by Gerald Gaus
2.5 Markets
Although D’Agostino is surely correct that markets employ “separation,” their place in the Open Society is rather deeper. They are not simply one more jurisdictional device. To be sure, agents in the market pursue their evaluative standards as each understands them, but markets also are a fundamental way in which different perspectives can interact without normalization. Consider again Muldoon’s insightful analysis of the way in which different perspectives can bargain about some social object, without agreeing about the object’s features, and so in a fundamental sense, without truly agreeing what they are bargaining over (§IV.1.3.1). While I do not believe the moral constitution should itself be conceived as such a bargain, in market transactions this is precisely what constantly occurs. Alf wishes to sell his 1998 Subaru Outback; he sees it as a worn-out symbol of sensible middle-class attitudes and seeks something new, flashy, and more fun to drive. Betty is an environmentalist, who recognizes that driving used but reasonably fuel-efficient cars decreases her carbon footprint; and as an untenured assistant professor in a college of humanities, she sees it as showing her colleagues that she is not captured by American car culture and takes only the most utilitarian view toward automotive transport. But despite this, Alf can sell the 1998 Subaru to Betty, and they can agree on the terms of the exchange—though they do not agree about what Alf is selling and Betty is buying, they do, as Muldoon wonderfully demonstrates, have overlapping projections, which make their dealing possible.
Markets thus provide bridges between different perspectives; each sees the object in different ways (indeed, as different objects), but participants typically share enough so that, from the perspective of each, they can agree on what is being traded, and that each is better-off. It is thus not at all surprising that those who would have the moral constitution establish the supremacy of their perspective often disparage markets. Consider the artist who bemoans that gallery owners view his art as an investment, the environmentalist who insists that forests are not to be viewed as productive resources, the feminist who is adamant that it is degrading for women to sell their sexual favors, the conservative who insists that American flags must be revered and never used as a patch for jeans, or the Catholic who demands that stem cells not be sold for scientific research. Of course most perspectives claim that their characterization of the features of the social world (WF) is the correct characterization, and so other perspectives, which do not view the objects as possessing the same properties, are misrepresenting them. However, when, on the basis of their perspective, some insist that the characterizations of others are illicit, and so trades based on them should be prohibited, they are then insisting that incompatible perspectives be normalized away (i.e., deemed illegitimate). Such views often take the form of criticizing markets, and the deep assumption that what is traded is up to buyers and sellers.
To be sure, in any given community perspectives may converge on their conception of some objects and their properties, and so a moral constitution could prohibit some trades. More importantly, it may be impossible to allow some trades without deleterious effects on third parties, and there is no practical way to include them in the bargain.118 But as the moral framework increases the class of “taboo trades”119—trades that are conceived of as not respecting the “true character” of the objects—it is normalizing the public perspective, assuming a correct understanding of the social object and its value. To the extent it does so, the moral constitution impairs the ability of markets to perform their critical role in facilitating exchange and cooperation between those who understand their worlds very differently.
The Open Society arose in the great commercial cities of Western Europe. The widespread extent of their markets facilitated cooperation among those with deeply diverse religious and cultural backgrounds, who saw the world in very different ways. Without great reliance on markets, diverse perspectives will be unable to work out the terms of their myriad interactions—in which they do not fully agree on how to characterize their exchanges or the benefits obtained from them. No set of social rules could ever itself define the rational bargains for the daily interactions of a large, diverse society.
None of this, however, is to say that the rules of the market are beyond the ken of the moral constitution. There is good evidence that concern with fair, nonexploitative transactions is something approaching a universal feature of diverse societies.120 An especially important finding concerns play in the widely replicated Ultimatum Game. An Ultimatum Game is a single-play game between two anonymous subjects, Proposer and Responder, who have X amount of some good (say, money) to distribute between them. In the simplest version of the game, Proposer makes the first move and gives an offer of the form, “I will take n percent of X, leaving you with 100-n percent,” where n is not greater than 100 percent. If Responder accepts, each gets what Proposer offers; if Responder rejects, each receives nothing. If players cared only about the amount of X that they received, it would be rational for Proposer to, say, take 99 percent, offering Responder 1 percent. Responder would be faced with a choice between 1 percent of X and nothing; if the Responder cares only about maximizing the amount of X to be received, she will accept the offer. Since Proposer knows this, and since Proposer also will not choose less over more, Proposer will make the “selfish” 99:1 offer. This is not the observed outcome. In the United States and many other countries, one-shot ultimatum games result in median offers of Proposers to Responders of between 50 percent and 40 percent with mean offers being 30 percent to 40 percent. Responders refuse offers of less than 20 percent about half the time.121 Strikingly, while those in market societies throughout the world play Ultimatum Games in these ways, there is much more variance in small-scale, nonmarket, societies. Indeed, in some small-scale societies (the Machiguenga of the Peruvian Amazon and the Mapuche of southern Chile) the game is played in the relentlessly “selfish” way, as figure 4-3 indicates.122 The Machiguenga are essentially without markets; the Mapuche have limited acquaintance with markets. Note that fair play in the Ultimatum Game seems characteristic of market, but not nonmarket, societies. A plausible hypothesis is that market societies have developed moral rules—parts of their basic frameworks—for fair terms of interaction with strangers. The Machiguenga, for example, do not seem to have social rules regulating anonymous transactions with strangers and thus do not see anything unfair about “selfish” Proposer offers. Machiguengan Responders seem to simply view it as bad luck that they were not chosen as Proposers. The Mapuche do see “selfish” offers as unfair but do not seem to think there is a basic moral rule to which they can hold others accountable.
Figure 4-3. Comparison of Ultimatum Game results
2.6 The Moral and Political Constitutions
I have been stressing the moral constitution of the Open Society, rather than the political constitution and the legal framework. Political philosophy has typically focused on the state and the law, often ignoring—indeed denying the existence of—the moral constitution. This has led to a state-legal-centric view of the regulation of the Open Society. A typical view in political philosophy is to take it as axiomatic that if there is an injustice the state should enact a law that prohibits it; if people have a human right to perform an act, or to a performance on the part of others, the law should make it so. The truth about morality dictates what justice demands, and the law must enforce it. Gerry Mackie has recently argued that this is an error. As he points out, there are hundreds of critical cases around the world in which deeply unjust practices—among them female genital cutting, caste discrimination, child marriage—have been widely criminalized yet continue to be practiced. Laws that depart from the basic moral and social norms of a society mostly likely will be ignored, often engendering contempt for the law. As Mackie, following Iris Marion Young,123 concludes, “Criminalization is an appropriate response to a criminal injustice, a deviation from accepted norms, its harmful consequences intended, knowingly committed by identifiable individuals, whose w
rongdoing should be punished. It is not an appropriate response to a structural injustice, in compliance with accepted norms, its harmful consequences unintended byproducts, and caused by everyone and no one. The proper remedy for a harmful social norm is organized social change, not fault, blame, punishment.”124
In recent years students of social change have come to something of a consensus that effective legal regulation cannot stray too far from the underlying informal social rules.125 One of the most striking “social experiments” based on this insight was that of Antanas Mockus, mayor of Bogotá in the late 1990s and early 2000s.126 In terms of the present analysis, Mockus’s aim was to harmonize the moral constitution and legislation; he recognized that unless supported by the underlying informal moral and social framework, attempts to induce change though the law would not succeed. For example, Bogotá was characterized by a very high rate of traffic fatalities in the mid-1990s, with widespread disregard for traffic regulations. Mockus distributed 350,000 “Thumbs Up/Thumbs Down” cards that drivers could display in response to dangerous driving by others, to drive home the message that such behavior not only was illegal but violated the informal normative judgments of other drivers. Along with related programs, Bogotá witnessed a 63 percent decrease in traffic fatalities between 1995 and 2003. Similar programs based on harmonizing the law with informal social normative expectations led to decreases in water usage and, critically, homicides.
Of course the law does have a fundamental place in the regulation of interactions in the Open Society. Gillian K. Hadfield and Barry R. Weingast have developed an enlightening model of legal coordination according to which the critical function of law is to provide shared classifications of prohibited behavior among those with, essentially, different perspectives.127 In this sense law is absolutely critical in creating a shared social world. Nevertheless, the fundamental point of this section stands: political philosophers and policy experts have too often conceived of the legal system as an autonomous center of social regulation, failing to realize that its efficacy depends on harmonizing its requirements with the underlying moral constitution.
3 RULES WE CAN LIVE WITH
3.1 On Choosing without Agreeing on the Best
In the previous section I sketched a set of institutions that are hospitable to diversity as such. We now turn to the critical question of justification: in what sense can we say that a perspective on justice could endorse the rules of the public moral constitution? That is, under what conditions can Alf, holding perspective Σ, endorse these rules and institutions in a way that allows him to see them as a practice of moral responsibility to which he holds others, and accepts that others appropriately hold him accountable? I first consider the conditions under which some single perspective can do this, and then extend the analysis to examine the range of perspectives that might accomplish it.
Recall the problem with which we began this chapter. Rawls seemed unable to build his insight about deep diversity of views about justice into his contractual model. One can see why he might have been reluctant to build diversity into a contract: it is easy to assume that a contractual model that yields no unique result may be one that simply ends in a deadlock, and a deadlock is of no help at all in deciding what is just. In A Theory of Justice Rawls modeled the parties in the original position as ranking the various competing conceptions of justice in pairwise comparisons.128 But the choice depends on unanimity about what is ranked best in the set: if in your ranking you hold that x is preferred to y, and in my ranking y is preferred to x, we seem stuck. As Sen saw it, the social contract’s unanimity requirement appears to be a “straightjacket” that leaves us unable to make any choice at all.129 As Rawls came to recognize the diversity of reasonable conceptions of justice, he appeared to allow multiple original positions, which would yield multiple conclusions. But this leaves us without a conclusion about what justice is. At the end of a fifty-year quest, we have no coherent theory of justice. For some partially normalized perspectives we have one conception (say, “justice as fairness”), for another normalization, another theory. But what is the theory of justice?
Let us commence with a version of this problem. Instead of a set of theories of justice {T1… Tn}, let us substitute a set of alternative rules {R1… Rn}, which are proposed as part of the moral constitution over some matter regulating the interactions of a set of perspectives {Σ1… Σn}, which includes Alf’s perspective. We suppose these are strict alternatives in the sense that for every such rule in the set, other members provide inconsistent deontic requirements or permissions over the core cases which R-type rules are intended to regulate.130 I assume for simplicity’s sake that all perspectives (or, alternatively, all representatives of a perspective) {Σ1… Σn} strictly rank all the rules in {R1… Rn} in terms of their justice, as that perspective judges it.
It might seem that we have simply recreated Sen’s problem: in order for the choice situation to make sense, the different perspectives must be ranking the same thing, but it is precisely in their identification of the features of objects on which perspectives disagree (§IV.1.2.3). Recall, however, that a rule simply is a common public system of classification: it is a socially constructed artifact. The question for the perspectival “contractors” is how some artificial system of social classification relates to their perspective on justice: given their view of justice, they ask how well this system of social classification scores.131 For now I assume that they have a very high level of agreement as to the characterization of a given rule R; I relax this assumption in §IV.4.2.
Granting for now, then, that the perspectives agree on how they understand any given proposed rule, and given that our concern is a contract among deeply diverse perspectives, we clearly see the dead end of what we might call optimizing (i.e., best in the set) unanimity: it is almost certain that the perspectives will not agree on the best rule in the set. Of course we can guarantee optimizing by full normalization—if Alf’s perspective is the unique normalized perspective, then his choice necessarily identifies the best in the set; his choice defines the Archimedean perspective (§IV.1.1). At this point in our analysis the attractions of normalization have hopefully faded. Now ironically enough, it is Sen’s pathbreaking work on choice sets, maximal sets, and choice under incompleteness that helps rescue the contractual project from the specter of optimizing deadlock. To simplify, suppose that there are three rules in the set {R1, R2, R3}, which each perspective ranks from best to worse, and no single rule is ranked best by all perspectives. The perspectives disagree in their rankings of this set, yet, as Rawls so often stressed, they require a public moral constitution—a public social world. It will not do for some to insist that R1 is the one and only correct rule; given that other equally eligible perspectives hold that R2≻R1, how can a proponent of R1 claim that all must endorse it over R2? As Sen recognized, it is ranked higher on one evaluative perspective, but, since we have supposed that it is not the only perspective that matters, its conclusions are not determinative for other perspectives. The parties need to devise a public social world, but they do not agree about what is best.
Our parties are in a position similar to “Buridan’s ass,” the donkey who was precisely midway between two haystacks and could not decide whether to turn right (x) to eat from one or left (y) to eat from other, and ended up dying of starvation (z). Sen writes:
The less interesting, but more common, interpretation is that the ass was indifferent between the two haystacks, and could not find any reason to choose one haystack over the other. But since there is no possibility of a loss from choosing either haystack in the case of indifference, there is no deep dilemma here either from the point of view of maximization or that of optimization. The second—more interesting—interpretation is that the ass could not rank the two haystacks and had an incomplete preference over this pair. It did not, therefore, have any optimal alternative, but both x and y were maximal—neither known to be worse than any of the other alternatives. In fact, since each was al
so decidedly better for the donkey than its dying of starvation z, the case for a maximal choice is strong.132
Our contract among diverse perspectives is in the same position. If z is {¬R1, ¬R2, ¬R3}, and all concur that R1≻z, R2≻z, and R3≻z, all three rules are best elements in the set {R1, R2, R3, z}; they are all ranked higher than z, but there is no unanimous ranking of any pair in {R1, R2, R3}. However, if the parties do not choose from {R1, R2, R3} they know they will end up with z, which all deem their worst choice. Theirs is Buridan’s ass’s predicament: at this stage of their deliberation they cannot, via unanimity, select any of the three rules as the best, but if they do not choose they end up with z. And, as Sen suggests, only an ass would do that. Thus the contract among diverse perspectives may well have an outcome—a choice set (a set of best options) from which all the contractors have reason to endorse a choice. The important thing is not to get stuck with z.
3.2 The Socially Eligible Set
We can divide the set of the rules {R1… Rn} into three parts. The critical divide is between the rules in {R1… Rn} that all perspectives agree are better than “z,” which we can define as no moral rule at all on this matter. If any of the perspective-contractors rank z above some proposed rule Ri that perspective deems it better to have the moral constitution silent on this matter than to include Ri. Another way of thinking of this is that Alf’s perspective cannot acknowledge Ri as the basis of a practice of accountability; Alf is not willing to have others hold him responsible for failing to act on Ri, nor is he willing to hold others accountable. Thus, faced with a choice between conforming to Ri and simply acting as his own perspective deems best, Alf would choose to act as he thinks best. Let us, then, define the socially eligible set of moral rules as all those rules that all perspectives in the contract rank as better than z.