is supported by the4intuitively plausible reflection Senses that a self-governing society need not be an especially These observations suggest not only that the second good society.3 The first two senses of self-government sense of self-government is the central one but also preserve that intuition, but the third sense overturns it. that it is largely independent of the third. This is a A society could be governed only by its own members pivotal point in my argument. One of the main claims (first sense), and it could know and pursue its own will I advance is that many of the difficulties involved in the (second sense), but from neither supposition does it notion of self-government can be removed by main- automatically follow that this society acts in accordance taining a clear distinction between the second and third with its common good. The social will might be mis- senses. That is, we should separate the question of how guided or even evil, and those citizens who wield power society can determine its own will from that of how we might be entirely corrupt. The third sense, however, might discern the public interest or society's common precludes these possibilities, because when policy runs good. Both classical and contemporary discussions of counter to the correct appreciation of the common self-government have systematically confused these two issues. Two characteristic forms of this confusion have particularly plagued these discussions. and self-government in modern empirical research, see Duverger 1954, 424: "No people has ever been known to govern itself and none ever will. All government is oligarchic: it necessarily implies the Confusion 1. Sometimes it is assumed that there is domination of the many by the few.... All government implies ultimately no way to specify the common good apart discipline ... by definition constraint is external to the constrained. A from somehow amalgamating citizens' judgments on people does not constrain itself, it is constrained. It does not govern some occasion about what the common good requires. itself, it is governed." 3 Note, however, that this is perfectly compatible with the claim that The effect of this move is to collapse the third sense of any authentically good society would have to be to a high degree self-government into the second: In this view, the self-governing. notion of the common good involved in the third sense 565 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions The Possibility of Self-Government September 2000 is identified with whatever society contingently nomi- my argument fails, it still does not follow that social nates as its common good through the kind of "will- choice theory puts in doubt the possibility of resolving revealing procedures" envisaged in the second sense. the second issue; determining the common good is a Social choice theorists are particularlyprone to making different problem, and there is little reason to believe this move, mainly because they subscribe to an atten- the apparatus of social choice theory is equipped to uated version of utilitarianism that views the public address it. Note, finally, that one can admit all this interest or common good as an aggregative function of without in the least casting doubt on the proviso that, the contingently expressed preferences of a society's properly understood, the common good is a function of constituent individuals. Therefore, they are tempted to the welfare of the individuals who make up the political claim that answers to questions about society's com- community. This unobjectionable proviso must be dis- mon good or the public interest stand or fall on the tinguished from the very different and more dubious results of voting procedures that aggregate citizens' claim that the common good is whatever citizens various views on the subject. Consider in this regard contingently judge it to be on some occasion. Riker's (1982, 137) claim that social choice theory "even leads us to suspect that no such thing as the Confusion 2. A second way in which the second and third senses of self-government are often conflated 'public interest' exists, aside from the subjective (and moves in just the opposite direction, by effectively hence dubious) claims of self-proclaimed saviors" (see collapsing the second sense into the third. This occurs also Plott 1976, 526-7). Riker seems to think this when one succumbs to the temptation to stipulate that follows from the fact that the impossibility theorems some putatively valid account of the common good show that outcomes of votes will systematically be (and its correct application to contingently arising ambiguous or meaningless. social and political problems) embodies the "real will" But this inference is surely invalid. We would not of the people. On this sort of view, in order to disclose normally conclude from the fact that individuals make the people's will on some occasion, we need only grasp contradictory or ambiguous claims about their own the correct account of their common good and deter- interests that there is no longer any basis on which to mine what it recommends under the relevant circum- make credible assessments of their best interests. Still stances. This proposal derives such plausibility as it has less does it show there is "no such thing" as an from the Socratic intuition that agents, to the extent individual's best interests. Perhaps the individual in they are rational, will their own good. When individuals question is simply confused or ill informed. The same or groups will what is counter to their best interests, point applies with equal force in the case of claims they must have failed to understand or grasp what they about the best interests of a collectivity (i.e., the "really"will. This assumption undoubtedly underlies a common good). The fact that citizens' amalgamated good deal of the mistrust of popular rule voiced over judgments about their common good are (even system- the centuries. If we suppose that the average citizen is atically) ambiguous or incoherent is not a reason to not competent to determine the common good, then suppose that no such thing as the common good or the "real will" of the people may be better served by public interest exists or can be determined. The salient taking power out of their hands and giving it to "the question in this context is: What is the common good, experts." and what does it require under circumstances A, B, C? Some political theorists have tried to defend popular But a claim about the ability or inability of a particular rule even within the terms of the Socratic assumption. agent or agency to enunciate a clear, unambiguous Rousseau (according to some influential interpreta- answer to this question does not amount to any sort of tions) is responsible for the most historically significant answer to that question itself. We would at the very version of this argument, and for that reason I refer to least need some additional argument to show that this it as the General Will argument. It confuses the social particular agency's opinion on the subject carries spe- will and the common good in exactly the way I am cial authority, so that its failure (say) to arrive at a trying to identify. In order to diagnose this second clear, coherent answer definitively establishes that the confusion and highlight its shortcomings, I will briefly question itself just is finally unanswerable. After all, the discuss some salient features of the General Will fact that some agent manages to enunciate a coherent, argument. unambiguous view of its own best interest does not in any way imply that that view is correct or even plausi- ble. THE GENERAL WILL ARGUMENT To my mind, these reflections highlight the funda- mental gap between two issues raised, respectively, by The common good requires that we give fair consider- the second and third senses of self-government. The ation to the best interests of all citizens, and the second sense raises the question of how we can estab- General Will argument suggests that the citizenry in lish that a particularjudgment about the common good full assembly is the appropriate "expert" on the com- can genuinely be regarded as society's own will. The mon good. To the extent that we entrust power to third sense raises the different issue of how we can groups smaller than the whole citizenry, we increase determine a correct or plausible account of society's the likelihood that government will unfairly privilege common good. Perhaps social choice theory puts in the interests of some partisan sector of the community. doubt the possibilityof ever resolvingthe first issue, This argument need not imply that the whole citizenry althoughthat is the claim I seek to refute. But even if is infallible, merely that a virtuous citizenry is more 566 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions American Political Science Review Vol. 94, No. 3 likely to identify correctly the general will than are any stances. Even if we concede that epistemic populist of the less inclusive alternative candidate groups. The voting procedures do indeed provide reliable evidence latter are more likely to be unduly influenced by their about the common good and its imperatives, is it narrow sectarian interests and so are likely to offer obvious that we establish the content of the popular correspondingly less reliable guidance as to the content will by applying them? Suppose a society conducts a of the general will or common good. vote along the lines recommended by the epistemic Today, political theorists appeal to versions of this populist. Citizens now have good reasons to believe argument in order to respond to the difficulties raised that the victorious policy conforms to, or is dictated by, by social choice theory. But if social choice theory calls the common good. But are they entitled to infer from into question the idea that we can identify a will that is this fact alone that the winning policy constitutes their society's own, it is not clear that the General Will own will? Only if they already know that the epistemic argument is capable of supplying an adequate re- populists' voting procedures are an appropriate barom- sponse. The reason is that this argument implausibly eter of their collective will. Yet, the mere fact that reduces the concept of the social will to that of the those procedures are built around a plausible view of common good. Consider a contemporary variant of the the common good does not establish that this is so: General Will argument that has been defended by a Perhaps there is a conflict between what the common number of theorists who sometimes describe them- good actually requires of a society and what that society selves as "epistemic populists" (Cohen 1986; Coleman might- on further reflection- choose to will for itself. and Ferejohn 1986; Grofman and Feld 1988; Radcliff Suppose, for example, that putting a man on the 1992). moon is a complete waste of valuable social resources Epistemic populists respond to social choice theory from the point of view of the common good: The by defining the general will as an "independent stan- money would be far better spent improving the condi- dard of correct decisions-that is, an account of justice tion of disadvantaged citizens. This fact alone hardly or of the common good that is independent of current establishes that a society has collectively decided to consensus and the outcomes of votes" (Cohen 1986, stop funding moonshots. After all, members of society 34). This is exactly the reduction of the social will to the might choose to sacrifice their common good to the common good to which I am objecting. Epistemic goal of putting a man on the moon. Perhaps they see populists then require that voters express their sincere themselves as responsible not just for their own com- beliefs about the content of the general will in specific mon good but for the progress of mankind in general. instances. Under this requirement, voters do not In such cases of a conflict between the common good merely assert whatever unreflective preferences hap- and the social will, the content of that will cannot pen to pop into their heads, so epistemic populists simply be read off the best available account of society's believe that "public deliberation is guided by the common good, even if we determine the latter through principles that define ... [the general] will" (p. 34). some sort of voting procedure. This transforms the context and purpose of voting. At best, then, epistemic populism justifies a kind of Citizens are now expected to vote on the basis of good social decision procedure that can be expected to (1) faith attempts to discern the general will. Clearly, this generate meaningful results and (2) provide reliable obviates social choice theory's search for an all-purpose guidance about what the common good requires on aggregation function that can reveal a social preference specific occasions. But these do not constitute answers in the absence of any restrictions on the preferences to the question which I am considering and which individuals are permitted to express. Moreover, under social choice theory claims is unanswerable. That ques- these conditions, the outcomes of votes can constitute tion is not: What is the common good, and what does compelling evidence about the content of the general it require? The question is: What is it that we, as a will or common good, or so the argument goes.4 society, actually want to do? To be sure, this latter As will become clear, my own argument shares many question requires that we identify some clear, unam- features of this approach. But as it stands, this epis- biguous result, but it does not require that the result be temic populist proposal does not satisfactorily explain mandated by the common good. Instead, it requires why the outcomes of these sorts of votes should that in arriving at a certain result, society must "know" automatically count as the social will. Epistemic popu- that it is being true to itself or must authentically lists simply recruit their specially configured voting express its own view (right or wrong) of what ought to procedures to the project of determining what some happen. If we are to respond to the social choice antecedently specified and independently valid account objection, then we have to show how this can be so, and of the common good requires under-specific circum- how citizens can come to know that this is so. To the extent that epistemic populists and other 4Arrow ([1951] 1963, 85) had already anticipated this proposal in his proponents of the General Will argument offer any original monograph, suggesting that it "has much in common with answer to this question, it comes in the form of the the statistical problem of pooling the opinions of a group of experts Socratic stipulation that the social will and the common to arrive at a best judgment; here individuals are considered experts good must ultimately be one and the same. Defining at detecting the moral imperative." If one assumes the existence of a the general will as the common good properly under- "right answer," and individuals are likely to arrive at the correct decision on their own more than 50% of the time, then the stood effects just this equation. But as an answer to our probability that a majority will be right increases as the size of the question about the possibilityof self-government,this relevant group increases. See Barry 1990, 292-3. stipulationhas at least two decisive drawbacks.First, 567 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions The Possibility of Self-Government September 2000 notions of the common good or the public interest are Ideally, of course, we might hope that members of a not exactly unproblematic. If we are seeking to dissolve fully self-governing society both organize themselves so the ambiguities and problems of the notion of a social as to secure their common good and know that in doing will, then an appeal for aid to the equally vexed notion so they are enacting a will that is authentically theirs. of a "correct account of the common good" is like When both conditions are met, we can say that society throwing out an anchor to save a drowning swimmer. is freely and willingly pursuing its true common good. There is every reason to demand that unravelling the But the means by which we determine that social mystery of the social will not await a solution to the practices and policies conform to the common good riddle of the common good. and that they fulfill society's will are, and should be, Second, stipulating an equivalence between the so- independent. If we conflate them in the manner of the cial will and the common good radically distorts the General Will argument, we risk losing our grip on the relation between the idea of self-government and the form of social freedom that lends moral weight to the notion of freedom. Intuitively, the concept of self- ideal of self-government. government commands our moral attention in large The lesson, then, is that the second and third senses part because it seems to be a condition for a certain of self-government should be kept apart. It is a mistake kind of social freedom. A society that is not self- to suppose that the absence of a social will implies that governing is in some way subjugated, enslaved, subor- there is no way to determine the content of the dinated to alien forces, and so on, and in these senses common good (the first confusion), but it is also wrong is unfree. By contrast, a society that acts on a will that to think that appealing directly to some conception of it knows to be its own is (to that extent) free in the the common good and defining it as society's "real will" sense relevant to the concept of self-government. But (the second confusion) allows us to supply a satisfac- such a society is surely no less free in this sense if its will tory answer to the social choice skeptic. Such a re- is misguided or runs counter to its common good. To sponse requires a direct and free-standing account of establish that a society is free in this sense, we need how a political community might identify a will as its only establish that social action is loyal to society's own own. Below I attempt to show that the results of social view (right or wrong) of what ought to be done. If we choice theory at least do not shortcircuit this project. follow the General Will argument, however, and stip- ulate that society's real will or general will is whatever THE ANALOGY BETWEEN INDIVIDUALAND the common good properly requires, then we com- SOCIAL WILL pletely invert these perfectly sound intuitions about freedom. If freedom is a function of acting in accor- The concept of collective self-government suggests an dance with the social will, and the social will is really analogy between individual agents and collectivities. the common good, then whenever a society opts for a According to this analogy, a collectivity can be an actor course of action inconsistent with the common good it in its own right, capable of formulating and determin- is acting unfreely, because it is not (really) acting in ing its own will, and thereby responsible for its own accordance with its (real) will. On this view, assuming choices and self-determination. Social choice theorists for the sake of argument that he understood the typically claim that their results "demonstrate" that any common good plausibly, Plato's Republic becomes the such analogy "involves an illegitimate transfer of the ideal of a free, self-governing political community, properties of an individual to the properties of a whereas the Athenian democracy that put Socrates to collection of individuals" (Plott 1976, 525). The social death becomes its antithesis. This is surely absurd, choice theorems largely derive their shock value from quite apart from Plato's idiosyncratic view of the their claim to expose such analogies as logically con- common good. We may agree with Plato that Athens fused. made a catastrophic mistake when it condemned This claim is mistaken, however. Understanding why Socrates, but it seems bizarre to claim that for this is the key to my strategy in this paper. To see this, reason Athens was acting unfreely or was somehow however, it is first important to grasp the overall subjugated to "alien forces" when it chose to convict structure of the argument that persuades social choice Socrates. But by stipulating an equivalence between theorists of this claim. This argument has four steps: the social will and the common good, proponents of the (1) We can formulate a number of conditions that any General Will argument are forced into this deeply social decision procedure capable of generating clear, counterintuitive account of social freedom.5 The price unambiguous results would have to satisfy in order for of reducing the social will to the common good is the us to certify its ability to enunciate a "will" that is sacrifice of the concept of freedom implicit in our authentically society's own. (2) It is logically impossible intuitive concept of self-government. to satisfy all these conditions in a way that permits meaningful, coherent results to emerge. (3) Therefore, 5 Just this line of argument has led a host of critics to demur at there is no possible decision procedure for identifying Rousseau's notorious claim about forcing citizens to be free. They a social will since there is no way to satisfy the think that Rousseau gave the game away when he claimed that the authenticity criteria and at the same time produce clear "General Will is never wrong" and that popular participation in results. To say that no such procedure is logically expressing it causes "the public to learn to know what it wants." Social choice theory tends to reinforce the suspicion that Rousseau's possible is tantamount to showing that there is no such democratic and voluntarist terminology disguises an essentially Pla- thing as the popular will. (4) Because we routinely tonic dogmatism. assume that when individualsdecide on a course of 568 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions American Political Science Review Vol. 94, No. 3 action or preference, those decisions do or at least may preference of society as between x and y depends reflect their own will, it follows that collectivities are to only on the preferences of individuals as between x this extent fundamentally unlike individuals. They are andy. not agents in the way that individuals are because it is meaningless to think of them as having their own will. In what follows, I will refer to this set of conditions on This argument is decisive only if we are sure that the both decision and aggregation as the TUPID condi- social choice theorist's conditions are reasonable con- tions. straints on purportedly will-revealing procedures. If The social choice theorems show that we cannot they are unreasonable or nonessential, then the argu- satisfy all the TUPID conditions at once: We must ment collapses, because the violation of those condi- choose between T and one or other of the remaining tions will no longer automatically imply that a coherent four. This claim is the second of the four steps in the result does not authentically represent the social will. I argument I set out above. But can this second step suggest that the conditions imposed by social choice establish the later claims in the sequence? Can it show theory are not necessarily essential or even reasonable that there is no such thing as the popular will and that in the context of attempts to identify a social will. If so, the analogy between individual and collective agency then will-revealing procedures may legitimately violate must be spurious? these conditions, and the idea that society can identify To see that the answer is "no," consider the follow- a will of its own is not threatened by the results of social ing two claims: (Al) Societies are potentially like choice theory. individuals because they can have their own will, and We need to understand, therefore, the conditions (A2) societies are potentially like individuals because imposed by social choice theorists and the implicit both are able to determine the content of their will in expectations they make of reasonable will-revealing analogous ways. procedures. I will focus on five conditions.6 The first Note that Al does not imply A2. It is possible that embodies the expectation that these procedures gener- both individuals and collectivities can have a will of ate transitive results: their own, but that there is no resemblance whatsoever between the respective procedures under which these T (Transitivity):For any alternatives x, y, and z, if x is two kinds of agencies determine the content of their preferred to y and y is preferred to z, then x is will. But A2 assumes Al: If we know that societies are preferred to z. able to determine their will in the same way that For our purposes, T is the only relevant expectation of individuals do, then we presuppose that societies can decisions. Preference orderings that violate T are said have a will. to be "cyclical."The problem with cyclical orderings is The difficulty facing the social choice argument is that they fail to communicate a clear, unambiguous this. Social choice theorists claim to falsify Al (there is preference. The relevance of this condition to the no such thing as the popular will), but to do so they present argument is clear: If the public is to be in a deploy an argument compatible with the truth of A2. position to communicate a will to governments ex- Their argument cannot decisively falsify Al, since it is pected to execute it, then acceptable will-revealing presupposed by A2. procedures must satisfy T. Otherwise, the procedures How is the social choice argument compatible with will enunciate cyclical judgments and thus fail to iden- the truth of A2? The answer is that social choice tify any clear instructions or decisions to the institu- theorists (rightly) acknowledge the possibility that the tions authorized to execute the social will. TUPID conditions might also apply to the case of The remaining four conditions embody expectations individual decision making (Hurley 1989, 226f; May of the amalgamation procedure that is supposed to 1954; Roemer 1996, 25-6; Steedman and Krause process the inputs of each citizen into a final judgment 1986). The standard way of making the case points out representative of society's own will. that when individuals are called upon to express pref- erence rankings over defined alternatives (say, candi- P (Weak Pareto Principle): For any alternatives x and y, dates for a professorship), they very rarely have the if all individuals prefer toy, then society prefers x to luxury of being swayed by a single dimension. Rather, Y. they must rank candidates relative to a range of D (Nondictatorship):There is no individual such that, emergent preferences along different dimensions: for every set of individual orderings, for all alterna- teaching ability, reputation, quality of written work, tives x and y, if that individual prefers x to y, then affability,compatibility of research interests with mem- society prefers x and y. bers of the department, and so on. Moreover, these U (UnrestrictedDomain): For any set of alternatives dimensions will very often generate widely divergent and any set of individuals, the domain of the social rankings. In principle, an individual seeking to arrive at welfare function includes all orderings of alterna- a final transitive ranking must find some way to process tives by individuals. these emergent lower dimensional rankings into an I (Independence of Irrelevant Alternatives): For any overall (all things considered) ranking, much as a alternatives x and y, if the preferences of all individ- society must find some way to amalgamate individuals' uals as between x and y remain the same, then the preferences into a social preference. If the TUPID conditionsdefine reasonablewill-revealingprocedures 6 These conditions are taken from Hurley 1989, 229. in both these cases, then A2 has been conceded. 569 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions The Possibility of Self-Government September 2000 It is relatively easy to stipulate equivalent TUPID ble of establishing that the ability to will belongs to conditions for the case of intrapersonal deliberation. I individuals but not to collectivities. will follow Hurley's (1989, 231) extremely helpful stip- Alternatively, the social choice theorist could simply ulation of TUPID analogues for individuals seeking a drop the idea that the TUPID conditions apply in the "coherence function" capable of processing several individual case and say that they are appropriate only criterial rankings of alternatives into an "all.things- in the collective case. This at least would allow one to considered" ranking. Apart from T itself, which re- claim that individuals can have wills but collectivities mains unchanged, Hurley amends the other four con- cannot. There are two difficulties, however. First, this ditions as follows. cannot serve as an argument to the proposition that it P* (Weak Pareto Principle): For all alternatives x and y, is futile to model collective decision on the analogy of if all criteria rank x above y, then x ranks above y all a single individual. The argument rests on the assump- things considered. tion that this analogy is misguided. This simply begs the D* (Nondictatorship): A coherence function must not question against the claim that A2 is false. Even if we give so much weight to one criterion that it out- certify independently that A2 is false, this still will not weighs any criterion that conflicts with it under any falsify Al, since Al does not require A2. This helps us circumstances. That is, it must not be the case that understand why the social choice framework is in no there is one criterion such that any one alternative's position to test the claim its practitioners often see it as superiority over any other according to this criterion proving: that the analogy between collective and indi- would always result in its superiority all things con- vidual agency is flawed because collectivities cannot be sidered, regardless of how other criteria might rank said to share the individual property of having a will. those alternatives. Plott's claim that Arrow's theorem "demonstrates" the U* (UnrestrictedDomain): For any set of alternatives fallacy of treating collectivities like individuals cannot and any set of criteria, the domain of the coherence be sustained. function includes ordering of all alternatives by Second, adopting this line of argument results in a criteria. crucial shift in the burden of proof. If one concedes J* (Independence of Irrelevant Alternatives): For all that the TUPID conditions need not apply to individual alternatives x and y, if the rankings of x and y by all will-revealing procedures, then the case for their sa- criteria remain the same, then the ranking of x and y lience in the collective case is weakened. If it is all things considered remains the same; that is, a permissible for individuals to override one or other of deliberator's ranking of a pair of alternatives all the TUPID conditions when they determine the con- things considered depends only on the ranking of tent of their will, then why should it be impermissible those alternatives by all the relevant criteria, not on for a collectivity to do so? We have no choice but to the rankings of other alternatives. take the individual will as the basic primitive unit of If it is plausible to claim that these conditions apply to comparison-there are no other places to look for individuals trying to process several emergent prefer- accounts of how the will operates. But if it turns out ences over given alternatives, then it would seem to that in determining and pursuing their own wills indi- follow that transitive decisions can be reached only by viduals routinely and legitimately violate the TUPID* violating one or other of the TUPID* conditions. conditions, then the case for expecting collectivities to Roemer (1996, 26) suggests that this conclusion is an satisfy them might be undermined. I will now explore "unexpected bonus" of Arrow's theorem. But it seems this possibility, although the findings are neither con- to me that it gives the game away: If these conditions clusive nor complete.8 apply in the individual case, then this undermines the ability of social choice theory to adjudicate the ques- tion of whether there is such a thing as the popular will. conception of the will: Many philosophers and theorists have de- Once we allow this possibility, we concede the possible fended views of the will that seem consistent with this one. Hobbes, for example, can be interpreted as suggesting that individuals are validity of A2, and we can no longer claim to be only able to arrive at decisions at all by violating the nondictatorship establishing that Al is false. condition. In Hobbes's view, deliberation is nothing other than the To see that this is right, suppose it is plausible to process by which a sequence of values (embodied by a series of depict the individual will along these lines. The viola- emergent appetites and aversions) compete for dominance within the mind of an agent. An individual finally wills an action when one or tion of one or other of the TUPID conditions then other of these competing values wins out. In this way, the victorious becomes a generic condition of all aggregation that appetite can be said to dictate the decision. That "dictated" decision results in transitive outcomes, whether individual or represents the individual's will, just as the arbitrary decisions of the collective. If we assume that this form of aggregation sovereign represent the will of the collectivity after the social captures essential features of any will-revealing proce- contract has instituted a Leviathan. So it is perfectly possible to defend a notion of the will that violates one or other of Arrow's dure (whether individual or social), then this would conditions. On this point generally, see Mayo 1956. imply either (1) that neither individuals nor collectivi- 8 I am not the first to suggest this general strategy of analysis. Hurley ties have a will or (2) that in both instances the (1989, 234-41) explores a similar line of argument. But I find her violation of the TUPID conditions is a precondition for argument unsatisfactory. Hurley accepts P* and D* as appropriate conditions on individual deliberation but rejects U* and I*. To my identifying such a will.7 Clearly, neither claim is capa- mind, U* is the one condition that does make sense in the individual case (see Steedman and Krause 1986, 208), so my own arguments will 7 The point here is precisely not that this results in an indefensible target P*, I*, and D*. 570 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions American Political Science Review Vol. 94, No. 3 RECONSTRUCTING THE INDIVIDUALWILL the intensity of preferences, or of behavior. Our intu- itive concept of the will allows, for example, that it is Thus far I have argued that in order to be self- often weak relative to intense desires, even that it can governing, societies and groups need some procedure be "betrayed" by intense temptations and actions for certifying that some policy fulfills or conforms to its prompted by them. This suggests that even if we had a own view of what should happen. I maintained,.that viable measure of intensity (itself doubtful), that vari- such a procedure must be independent of procedures able would not predict, or stand in any determinate designed to determine the content of the common relation to, the content of one's judgments about one's good. Correctly discerning the common good is one will. thing, and fidelity to society's own judgment is another. Second, we must remember that we are investigating The latter matters more in the context of self-govern- the individual will not for its own sake but for the ment. To be self-governing, a group's will must authen- purposes of illuminating possible collective will-reveal- tically be its own. I will call this the requirement of ing procedures. Taking into consideration the intensity reflexive authenticity. of citizens' preferences will only be useful in the latter Social choice theorists believe that Arrow's theorem context if we have some interpersonally applicable undermines the possibility of such a will because they standard of intensity. But there is overwhelming agree- take the TUPID conditions to define a minimal, but ment among economists and social choice theorists essential, standard of reflexive authenticity for collec- that interpersonal cardinal comparisons of utility are tive will-revealing procedures. This is clear from the meaningless (indeed, many think the same is true of justification social choice theorists normally offer for intrapersonal comparisons). This consensus might be the critical P, I, and D conditions. They argue that misguided, but relaxing I's requirement of ordinality when these conditions are violated, social choice is not will help us reconstruct an account of the collective will determined by voters' own judgments but is instead an only if we can defeat the argument that interpersonal artifact of the aggregation method (Riker 1982, 117-9). comparisons of intensity are meaningless. This route This is a technical way of expressing the demand that seems unpromising, and it is certainly unlikely to procedures for determining social choice should not persuade a social choice theorist. preempt citizens' own choices. P, D, and I comprise In what follows, I sketch a more promising line of social choice theory's test of reflexive authenticity. argument against P, D, and I. It hinges on the idea that But do these conditions represent the only reason- practical deliberation about how to be true to oneself able standard of reflexive authenticity in the context of has a characteristic depth structure that is incompatible self-government? Perhaps there exist other reasonable with the TUPID model of the will. I follow a number of interpretations of reflexive authenticity that are incon- philosophers in using the hierarchical metaphor of sistent with P, D, and I. I have suggested that we might orders of desires, preferences, volitions, and reasons to seek such alternative conceptions by questioning the explicate this important dimension of will formation. claim that analogous conditions (P*, D*, and I*) are To elaborate the point, I make extensive reference to essential constraints on individual will-revealing proce- Frankfurt's (1988, 11-26) classic, "Freedom of the Will dures, although many theorists have considered this and the Concept of a Person." I use Frankfurt largely claim (Roemer 1996, 25-6; Steedman and Krause for reasons of convenience and ease of exposition; I do 1986, 207-10). But if it is wrong, understanding why not necessarily endorse his views on the definition of might allow us to reconstruct a notion of a popular will persons or the subject of free will. that is immunized against the threat posed by Arrow's Consider the case of the tempted spouse, who finds theorem. We might learn why violating one or other of herself contemplating infidelity because of three dis- these conditions is consistent with an appropriate test tinct emergent desires. First, her husband no longer of reflexive authenticity, thus undermining the basis for satisfies her appetite for sexual experimentation and the social choice objection to the idea of a social or variety; sex within her marriage has become routine popular will. and has lost its earlier sparkle. Moreover, she is How might one call P*, D*, and I* into question? It strongly attracted to a colleague at work, who has given is tempting to focus on I*. As is well known, I* strong indications of a reciprocal interest. Second, her prohibits (among other things) cardinal rankings of husband cheated on her three years ago, and she has alternatives and thus rules out claims about the relative harbored a latent desire for revenge ever since. Third, intensity of several emergent preferences. But surely the colleague is the tempted spouse's boss, and she has individuals should be permitted to take such informa- reason to believe that a sexual relationship with him tion into account when they deliberate about their will advance her career. Although the tempted spouse choices? This is certainly right, but in the present experienced emergent desires in the past (e.g., over- context the objection is insufficient. Relaxing the re- whelming sexual interest in her husband) that would quirement of ordinality is unlikely to help us recon- rank fidelity over infidelity, no such desires are present struct a viable account of will formation for two now. Her emergent desires unanimously recommend reasons. infidelity, but the tempted spouse still might decide, in First, intensity of preference may predict behavior, the face even of this unanimous recommendation, to but the content of one's will is another matter. Prima refrain. The feature in virtue of which this is possible is facie, there is no reason to expect that the content of the capacityfor formingwhat Frankfurtcalls "second one's will ought to be considereda simple functionof ordervolitions." 571 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions The Possibility of Self-Government September 2000 Frankfurt (1988, 16) claims that we do not merely fulfilling any personal ideals of the sort sketched above. experience emergent first-order desires of the sort that Second, as illustrated by the case of the tempted spouse dominate the mind of the tempted spouse. We also or Frankfurt's (1988, 16-9) example of an unwilling have the ability to form second-order desires, that is, addict, an agent may be capable of forming a will, but desires to have or not to have certain desires. A it does not follow that the will is effective in generating second-order desire becomes a second-order volition action. when a person desires that a certain (first-order) desire The cornerstone of a Frankfurtian will-revealing move, or not move, them to act. A second order procedure, then, is a higher order review of currently volition is, therefore, a desire to be moved or not to be emergent first-order preferences. Agents seeking to moved to action by a certain desire or set of desires.9 arrive at an all-things-considered judgment as to the By virtue of this capacity, the tempted spouse may content of their will, given a variety of rankings based choose not to identify her will with her emergent on emergent first-order desires, must ask a version of first-order desires. It is important to emphasize that the following question: If I allow my will to be deter- this does not imply she will successfully battle these mined by emergent preferences of this particular char- temptations; she may give in to them. The point, acter, will I be able to think of myself as living up to however, is that even if her unanimous first-order some personal ideal with which I identify? If the desires cause an act of infidelity, this action does not answer is "no," the agent may disregard the recommen- necessarily receive the blessing of her will. It is more dation. There are two things to note about this test. accurate to say that temptation caused her to act First, it can legitimately claim to be one of reflexive against her will, which will be reflected in the self- authenticity: It reveals whether agents are being true to recrimination, shame, and guilt likely to follow her themselves by following one preference or another. capitulation. Second, it does not evaluate emergent preferences by What sort of reasons might agents have for counter- directly assessing their recommended rankings. In- manding or even disowning their first-order desires by stead, it assesses the character of the values generating refusing to identify them with their will? Very roughly, these rankings and their relation to some operative the most obvious can be collected under the general personal ideal. Agents can disregard an emergent category of the personal ideal. A personal ideal here preference on this basis without thinking that they refers to any sustained intention, aspiration, hope, or should ultimately repudiate the ranking it recom- plan to become a certain kind of person or to live up to mends. a certain exemplar. The tempted spouse may refuse to This reconstruction of a Frankfurtian will-revealing identify her will with her emergent desires because she procedure raises basic questions about the credentials is antecedently committed to "being a good spouse" or of P*, 1*, and D* as necessary conditions for such a "being a good Christian." If those ideals demand that procedure. These conditions conflict with the Frank- agents not be moved by lust, vengeance, greed, and so furt model because they interact in a way that forces on, then people who identify with those ideals may try one to treat emergent preferences as social choice to renounce the disqualified desires. One reason to theorists propose to treat voters: Each is entitled to affiliate such ideals with one's "will" is to encourage equal voice and weight in the amalgamation process. vigilance toward motivations that might deflect or But the Frankfurt model reveals difficulties with this undermine one's resolve to be a certain kind of person. assimilation. Consider again the predicament of the By refusing to identify her first-order desires with her tempted spouse. If she believes that P* and J* are will, the tempted spouse is trying to do just this; and her appropriate expectations to place on the process by will to realize whichever personal ideal is operative in which she determines the content of her will, she her case is instantiated in the formation and pursuit of presumably will have to identify with the course of certain second-order volitions about emergent desires. infidelity, since this is what her emergent desires unan- * Frankfurt's account of second-order volitions allows imously recommend. one to distinguish between two different ways in which Moreover, strictly speaking, considering second-or- agents may fail to generate or be in a position to pursue der assessments of the desirability of living in accord their own will. First, they may fail to form second-order with currently emergent desires is ruled out by J*. The volitions at all; Frankfurt calls them "wantons." It is reason is that when the tempted spouse invokes her vital to note that wantons may have second-order second-order volition as a reason for disregarding her desires, and they may have the capacity to deliberate. first-order rankings, she is choosing between fidelity The distinctive feature of a wanton is "mindless indif- and infidelity precisely on the basis of a preference ference to the enterprise of evaluating his own desires between alternatives that J* deems "irrelevant." J* and motives"; the wanton may experience the second- requires that agents' final orderings be independent of order desire to suppress his first-order desire for sex or information concerning alternatives not in the set of drugs, but he is indifferent as to which should win out. alternatives under direct consideration. In the Frank- It follows that a wanton could not be self-governing in furtian procedure, the tempted spouse reaches a final Frankfurt's sense because he could not care about judgment about fidelity because of the way she ranks two quite different alternatives: a life led in accordance 9 It seems possible in principle that second-order volitions may be with greed, lust, and vengeance versus a life organized capable of producing, or causing one to have, particular first-order around different motivations. But these two possible desires. See, for example, Raz 1986, 32-3. modes of life are not themselves members of the set of 572 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions American Political Science Review Vol. 94, No. 3 alternatives under direct consideration, a set that in- in the tempted spouse example, because it is difficult to cludes only fidelity and infidelity on this specific occa- imagine "spouseworthy" motivations for infidelity, but sion. not in other cases. John may choose not to date Sarah It is tempting to object that the tempted spouse's because he thinks his desires for doing so are suspect second-order volition to disregard her first-order de- (lust, a desire to get back at her current husband, sires can itself simply be counted as an emergent wanting to access Sarah's millions), but he remains preference for fidelity over infidelity, on a par with the open to the possibility that other desires for going out other first-order emergent preferences. On this suppo- with Sarah (e.g., true love), if they emerge, could be sition, it would no longer be true that the tempted allowed to determine his will. spouse's emergent desires unanimously recommend This does not mean that the tempted spouse's even- infidelity over fidelity, since there is now an additional tual will cannot be seen as expressing a certain kind of preference for fidelity. This objection misfires in two desire to refrain from infidelity. That description re- ways, however. First, stipulating that the second-order mains available, subject to an important qualification: volition counts as an emergent preference for fidelity The tempted spouse's eventual desire to refrain from may allow the Frankfurtianprocedure to satisfy P*, but infidelity is the result of a deliberative assessment of only by raising a new question about D*. D* requires her emergent desires on this occasion, not one of the that there be no criterion or emergent preference initially emergent desires for or against fidelity. Ap- dimension such that it determines the final ranking pealing to the ideal of spouseworthiness and associated regardlessof the rankings recommended by other emer- second-order volitions is a method for arrivingat a final gent preferences (see Riker 1982, 118). But if we count judgment, and the outcome of this appeal can legiti- the second-order volition as a direct preference for mately be described as a desire to refrain from infidel- fidelity over infidelity and the tempted spouse opts for ity. But this does not mean that in the first instance the fidelity over the objections of all her other preferences, second-order desire is itself a desire for or against this requirement is violated. Given its special impor- fidelity. Rather, it is essentially a desire not to be the tance in the tempted spouse's self-image, the personal kind of person who acts on the sorts of motivations ideal of "spouseworthiness" functions in just such a implicated in the first-order rankings of fidelity and "dictatorial" way in the formation of her will. But in infidelity. combination with J* and P*, D* denies this sort of This is not to deny that the desire to live up to the special privilege to particular preferences. Instead, operative ideal is itself an emergent desire, but it only these conditions hold will formation hostage to a ranks fidelity over infidelity indirectly, by rejecting the principle of "one desire, one vote." As the tempted motives on which the initially unanimous first-order spouse illustrates, however, the process of will forma- recommendation is based on this occasion. Assessing tion is not necessarily well captured within the terms of the relation between one's currently emergent first- this principle. Deliberation about the content of one's order rankings and some privileged personal ideal is will and about how to be true to oneself often is surely a perfectly legitimate way to process one's precisely and legitimately about which preferences to initially emergent preferences into a final judgment privilege over others and about which alternatives are about one's will on a given occasion. As long as we take relevant in a particular instance. P*, D*, and J* take P*, I*, and D* seriously, however, we must regard such these issues off the deliberative agenda by deciding a procedure with suspicion, for these conditions pro- them in advance according to an arguably inappropri- hibit agents from assigning any greater weight to ate principle of equal consideration. second-order volitions than to any other ranking. Yet, This hints at the second way in which the objection flattening out the will in this fashion arguably misrep- misfires. The objection simply stipulates that we assim- resents the special character and legitimate role of ilate second-order volitions and first-order emergent privileged second-order volitions in the formation of rankings of the alternatives. It may well be possible to the will. Substantive doubts of this sort cannot be retrieve a rationale for the TUPID* conditions once we removed simply by stipulating a priori that this com- make this stipulation, but it begs the critical question of pressed model of the will is the correct one. whether we must view second-order volitions in this way. If this stipulation is independently implausible RECONSTRUCTING THE COLLECTIVEWILL (because it fails to detect relevant differences between these two kinds of emergent preference), then so much I have called into question the appropriateness of P*, the worse for the TUPID* model of the will. And the I*, and D* as conditions under which individual agents stipulation is implausible for this reason. Second-order determine the content of their will, given emergent volitions cannot simply be assimilated without qualifi- desires or criterial rankings of the alternatives under cation to other first-order rankings of the alternatives. consideration. Can the lessons of this analysis be A second-order volition not to be moved to X (where Y extrapolated to the case of collective self-government is some alternative) when prompted by first-order and, in particular, democratic self-government? At first desires A, B, and C is not adequately described as a sight, this seems unlikely, because I have claimed that preference for Y over X. There may be other first-order it is problematic to treat agents' emergent preferences desires (P, Q, R) that, if emergent, also would recom- as if they were voters expressing judgments entitled to mend X but to which the operative second-order equal consideration. But this is not so problematic in volition has, as it were, no objection. This is hard to see the collective case, in which we are dealing precisely 573 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions The Possibility of Self-Government September 2000 with voters' judgments, and to which the democratic ground that the later unanimity immediately preceded principle of equal consideration seems entirely appro- the moment of resolution and so was contingently priate. Perhaps the argument I have given simply shows decisive. But the jury need not have changed its mind. that the individual psyche is typically not organized Had it not done so, by the lights of P, D, and I, the according to democratic principles. But if a democratic earlier decision could have stood equally well as the polity is just a group of individuals whose ownsviews jury's will. about public policy must be weighted fairly, then a It is not obvious that this is the only conceivable or presumption in favor of equal consideration may seem reasonable standard of reflexive authenticity for juries. to follow very naturally, and with it a cogent rationale From another point of view the first unanimous verdict for P, D, and I. It is because these conditions seem to represents the jury's will less successfully than the express a norm of democratic reflexive authenticity that second, but this perspective will elude us as long as we social choice theory is so disturbing to conceptions of think of a jury as nothing more than a collection of democratic self-government. separate individuals free to use whatever criteria they I will now show how the Frankfurtian argument wish to determine their own view of the appropriate offered above helps us see why P, D, and I may not be verdict. If this is what a jury is, then P, D, and I will essential requirements for any reasonable standard of seem perfectly appropriate, and the failure of these democratic authenticity. The discussion of second- conditions to discriminate between the two verdicts order volitions questioned the requirement that the becomes unproblematic. For on this understanding, a individual will depend exclusively on the profile of jury is simply an empty vessel whose character is currently emergent preferences without any assess- determined entirelyby the preferences of the individual ment of their character in relation to operative ideals. jurors who happen to make it up at a given moment. Does this proviso carry over to the collective case? Since both unanimous verdicts in the example repre- Consider the case of a jury in a criminal trial, as that sent them equally well (at different times), there is no institution is currently understood in common law reason to expect that one verdict represents the jury's jurisdictions. Suppose that initially the jurors are unan- will more successfully than the other. imously in favor of conviction, but upon deliberation it Clearly, this is not the only available conception of a emerges that this unanimity is largely an artifact of jury. Another defines it as a form of association that prejudice, tendentious interpretation of the evidence, demands that its members be moved by certain consid- and disregard for the reasonable doubt standard. Un- erations rather than others. A group could completely der these (hardly exotic) conditions, the jurors might disregard the appropriate considerations and perhaps reasonably decide that to permit their joint will to be reach joint decisions, but it could not reasonably regard determined by these initially emergent preferences them as reflecting a will that is theirs qua jury. It would would fall short of the ideal of juryhood that defines be basing decisions on considerations that are, in effect, their association. In that sense, the verdict implied in unworthy of a jury. Indeed, in a sense they are failing to the initial consensus could not represent their will qua be a jury, or at least failing to behave like one. jury. But at this point the jurors need not conclude that Adopting this more restrictive conception does not in they will eventually acquit the defendant. Their rejec- any way show that the ideal of juryhood is itself worthy; tion of the original recommendation to convict was not jurors might have their own excellent reasons for based on a direct judgment of the proper verdict but on wanting to subvert conventional legal practice. But as an indirect assessment of the way it was reached. No long as that ideal is partly definitive of a jury (for good definitive conclusion about the proper verdict can be or ill), a standard of reflexive authenticity for juries can inferred at this stage. The jury simply resolves to reasonably demand that jurors take that ideal seriously. continue the discussion until it reaches a verdict on Moreover, it seems clear that the jury in the example terms compatible with a certain ideal of how juries fulfills this expectation. When they rejected the initial ought to make decisions. Suppose that eventually it verdict, jurors did so because they reflected upon their votes unanimously for acquittal after a sincere effort to initially emergent preferences and found their charac- apply the reasonable doubt test. ter is inappropriate given their self-image as a jury. Keeping this example in mind, suppose that we must Insofar as they aspired to juryhood, they had to con- determine rules and conditions under which the votes clude that their first verdict could never count as their of a jury can genuinely be described as expressing a will and must be revised, even though unanimous. jury's will. Must we accept P, D, and I as appropriate Although P, D, and I do not disallow this revision, they conditions? If so, we could not make any clear distinc- also do not require it; these conditions could allow the tion between the two votes reached by the jury in the first verdict to count as the jury's will. But a standard of example. Both the initial vote for conviction and the reflexive authenticity based on the alternative concep- later vote for acquittal are unanimous, and accepting tion of juryhood might well require the jurors to either is compatible with P, D, and I. From the disallow the first verdict, so that alternative view must viewpoint of the TUPID model, the only salient differ- exclude P, D, and JL1OIt follows that if this alternative ence between the two votes is the time at which they occurred and the fact that at those moments the 10Suppose, for example, that after the first unanimous vote, the jury pattern of emergent preferences was different. But P, holds a second vote on whether to discuss the verdict further or let D, and I by themselves give us no basis on which to the recommendation stand. In this second vote, all but one agree to prefer one to the other, apart from the Hobbesian let the conviction stand, because they want their jury duty to be over 574 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions American Political Science Review Vol. 94, No. 3 conception is reasonable, then we have found at least who advocate them (e.g., skin color, ethnic origin, and one collective will-revealing procedure in which relax- religious affiliation) and determining one's vote on the ing the TUPID conditions is innocent. basis of criteria that give absolutely no weight to, or One might object that there are fundamental differ- completely discount, the values of free and open ences between a jury and the sovereign electorate of a debate, compromise, accountability, equal consider- purportedly self-governing democracy. There is some ation, and transparency. These restrictions neither truth to this: Directing democratic debate about public amount to nor necessarily rest on a particular view of policy along the narrow lines defined by a judge who the common good or the substantive goals of public instructs a jury to disregard all criteria apart from the policy, but a society that ignores these desiderata reasonable doubt standard would surely impinge un- clearly fails to meet expectations implicit in the idea of duly upon democratic freedoms. Moreover, in the jury democratic agency. As in the jury example, citizens example constraints are imposed externally by the might reasonably conclude that when a society acts on judge and the wider legal apparatus in which the jurors the basis of such motivations and values it fails to live find themselves immersed, but the idea of democratic up to the ideal of an open, inclusive democratic self-government seems to require that any relevant society.12Unlike the jury example, it is not obvious that restrictions be imposed from within. these standards are necessarily "externally imposed" The force of these points is easily exaggerated, upon citizens. They simply may be embedded deeply in however, particularly if we draw an unnaturally sharp the institutional history and discursive practice of a line between procedure and substance in the context of given polity, such that it would seem very unnatural to democratic practice. Stressing the way in which democ- regard citizens who apply them as appealing to any- racy asserts the priority of certain decision procedures thing alien or extraneous.13 over substantive political viewpoints can obscure the The central point is that practices of compromise, way in which democratic procedure can itself function open debate and inclusion, and the associated deliber- as a substantive social ideal. That social ideal need not ative commitments to values such as fairness, consid- specify any particular view about the goals of public eration, and freedom are partly constitutive of demo- policy. But, as in the jury example, it unavoidably cratic agency.14To the extent that restrictions essential defines a certain ideal about the terms on which certain to these practices fail to operate, one may legitimately policy alternatives are to be identified as the will of a doubt whether the results of the process can qualify as democratic public. To the extent that citizens represent the will of a group with the distinctive character of a themselves as members of a collectivity with the dis- democratic public. By contrast, P, D, and I involve an tinctive character of a democratic public, they may interpretation of democratic authenticity that pays no reasonably impose restrictions on the range of first- attention to the character of the motivations behind order values and motivations that could in principle individual votes and preferences. The only information determine their joint will. Moreover, the justification for such restrictions need not be based upon an 12 The relevant restrictions would therefore be what Goodin (1986, elaborated and defensible theory of the public's best 78) helpfully calls "input filters." 13 This may imply a certain conservative bias, in the sense that interests, or of the common good. To establish that embedded social ideals (such as democracy) require citizens to such a democratic public is self-governing, it would be acknowledge an overriding allegiance to the extant practices, cul- necessary only to justify restrictions on the following tures, and institutions in which such ideals are latent. But perhaps more slender ground: They are a necessary condition that is necessary in order for a political group to have a persisting for plausibly ascribing to the authors of the social will identity over time such that it can be in principle the subject of its own destiny. This tends to vindicate the Aristotelian idea that a the distinctive characteristics of a democratic public.1" society's constitution is both the source of a local consensus on If citizens see themselves as members of such a public, certain social values and a political society's essential identifying they would be untrue to their collective character if attribute (Aristotle 1981, 176). Perhaps the real lesson of the they were to ignore such restrictions. impossibility theorems is that a social will cannot emerge in a social These restrictions will vary according to the content vacuum: There must be at least some nonarbitrary constraints informing citizens' joint self-understanding as a society organized of the relevant democratic ideal, but it is not difficult to around certain ideals and values. Only then can they see themselves imagine likely candidates. Excluded motivations might as participants in an overriding political identity to which their joint include, minimally, and formally: determining votes by decisions and actions may (or may not) be loyal. But note that all consulting the same oracle and following its direc- constraints of this kind are crucially different from those imposed by the various forms of the General Will argument we considered tives or by tossing a coin. More substantively, they earlier. Under that argument, the constraints are imposed by some might include ranking alternatives entirely on the basis independently authoritative account of the common good so as to of judgments about irrelevant features of the people increase the likelihood that the right answer will emerge from appropriately designed decision procedures. 14 This does not mean that democratic publics must consciously take and done with. In our example the lone holdout is right to think that these values as their concrete goals, only that consciously or not, the initial vote, albeit unanimous, may not represent a will that is these restrictions must in fact shape the character of the political theirs qua jury, so a will-revealing procedure based on our alternative process. The property of acting like a democratic public may, for understanding of juryhood might require that his single vote deter- example, be what Elster (1986, 122-7) calls a by-product. This mine their will as a jury at that moment-in this case to continue highlights another difference between the line of argument devel- deliberation. But this seems to require relaxing D. oped here and epistemic populism. In the latter view, citizens are 11 Or they are characteristic of some polity that aspires to some more required to embrace consciously a particular collective goal (some substantive social ideal that is nested within and compatible with substantive view of the common good) as part of the political democratic ideals and procedures (e.g., "the American way"). process. In my view, this need not be so. 575 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions The Possibility of Self-Government September 2000 that is allowed to count concerns patterns in the list of much less disturbing to the idea of democratic self- individual rankings offered by citizens. This is the same government. The claim that the TUPID conditions "empty vessel" interpretation of reflexive authenticity cannot result in a coherent will can now be interpreted discussed earlier: The character of a democratic public simply as a formalization of the claim that wanton can be whatever the preferences of its members make collectivities, like wanton individuals, are systemati- it. From a Frankurtian viewpoint, democratic authen- cally unable to form a will. ticity requires second-order restrictions on the sorts of motivations and values upon which voters' judgments are based. The standard objection to such restrictions CONCLUSION is that when they are applied, "social choice is based as Clearly, these arguments are not intended as a solution much on the restrictions as it is on individual judg- to Arrow's theorem on its own terms, if by that one ments" (Riker 1982, 117). The response is to insist that means an all-purpose algorithm for guaranteeing that although such restrictions may not be based on indi- voting procedures will not result in cyclical preferences. vidual judgments on specific occasions, observing them My argument also is not meant as a refutation of the may nevertheless be a necessary condition for certify- theorem; I do not believe that Arrow's theorem can be ing that the individuals expressing those judgments refuted. Rather, I have attempted to put the impossi- comprise a group with the distinctive character of a bility theorem in proper perspective within the context democratic public. The criteria for attributing to a of self-government. Social choice theorists have made collectivity a democratic character automatically entail strong claims that Arrow's theorem establishes the various restrictions on the range of motivations allowed logical absurdity of the notion of a popular democratic to constitute its will. Once we see this, we can dismiss will. That is only true if the expectations imposed by the charge that such restrictions necessarily preempt social choice theory are necessary conditions for any democratic authenticity because they are "imposed reasonable will-revealing procedure. I have suggested upon" a democratic public as if from outside. Frank- that this assumption is dubious. We may have been furtian restrictions are presupposed in the very concept looking in the wrong place for democratic will-reveal- of a democratic public itself. In a sense, they cannot be ing procedures and tacitly relying on an unreasonably imposed; if they are not operative, there simply will be stringent criterion of collective self-government. no democratic public on which to impose. That is why The specification and application of my broader they can legitimately be regarded as essential to an understanding of will formation are complex matters appropriate test of reflexive authenticity for demo- and face many difficulties. The result may not be an cratic publics. understanding of the social will with the formal ele- Given this, citizens who think of themselves as gance and rigor offered by the social choice model. members of a democratic public have reasons to reject That would be a real loss. But there is also the potential P, D, and I as criteria of democratic authenticity. for compensating gains. A wider notion of the social Frankfurt's terminology helps illuminate what is at will may illuminate a number of issues that the social stake: For a democratic public to endorse P, D, and I as choice model leaves obscure. In closing, I offer three its criteria of reflexive authenticity would institutional- thoughts along these lines. ize wantonness. In virtue of its accepted self-image, First, many insights are offered by social choice such a public would have a second-order desire to fulfill theorists. I merely claim that the model of will forma- its aspiration to be genuinely democratic. But the effect tion presupposed by Arrow and his successors is not of endorsing P, D, and I would be to deny any special uniquely reasonable and appropriate to all choice privilege to that ideal in the formation of its will. Such situations. Furthermore, some choice situations may an authorization would instantiate wantonness by un- not be governed by relevant second-order volitions. In dermining the basis for any plausible discrimination such cases, the TUPID conditions may yet be appro- between the fulfillment of society's aspiration to be priate, and it will follow that there is no way to democratic and those first-order preferences or emer- determine the content of the social will with regard to gent desires that might signify its betrayal. the specific alternatives under consideration. Con- Replacing P, D, and I with a more restrictive crite- versely, the presence of second-order volitions may rion of democratic authenticity does not itself guaran- cause preferences over alternatives to have the prop- tee that such betrayals will not occur (although it might erty of single-peakedness (Riker 1982, 128). It has long imply institutional-perhaps in particular constitution- been understood that where this property exists, the al-mechanisms to disable or impede them). But this Arrow problem does not arise. Integrating social does not affect the key theoretical point that from the choice theory into the broader perspective proposed perspective of an ideal of democratic self-government here may allow us to define more sharply the precon- it is irrational to institutionalize wantonness. Surely, ditions for the formation of a social will. there is a fundamental difference between a society Second, the perspective offered here brings into the whose aspiration to be genuinely democratic has been foreground questions about the interface between sec- betrayed in practice and a society that has institution- ond-order volitions that define citizenship and those alized a license for this kind of betrayal. It is the that operate in other arenas. For example, individuals difference between Frankfurt'swanton and Frankfurt's have personal ideals, and the ideals of democratic unwilling addict (or the tempted spouse). From the self-government also make demands on them. Differ- Frankfurtian perspective, Arrow's theorem becomes ent personal second-order volitions might threaten or 576 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions American Political Science Review Vol. 94, No. 3 facilitate the organization of a democratic will, and it is Barber, Benjamin R. 1993. "Reductionist Political Science and important to know which. Moreover, semipublic insti- Democracy." In Reconsidering the Democratic Public, ed. George tutions such as political parties form wills using their E. Marcus and Russell L. Hanson. University Park: Penn State University Press. Pp. 65-72. own characteristic second-order volitions. Perhaps per- Barry, Brian. 1990. PoliticalArgument. Hemel Hempstead: Harvester sonal identification with such ideals being a "loyal Wheatsheaf. Republican" or a "faithful trade unionist," and the Cohen, Joshua. 1986. "An Epistemic Conception of Democracy." sanctions imposed by such institutions for disloyalty, Ethics 97 (October): 26-38. may help reduce or increase conflicts between personal Coleman, Jules, and John Ferejohn. 1986. "Democracy and Social ideals and ideals of citizenship. Understanding the role Choice." Ethics 97 (October): 6-25. Duverger, Maurice. 1954. Political Parties, trans. Barbara and Robert of semipublic institutions might tell us much about the North. London: Methuen. conditions under which a public will can be organized. Elster, Jon. 1986. "The Market and the Forum: Three Varieties of There is also the related question of the range of Political Theory." In Foundations of Social Choice Theory,ed. Jon motivations affiliated with particular parties or groups Elster and Aanund Hylland. Cambridge: Cambridge University that a given ideal of democratic citizenship will toler- Press. Pp. 103-33. ate: Unfortunately, the people of Austria and Northern Frankfurt, Harry. 1988. The Importance of "hat We Care About. Cambridge: Cambridge University Press. Ireland have recently found themselves embroiled in Goodin, Robert. 1986. "Laundering Preferences." In Foundations of just this issue. Arguments to the effect that certain Social Choice Theoty, ed. Jon Elster and Aanund Hylland. Cam- kinds of parties, and the values that identify them, are bridge: Cambridge University Press. Pp. 75-101. "beyond the pale" or inimical to the conduct of dem- Grofman, Bernard, and Scott Feld. 1988. "Rousseau's General Will: ocratic decision making may be essential to the practice A Condorcetian Perspective."American Political Science Review 82 of democratic self-government. (June): 567-76. Third, the Frankfurtian position preserves the idea, Hurley, Susan. 1989. Natural Reasons: Personalityand Polity. Oxford: Oxford University Press. deeply embedded in ordinary usage, that wills, once May, Kenneth. 1954. "Intransitivity,Utility, and the Aggregation of identified, may nevertheless be betrayed in practice. Preference Patterns." Econometrica 22 (January): 1-13. Drawing on Aristotle, Hurley (1989, 136-7) gives a Mayo, Bernard. 1956. "Is There a Case for the General Will." In fascinating illustration of how this idea can have col- Philosophy, Politics and Society, ed. Peter Laslett. New York: lective applications. Members of a democratic public MacMillan. Pp. 92-8. may "will" an end to racial discrimination (perhaps McClean, lain. 1991. "Rational Choice and Politics." Political Studies 39 (September): 496-512. because they take this to be mandated by democracy McKelvey, Richard. 1979. "General Conditions for Global Intransi- itself), yet find that their own will is frequently betrayed tivities in Formal Voting Models." Econometrica 47 (September): in their interactions as private citizens (de facto dis- 1085-111. crimination, powerful groups using their favored posi- Michels, Robert. 1916. Political Parties: A Sociological Study of the tion to circumvent or exploit loopholes in relevant Oligarchical Tendencies of Modem Democracy. London: Jarrold. statutes). The line of argument opened here provides a Plott, Charles. 1976. "Axiomatic Social Choice Theory: An Overview and Interpretation." American Journal of Political Science 20 framework for analyzing this kind of phenomenon and (August): 511-96. can help us perceive a politics of the will in the Radcliff, Benjamin. 1992. "The General Will and Social Choice characteristic discourse that develops around the at- Theory." The Review of Politics 54 (Winter): 34-49. tempt to impose legal and social controls on citizens' Raz, Joseph. 1986. The Moralityof Freedom. Oxford: Oxford Univer- behavior. sity Press. These suggestions are sketchy and present serious Riker, William. 1982. Liberalism against Populism: A Confrontation between the Theoryof Democracy and the Theoryof Social Choice. difficulties, but there is no reason to suppose a priori San Francisco: W. H. Freeman. that these difficulties are impossible to overcome. The Roemer, John. 1996. Theoriesof DistributiveJustice. Cambridge, MA: social choice theorists suggested that the notion of a Harvard University Press. collective will is indeed an a priori impossibility. If the Schelling, Thomas. 1984. Choice and Consequence. Cambridge, MA: argument of this article has succeeded in dispelling Harvard University Press. these alleged impossibilities and in opening up a viable Schofield, Norman. 1985. Social Choice and Democracy. Berlin: line of argument, the progress, though modest and Springer-Verlag. Schumpeter, Joseph. 1942. Capitalism, Socialism, and Democracy. incomplete, will nevertheless be real. New York: Harper & Row. Sen, Amartya. 1970. Collective Choice and Social Welfare. San Francisco: Holden-Day. REFERENCES Steedman, Ian, and Ulrich Krause. 1986. "Goethe's Faust, Arrow's Aristotle 1981. The Politics. London: Penguin Possibility Theorem and the Individual Decision-Taker." In The Arrow, Kenneth J. [1951] 1963. Social Choice and Individual Values. Multiple Self, ed. Jon Elster. Cambridge: Cambridge University New Haven, CT: Yale University Press. Press. Pp. 197-233. 577 This content downloaded from 128.143.1.88 on Thu, 5 Sep 2013 09:51:02 AM All use subject to JSTOR Terms and Conditions