Chapter 12
THE PREVIOUS CHAPTER argued that even if one does not adopt a metaphysical explanation of what moral obligation and excellence are, it remains reasonable to view ethical discourse as an objective affair. I will now give additional support to this conclusion by reflecting on Brandom’s account of the ways in which we all keep track of the beliefs and intentions we undertake and attribute when conversing with other people. A game, as Wittgenstein realized, is a relatively perspicuous example of a social practice—a relatively simple species of the genus to which ethical discourse belongs. Brandom develops the analogy between discursive practices and games in a way that sheds light on the objective dimension of ethics as a social practice. My main purpose in discussing Brandom is to show how pragmatism can best resist degenerating into something like conventionalism, according to which the tyranny of the majority effectively determines what obligation and excellence are.
OBJECTIVITY IN THE CONTEXT OF A SOCIAL PRACTICE
In sandlot baseball, there are no umpires. In street soccer, there are no referees. The players keep track of runs or goals and how well everyone played. Similarly, in forms of discourse that include a dimension of objective inquiry, players make commitments and attribute commitments to one another. They give credit to one another for being entitled to commitments and occasionally blame one another for commitments undertaken irresponsibly. They also award points, so to speak, for commitments they deem correct. Any participant can withdraw a point once awarded if he or she thinks that subsequent developments in the practice of inquiry warrant such a change in the awarding of endorsement status to propositions. The way to withdraw a point once awarded is to say something like, “I thought Smith’s belief about X was true, but it isn’t.” When this happens, no aspersions need be cast on how Smith played the game, on his entitlement to his commitment. He may still count as a responsible inquirer or even be ranked among the best. A person who thinks so will be inclined to excuse Smith for failing to commit himself to views that we now accept. The way to excuse him is to say something like, “What Smith believed about X wasn’t true, but he was justified in believing it.”
It would not improve the rules of baseball if we appended to them a chapter on the nature of scoring runs, explaining that a player who touches home plate after rounding the bases has scored only if he or she has really touched it. We would be no more enlightened if we were told that scoring a run within the game corresponds to the physical state of affairs in which the player actually touches home plate. But sandlot baseball would be a much different game if players did not feel constrained to award runs by adhering to a rigorous discipline. At a minimum this discipline involves attentiveness to evidence and an attempt to avoid being wrongly influenced by wishful thinking. These non-narcissistic features of scorekeeping discipline contribute to the objective dimension of baseball as a practice. Objective scientific inquiry includes similar features, and endeavors to extend and perfect them. Its objectivity is a matter of the constraints the practice imposes on its practitioners when they make commitments and when they keep track of normative statuses involving other practitioners and the propositions to which they commit themselves. The constraints are object-directed in the sense that they involve attentiveness to something being investigated as well as disciplined avoidance of wishful thinking, rationalization, and related intrusions of “merely subjective” factors.
Ethical discourse has an objective dimension insofar as it involves constraints of this kind. Why does it make sense to draw a distinction between being justified in believing a moral proposition and the truth of that proposition? Because our way of keeping track of one another’s commitments and entitlements pertaining to ethical topics includes the same implicit distinction between the “how-you-play-the-game” factor and the correctness of commitments. The fact that ethical topics are themselves often practice-dependent normative statuses (and not natural properties) does not diminish the need to earn entitlement to one’s commitments about them by attending to matters other than one’s own subjective states. And whether one’s commitments, attributions of commitments, and attributions of entitlements in ethical discourse are correct is not a matter of willing or wanting them to be so.
A divine-command theorist might want to insist that an adequate model for tracking the attitudes of our interlocutors in ethical discourse would have to include a single ultimate authority figure. The rules of major league baseball and major league soccer designate the head umpire or the referee as the only scorekeeper. A run or a goal can then be defined as having been scored when and only when the officially designated scorekeeper says so. Brandom’s theory shows that a discursive practice can be objective in the sense at issue here without being construed on an authoritarian model of scorekeeping. If Brandom is right, in a democratic society, where no monolithic authority is recognized by all those engaging in ethical discourse, it should still be possible in principle to make sense of being entitled to commitments and of making commitments that are correct in content. For the same reasons that baseball can be played on the sandlots and soccer can be played in the streets, ethical discourse can retain an objective dimension without there being a single authority on questions of truth and falsity. In ethics, as in most other forms of objective discourse, we are all keeping track of our interlocutors’ attitudes, as well as our own.1
Let us consider the example of soccer a little more closely. Before human beings invented this practice, there was no such thing as the normative status that soccer people refer to as “having committed a foul.” This normative status is a creature of a social practice in which people take one another to have committed a foul or not when competing with their opponents on the playing field. But this dependence of a normative status on a social practice does not in the least deprive claims about fouls in soccer of their status of being true or false. Everybody who cares about soccer, and understands how it works, behaves as if claims about this status have a truth-value. They treat inquiry into the truth-value of such claims as an objective affair, to be settled by the testimony of trustworthy eyewitnesses, the evidence of instant-replay videos, and so forth.
Before the officials of British public schools began formalizing the proprieties of soccer in explicitly stated rules, the norms governing the sport were entirely implicit in what soccer players did. By the middle of the nineteenth century, some people hoping to reduce the mayhem of their recreational activities had become disposed (a) to stop play when an especially brutal “hacking” occurred in a game of soccer and (b) to award the ball to the side that had been “hacked.” The implicitly recognized impropriety of hacking was later made explicit in a rule against fouling, which formally mandated (a) and (b) as a penalty for fouling. Once the “Laws of the Game” had been formulated, this rule was subject to critical review and rational revision. The tackle from behind has recently been formally prohibited as a result of this critical process. The reasons for introducing this particular rule happen to be good ones. Just as the application of established rules is an objective affair, so, too, is the rational revision of the rules in light of the pleasures, excitements, and other goods to be enjoyed when playing (or watching) the game.
“Norms,” according to Brandom, “are not objects in the causal order” (MIE, 626). He immediately adds, however, that normative attitudes are in the causal order. Kant was right, according to Brandom, in holding that normative attitudes institute normative statuses. Hegel agreed with this central Kantian thesis, but he criticized Kant for explaining the process of norm-institution as a matter of self-legislation and for modeling society on a contract among independent rational agents. Hegel argued that social-contract theory is insufficiently sociological or historical to account for the institution of ethical norms. Norms do not arise because solitary individuals, already in full possession of practical rationality, commit themselves to a social contract. They emerge out of the mutually recognitive activities through which a people comes to share a culture. Ethical norms are instituted, according to Hegel, not by something analogous to a contract (among already rational independent operatives) but by a form of ethical life (which shapes the subjectivity and rationality of the individuals who participate in it). This form of life gives birth simultaneously to ethical norms and to the rational individuals who use, scrutinize, and revise such norms.
Democratic ethical norms were instituted, then, in the same way that soccer players instituted the normative statuses associated with fouling. Before there was a rule against fouling, soccer players acquired a habit of stopping play in response to instances of hacking. By the same token, democratic ethical norms originally took shape in shared dispositions to respond in certain ways to behavior of certain kinds. It is a small but momentous step from this humble beginning to the explicit statement of rules and then to the critical scrutiny and rational revision of rules in light of the purposes they serve (the rationally self-conscious form of life they make possible). This conception of ethical norms as creatures of a social practice may originate in Hegel, but it is reformulated, without the obscurities of Hegelian diction, in American pragmatism. While this conception stops short of the objectivist picture of norms as wholly independent of human activity, it nonetheless allows us to affirm that ethical discourse is an endeavor in which truth-claims are put forward and entitlements to them are assessed by attending to objective considerations.
The normative claims soccer players make about who fouled whom can clearly be true or false. Whether it is true that Beckenbauer fouled Charlton in the fifty-third minute of the 1966 World Cup Final depends on the physical relationship obtaining at that time between the relevant body parts of Beckenbauer and Charlton. Clearly, this is not a question about our subjective states. If we set out to determine responsibly whether Beckenbauer fouled Charlton, we will need to position ourselves advantageously within the physical world and interact causally with such physical objects as the shins, ankles, feet, and elbows of the two players. Of course, in doing so we will be taking for granted, at least for the time being, what it means for something to be a foul. We will be taking for granted that the proprieties of soccer are what they are (in a way that makes us complicit in the valuations of the relevant social practice). But nothing prevents us, a moment later, from subjecting the Laws of the Game, as they now stand or as they stood in 1966, to critical scrutiny. Whether a particular rule should be changed is itself obviously a question on which objective considerations have a bearing. So there is every reason to suppose that talk about soccer fouls and norms has a truth-claiming, objective dimension.
There is also much talk in soccer about excellence, and this too has an objective dimension. Beckenbauer and Charlton are both universally acknowledged as excellent players, and Beckenbauer is admired as one of the four or five greatest players the sport has seen. The esteem these players enjoy is either merited or not. Either they actually possessed the excellence attributed to them or they did not. Anyone who spoke of Charlton as a poor or mediocre player would be speaking falsely—and also exhibiting either ignorance of Charlton’s play or incompetence as a judge of such matters. Truth and falsity are at issue here. In declaring Charlton excellent, one makes a normative judgment. It is possible to make the relevant norms explicit by stating a standard of excellence for soccer players or for players of a certain type. In some cases this will involve treating a particular player as an exemplar. In other cases it will involve projecting an ideal in which the virtues of several good players are imaginatively enhanced and combined in a way not yet instantiated. But the standards change over time. One thing that counts in favor of Beckenbauer’s greatness is the way in which his play as a defender transformed the standards by which defenders have subsequently been judged.2 The standards changed in response to specific features of Beckenbauer’s play—his skill as a passer, his ability to join in the attack without weakening the team’s defensive configuration, and so forth. These were objective considerations. The need to rely on conditioned norms, which evolve along with the game itself, does not make judgments about excellence a merely subjective matter.
Needless to say, ethical norms are much more important in most contexts, and much harder to assess critically, than the proprieties and ideals of soccer are. But if ethical norms and soccer norms are, from a point of view like Hegel’s, Dewey’s, or Brandom’s, instituted in roughly the same way, then there is no reason to suppose that the adoption of this form of pragmatism deprives ethical discourse of its title to objectivity.3 Pragmatism does not reject the objective dimension of ethics, but it does doubt the adequacy of two sorts of philosophical theorizing about it. The first of these are objectivist accounts of moral objectivity, which fail to do justice to the responsibility we bear for the object-oriented norms we apply in ethics. The second are a series of subjectivist reactions against objectivism, which either try to pull the objective rabbit out of a subjective hat or abandon the ideal of objectivity altogether. What I am calling pragmatism here is not to be interpreted as a form of subjectivism, but as a third way. It offers a social theory of moral objectivity—according to which both objective ethical norms and the subjectivity of those who apply them are made possible in part by social interactions among individuals. But if it is crucial to distinguish pragmatism from subjectivism, it is no less crucial to specify what kind of social theory it involves.
Consider the impropriety, other things being equal, of hitting somebody over the head with a sledgehammer. On the account being presented here, this moral impropriety was initially implicit in the dispositions of people to respond negatively to one sort of harmful behavior, to take it as having harmed or breached the relationship among persons, as well as the person harmed. There are various ways in which a community can make this normative status explicit. We might, for example, attribute to everyone the right not to be treated in this way. Once this right is recognized, we are in a position to ask whether, in any given case, it has been violated. This question is obviously every bit as objective as the question of whether Beckenbauer fouled Charlton in a soccer match. But what about the question of whether individuals really have the right we have attributed to them?
This question gets us on the wrong track if we take it as an invitation to do an inventory of everything there really is. Pragmatism construes rights and obligations as creatures of social practices. Does not anything created by a social practice exist? Of course it does, but some thinkers have wanted to contrast social phenomena with, say, physical objects, concluding that the former are merely subjective or ontologically queer, whereas the latter alone should be counted as first-class entities. Pragmatism involves no such contrast. The use of “really” in the question, “Do individuals really have this right?” is cautionary. First, it helps us raise the possibility of a revision in our norms or even in the vocabulary in which we state our norms. This is not a question to be settled by doing an ontological inventory or a typology of first-class and second-class ontological categories. Second, the term helps us focus on the possibility that we might all be wrong to attribute this particular right or to embrace the underlying norm.
For practical purposes, we can often replace the previous question by asking, “Do we (really) have sufficiently good reason, all things considered, to attribute the right not to be smashed on the head with a sledgehammer?” The question is analogous to the question, “Do we have sufficiently good reason to remain committed to the rule against tackling from behind in soccer?” Both of these questions have the merit of directing our attention explicitly to reasons, to rational considerations that would count for or against the rule being discussed. The answers to both of these questions happen to be clear, for we have every reason to remain committed to our conception of these particular proprieties. We know this because we can easily survey all of the considerations that are likely to prove relevant. When we do so, we find that they all line up on one side of the question. There is nothing arbitrary about our answers at all.
In controversial cases, however, we sometimes feel tempted to say that the decision of whether or not to revise a norm is a matter of subjective choice. This feeling can induce the worry that the entire practice is, at bottom, a reflection of arbitrary will, a subjective affair. But in such cases what we are actually faced with is a conflict or balance of rational considerations, not an absence of such considerations. The endeavor of weighing those considerations can be an objective affair (in the only sense worth bothering with in this context), even if it does not lead to an unambiguous result. Why? Because the considerations we are weighing direct our attention away from our subjective states to how things and persons are in the world and how things and persons would be if we revised our norms. As responsible subjects we must still weigh these considerations in light of what we have come to care about as a result of participating in our evaluative practices. And the considerations themselves, the reasons that guide us when we undertake our ethical commitments, cannot be understood apart from the norm-instituting social practices in which we offer and ask for them. For it is those practices, not a lonely subject’s arbitrary act of will, which confer on our reasons whatever content and force they have.
At this point in the argument, worries over the distinction between truth and justification are bound to resurface. For it might appear that even if my pragmatism accounts successfully for the relativity of justification to social context, the nonrelativity of truth could never be made consistent with the priority of social practices. Does not my emphasis on social practices require me in the end to view truth as a matter of communal agreement? Am I not proposing to substitute the community of competent ethical judges for the eternal law and the self-legislating Kantian will as an indefeasible authority on ethical truth? And would this proposal not just collapse truth into justification all over again, thus threatening the social critic’s cautionary use of “true”?
The answer to all three of these questions is no. To see why, consider the following passage from Lovibond’s defense of Wittgenstein against similar charges:
[A]lthough (in Wittgenstein’s view) it is an agreement, or congruence, in our ways of acting that makes objective discourse materially possible, this agreement does not itself “enter into” the relevant language-game: when we ask a question about some aspect of reality, we are not asking for a report on the state of public opinion with regard to that question, we are asking to be told the truth about it. … The idea of rationality as resting upon a consensus, then, does not imply that the fact of consensus need carry any weight with us in any particular piece of thinking about the objective world: a point which is demonstrated by the absence of any logical (or “grammatical”) objection to statements of the form: “I’m right and everyone else is wrong.”4
There is indeed a sense in which our use of the term “true” rests on agreement within a social practice. But the agreement that matters in this context is not agreement specifying which substantive ethical claims are true, as if the community as a whole could decide such things by fiat. It is rather a practical agreement that we exhibit when we use the terms “true” and “justified” differently. What we have agreed to do, in effect, is to treat truth in practice as something that cannot be settled simply by communal agreement. It is this underlying social agreement on the use of certain words in the process of self-criticism that gives the term “true” its nonrelative sense. When I insist on the importance of the underlying social agreement, I am not saying or implying that in deciding which ethical claims to accept as true, one must ultimately bow to the will of the community as a whole or to the majority of its members.5 The agreement at issue here operates on a different, more basic level.
Brandom makes the same point by denying the adequacy of what he calls “I-we construals of the social practices in which conceptual norms are implicit.”
Such construals fund a distinction between what particular individuals treat as or take to be a correct application of a concept, on the one hand, and what is a correct application, on the other, by contrasting individual takings with communal ones. This is the standard way of understanding objectivity as intersubjectivity. The cost of adopting this way of understanding the significance of the social dimension of discursive practice is, unacceptably, to lose the capacity to make sense of the distinction between correct and incorrect claims or applications of concepts on the part of the whole community. (MIE, 593f.; emphasis in original)
It is very important to see that in endorsing the view that norms are creatures of social practices, I am not endorsing a reduction of truth or objectivity to what some community takes to be true or objective. Our socially instituted conceptual norms do not in fact permit such a reduction.
Brandom specifically rules out what he calls an “I-we” account of discursive social practices for the following reason:
I-we accounts mistakenly postulate the existence of a privileged perspective—that of the “we,” or community. The objective correctness of claims (their truth) and of the application of concepts is identified with what is endorsed by that privileged point of view. The identification of objectivity with intersubjectivity so understood is defective in that it cannot find room for the possibility of error regarding that privileged perspective; what the community takes to be correct is correct. (MIE, 599; emphasis in original)
According to an I-we theory of discursive social practices, ethical truth can be defined as that to which one’s community agrees. It “is settled in advance that any perspective from which a distinction appears between how things seem from [the community’s] privileged point of view and how things in fact are is itself without any authority at all” (MIE, 600). Brandom’s I-thou theory, in contrast, treats each individual’s perspective as “at most locally privileged in that it incorporates a structural distinction between objectively correct applications of concepts and applications that are merely subjectively taken to be correct.” As Brandom puts it, what is shared by the discursive perspectives of each “I” and each “thou” is “that there is a difference between what is objectively correct in the way of concept application and what is merely taken to be so” (MIE, 600; emphasis in original). Interlocutors do not, however, necessarily share a view of what this difference is. What they share is a structural distinction implicit in the cautionary use of “true,” not the content of any particular belief about what is true.
It should be clear, I hope, that this is the same point Lovibond was making when she said that the agreement that makes objective discourse materially possible “does not itself ‘enter into’ the relevant language-game.” What Brandom adds to this point is an analysis of the perspectival structure of objectivity.
INDIVIDUAL AND COMMUNITY
Brandom’s account of the social dimension of discursive practice “focuses on the relation between the commitments undertaken by a scorekeeper interpreting others and the commitments attributed by that scorekeeper to those others.” The “crucial feature” of this relation, he says, “is the symmetry of state and attitude” between I and thou—between the person who is ascribing commitments and the one to whom those commitments are ascribed (emphasis in original). Conceptual contents, Brandom says, “can be specified explicitly only from some point of view, against the background of some repertoire of discursive commitments, and how it is correct to specify them varies from one discursive point of view to another.”
Mutual understanding and communication depend on interlocutors’ being able to keep two sets of books, to move back and forth between the point of the view of the speaker and the audience, while keeping straight on which … commitments are undertaken and which are attributed by the various parties. Conceptual contents … can be genuinely shared, but their perspectival nature means that doing so is mastering the coordinated system of scorekeeping perspectives, not passing something nonperspectival from hand to hand (or mouth to mouth). (MIE, 590)
For this reason, each scorekeeper must exercise his or her own authority, on a strictly local basis, to distinguish between what is correct and what is merely taken by someone else to be correct. For any two interlocutors, this relation will be symmetrical. I draw the distinction from my point of view, in light of my collateral commitments, taking my specifications of the conceptual content of commitments as authoritative. You, as my interlocutor, draw the distinction from your point of view, in light of your collateral commitments, taking your specifications of the conceptual content of commitments as authoritative. The authority each of us exercises in this process is defeasible, in the sense that we are both free to change our minds with respect to these specifications and also in the sense that neither of our perspectives holds sway over the discourse that transpires between us. But we cannot get by without taking a specific point of view, and when we do this, we are bound to draw a distinction between “what is correct and what is merely taken to be correct, between objective content and subjective view of it” (MIE, 601). If I did not draw such a distinction, I would be utterly incapable of keeping track of how other people’s commitments differ from mine, in content and in entitlement, and thus incapable of participating in the social exchange of reasons that confers significance on my own attitudes. Once I do draw this distinction, however, I am already well on the way toward being able to grasp the difference between truth and justification.
Democratic ethical discourse is social, then, not in the sense that the community of committed democrats functions as an ultimate authority, which no individual can in principle oppose. Rather, it is social in the sense that it needs to be understood in terms of what the individual members of a group do when they keep track of their interlocutors’ commitments from their own perspectives. “This symmetric pair of perspective types, that of attributor and attributee, each maintaining this fundamental normative distinction [between what is correct and what is merely taken to be correct] is the fundamental social structure in terms of which communities and communal practice are to be understood” (MIE, 601). As in street soccer or sandlot baseball, all of the participants have the authority to “keep score,” and each of them necessarily does so in light of his or her already-adopted commitments. That I have the authority to track commitments and entitlements, and thus to draw the fundamental normative distinction from my own point of view, does not make my commitments correct; nor does it make me entitled to them, in the sense that entails being epistemically justified in holding them. It simply puts me in the democratic game of giving and asking for reasons. “Sorting out who should be counted as correct, whose claims and applications of concepts should be treated as authoritative, is a messy retail business of assessing the comparative authority of competing evidential and inferential claims” (MIE, 601).
What does an ethical community share that makes it an ethical community? My answer thus far has been: a way of thinking and talking about ethical topics, or more precisely, a discursive social practice. One danger in this answer is that it can easily be interpreted in communitarian, conventionalist, or contractarian terms—that is to say, in terms that ascribe perfect, unassailable authority to something communal that the individual must obey on pain of rational incoherence or unreasonableness. We can avoid this danger by adding to my initial answer a claim about what makes a discursive social practice social. It is not the indefeasible authority of the community, its conventions, or a social contract that makes such a practice social. It is rather the need for each participant in the practice to keep track of other participants’ commitments and to assess those commitments from his or her own point of view.
No ethical community could sustain a discursive practice without imposing on each of its members the necessity of keeping track of the normative attitudes and entitlements of their interlocutors, because without this there would be no communication—and therefore no exchange of reasons—among them. But, as we have seen, ethical communities have different ways of going about their discursive business. They employ different concepts, which implicate them in different material inferential commitments and dispose them to have different noninferential responses to things going on around them. They also distribute discursive authority very differently. Adopting an I-thou model of sociality for all discursive practices allows us to make room for the full range of variation in the “default-and-challenge structure of entitlements” in different sorts of ethical communities. Where the default position in a given community is that the ethical judgments of those in ecclesial or political office are correct, we have a pattern of deference to one kind of authority. The authoritarian extreme is reached whenever such a default position is treated as indefeasible and the authority is treated as self-interpreting. As one approaches this extreme, ethical truth is reduced in practice to what the highest authority says. When the reduction is total, it no longer makes sense to claim that the highest authority says P, but P might not be true.
One should not assume, however, that all religious communities are huddled near the authoritarian extreme of the spectrum. Many religious traditions and movements have developed relatively flexible structures of authority, and even those best known for their official rigidity are rarely able in practice to stamp out critical questioning of allegedly indefeasible authorities. In most cases, the indefeasibility asserted in doctrine reflects the wishes of a particular well-placed faction more accurately than it reflects the practice of the rank and file. Roman Catholicism, for example, is not only the tradition of Vatican I, Pius IX, and Cardinal Ratzinger, but also that of Vatican II, John XXIII, and Lord Acton. The official doctrine of papal infallibility is itself treated by many Catholics as a defeasible teaching—as something to be subordinated to the Spirit that is manifest in the life of the church as a whole. Many Protestants, Jews, and Muslims who begin by attributing indefeasible and unique authority to a revealed law end up interpreting that law in a highly flexible way. All of these traditions have been more pluralistic and less authoritarian in practice than some of their officials and intellectuals have wanted them to be. Religious recognition of the faithful as a common body and of the need to conform oneself to the best available understanding of what membership in that body involves can be fleshed out in many ways, only the most extreme of which deserve to be impugned as inflexible and uncritical. What Brandom calls the “I-we” model of sociality applies only to communities at the authoritarian limit where the distinction between what the group holds and what is true melts away, and individuals disappear without remainder into the collectivity. The “I-thou” model is preferable because it allows us to see how the “notion of a discursive community—a we—is built up out of” the actions and reactions of the individuals involved (MIE, 508; emphasis in original). Deferential activity on the part of individuals is what confers authority as a normative status on those who exercise it. Communities take shape only insofar as their members perform the work of mutual recognition, whether they are aware of this or not.
Many critics of democracy portray it as a tendency to attribute indefeasible authority to the ethical opinions of the citizenry taken as a whole or on average. That modern democratic culture can devolve into this sort of authoritarianism is both clear and cause for grave concern, especially when the people are corrupted by fear and resentment of their enemies and by a relentlessly homogenizing stream of mass entertainment, manipulative advertising, and political pandering. Perhaps an essentially conformist type of sociality—absorption into the demos—is the one most likely to result from the dissolution of older communities in the age of capitalism and counterterrorism. But this is precisely what Whitman was warning against, not arguing for, when he spoke of the people as “canker’d, crude, superstitious, and rotten.”
The only defensible form of democratic community is one in which ethical authority is treated as an entitlement (to deference) that one must earn by repeatedly demonstrating one’s reliability as an ethical judge. This is not achieved by conforming one’s opinions to those of the majority, but rather by expressing judgments that withstand critical scrutiny in a discussion open to all. The “default-and-challenge structure of entitlements” in a genuinely democratic ethical community is not to be understood in purely procedural terms. It is intimately connected with substantive commitments to mutual respect, equal voice, and the value of critical discussion. But aren’t these substantive commitments held by democrats to be indefeasible? No, they are not; nor need they be. They certainly enjoy a default status as presumed correct until or unless devastating considerations are brought to bear against them, and no committed democrat expects such considerations to emerge. But this does not mean that these commitments have been removed from the agenda of discussion. To the contrary, the tradition of democratic reflection subjects them to relentless critical scrutiny. The question of whether central democratic commitments can with-stand such scrutiny is always in order, and has been discussed in earnest in every modern democracy throughout the last two centuries. If this question had been taken off the agenda, Whitman could easily have ignored the challenge of Carlyle’s Shooting Niagara in the 1860s, Baldwin and Ellison could have ignored the challenge of Black Nationalism in the 1960s, and we could ignore the challenge of the new traditionalism today.
A democratic ethical community at least permits and at best encourages its members to express their normative stocktaking in public and to include the wealthy and powerful among those who are publicly held responsible for what they say and do. Ideally, it also invites its members to resist their own absorption into the social mass and to cultivate whatever virtues are required to foster the development of novel forms of action, speech, association, and selfhood. Whitman calls this the “principle of individuality.”6 A self-consciously democratic ethical community is aware of itself as a community of individuals: each of whom has evaluating to do that no one else can do on his or her behalf; each of whom stands in a potential relation of dialogical exchange with each other; each of whom can be challenged to offer reasons to the rest; and each of whom is ultimately responsible not only for actions taken and commitments undertaken, but also for the self he or she has become through the exercise or neglect of expressive freedom.7 It is no less a “we” for that.
The notions of democratic individuality and expressive freedom bring us back, once again, into the area in which Emersonian perfectionism has staked out its strongest claims in ethics and political theory. Modern democracy began in a series of protests and revolutionary upheavals driven by resentment of specific injustices and the reasoned insistence on holding rulers discursively and politically responsible to the people for their actions. This was primarily what philosophers call a deontic matter, an issue of principle. It involved the assertion of entitlements against monarchical and theocratic orders intent on denying them. Its first preoccupations were expressed in terms of negative freedom, of rights against the established order.
But the political consequences of democratic moralizing eventually transformed the received ethical culture by opening up a space in which the ideals of democratic individuality and expressive freedom could be self-consciously pursued. Writers like Emerson and Whitman pursued these ideals in such a way that both democratic moralizing and democratic political arrangements came to seem subservient, in their eyes, to the positive freedom that results from self-cultivation. They were therefore attracted to the idea that the authority of any state to exercise political constraint depends on the success with which it protects or cultivates the expressive freedom of its citizens.8 This thesis in political theory has an ethical counterpart, which is the notion that how to make appropriate use of expressive freedom in the development of one’s character and individuality is the central question of ethics. Preoccupation with rights and wrongs, however important this may be when horrible injustices need to be opposed, threatens to engulf individuals in resentment and to distract them from taking full responsibility for what they have become as human beings. The most challenging democratic thinkers care at least as much about character and selfhood, about what the philosophers call “aretaic” questions, as they do about the rules of proper conduct. As theorists of virtue they are anything but levelers. What they want to promote in society generally and hope to exemplify in their own lives is excellence—and, if possible, spiritual greatness. The mere conformity to custom or convention exhibited in the life of the average citizen is something they abhor, and seek to disturb, in the name of democracy.
APPENDIX: METHOD IN COMPARATIVE ETHICS
The process of reason-exchange in any discursive community is constrained by socially engrained (but defeasible) assumptions about discursive authority and responsibility. These assumptions specify the circumstances in which one is, by default, entitled to a commitment or entitled to raise a challenge. They also specify the circumstances in which entitlement to a commitment or a challenge must be earned through reason-giving. Against the background of these assumptions, we engage in the inferential activity of reason-giving. That is, we make claims and assemble them into argumentative form by treating some claims as premises and others as conclusions. And by exchanging claims and arguments with one another, we bring about changes in the normative statuses—the commitments and entitlements—we have reason to attribute to one another.
The inferential activity of reason-giving is hardly the whole of ethics as a social practice, but it is central to it. Indeed, it is because the claims in which we express our commitments can play the role of premises or conclusions in inferences that they have conceptual content (MIE, 214). If we did not have inferential commitments, if we were not committed to treating one claim as a reason for another, the claims themselves would lack significance. To mean something in particular, they need to be taken to imply something and not to imply something else. Their inferential relations confer significance—conceptual content—upon them.
How do we know, for example, that a given community possesses the concept of cruelty that I have attributed to modern democracies? One bit of positive evidence would be acceptance of a constitution that treats the cruelty of a form of punishment as sufficient reason for declaring it unacceptable. A group—for example, a gang, a wayward military platoon, or a sports team—that never treated the claim that an act is cruel as a reason against performing the act would not attribute the same significance to cruelty that modern democratic constitutions do. In other words, they would not endorse the modern democratic concept of cruelty. The claim that an act is cruel, then, has the conceptual content it has within a discursive practice at least in part by virtue of what can be appropriately inferred from it, according to the norms implicit in that practice. Clearly, it is not formal logical relations that are at issue here. No one supposes that “I shall not do X” follows from “X is cruel” simply by virtue of logical form. What matters is that participants in the discursive practice treat the inference from “X is cruel” to “I shall not do X” as materially sound in the sense introduced in chapter 8. Material inferential commitments involving moral expressions like “cruel” confer conceptual content on those expressions.
Jewish, Muslim, and Christian communities have often treated anyone who is committed to “X is against God’s commandments” as entitled by default to infer “X is morally wrong.” It is beside the point to object to this practice by saying that no “ought” statement can validly be inferred from an “is” statement, for the inferential commitment at issue here is obviously a material one. No one supposes that the inference from “X is against God’s commandments” to “X is morally wrong” is good by virtue of its logical form. Theologians can make their material inferential commitment explicit in the form of a claim by asserting the conditional, “If X is against God’s commandments, then X is morally wrong.” Or they can achieve a comparable degree of explicitness by asserting the principle, “Morally speaking, thou shalt not act against God’s commandments.” The conditional and the principle codify the underlying commitment and reflect a conceptual connection, in the linguistic practice of these communities, between God’s commandments and moral wrongness. One type of divine-command theory in ethics consists of the claim that “morally wrong” means “against God’s commandments.” If material inferential commitments do confer conceptual content on the expressions caught up in them, as Sellars and Brandom argue, then this sort of divine-command theory can be seen as a perfectly legitimate attempt to make explicit the conceptual content of the relevant terms.
But two points need to be kept in mind here. First, the theory accounts semantically only for communities where the underlying material inferential commitments are already operative. It does not give an accurate account of how the expression “morally wrong” behaves in religious communities where God is not conceived as a commander or in secularized discursive practices where nothing can be taken for granted about the existence, nature, or acts of God. Hence, it does not give an account of what “morally wrong” means simpliciter. According to the approach I am defending, the meaning of a concept is always the meaning of a concept in a particular discursive practice. The semantics of ethical expressions must be done for one community at a time. Second, the role of a semantic explication like the one we are considering is expressive. It makes explicit in a semantic idiom precisely what conditionals and principles make explicit in slightly different ways—namely, the underlying material inferential commitments of the community being investigated. It does not tell us what the underlying commitments ought to be. So even if the semantic analysis correctly explicates a given community’s inferential commitments, this does not show that some other community should change its inferential commitments to conform to the expectations of a divine-command theory. Nor does it show that a Jewish, Muslim, or Christian community should remain committed to its current inferential habits.9
If I am right in affirming the expressive role of principles, then it is a mistake to think that any community’s ethical discourse can be reduced to the most general norms accepted by its members. My approach does not assume, as David Little and Sumner Twiss did in a book that has become the standard methodological text in comparative ethics, that “all codes of conduct can be reduced to a basic norm, or set of norms.”10 This sort of reduction could be carried out only if ethical reasoning proceeded in but one direction—as Little and Twiss originally put it, in an “appellate pattern,” with the most basic, general rules serving as the final court of appeal.11 But in fact, as Little now grants, much ethical reasoning uses considerations other than basic norms to call such norms into question.12 If “basic norms” are held accountable to other sorts of ethical commitments, as well as vice versa, there is no way to reduce codes of conduct to basic norms, as Little and Twiss originally proposed. Little’s subsequent admission that ethical reasoning is not unidirectional should not, then, be viewed as a minor adjustment to the method for which he and Twiss are known. It opens the door to an inferentialist pragmatism of the sort I am defending here.
Another point worth keeping in mind is that the explicit codes and semantic theories officially promulgated in a given community do not always accurately reflect the actual inferential commitments of its members. There are several reasons for this. First and most obviously, not all official spokespersons possess the skill, time, or conceptual resources to explicate the inferential commitments of their communities in an accurate or revealing way. Second, actual inferential commitments are constantly being revised in response to new circumstances, while officially promulgated codes change slowly or not at all. So even in the best of circumstances, there is often a time lag to be taken into account. Third, the inferential, cognitive, and practical commitments of a community can have implications that escape the notice of most or all of its members. Where such implications are at odds with the official codes or semantic doctrines of the community, the resulting contradictions or tensions are often of great interest to ethical inquiry. In some cases they are the as-yet-unnoticed seeds of revolutionary change. They can also provide evidence of wishful thinking, rationalization, or other forms of ideological distortion that effectively insulate the group from critical questioning. Finally, the codes and doctrines of a community are typically the work of elites that have their own axes to grind as explicators. Any elite has an interest in projecting its own commitments as normative for the larger community for which it claims to speak. Such an elite can easily be tempted into declaring that a particular office holder’s edicts or a particular text—as interpreted by the elite itself—possesses indefeasible authority. It is one thing for such authority to be asserted, and another thing for the members of the community to defer accordingly. We should therefore always be wary of the biases that are expressed in codes and doctrines, as well as the possibility that codes and doctrines represent unsuccessful attempts to enforce uniformity against the grain of communal practice. The question “Whose commitments are actually being expressed here?” is always worth asking.
For all of these reasons, then, it is unwise to take established moral codes, the officially endorsed “action guides” of ethical communities, at face value (as Little and Twiss appear to do in Comparative Religious Ethics). I am arguing for an approach that interprets codes expressively, contextually, and suspiciously. The ethical concepts of a community reside largely, according to my approach, in the material inferential commitments of the group’s members, and to get at these we need to treat codes as a limited, sometimes severely misleading, source of evidence. If codes of conduct are understood in relation to the material inferential commitments they make explicit, then it is a mistake to identify norms (as Little and Twiss tend to do) with rules or law-like principles stated in the form of propositions or prescriptions. The pragmatic approach being pursued here assumes, on the contrary, that norms are initially implicit in practice. Material inferential commitments are fully capable of guiding action, of functioning normatively, without taking the explicit form of a rule. Hence, norms are not to be identified with rules.13 Rules do not have the primacy in my approach that they have for Little and Twiss, who see the ethical reasoning of a community as essentially an expression of commitment to an explicitly stated basic rule or set of such rules. From my pragmatic point of view, their approach not only pays insufficient attention to the criticism of basic rules in terms of other ethical considerations, it also gets the expressive relations between inferential practice and rules exactly the wrong way around. Hence the importance of focusing on discursive practices rather than simply on codes in comparative ethics.