Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

date: 17 February 2020

Normative Methodology

Abstract and Keywords

This article shows that one should start social science inquiry with individuals, their motivations, and the kinds of transactions they undertake with one another. It specifically discusses four basic schools of social theory: conflict, shared-values, exchange, and coordination theories. Conflict theories almost inherently lead into normative discussions of the justification of coercion in varied political contexts. Religious visions of social order are usually shared-value theories and interest is the chief means used by religions to guide people. Individualism is at the core of an exchange theory. Because the first three theories are generally in conflict in any moderately large society, coercion is a sine qua non for social order. Coordination interactions are especially important for politics and political theory and probably for sociology, although exchange relations might be most of economics, or at least of classical economics. Shared-value theory may possibly turn into the most commonly asserted alternative to rational choice in this time as contractarian reasoning recedes from center stage in the face of challenges to the story of contracting that lies behind it and the difficulty of believing people actually think they have consciously agreed to their political order.

Keywords: social theory, conflict theory, shared-values theory, exchange theory, coordination theory, political theory, politics

Modern political philosophy begins with Thomas Hobbes, David Hume, and others who train their focus on the individual and on interactions between individuals. The purpose of politics in their view is to regulate the behavior of individuals to enable them to be peaceful and productive. They treat of behavior and virtually ignore beliefs. They are interested in social order and its maintenance, not in the salvation of the soul, the creation of a heavenly city, or the ideal society. Hobbes’s (1642; 1651) great works of political theory, De Cive and Leviathan, were published in the first and last years, respectively, of the English Civil Wars, one of the most devastating periods of English history. Against this background, his view of the role of political theory is the explanation and therefore the enablement of social order, a focus that continued through Locke and Hume, although they are increasingly concerned with the working of government and the nature of politics. If any of these three theorists were concerned with “the good society,” they would have meant a society that is good for individuals. In an important sense, they are normatively behaviorist. That is to say, they attempt to explain rather than to justify political institutions and behavior. They are also forerunners of the modern self-interest and rational-choice schools of social thought. They are normative theorists only in the very limited sense of explaining what would get us to better states of affairs, in the sense of those states’ being de facto in our interest or better for us by our own lights. From this vision, the main contemporary approaches to explanation derive. In contemporary normative social theory, there are three main schools—conflict, shared-value, and exchange theories—based, respectively, on interests, shared values, and agreement (as in contractarian theories of both explanation and justification).

(p. 90) The first move in much of normative social science, especially in normative political theory, is to establish a background of self-interested motivation and behavior. Indeed, the transformation of political theory by Hume in his Treatise of Human Nature is based on an account of normative issues that is not specifically a theory of those issues and how we should deal with them but is rather an account of how we see them and why we see them that way (Hume 2000 [1739–40], book 3; Hardin 2007, ch. 5). His account is essentially psychological. The way we see normative issues is to fit them to our interests. In keeping with their program to explain, not to justify, Hobbes and Hume are naturalists. Their explanations are grounded in the assumption that people are essentially self-interested and that their actions can be explained from this fact. From their time forward, the development of normative social science has depended heavily on the assumption that individuals are relatively self-interested.

1 Self-interest

One need not suppose that people are wholly self-interested, but a preponderance or a strong element of self-interest makes behavior explicable in fairly consistent terms. Consistency of individual motivations is central to the task of general explanation of behavior. Many normative or moral theories might yield explanations of behavior but only idiosyncratically, so that we can explain much of your behavior and commitments but not those of your neighbor. No standard moral theory comes close to the general applicability of self-interest as a motivation for large numbers of people.

Hobbes and Hume are not alone in this view. Bernard Mandeville,1 Adam Smith, and Alexis de Tocqueville, among many others, conclude that self-seeking behavior in certain very important and pervasive contexts promotes the good of society in the—to them—only meaningful sense, which is promoting the good of individuals. Consider Tocqueville (1966 [1835 and 1840], ii, ch. 8) who, with his characteristic clarity, justifies the interest-based normative program in a forceful chapter on “Individualism and the doctrine of self-interest properly understood.” He says that the doctrine of self-interest properly understood is the best moral theory for our time. He comes from a background in which French Catholic virtue theory was the dominant strain of moral judgment. He notes that in the United States, where he famously toured as de facto an ethnographer, there was no talk of virtue. Clearly he approves of this fact. In virtue theory, he says, one does good without self-interest. The American trick combines interest and charity because it is in the interest of each to work for the (p. 91) good of all, although they need not know or intend this. This is Smith’s argument from the invisible hand and it leads us to a resolution of the logic of collective action in the provision of large-scale public benefits. I seek my own good, you seek yours, and all together we promote the good of all. The happiness of all comes from the selfishness of each (ii. 376). Recall one of Smith’s most quoted aphorisms, that it is “not from the benevolence of the butcher, the brewer, or the baker, that we expect our dinner, but from their regard to their own interest” (Smith 1976 [1776], 1.2.2, 26–7). Arguably, Tocqueville’s central thesis is that, if you give democratic peoples education and freedom and leave them alone, they will extract from the world all the good things it has to offer (Tocqueville 1966 [1835 and 1840], ii. 543). This is, of course, a collective achievement based on individually motivated actions.

This view is not strictly only a modern vision. Aristotle states a partial version of it, in passing, in his praise of farmers as especially good citizens for democracy: “For the many strive for profit more than honor” (Politics, 1318b16–17). Aristotle says this with approval. If his claim were not true, he supposes that society would not cohere, because it is founded on the generality and stability of the motivations of farmers, whose productivity is fundamentally important for the good of all in the society. The scale of the contributions of farmers to the good of society remained relatively constant from the time of Aristotle until roughly two or three centuries ago in Europe when industrial production began to displace it as the main locus of employment. Today 2 or 3 percent of the workforce in the advanced economies suffices for agricultural production. It is an extraordinary fact that all of our main strands of political theory originate in the earlier era, when social structure was radically different.

A slight variant of the Aristotle–Hobbes–Hume view of the role of interest in the ordering of society is an assumption at the foundation of John Rawls’s theory of justice. Rawls (1999, 112 [1971, 128]; see also Hardin 2003, 3–8) supposes that citizens are mutually disinterested. By this he means that my assessment of my own benefits from the social order established under his theory of justice does not depend on your benefits from that order. For example, I do not envy you and you do not envy me. Our social order has been established as just and there is no alternative that is similarly just and that would better serve my interests.2 If we are mutually disinterested, then we have no direct concern with the aggregate outcome, but only with our own part or share in that outcome. This is a fundamentally important assumption in Rawls’s theory, without which the theory would not go, but it is not often addressed in the massive literature on that theory. But even that theory, put forward in a nonagricultural world, builds on earlier visions of society.

(p. 92) 2 Three Schools of Social Theory

One can do normative political analysis without starting from rational choice principles and, indeed, such analysis is often done as an alternative to rational choice theories. But one cannot do very systematic, coherent political analysis without a clear delineation of basic principles on which the analyses are to be built. So for example, there are three grand theories—or schools of theory—on social order, each of which is based on a systematic set of theoretical assumptions. First are conflict, as represented by Thrasymachus (in Plato’s Republic), Karl Marx, and Ralf Dahrendorf (1968; also see Wrong 1994). Hobbes is also commonly considered strictly a conflict theorist, but I think that this is wrong; that, as noted below, he is largely a coordination theorist. Conflict theories commonly turn to coercion or the threat of coercion to resolve issues. Hence, they almost inherently lead us into normative discussions of the justification of coercion in varied political contexts (Hardin 1990). They can also lead to debates about the nature of power and compliance as in Machiavelli, Marx, Gramsci, Nietzsche, or Foucault.

Second are shared-value theories, as represented by John Locke, Ibn Khaldun, and Talcott Parsons (1968 [1937], 89–94). Religious visions of social order are usually shared-value theories and, as Tocqueville notes, interest is the chief means used by religions to guide people. Religious and theological theories and justifications once held sway but are now of little import in Western social science. Now religious commitments and beliefs are merely social facts to be explained. Many contemporary shared-value theorists in the social sciences in the West are followers of Parsons. These followers are mostly sociologists and anthropologists—there are virtually no economists and there are now few political scientists in the Parsons camp. There was a grand Parsonian movement in political science from the 1950s through some time in the 1970s. The most notable and creative example of this movement is the civic culture of Gabriel Almond and Sidney Verba (1963) and others. Although there is not much of a grand-synthesis view of norms that remains in political science or even in much of sociology, there are still ad hoc theories of norms. For example, political scientists often explain the voting that occurs as public spirited, altruistic, or duty driven. And there is today a rising chorus of political scientists who take a more or less ad hoc stand on the importance of a value consensus, as represented by those concerned with the supposed declines in trust, family values, and community (e.g. Putnam 2000).

Contractarians in social theory are typically shared-value theorists. This may sound odd, because legal contracts typically govern exchanges. But social contract theory requires a motivation for fulfilling one’s side of a contractual arrangement and a social contract is not analogous to a legal contract in this respect. Because there is no enforcer of it, a social contract is commonly therefore seen to require a normative commitment—essentially the same normative commitment from everyone (see Hardin 1999, ch. 3). For example, in the view of Thomas Scanlon (1982, 115 n.; 1999; see further Barry 1995 and Hardin 1998) the motivation to keep to a social (p. 93) contract is the desire to achieve reasonable agreement on cooperative arrangements. This appears to entail a straightforward factual issue about the existence of this desire. Is this desire prevalent? Because of the difficulty of defining reasonable agreement, it seems unlikely. The methodological task of demonstrating the prevalence of such a desire seems simple enough, but the reasonable agreement theorists have not bothered to test their assumption. It seems very unlikely that there is such a desire, so that Scanlon’s contractualism cannot undergird social cooperation or, therefore, social theory. Contracts for ordinary exchanges are backed by various incentives to perform, especially by the threat of legal enforcement, by the interest the parties have in maintaining the relationship for future exchanges, or in maintaining their reputations. Social contracts have none of these to back them.

And, third, there are exchange, which are relatively more recent than the other two schools, with Bernard Mandeville and Adam Smith among the first major figures, and, in our time, George Homans and many social choice theorists and economists.3 At the core of an exchange theory is individualism. Tocqueville (1966 [1835 and 1840], vol. ii. 506–08), writing in the 1830s, says “individualism” is a new term. It turns on the calm feeling that disposes each to isolate himself from the mass and to live among family and friends. It tends to isolate us from our past and our contemporaries. The rigorous, uncompromising focus on individuals is a distinctive contribution of Hobbes, the contribution that puts us on the track to modern political philosophy and that makes Hobbes at least partially an exchange theorist. For him, the assumption of individualism is de facto a method for focusing on what is central to social order. It is also, of course, a descriptive fact of the social world that he analyzes. It becomes Tocqueville’s assumption in analyzing American society two centuries later, when it is also the basis for criticizing his own French society. He says that, at the head of any undertaking, where in France we would find government and in England some territorial magnate, in the United States we are sure to find an association (513). These associations are made up of individuals who voluntarily take on their roles; they are not appointed to these roles, which are not part of any official hierarchy. Tocqueville has a forceful method: go to the core of any activity to explain the form of its successes and failures. And when we do that for America in the 1830s, we find individuals motivated by their own interests. When we do it for France, we find government agents and regulations. Anyone who has lived in both France and the United States might reasonably conclude that the two societies have moved toward one another in this respect, but that they still differ in the way Tocqueville finds nearly two centuries ago.

Note that these three sets of assumptions—individualism, self-interest, and the collective benefits of self-seeking behavior—are the assumptions of both positive and normative theories. This should not be a surprise because the world we wish to judge normatively is the same world we wish to explain positively. Moreover, all of the normative theories we might address are likely to have positive elements that (p. 94) we could analyze from the perspective of relevant positive theories. For example, to argue persuasively for shared-value theories we must be able to show that there are shared values. This is often not done very well or even at all, but is merely assumed as though it were obvious. A fully adequate normative theory must therefore fit both positive and normative assumptions and must depend on both positive and normative methodologies. Often this must mean that the methodological demands of normative claims are more stringent than the methodological demands of any parallel positive claims. Normative claims must pass muster on both positive and normative methodological standards.

Given the pervasiveness of shared-value theories in contemporary social and political theory, we should consider whether there are shared values of the relevant kind and force. This is, again, a positive issue and it should not be hard to handle. Once we establish that there are or are not relevant shared values, we can go on to discuss how they are constructed and what implications they might have for social theory, actual institutions, and political behavior.

3 Shared Values

Suppose it is established that we do share some important set of political values, X, Y, and Z. What follows? Our shared values do not directly entail any particular actions because acting on those values might conflict with our interests in other things, and acting on our shared values might cost you heavily enough to block you from acting in our common interest. Superficially it might seem that interest, for example in the form of resources, is merely another value, or rather a proxy for values that could compare to X, Y, and Z. But this will commonly be wrong. For an important political example, suppose we are all or almost all patriotic. Your patriotism benefits me if it motivates you to act in certain ways, but acting in those ways likely has costs for you, so that although we share the value of patriotism we may not have incentives to act in ways that benefit each other. Given that we share the value of patriotism to a particular nation, we might want to ask on what that value commitment is founded. It could be founded on interests, identity, or bald commitment to our nation, right or wrong. It might not be easy to establish which, if any, of these plays a role. Tocqueville supposes that patriotism founded on interests must be fragile, because interests can change (Tocqueville 1966 [1835 and 1840], i. 373). We might also suppose that our interests in patriotism here could be compromised in favor of other interests.

Perhaps our commitment turns on our ethnic identity, as is commonly claimed for nationalist commitments. There can typically be no compromise on ethnicity and the costs of defending one’s ethnicity may be discounted heavily for that reason. You cannot trade half of your ethnic commitments for half of mine. Of course, (p. 95) the next generation might do exactly that. They might marry across our ethnic divide, engage in joint corporate activities, and have friendship groups that straddle ethnic lines. Sadly, such actions and even merely their possibility might be sources of deep conflict between our groups. On economic issues, there commonly is some possibility of compromise that lets the parties split differences to allow all to gain from staying involved with each other, even cooperating together and coordinating on many fundamentally important activities. This is, for Smith and many other political economists, a major unintended benefit of the market for exchange.

Contract or agreement theories suggest a need or at least an urge to explain why we agree, and the answer often must be that it is in our interest to agree on some particular social arrangement or that we share the values on which we are to contract. Hence, agreement theories threaten to reduce to simple interest or to shared-value theories or explanations. But even then they have a strength that shared-value theories often lack. Once your interests, pro and con, are established, there is likely no further need to explain why you act in relevant ways. Motivations and interests tend to collapse into each other if they are fully defined. Unless someone’s commitment to some value translates in standard terms into their interests (hence, the odd locution “can be cashed out” as), we still face the task of determining how that value commitment will motivate action, if at all. In sum, interest is both a value and a motivation. Shared-value theories must first establish what values are shared and then give an account of how commitment to them motivates action. Both steps here may be very difficult. Indeed, each of these steps might challenge some of our standard methodologies for establishing social and psychological facts.

An important subcategory of shared-value theory is the body of norms that regulate our behavior in social interaction. The category of norms is much broader than that for social order, but it is these that matter for political theory. We may parse the category of norms in many ways. The most common move is simply to list many norms and to apply them to particular problems, as with the putative norms on voting. In a far more systematic approach, Edna Ullmann-Margalit (1977) lays out several categories as based on the game-theoretic structure of the underlying problems that the norms help to resolve or at least address. Her deep insight is that norms must handle the strategic structure of the incentives people face if the norms are to get them to behave cooperatively. Her modal strategic categories are prisoner’s dilemma, coordination and unequal coordination, and conflict. Some of Ullmann-Margalit’s norms help us, respectively, to coordinate, to cooperate, or to manage conflict in these contexts.

It is striking that Ullmann-Margalit’s book from only four decades ago is among the first serious efforts to bring strategic analysis systematically to bear on normative theory and problems. Indeed, we might well speak of the strategic turn in social theory, a turn that has been heavily influenced and even guided by game theory, which was invented roughly during the Second World War (Neumann and Morgenstern 1953 [1944]). That turn has influenced both positive and normative theory. There are standard norms that address all of Ullmann-Margalit’s strategic categories and those norms have vernacular standing in ordinary life contexts. But Ullmann-Margalit (p. 96) shows that many norms are strategically related and thereby shows how they are grounded in incentives. In political theory, the norms that most interest us are those that regulate social order (Hardin 1995, chs. 4 and 5).

4 A Fourth Theory: Coordination

Because there generally is conflict in any moderately large society, coercion is a sine qua non for social order. But it is only one sine qua non. Two others are exchange and coordination. All are needed because the strategic structures of our potential interactions are quite varied, and we need devices for handling all of these reasonably well if we are to have desirable order and prosperity. In a subsistence agricultural society, coercion might be very nearly the only point of government. But in a complex society, coercion seems to be a minor element in the actual lives of most people, although the threat of it might stand behind more of our actions than we suppose. In such a society, exchange and coordination loom very large, radically larger than in the subsistence economy.

The three grand, broadly established schools of political thought—conflict, shared values, and agreement or exchange—are right about particular aspects of social order. But they miss the central mode of social order in a complex modern society, which is coordination (Lindblom 1977; Schelling 1960). We do not necessarily share values but we can coordinate to allow each of us to pursue our own values without destructive interaction or exchange. To grossly simplify much of the problem of social order in a complex society, consider the relatively trivial problem of maintaining order in traffic on roads. There are two main coordinations at stake. The first is the obvious one of merely getting all drivers to drive on the same side of the road—either all on their left or all on their right—in order to prevent constant accidents and difficult problems of negotiating who gets to go first. The second is the problem of controlling the flow of traffic at intersections, for which traffic signals and signs are used when the traffic is heavy enough. Two striking things about the collection of drivers are that they are not genuinely in conflict and that they do not typically have to share any general social values in order for these coordinations to work well. Furthermore, there is no exchange that they can make to solve the problems arising from their interactions. I have my purposes, you have yours, and we want merely to avoid getting in each other’s way while going about our own affairs. The seeming miracle is that often we can do all of this spontaneously. For example, some coordinations can be managed by relying on focal points (Schelling 1960) that make a particular solution obvious or on institutions, which can define a resolution. Getting everyone to drive right is an instance of the first of these devices; managing traffic flow at intersections is an instance of the second.

(p. 97) As are conflict theories, coordination is an interest theory. Hobbes is perhaps the first major coordination theorist.4 But David Hume (2000 [1739–40], book 3), Adam Smith (1976 [1776]), and C. E. Lindblom (1977) see much of social order as a matter of coordinating the disparate interests of many people. A shared-value theory could be essentially a coordination theory if the values motivate coordinated actions, but coordination does not require broadly shared values. This is the chief reason why coordination is fundamentally important in modern social and political theory. Shared-value theories typically make adherence to relevant values a matter of overriding one’s interests and, when put into political power, overriding the interests of many citizens. For example, I help to defend my community despite the risks that such effort entails, I submerge my identity in the collective identity (whatever that might mean), or I vote despite the burden to me of doing so and despite the virtual irrelevance of the effect of my vote on my interests. But against the strenuous and implausible view of Parsons, a collection of quite diverse pluralists can coordinate on an order for the society in which they seek their diverse values. In sum, coordination interactions are especially important for politics and political theory and probably for sociology, although exchange relations might be most of economics, or at least of classical economics. In a sense, the residual Parsonians are right to claim that conflict relations are not the whole of political order, although not for reasons that they might recognize. They are right, again, because the core or modal character of social order is coordination.

While at a commonsense level the problem of coordination is typically not difficult to grasp, its general significance and its compelling nature have not been central understandings in the social sciences or in political philosophy. Hobbes had a nascent coordination theory in his vision of our coordinating on a single sovereign (Hardin 1991). Had he been more supple in his views, he might have recognized that the dreadful problem of civil war in his England was a matter of multiple coordinations of various groups in mortal conflict with each other. There was no war of all against all but only war between alternative factions for rule, each of which was well enough coordinated to wreak havoc on the others and on nonparticipant bystanders. Hume made the outstanding philosophical contribution to understanding coordination problems, but his insights were largely ignored for two centuries or more after he wrote and they are still commonly misread.5 Thomas Schelling (1960, 54–8) gave the first insightful game-theoretic account of coordination problems and their strategic and incentive structures. But their pervasive importance in social life is still not a standard part of social scientific and philosophical understanding.

In social life, coordination occurs in two very different forms: spontaneously and institutionally. We can coordinate and we can be coordinated as in the two-part coordination of traffic. In Philadelphia in 1787 a small number of people coordinated spontaneously to create the framework to organize the new US nation institutionally. (p. 98) Once they had drafted their constitution, its adoption was beneficial to enough of the politically significant groups in the thirteen states that, for them, it was mutually advantageous (Hardin 1999). Therefore, they were able to coordinate spontaneously on that constitution to subject themselves to being coordinated institutionally by it thereafter. This is the story of very many institutional structures that govern our social lives, and the more often this story plays out in varied realms, the more pervasively we can expect to see it carried over to other realms and to organize our institutions, practices, and even, finally, our preferences, tastes, and values. As it does so, it might be expected then to drive out or to dominate alternative ways to create and justify our social organization.

5 Concluding Remarks

In the era of Hobbes, writing during the English Civil War, the first focus of political theory was social order in which individuals might survive and prosper. Success in managing order has pushed worry about social order out of its formerly central place, even virtually out of concern altogether for many political theorists. The meaning of justice has changed to match this development. Through Hume’s writings, justice is commonly conceived as “justice as order,” as in Henry Sidgwick’s (1907, 440) somewhat derisive term. This is more or less the justice that legal authorities and courts achieve in the management of criminal law and of the civil law of contracts and property relations. By Sidgwick’s time, it begins to be conceived as, or at least to include, distributive justice, as in the theory of John Rawls (1999 [1971]). Hume and John Stuart Mill (1977 [1861]) also shift the focus toward the institutions of government, which in large modern societies entails representative government. This move brings back classical and Renaissance political thought. It also makes great demands on causal understandings and therefore on positive theory and methodology, again tying the normative and the positive tightly together in a single account. Rawls’s theory also requires massive positive understandings when he says that now the task is to design institutions capable of delivering distributive justice, a task that he leaves to others, who have so far generally failed to take it up. Rawls’s and Hobbes’s theories are relatively holistic and general; Hume’s and Mill’s are relatively piecemeal and specific. Perhaps no methodology gives us serious entrée to handling holistic social and political theory at the level and scale required by Hobbes and Rawls. Eventually, therefore, we must want to break down the institutional moves entailed by Rawls’s theory to make them piecemeal and manageable.

It is an interesting fact that normative methodologies have changed substantially over the past several decades. Methodologies in many fields of social theory and explanation have been refined extensively during that period, especially under the influence of rational choice and game theory but few if any of them have been dropped or (p. 99) developed de novo. Today’s three leading normative methods have come into their own during that period, so much so that it is hard to imagine what normative theories would be like today without those methods driving their articulation and refinement. Developments have not been equally dramatic in all three methods. Two of the methods, shared-value and contractarian arguments, threaten to be narrowed down to use by academic moral theorists with little resonance beyond that narrow community. Any method that becomes as esoteric as much of contemporary moral theory has become is apt to be ignored and even dismissed by the overwhelming majority of social theorists as irrelevant. That would be a profoundly sad separation of normative from positive theory, the worst such separation in the history of social theory, worse than the separation of economic from utilitarian value theory wrought by G. E. Moore (1903, 84) a century ago, when he literally took utility into the vacuousness of outer space.

The theorists who work in the normative vineyard often seem to strive more for novelty than for comprehensiveness or even comprehension. Great novelty cannot generally be a worthy goal for us in social theory. The occasional major novel invention, such as Hobbes’s all-powerful sovereign as a form of institutionally enforced coordination, Hume’s convention as a form of spontaneously enforced coordination, Smith’s classical economics, Vilfredo Pareto’s (1971 [1927]) value theory, John von Neumann and Oskar Morgenstern’s (1953 [1944]) game theory, or Schelling’s (1960) coordination theory takes a long time to be incorporated into the main stream of theory and explanation of social institutions and practices. A flood of supposedly novel contributions is apt to be ignored or openly dismissed. Creativity in social theory is not likely to depend on such major innovations except on relatively rare occasions. Most of the creativity we see is in the application of well-established innovations across many realms.

Over the past four or five decades, rational-choice normative theory, the third major branch of contemporary normative methodology, has become a vast program that increasingly leaves the other two branches behind in its scope and sheer quantity of work. This development is made more readily possible by the clarity and systematic structure of game theory and game-theoretic rational choice. Game theory and rational choice methodology are very well laid out and easily put to use. Perhaps at least partially because of that fact, rational choice methods are taking over normative theorizing and theories. Early steps along the way in this seeming conquest include Richard Braithwaite’s (1955) use of game theory in moral reasoning, David Lewis’s (1969) analysis of convention in the spirit of Hume, Ullmann-Margalit’s (1977) theory of norms, and a flood of other works from the 1980s on.

In this program, method and theory tend to merge. One might wonder whether this is a typical tendency for relatively developed theories and the methods successfully associated with them. Shared-value theory is perhaps becoming the most commonly asserted alternative to rational choice in our time as contractarian reasoning recedes from center stage in the face of challenges to the story of contracting that lies behind it and the difficulty of believing people actually think they have consciously agreed to their political order, as long ago noted by Hume (1985 [1748]). But it faces a (p. 100) harder task than rational-choice normative theory because it has barely begun at the basic level of establishing a set of demonstrably shared values other than own welfare. Own welfare is, of course, the shared value that shared-value theorists most want to reject, although one wonders how many of the most ardent opponents of that value as a general guiding principle in social theory would actually reject that value in their own lives.


Almond, G. and Verba, S. 1963. The Civic Culture: Political Attitudes and Democracy in Five Nations. Boston: Little, Brown.Find this resource:

Aristotle 1958 [4th century BC]. Politics, trans. E. Barker. New York: Oxford University Press.Find this resource:

Barry, B. 1995. Justice as Impartiality. Oxford: Oxford University Press.Find this resource:

Braithwaite, R. 1955. Theory of Games as a Tool for the Moral Philosopher. Oxford: Oxford University Press.Find this resource:

Dahrendorf, R. 1968. In praise of Thrasymachus. Pp. 129–50 in Dahrendorf, Essays in the Theory of Society. Stanford, Calif.: Stanford University Press.Find this resource:

Hardin, R. 1990. Rationally justifying political coercion. Journal of Philosophical Research, 15: 79–91.Find this resource:

——1991. Hobbesian political order. Political Theory, 19: 156–80.Find this resource:

——1995. One for All: The Logic of Group Conflict. Princeton, NJ: Princeton University Press.Find this resource:

——1998. Reasonable agreement: political not normative. Pp. 137–53 in Impartiality, Neutrality and Justice: Re-Reading Brian Barry’s Justice as Impartiality, ed. P. J. Kelly. Edinburgh: Edinburgh University Press.Find this resource:

——1999. Liberalism, Constitutionalism, and Democracy. Oxford: Oxford University Press.Find this resource:

——2003. Indeterminacy and Society. Princeton, NJ: Princeton, University Press.Find this resource:

——2007. David Hume: Moral and Political Theorist. Oxford: Oxford University Press.Find this resource:

Hobbes, T. 1983 [1642]. De Cive, ed. H. Warrender, trans. from the Latin. Oxford: Oxford University Press.Find this resource:

——1994 [1651]. Leviathan, ed. E. Curley. Indianapolis: Hackett.Find this resource:

Hume, D. 2000 [1739–40]. A Treatise of Human Nature, ed. D. F. Norton and M. J. Norton. Oxford: Oxford University Press.Find this resource:

——1985 [1748]. Of the original contract. Pp. 465–87 in David Hume: Essays Moral, Political, and Literary, ed. E. F. Miller. Indianapolis: Liberty Classics.Find this resource:

Lewis, D. K. 1969. Convention. Cambridge, Mass.: Harvard University Press.Find this resource:

Lindblom, C. E. 1977. Politics and Markets: The World’s Political-Economic Systems. New York: Basic Books.Find this resource:

Mill, J. S. 1977 [1861]. Considerations on Representative Government. Pp. 376–613 in Mill, Essays on Politics and Society, vol. xix of Collected Works of John Stuart Mill, ed. J. M. Robson. Toronto: University of Toronto Press.Find this resource:

Moore, G. E. 1903. Principia Ethica. Cambridge: Cambridge University Press.Find this resource:

Neumann, J. von and Morgenstern, O. 1953 [1944]. Theory of Games and Economic Behavior, 3rd edn. Princeton, NJ: Princeton University Press.Find this resource:

Pareto, V. 1971 [1927]. Manual of Political Economy, trans. from French edition. New York: Kelley.Find this resource:

Parsons, T. 1968 [1937]. The Structure of Social Action. New York: Free Press.Find this resource:

(p. 101) Putnam, R. D. 2000. Bowling Alone: The Collapse and Revival of American Community. New York: Simon and Schuster.Find this resource:

Rawls, J. 1999 [1971]. A Theory of Justice. Cambridge, Mass.: Harvard University Press.Find this resource:

Scanlon, T. M. 1982. Contractualism and utilitarianism. Pp. 103–28 in Utilitarianism and Beyond, ed. A. Sen and B. Williams. Cambridge: Cambridge University Press.Find this resource:

——1999. What We Owe to Each Other. Cambridge, Mass.: Harvard University Press.Find this resource:

Schelling, T. C. 1960. The Strategy of Conflict. Cambridge, Mass.: Harvard University Press.Find this resource:

Sidgwick, H. 1907. The Methods of Ethics, 7th edn. London: Macmillan.Find this resource:

Smith, A. 1976 [1776]. An Inquiry into the Nature and Causes of the Wealth of Nations. Oxford: Oxford University Press.Find this resource:

Tocqueville, A. de 1966 [1835 and 1840]. Democracy in America, trans. G. Lawrence. New York: Harper and Row.Find this resource:

Ullmann-Margalit, E. 1977. The Emergence of Norms. Oxford: Oxford University Press.Find this resource:

Wrong, D. 1994. The Problem of Order: What Unites and Divides Society. New York: Free Press.Find this resource:


(1) Mandeville’s subtitle is “Private Vices, Publick Virtues.”

(2) There could be two equally qualified just orderings, in one of which I am better off than I am in the other. It does not follow that a society of people who are committed to justice would rank the one of these equally qualified orderings in which I am better off above the other, because someone else will be worse off in that ordering. Hence, there would be no mutual advantage move that would make both of us better off.

(3) There are also many theories and assumptions, such as structural theories as represented by Marx and articulated by many structuralist sociologists in our time, that are much less broadly applicable, both positively and normatively.

(4) Not all Hobbes scholars would agree with this assessment. For an argument for understanding him as a coordination theorist, see Hardin (1991).

(5) Hume’s arguments may have been overlooked because they are chiefly in a series of long footnotes in Hume (2000 [1739–40], n–11 n).