To have "a point d'appui" entails assigning to representation the character of "advocacy." Advocacy combines two elements: a "passionate" link of the representative with the electors' cause, and a certain autonomy of judgment on the part of the representative. On the one hand, it engenders a politics of conviction, and nurtures a spirit of controversy (Mill spoke of sympathy linking "friends" and "partisans" against their "opponents") (Mill [1861] 1991, 282). On the other hand, it pursues democratic deliberation, and thus respects consented procedures and favors outcomes that are congruent with the basic principle of political equality. The tension between commitment and autonomy that animates advocacy exemplifies the character of representative democracy, whose condition of existence and persistence rests on the acknowledgment that it cannot square the circle between the "general will" and the "will of all." This means not simply that dissent and conflict cannot be avoided, but moreover than any consent is permanently open to revision, and finally that a deliberative politics is also a politics of controversy.
The difference between direct democracy and representative democracy pertains mainly to the way of looking at consent. On the one side, Rousseau's model entails a progressive overcoming of sources of disagreement, because it is indicative of a pluralism of interests that obstructs the attainment of the general good. On the other, the representative model stresses the moment of discussion. In On Liberty, Mill went so far as to claim that the debating process is good in itself. He stylized the tension between the two models of democracy with the names of Plato and Descartes on the one side and Socrates and Bacon on the other (Mill [1866] 1978, 377-405). As we shall see in a while, he referred to the former as a mathematical and the latter as a rhetorical kind of reasoning. Finally, he thought that only within a searching perspective could political deliberation make sense.
Contemporary debate on the meaning and significance of deliberative democracy is once again dealing with more or less the same divide. On the one hand, some theorists propose a conception of deliberation that recalls a Platonic dialogue in which the interlocutors are allowed to hold wrong ideas but not passions, which in fact would hamper the rectification of the wrong (Thrasymachus had no alternative but leaving the stage). Here deliberation is seen from the point of view of the outcome, and is expected to reduce differences by amending "distorted" interpretations of the public good (26). I would call this a hegemonic model of deliberative democracy. On the other hand, other theorists propose a conception of deliberation that stresses the process of communication and debate, and are not troubled by the persistence of differences and disagreement, which they actually see as the very condition for deliberation to occur. Here the rationalist vice is circumvented through a rejection of the "dichotomy between reason and desire," while the emphasis is placed on the critical moment more than on a rational consensus over an impartial definition of general good (Young 1990, 102-11; Fraser 1997, 121-30). I would call this an agonistic model of deliberative democracy (Manin 1987, 338-68).
Mill's theory of representative democracy belongs to this model. Indeed, while he shared a sincere aspiration for the general good, he also interpreted it as a regulative principle. Against the two "bad" visions of democracy that I mentioned in the preceding chapter, Mill's presumption was not that there is a necessary "identification of interest between the rulers and the ruled;" but that such an identification does not exist. Were the identity real, not only representation and deliberation but, moreover, government would be useless. "Identification of interest between the rulers and the ruled, being, therefore, in a literal sense, impossible to be realized, must not be spoken of as a condition which a government must absolutely fulfill; but as an end to be incessantly aimed at, and approximated to as nearly as circumstances render possible, and as is compatible with the regard due to other ends" (Stuart Mill [1835] 1977, 22-3). Within this theoretical frame, the democratic process of deliberation becomes meaningful.
Now, the relevance of Mill's political thought in contemporary debate rests on the strategy he adopted to justify the agonistic model of deliberative democracy, that is the theory of representation as "advocacy." The seminal idea Mill's political writings convey on this matter is that it is a mistake to oppose advocacy to deliberation, as if the former meant simply disagreement, and the latter simply consent. Not more helpful would be to depict the political scenario as an arena in which there would be either a split between advocacy and deliberation, or a compromise between the two. The splitting scenario would be "a disaster" from whatever point of view one would look at it; the compromise scenario would be more reliable, but it would bear the inconveniences of a mediation among opposite things. What we learn from Mill is that these readings rely upon a wrong premise: the assumption that the advocate excludes the deliberator, and vice versa (27).
In defining the representative as an advocate, we ought to ascribe to him or her not only the character of the partisan, but also that of the deliberator. In Mill's description, the assembly worked like a court whose protagonists played both the role of advocacy (proposing, defending and opposing a law) and judgment (28). Even though they were not deliberating when acting as advocates, nonetheless their speech was consciously performed in view of a deliberation. The two functions should not be seen as opposite, because without both of them there would be no advocacy to begin with. Thus, it is not correct to say that the good representative should combine advocacy and deliberation, because this would imply that it is possible to have them separately, and, moreover, that they exclude each other. Quite the contrary, the figure of the advocate presupposes deliberation, indeed he or she is expected to be both a passionate defender and an intelligent defender as well. Advocacy is not blind partisanship. Rather, it is a partisanship that is structurally projected toward deliberation. Hence, to recast the point made by Anne Phillips, I would say that the problem is not that in our democracy we "need advocates as well as deliberators," but that we need advocates in the first place (Phillips 1995, 162). An advocate who is "exclusively" a partisan is not an advocate, even if he acts as an advocate. And a deliberator who is "exclusively" a Platonic rationalizer is not a deliberator, even if he produces rational justifications. To rephrase Mill's words, I would say that what a good representative democracy needs are neither fanatics nor bureaucrats of representation, nor finally philosopher kings, but rather deliberators who judge causes "passionately" pleaded (and who plead causes in their turn). In fact, democratic deliberation comes to make sense only if representatives are like "advocates."
What is an advocate? In On Liberty, Mill advances a comparison between two kinds of "understanding," one which excludes objections, and another which develops out of objections. The "peculiarity" of the evidence of a mathematical truth, he claims, is that "all the argument is on one side. There are no objections, and no answer to objection" (Mill [1859] 1991, 41). By contrast, the kind of "understanding" belonging to moral and political decisions hardly allows for an answer with no objections. The peculiarity of this kind of judgment is that on "every subject" there are different opinions and no truth is "on one side." In such a case, the outcome rests on the ability to produce an argument that can convince the interlocutor. The argument will not make any claim to truth, for "the truth depends on a balance to be struck between two sets of conflicting reasons." The "balance" between two claims or visions is the only kind of impartiality a representative assembly can reach, while it is the deliberative outcome at which a good advocate should aim (29).
Now, a rationality that is not aimed at demonstration but rather at deliberation is not, properly speaking, a cognitive kind of reason, but a practical (or pragmatic) one (30). Practical reason does not discover truth but indicates how to justify or challenge a given assertion, a custom, and a law. "To reason is not merely to verify and demonstrate, but also to deliberate, to criticize, and to justify, to give reasons for and against --in a word, to argue" (Perelman 1980, 59). And practical reason is required when choices are to be made, and actions to be taken. To will to justify everything is senseless, wrote Aristotle. Justification --and the need for it-- occurs when what we are claiming or seeking is open to interpretative disagreement; its function is to meet others' claims, and does not entail necessity but only probability (Aristotle 1994, I.ii 1358b).
Thus, an "advocate" is not asked to be impartial like a judge, nor to be a solitary reasoner like a philosopher. Unlike a judge, he has some ties with the contender; his job is not that of applying the rule to his case, but that of defining whether the facts fit or contradict the existing rule or whether the existing rule reflects principles that society shares or a "good" government should contemplate. His point of departure is not an impartial stand, but the specific condition of the person. An "advocate" is not a philosopher either. Indeed, unlike the philosopher, and like the politician, she has "to conform to the wishes of an electorate in order to win" her cause or a vote (Perelman 1980, 66). The philosopher does not owe any justification to anybody, but only to his principles, nor he is seeking external consensus. "A philosophical justification cannot refer to the interests and passions of a particular group;" it "must be rational, or at least reasonable" (p. 59). On the contrary, the relation of the candidate to his electors, like that of the lawyer to his client, does "not require the electors to consent to be represented by one who intends to govern them in opposition to their fundamental conviction" (Mill [1861] 1991, 382).
Mill alluded to these maxims of practical reasoning when he described the public arena as a place for "compromise between right and wrong." In politics, "principles themselves are neither true nor false, but half-and-half;" a kind of "juste-milieu mosaic" not a "square of black and a square of white alternative." In an assembly, as in any meeting where persons discuss topics pertaining to their good, no opinion can claim a superior truth, no matter how just and reasonable the opinion may be. "But on every subject on which difference of opinion is possible, the truth depends on a balance to be struck between two sets of conflicting reasons" (Mill [1859] 1991, 41). In the agora, even the Newtonian system would probably find some opponents were its "advocates" unable to convince their audience of its truthfulness. "It is as if the Newtonian system had to be voted by show of hands; and the parties being unable to convince one another, agreed to a resolution that there was much truth in Newton's principles, but that Ptolomey also had a good deal to say" (Mill [1832] 1986, 424) (31).
This "balance" is the kind of outcome that deliberative reasoning can attain. Far from transcending the specific situation of the citizens, it rests on the premise that that specificity needs to be known and acknowledged. Hence, "understanding" and "seeing" are the faculties at work in forensic as well as deliberative speech (32). They express the complex nature of advocacy, which needs to adhere to its cause, but not to be driven by it. A good advocate has to believe in his cause, but also to understand the reasoning of others to the point of being able to reconstruct it in his mind. He has to "feel" its force to be able to envisage the right rhetorical strategy. This is the basic maxim that oratory has been teaching since Aristotle's and Cicero's times. Mill revived it thoroughly.
In On Liberty, Mill elaborates from the same passage of Cicero's "On the Character of the Orator" that had inspired Hobbes in his definition of the representative as a "persona" (33). Mill resorted to Cicero's precepts of good oratory to develop the concept of representation as advocacy.
Now, Cicero made two arguments: first, that an orator is good in proportion as he is able to traverse, as it were, identities and "minds;" and, second that the orator is able to traverse identities because (not in spite of) he believes in the cause he has accepted to carry on. One may say that the more he is in sympathy with his cause, the more he is able to transcend the limits of that sympathy and penetrate the reasoning of his adversary (34). Cicero then indicates the three steps an orator ought to follow to become a good orator.
First, the advocate has "always [to take] great pains to be instructed by [his] client himself in the nature of his cause" (Cicero 1808, vol. 1, 173-74). Thus, the client has to have "the greater freedom in speaking," and the chance to communicate directly with his advocate. Parallel to this is Mill's claim that freedom of speech is the main precondition of a representative government, and his view of the relation between good representation and citizen participation in electoral campaigns. He thinks that citizens and representatives should hold "free and public conferences." Interaction is needed because choice is what makes for representation. And, on both sides, the choice is always reversible: representation, Mill writes, describes a "course of action" not a "single act" (Mill [1861] 1991, 370) (35). Hence, like an advocate with his client, a representative should entertain a continuous and frank communication with his constituency. Language, Cicero wrote, "has no force when it is not supported by a proper solidity of sentiment" (Cicero 1908, vol. 2, 260). If a representative were to set the "aristocratic atmosphere" of the House, Mill cautioned, and "neglect those free and public conferences with his constituents," he would be liable "to tone down any democratic feeling which he may have brought with him, and [this, in turn, would] make him forget the wishes and grow cool to the interests of those who chose him" (Mill [1861] 1991, 370). This aspect of advocacy runs conter to Pitkin's theory of representation as a "description," and of proportional representation as a more detailed and passive mirror-like reproduction of a given social being. According to her reading, the representational relationship lacks both authorization and accountability because the moment of choice is missing.
The second step the good orator should take pertains to the transition from partiality to 'impartiality' as one from partisanship to the reflection on the overall deliberative process which entails both the client's cause and that of the rival as well as the judge's point of view. This step occurs in the solitude of the orator's mind, and presupposes his autonomy of judgment and a certain liberty from his own cause. Cicero offers a true phenomenology of advocacy, which Mill faithfully adapts to representation and, more in general, to citizenship (as we have seen in the first chapter, the deliberative process entails both speech and silence, while excluding the interpretation of autonomy of judgment as predicated on isolation). "Then, after he [the client] is gone, with the utmost impartiality, I assume three characters, my own, that of my antagonist, and that of the judge" (Cicero 1808, vol. 1, 173-74). The orator tests the strength of his case against both that of his rival and the interpretation of the law from the point of view of an impartial judge. This is also, Mill thought, what the good representative does when he checks his cause against both that of his adversary and the opinion of the assembly. Like the advocate, the representative has to study the "adversary's case with a great, if not with still greater, intensity than even his own" (Mill [1859] 1991, 42). Reflection entails a stepping back from factuality and "passionate" partisanship, an effort to see beyond them, to look for connections and comparisons in order to envisage a possible compromise or solution. It requires the ability to transcend (not to relinquish) one's sympathy with one's cause and to enter into a sympathetic relation with the adversary's (36). This is in Mill's view the rationale of proportional representation, which indeed presumes the "advocate" of a minority cause able to transcend his or her allegiance to the minority cause and thereby to broaden support and win consideration. The main difference between group and proportional representation rests here (Phillips 1995, 27-56; Young 1997, 361-69) (37).
The third and final step pertains to the event of public performance, the conclusive check of the work of the advocate and the representative. In Mill's accurate perception, the task of the "advocate" is not simply that of imagining the greatest number of possible arguments against the adversary abstractly, not even that of simply managing logical tools skillfully. The advocate should also be able to experience and directly to "feel" the reasoning and logic of others, to face the partisanship of the adversary. His public performance shows that the elaboration of good reasons is far from enough. Like an advocate, the representative "must be able to hear them [reasons] from persons who actually believe them ... he must feel the whole force of the difficulty which the true view of the subject has to encounter and dispose of" (Mill [1859] 1991, 42-3).
Thus advocacy places high degree of importance to personal ability and character. As a consequence, while in theory and de jure every citizen can become a representative, citizens nonetheless select those whom they judge to be better advocates. They do not choose at random nor do they feel it is enough that the candidate belongs to their "group" to elicit their consent (indeed they discriminate within their own very group). As Ronald Dworkin puts it, in the very moment we make a claim for an equal political say of all citizens, we are also forced to admit that people are different in their political performance, some having more ability or more passion than others and more chances to pursue their preferences than others (Dworkin 1988, 15). Advocacy, like representation, implies a selection, because we seek to get the best defendant, not a copy of ourselves (38). Hence, when Mill proposed proportionality (not only for the "higher class," but also for workers and women) he thought that representation should further people's claims --as second-rate citizens or non-equally treated citizens--, not reproduce their social groups. Minorities claimed an advocate not a post-mark, because their goal was political equality or equal consideration. Thus, proportional representation excludes an organic conception of representation, according to which representation would be the transcription of a pre- or non-political identity; it also excludes a vision of the society as a corporate federation of groups in relation to which proportionality would work as a preservation device (39).
It is within a rhetorical (that is agonistic) perspective that the role of the representative as an "advocate" acquires its full relevance, plainly contradicting both Pitkin's interpretation of proportional representation and its recent rationalist adaptation, according to which proportional representation becomes a supplement of information for the achievement of impartial or true outcomes.
Pitkin defines Mill's assembly as a passive congregation because her notion of activity, like Rousseau's, does not include speech, but only "doing" (40). When, like in a representative democracy, a direct "doing" can no longer be the source of political autonomy, then the only solution is to make the assembly a copy of the society. So, the descriptive accuracy of the "being" matters, not speech, and the representative assembly ends up being filled by mute doubles. Pitkin's reasoning rests on the assumption that I tried to criticize in the first part of the paper, that is that democracy, in its ideal model, excludes any form of indirectness --as for instance speech-- and, as a consequence, that representative democracy ought to be judged by reason of its ability to mimic the social composition (41). On the one hand, representation is supposed to be an objective description of the traits belonging to each segment of society, while on the other, the representatives are supposed to transmit information with their very presence, to supplement the data the assembly needs to reach a non-sectarian deliberation. In this framework, representation appears to mean testimony.
This reading has recently been used to defend a rationalist justification of proportionality, according to which a plurality of opinions has an instrumental meaning; as a device which is functional to the achievement of the truth. The lack of "full representation," writes Roberto Gargarella, thwarts deliberation because it thwarts impartiality, which is the object of deliberation and the outcome of the use of cognitive reason. Restating somehow the James Mill model, this theory of deliberation as a gradual unveiling of rationality maintains that what produces unjust or "non-neutral" outcomes is simply a lack of knowledge, not opposite interests: "Non-neutral decisions may be reached not because of a misinterest or partiality of the decision makers, but because of a misunderstanding of the way other people evaluate choices" (Gargarella 1998, 261). Whereas ignorance can sometime explain injustice, in a society shaped by a variety of means of information and communication, it seems hard to maintain that rulers bare no responsibility for injustice, except for a lack of knowledge. It would be difficult to prove that when in Mill's time the British parliament opposed female suffrage or workers' right to associate, it was actually acting out of mere ignorance and a lack of information about women and workers' living condition.
Proportional representation brings into politics a sense of partisanship that conflicts with a rationalist interpretation of deliberative democracy, and the disassociation of advocacy and deliberation, rhetoric and reason, on which this interpretation is grounded. Mill's writings offer an interesting reading because they acknowledge the role of practical reasoning in politics and speak to the complex nature of advocacy. "And certainly, all interests or classes of any importance ought to be represented, that is, ought to have spokesmen, or advocates, in Parliament." Nonetheless, the representative should also be able to judge the "'real' interest" of the whole country and "subordinate to reason, justice, and the good of the whole" the claims of his or her constituency (Mill [1861] 1991, 295, 300,323).
Mill's theory of representation as advocacy entails a notion of citizenship that is egalitarian in character, and that pertains to individuals' relations of power. Because its normative principle is reciprocity, its aim is not that of reproducing social identities, but that of giving voice to political relations of power that place individuals belonging to some groups in a position of subordination. Thus, the theory of advocay rests on a conception of citizenship that unifies the two basic equalities that have belonged to democracy since the classical age: isopsephia, or equality of voting --which gave to all citizens the right to an equal participation--, and isegoria, or the equal possibility of speaking --which gave to all citizens the opportunity to express their opinions publicly and be heard (or represented) equally (42). While the former implies a simple conception of equality (one-citizen-one vote), the latter does not exclude the use of diversified devices. One might say that thanks to the latter, difference gives substance to equality. Proportional representation is a "special manner" by which citizens try to redress an unbalance of political power (a form of domination, no matter how large the dominant class is) (Mill [1861] 1991, 323). It is this democratic understanding that gives salience to Mill's recognition that deliberation is a form of advocacy.