2. Representation and the Agora Model
The acknowledgment of a link between representation and deliberation was the premise that brought Mill to maintain that in a "good" democracy the assembly needs to be like an agora in which citizens' "voices" are represented proportionally.
Mill assigned to the assembly two powers, that of control, and that of discussion. Concerning the former, the assembly had to check the executive, to "throw the light of publicity" on its acts, compel their "full exposition and justification," and finally to censure and dismiss politicians from office. The power of control aimed at securing "the liberty of the nation" (Mill  1991, 282). Mill cautioned that this power would decrease in proportion as the assembly would identify itself with the majority leading the executive. Thus, in order to preserve its checking power, the assembly needed to function as the public forum of the whole country, in which "not only the general opinion of the nation," but "every interest and shade of interest" had its say and cause "passionately pleaded" so as to compel others to listen to and produce justifications. The agora had to "indicate wants" and be "an organ of popular demands," a place in which "adverse discussion" on questions pertaining to "public matters" could "produce itself in full light." The logic leading Mill to proportional representation was the anti-monopoly proviso divide et impera. Through proportional representation, he translated the democratic principle of equality into an argument for political liberty, and achieved a notion of democracy that was quite new in his time, when democracy meant mainly a regime dominated by the blind "passion of equality," as according to Tocqueville.
The anti-tyrannical argument of liberty as "security for good government" --an expression that aims to underline the idea that political liberty entails self-government-- was the weapon Mill used against a majoritarian interpretation of democracy, in the form of both a "pure" or direct democracy (Rousseau's model) and of a representative democracy "by a mere majority of the people" (James Mill's model) (Mill  1991, 302) (14). The core of Mill's theoretical qualification of representative democracy lies in his objections to these two models.
His first and more radical criticism pertained to the deductivist structure of Rousseau and his father's system, a Cartesian inference from the axiom which assumes political liberty as coincident with the identity of the body politic. While naturally in agreement with direct democracy, that theorem would have devastating effects on James Mill's strategy, the goal of which was that of defending representative government.
James Mill's theorem claimed that the interest of a democratic government coincided necessarily with the general interest; that the general interest was identical with the interest of the majority, because the "laborious many" were more immune from a bad use of political power than were the few (at any event, their misgovernment would be caused by ignorance and not by sinister interests); and, as a consequence, that each could represent the interests of others without neglecting or abusing and exploiting them (James Mill  1992, 7) (15). The theory of representation as a "mirror" --which Pitkin applied indifferently to majoritarians and proportionalists alike-- fits James Mill's model perfectly well. In his vision, the parliament was a faithful reflection of a uniform compound that chose a "certain number of themselves to be the actors in their stead" (p. 8, italics added; Pitkin 1967, 60-91). The benefits of representation would vanish "in all cases in which the interests of the choosing body [were] not the same with those of the community" (James Mill  1992, 27). Representation did not have to represent claims or opinions, but rather what people held in common, that is their potential to pursue their well-being. Moreover, it worked as a simplifier of interests and an assimilator of subjects, because the more industrious were assumed to promote the interests of others by promoting their own. This Chinese box model worked particularly well with dependent people: women's interests were included within their husbands' and fathers' interests, and workers' interests within their employers'. De facto, voting was a means for protecting the security of the majority, not for promoting political emancipation of all citizens. Finally, nobody could act as an "advocate" or even a "mouthpiece" for anybody, because "the laborious people" could not allow any segmentation if "sinister interests" had to be avoided. In spite of his defense of representative government, James Mill ended up restating Rousseau's theory, with the crucial difference that, now, the sovereign people were the majority (16).
James Mill's model undermined representation, because it defined it by aggregate interests instead of individuals' ideas and claims, and linked it to objective truth instead of opinions. His assembly was not a dissenting body, but a place in which those standing in the place of the "laborious people" assessed an objective estimate of their interests. A difference of opinions entailed that the representatives either lacked knowledge or defended "sinister interests " (17). Properly speaking, in James Mill's assembly disagreement, as well causes "passionately pleaded," were out of place. A defense of the assembly as an agora would need to restore that which majoritarianism dismissed, that is the idea that representation is personal and that the representative is an "advocate."
Concerning the first mistake, John Stuart Mill objected to it by maintaining an individualistic foundation of democracy, and arguing that democracy does not mean that people are involved en-mass, but that they are involved as individuals, that they have --as individual citizens-- an equal political liberty. Hence, his conception of representative democracy could contemplate both the principle of equality and the principle of individual expressiveness or liberty. Mill revived Aristotle's idea that the principle of democracy "is that each citizen should be in a position of equality," which means that it is the "position" of each citizen that needs to be considered, not that of the masses (Aristotle 1995, VI:1317a49-50). This was also Mill's intuition, who claimed that the normative distinctiveness of democracy does not rest solely on the fact that the majority --"a flock of sheep innocently nibbling the grass side by side"-- rules, but on the conscious share of each in the life of the country. Democracy is a regime in which the institutional power of control is vested in a collective body which (through its representatives) ought ideally to profit from the voice of "every citizen" (Mill  1991, 244).
The personal dimension of the vote makes the debating character of the assembly predictable, while advancing the belief that every voice should find a way to be heard. Indeed, whereas universal suffrage guarantees that all citizens are treated equally, proportional representation guarantees that the specific condition of the citizen is not ignored. The former needs to be blind, while the latter is conscious of differences. However, proportional representation is not a form of differential treatment, because it does not distribute political voice unequally to unequals, but it gives to each the same chance of choosing, to those who believe A as well as to those who believe B. Its regulative principles are thus equality and the intensity of individual preferences. Hence, proportional representation is able to favor minorities not because it enacts a policy of "favoring," but because it takes away any policy of favoring. It does not give to the minority more of their respective numerical due, but it gives them the same chance of choosing representatives as it gives the majority. Thus, it does not use a "compensatory" logic, because compensation presumes that the stronger is and will remain stronger, while treating the weaker with a benevolent charity (18). It is therefore a mistake to think that proportional representation follows the logic of "equitable treatment," or the Aristotelian principle of proportional justice (19). Proportional representation takes seriously the principle underlining universal suffrage: the individual right to an equal vote. Hence it assumes that citizens have different opinions, ideologies, ideas, claims, and that they ought to be counted. Proportional representation recognizes pluralism in its entirety, while majoritarianism is a devise that first recognizes the majority, and then tries to solve the presence of pluralism through a "compensatory" treatment. Only the former reflects a philosophy that is radically inimical to privilege, and takes equality more seriously.
To John Stuart Mill, the legitimization to obey a majority decision rested on the stipulation that people should have the chance to express themselves in order to have the opportunity of both influencing and overturning legislation. By making themselves heard minorities would make the majority aware of the fact that it is a majority.
Thus, Mill's agora model implied proportional representation because it implied a vision of democracy as a system whose political process is to be judged from the point of view of "all," both those who happen to be in the majority and those who happen to be in the minority, and it presumes that the final decision is achieved through a deliberative trial whose actors ought to be the "whole" of "every opinion which exists in the constituencies" and "obtains its fair share of voices" (Mill  1991, 305 italics added). For this reason, in criticizing the majoritarian model of democracy, Mill spoke openly of the "slavery of the majority" and advanced a conclusion that effectively captured the link between representative democracy and proportional representation. De facto, he argued, a majoritarian democracy is a "government of privilege," and, as such, contradicts the democratic principle of equality (p. 303). In a government in which the majority "alone possess practically any voice in the State," the political counting of voices is deemed identical with the arithmetical counting of votes, which means that only the majority counts while "every single individual" in the minority does not count as much as "any other single individual" in the majority (p. 302, 304). An arithmetical democracy pays attention mainly to the formation of the majority, because it stresses only the moment of decision while neglecting the whole deliberative process. Mill did not contest that "the minority must yield to the majority, the smaller number to the greater" when decisions are to be taken. What he forcefully opposed was the idea that counting should simply mean that the majority is counted. When the representative body votes, "the minority must of course be overruled" (p. 303). But a representative body does not limit itself to voting, while, on the other hand, debate cannot occur without a plurality of opinions.
The characterization of the assembly as not a silent body that only votes has been one of the main contributions that the eighteenth and the nineteenth century theory of representative government gave to the evolution of democratic theory (20). It is also one of the central themes of Considerations on Representative Government. Mill overturned a long and authoritative tradition that cut across ideological boundaries and enjoyed a solid reputation within modern political thought. Disdain for rhetoric and the admiration of the Spartan assembly went along with the decline of the humanist tradition. Speech and the art of disputation were esteemed in Machiavelli's times, not in Descartes' and Hobbes's. Sparta, and its silent assembly, was a model for James Harrington as well as for Rousseau, not for Machiavelli (21).
The dichotomy between a deliberative republicanism and a rationalist republicanism bore its fruits in the post-Revolution era, when the conceptualization of representative government was perfected. In Mill's time, the English conservatives who opposed a democratic transformation of the state referred explicitly to the rationalist tradition. In spite of his anti-republican stand, for instance, the "reactionary" Willian Mitford relied on Rousseau who, like Harrington and unlike Machiavelli linked a well-ordered republic to a silent assembly, to corroborate his anti-democratic ideas (22).
I would say that Mill was the thinker who contributed in reviving the deliberative tradition of early modern republicanism by stressing the role of a talking assembly in a representative democracy. Moreover, he understood that it was the institution of representation that made the agora necessary in so far as it would give the citizens the chance to participate somehow in the political debate, to identify with their representatives, and finally to judge their behavior. A symbolic presence would make up for physical attendance, while expanding political debate beyond the parliament (23).
Mill perceived that representative democracy was not to be defined as a system in which people govern indirectly, but as a regime in which political action has to pass public scrutiny and control. While it is the majority that makes the laws, it is debate and judgment that give those laws a moral legitimacy and make people feel protected from the tyranny of the majority. In the course of the debate, each representative can contribute to amending or changing a proposal; and even if the final vote will decree a majority/minority divide, nonetheless the dialectic process would make it possible for both the majority and the minority to take part in the legislation process. In this sense, it is debate, more than simply majority rule, that gives legitimacy to the democratic process of decision-making. As Bernard Manin has recently argued, for the theorists of representative government, discussion and disagreement were the consistent outcomes of an egalitarian premise, the belief that discord among opinions could not terminate "through the intervention of one will that is superior to the others" (Manin 1997, 188-90). Hence, it was thanks to deliberation that the common good could be seen as a construction of the community itself, and as the outcome of ceaseless work of persuasion and compromise ending in a majority vote, not an absolute verdict.
In conclusion, Mill linked the claim for a talking assembly to the two main principles of democracy: control (and thus security) and equality. Control implies the anti-Platonic stand that no one holds the right solution in political decisions, and that human knowledge is fallible (24). Fallibility implies the recognition of both diversity of opinions and equality of consideration and opportunity. Hence, as Anne Phillips has recently argued, popular control is not simply grounded on prudence (self-protection from the monopoly of power) but it is also a value in itself, in so far as it is predicative of equality (Phillips 1995, 27-8).
Control and equality imply, Mill thought, that every voice should have the chance to be heard in the assembly, and moreover, that every citizen should have the chance to count upon "a point d'appui, for individual resistance to the tendencies of the ruling power; a protection, a rallying point, for opinions and interests which the ascendant public opinion views with disfavor" (Stuart Mill  1991, 316). In this sense, in representative democracy political exclusion would take the form of silence, of not being heard and considered. In the domain of representation, the principle of liberty (or "security for a good government") can be better attained by ideally giving to every one both a vote and an "advocate." This is what makes representation not simply instrumental: "Including those previously excluded matters even if it proves to have no discernible consequences for the policies that may be adopted" (Phillips 1995, 40) (25).