Justice is a basic element of social life. It is a central moral standard in human affairs that involves the necessity of ”assuring that each person receives what she or he is due” (Cohen 1986, p. 1). Distributive justice is in the eye of the beholder, and debate usually surrounds the questions of what each person is due and what principles and procedures should be used to decide this. These differences not only occur among persons and social categories within a society but also vary across time and across cultures. A range of competing principles— rights or entitlements, equality of outcomes, equality of opportunity, equity or proportionality of rewards, and the satisfaction of basic needs—are prevalent standards of justice in most realms of social existence. These principles compete for recognition and application in human affairs, and there is considerable interest in this subject among philosophers, social and behavioral scientists, and others; substantial effort has been invested in the understanding of justice in human life.

Questions of justice, or fairness, arise in virtually all aspects of social life, and the topic of social justice covers a vast array of subjects. The social goods that are of concern in questions of distributive justice include a wide array of things that people want, usually referred to as primary goods, including basic freedoms, political enfranchisement, power, authority, status, income and wealth, education and employment opportunities, housing, and health care. In most discussions, it is assumed that those things are scarce, although in some cases that is clearly not always the case; for example, there should be an unlimited supply of basic freedoms in a well-ordered democracy. However, even if social goods are abundant, they have inherent satisfaction value—they are things that bring both extrinsic and intrinsic satisfaction to the individual—and their distribution is governed in part by principles of justice.

A major form of justice that concerns sociologists as well as legal scholars is criminal or legal justice, but justice issues pervade many other types of social relationships as well. Justice issues arise very often when inequalities of outcomes exist, but equalities of outcomes also raise questions in regard to justice. In Nichomachean Ethics, Aristotle (1953) stated: ”For if the persons are not equal, they will not have equal shares; it is when equals possess or are allotted unequal shares, or persons not equal, equal shares that quarrels and complaints arise.” Thus, in certain circumstances, equality of outcomes may be perceived as unjust while inequality of results may be seen as perfectly just.

The experimental literature in social psychology indicates that when persons perceive ”inequitable” inequalities, they frequently experience cognitive tensions and a drive to reduce those tensions by changing their judgments about relative investments and contributions or changing their values in regard to the importance of reward-relevant criteria. Research also shows that when experimental subjects in task-oriented settings have well-defined expecatations linked to objective indicators and contributions and investments, they find reward inequalities more acceptable (Brickman 1977; Cook 1975). Inequalities in the distribution of rewards also result from power differentials, and those in disadvantaged positions are more likely to view such inequalities as unfair (Cook and Hegtvedt 1986; Molm 1991).

There are several overlapping spheres of equality/inequality in which justice issues are especially important. These spheres concern the legal, political, economic, and social realms of existence and cover a broad range of human social behavior. They can be summarized as follows: (1) legal justice: the application of laws and procedures to individuals and organizations through a system of rules laid down or established, whether by custom or through the will of the state, for which penalties exist for disobedience, (2) political justice: other aspects of social life in which issues of dependence/independence arise from interdependence and from the extent of power and influence held by actors and other parties to the relationship, (3) economic justice: the distribution of the material outcomes of existence, where the economic well-being of parties to the relationship is at issue, including access to basic needs and shelter, and (4) social justice: the realm of status, respect, and the sense of worth given and received in social interaction or in relation to society.

Because questions of justice in society are so pervasive, most social sciences claim to understand the ways in which human societies deal with them. Thus, the literature on justice is massive and is perceived differently from a variety of perspectives. Extensive scholarship exists in philosophy (Buchanan and Mathieu 1986), anthropology (Nader and Sursock 1986), economics (Boulding 1981; Solo and Anderson 1981; Worland 1986), psychology (Deutsch 1975, 1986; Folger 1984; Furby 1986; Greenberg and Cohen 1982; Mukula 1980; Messick and Cook 1983), and political science (Barry 1981; DiQuattro 1986; Elster 1989; Hochschild 1981; Rae 1981). Justice is also a prominent theme in many traditions within sociology (Alwin 1987; Hamilton and Rauma 1995; Hegtvedt and Markovsky 1995; Kluegel et al. 1995; Markovsky 1985; Rytina 1986; Jasso 1980; Jasso and Wegener 1997), but the issues ofjustice have been studied primarily by those working in the tradition of social psychology; this article reflects that emphasis.


As was noted above, justice sentiments occur with respect to what each person receives relative to what he or she is expected to receive in regard to an important social good in which expectations may be governed by the application of a principle of justice. People’s expectations and perceptions of justice focus on a wide range of phenomena, such as sentiments about the fairness of social exchange and contracts, fairness in interpersonal relationships, and the treatment of themselves and others by a social group or by society as a whole. Justice concerns thus represent a ubiquitous aspect of social life at many different levels and in many different spheres.

Expectations are formed not only by reward recipients but by others as well, including both people who are involved in the relationship and observers. Justice sentiments thus derive from comparisons of what is received with what one believes should be received, that is, a comparison of the real with the ideal in a particular context. Those evaluations of differences between these two entities or quantities engage human faculties of perception, cognition, and emotion. Even if one knows little else about justice evaluations, one knows that they are subjective, and justice almost always refers to ”justice in the eyes of the observer” (Walster et al. 1973). Of course, these facts complicate the application of justice principles because, even setting aside the issue of which principle of justice to invoke in a particular situation, actors may not agree on what is real, that is, what the true outcomes are. However, to paraphrase a theme in interactionist sociology (from W. I. Thomas), what is real in the perceptions of humans is real in its consequences. Thus, if individuals perceive the social mechanisms for allocating scarce social resources and/or rewards as just, presumably the resulting distribution of outcomes also will be perceived as just.

From a social psychological viewpoint, then, a focus on justice is a focus on beliefs about inequality and perceptions of justice. Justice perceptions are pervasive in interpersonal interaction and exchange as well as in the nature of the relationship of the individual to the larger social collective in the macro-social realm. One way to simplify the vast-ness of this sociological terrain is to separate questions of justice and its evaluation posed at the micro-social level from those which occur at the macro level of society or the state (Brickman et al. 1981; Markovsky 1985; Hegtvedt and Markovsky 1995). Micro justice concerns justice evaluations at the individual level with regard to a person’s immediate circumstances. Macro justice involves the evaluation of justice above the individual level, for groups or for society as a whole. Clearly, justice evaluations take place across several domains at different levels and focus on the abstract or the concrete. Within each level, one also may distinguish between beliefs about inequality and perceptions of justice at an abstract level, such as how things should be distributed in a fair world, and the evaluation of justice in the ”real world,” that is, assessments of how fair things are in reality (Kluegel and Smith 1981). In most cases, these various levels and degrees of abstraction cannot be easily separated, for example, in the evaluation of the fairness of child support payments (Schaeffer 1990), the fairness of child custody resolutions (Elster 1989), and the fairness of economic rewards (Sennett and Cobb 1972), but it is useful to draw many of these conceptual distinctions for research purposes.


The historical roots of Western conceptions of justice lie in classical philosophy, the Judeo-Chris-tian religious traditions, and the theoretical and ideological underpinnings of legal, economic, and political arrangements. Aristotle’s Nichomachean Ethics provides the classical formulation of the problem of justice. Aristotle’s work sought to clarify principles of distributive and retributive justice and formulate the rules for the regulation of social exchange. The Aristotelian logic of justice in market relationships stressed the proportionality principle with regard to expected reward given considerations of merit and indicated that most contracts and purchases comply with the going rate of exchange (Cohen and Greenberg 1982, p. 4). Of course, it is not always easy for parties to agree on the ”going rate.” Worland suggests a crucial dilemma is revealed in Aristotle’s work involving competition between justice principles:

It becomes clear that the reference to two different kinds of justice and two different rules of proportions poses a crucial dilemma. If commodities sell at their fair price-or at a price that reflects the “fair” rate of exchange-then how can society guarantee that exchange at such prices will also provide society’s participants with an income proportionate to their relative “merit” or their standing in the community ? How is the rule requiring distribution of common goods in proportion to “merit” to be reconciled with the rule requiring “reciprocalproportionate equality” in the contractual, private exchange of commodities ? (1986, p. 48)

Eighteenth- and nineteenth-century philosophers and social theorists resolved this dilemma in a number of competing ways. Marx’s labor theory of value (see Buchanan and Mathieu 1986, p. 12) suggested that the need for principles of justice provided evidence that social institutions would be restructured, abolishing the market system and private goods. Although the notion of communal ownership of the means of production can be traced back to early Greek philosophy and before, this has not been viewed as the most adequate solution to Aristotle’s dilemma. Indeed, communism poses its own dilemmas, and the recent unpopularity of socialism as a political and economic system may provide some evidence of this (see Alwin et al. 1996, p. 124). Markets exist even under state socialism, and from the point of view of studying justice, the existence of a market for social exchange, as well as its comprehension and acceptance, seems to represent an important component in understanding the fairness of social interaction (Thiabaut and Kelley 1959; Lane 1986, p. 384).

Adam Smith, who is considered the philosophical father of modern capitalism, posed the Aristotelian dilemma differently. He observed that the modern economic system had become specialized in the sense that the production, distribution, and exchange of social goods had evolved to a stage in which economic activity was highly differentiated as an institution (Worland 1986, p. 50). Smith believed the ”natural” rules by which markets developed and the moral issues of justice were resolved could be determined empirically. What is just, then, was clearly a question of what individual actors considered just not only at the subjective level but also at the macro-social level. The operation of principles of supply and demand to set a going rate was seen as providing the answer to the moral question of justice. Thus, Smith propounded an early example of a principle that later was articulated in the social psychology literature: the principle that ”what is” determines ”what ought to be” (see Homans [1961] 1974, p. 250; Heider 1958, p. 235). This, however, falls short of resolving the Aristotelian dilemma because it gives too great a role to existential factors in defining justice.

As Worland (1986, p. 57) suggests, Marx’s theory of surplus value, in which Marx identified the real sources of profit and nonwage income as exploitation of working people by the capitalist class, did much to clarify what can now be seen as a condemnation of capitalist economies in Aristotelian terms. The twentieth-century response to Marx’s critique of capitalism in economics is known as marginal utility theory, which explains the issue of injustice in terms of market imperfections and the existence of monopolistic forms of capitalism instead of real free enterprise. According to Worland (1986, p. 81), the neoclassical response is able only to isolate and clarify ”the rules that a market society needs in order to comply with the Aristotelian moral imperative.” It does not answer the deeper question of moral psychology concerning ”whether such a society would be able to achieve the social consensus necessary for the practical implementation of the rules.”

The perspective offered by neoclassical economics on moral issues of justice often is seen more as a justification for social inequalities than as a ”natural law” of justice. It is reminiscent of Cohen and Greenberg’s (1982) suggestion that each social and economic system evolves its own unique concept of justice. For neoclassical economics, the idea that there may be merit in nonproductive activities or considerations is foreign. By contrast, contemporary political philosophers have attempted to address the broader question of merit. Rawls (1971) evaluates the Aristotelian imperative from a nonmaterial, rational perspective. Rawls’s principles of justice are as follows (quoted from Buchanan and Mathieu 1986, p. 27):

1. The principle of greatest equal liberty: Each person is to have an equal right to the most extensive system of equal basic liberties compatible with a similar system of liberty for all.

2. The principle of equality of fair opportunity: Offices and positions are to be open to all under conditions of equality of fair opportunity-persons with similar abilities and skills are to have equal access to offices and positions.

3. The difference principle: Social and economic institutions are to be arranged so as to benefit maximally the worst off.

According to Rawls (1971), these principles are ordered in terms of their primacy; that is, if principles conflict, the one listed first takes precedence. These principles obviously refer not only to social contracts and exchanges but also to the basic structure of society, including legal, political, economic, and social institutions. In most societies, that includes charters and constitutions, the means of production, competitive markets, the family, the legal system of laws and procedures, and so forth. The basic structure of society is said to specify how and by what principles society should distribute primary goods: basic liberties, powers, authority, opportunity, income, and wealth. The principle of greatest equal liberty specifies Rawls’s theory of how basic liberties are distributed, the principle of equality of fair opportunities regulates the distribution of life chances or prospects in the domains of power and authority, and the difference principle governs the distribution of income and wealth (see Buchanan and Mathieu 1986, p. 28).

It is beyond the scope of this article to review Rawls’s theory in full or summarize the critical comment that followed, except to note that his theory is utopian in the sense that it describes an ideal society run on the basis of just principles. Rawls does not spell out how one moves from states of injustice such as those found in contemporary society to the ideal situation. Still, no single book has generated more discussion and critique than his A Theory of justice. A prominent contemporary critique is Nozick’s (1974) entitlement theory, which departs radically from Rawls by stressing a libertarian view that a person is entitled to the ownership of a social good if it was acquired through just principles. Walzer (1983) argues that it is not possible to write a general theory of justice in the abstract without attempting to assay the ”substantive ways of life” of different cultures. To Walzer, ”justice” is a relative term, and no abstract theory of the ideal society can really work because justice does not exist in an abstract theoretical sense but only in terms of social meanings ordained by a particular way of life.


Distributive justice issues arise when one considers two sets of questions: Who gets what, and how? and Who should get what, and how? Some writers have suggested that the distinction between the concepts of procedural justice and distributive justice is critical to a complete understanding of the ways in which people evaluate justice (see Cohen 1986; Hegtvedt and Markovsky 1995; Thibaut et al. 1975; Tyler 1984, 1986). Procedural justice refers to the mechanisms or decision rules by which reward allocations of social goods are made, while distributive justice is concerned with the resulting allocation. These are two aspects of the same process and are clearly related, but conceptually they purport to refer to two distinct features of justice. Typically, distributive justice issues are thought of in terms of the comparison of the rewards received by a person or a group with a standard of fairness or deservedness, whereas procedural justice issues refer to the ”mechanics” of the system that regulates the process of distribution.

It is useful to distinguish three components of the distributive justice process: (1) the principles for the allocation of goods, (2) the system that governs the application of those allocative principles, and (3) the resulting distribution. While separable in this sense, these are all components of what is best thought of in terms of distributive justice, which is the overriding concept. This view is in agreement with the work of Deutsch (1986, p. 35), who states that ”procedural justice is a key aspect of distributive justice,” not something that is necessarily separate or separable. Procedural justice matters come to the forefront, suggests Deutsch, and arouse complaints of injustice more often than do the principles of justice, primarily because justice principles are often taken for granted, whereas procedural matters are not.

From this formulation, one can see that these three components—principles, procedures, and distributive outcomes—are likely to be confounded and confused in social life. If there is consensus on evaluative principles and if a clear and just set of procedures can be said to exist to implement those values, distributive justice presumably will follow. Procedural justice thus is an important component in the evaluation of distributive justice, and in the pure case the evaluation of distributive justice is not problematic. If, however, there is no consensus on the principles for allocating rewards in a social group or society or if there is a consensus but procedures are seen as ineffective or corrupted, distributive justice will be called into question. In such situations, the evaluation of distributive justice focuses on the evaluative principles that should be used, the application of the principles, or both. Therefore, the principles of allocation and the procedural aspects of allocation may be intrinsically inseparable, and it may not be clear whether unjust outcomes result from the ”wrong” principle being used or from misapplication of the ”right” principle.

In summary, while it seems useful to distinguish procedural justice from other components of the distributive process, it is not at all clear that procedural and distributive justice are really different forms of justice. Instead, they are different aspects of a common process. Clearly, distributive justice issues arise when persons perceive the allocative mechanisms to be unjust or perceive imperfections in the application of just mechanisms to real life. It can be seen, then, that overall evaluations of justice at the micro-social level or the macro-social level may be influenced by perceptions of the fairness of procedural or distributive justice issues. However, this distinction ultimately becomes the empirical issue of whether and under what sets of conditions distributive versus procedural injustice is perceived by the actors involved.


Much current research on distributive justice traces its theoretical roots to Homans’s (1976, p. 249) discussion of the rule of distributive justice, which hypothesizes that unless persons’ perceived inputs (contributions, investments, resources, etc.) are equal, some inequality of outcomes or rewards can be expected, and that rewards generally are expected to be allocated in proportion to inputs (see also Heider 1958, p. 288). Responding to what they saw as the need for a general theory of social behavior, Walster and her colleagues (see Walster et al. 1973, 1978; Berkowitz and Walster 1976) formulated equity theory to integrate the insights from a variety of social psychological theories, including reinforcement theory, cognitive consistency theory, psychoanalytic theory, and exchange theory (Walster et al. 1978, p. 2). Building on the work of Homans, Lerner, and others, equity theory as formulated by Walster et al. (1978, p. 6) contains four basic propositions:

Proposition I: Individuals will try to maximize their outcomes.

Proposition IIA: Groups can maximize their collective reward by evolving accepted systems for equitably apportioning resources among their members. Thus, groups will evolve such systems of equity and attempt to induce their members to accept and adhere to those systems.

Proposition IIB: Groups generally will reward members who treat others equitably and punish members who treat others inequitably.

Proposition III: When individuals find themselves particpating in inequitable relationships, they will become distressed. The more inequitable the relationship, the more distress individuals will feel.

Proposition IV: Individuals who discover they are in an inequitable relationship will attempt to eliminate their distress by restoring equity. The greater the inequity that exists, the more distress they will feel and the harder they will try to restore equity.

To define justice, equity theory distinguishes inputs and outcomes, both of which are expressed in the same units, say, dollars, points, or another unit. Assuming that inputs are positive (for simplicity), the equity principle states that under conditions of justice, there is an equality of relative gains. For two actors (persons, groups, nations, etc.) engaged in social exchange, equity is said to exist when the ratio of profits (outcomes minus inputs) to inputs is the same for both actors involved in the exchange.

While it is clear that rational self-interest is a strong motive for behavior in many types of situations and may be a safe assumption in competitive situations, there are other motivations for behavior (Becker 1996; Elster 1984, 1989). This can beseen as one of the unnecessarily restrictive assumptions of equity theory; the prevalence of altruism, cooperation, and other forms of prosocial behavior strongly questions this basic assumption. There is plenty of evidence that competitive behavior in mixed-motive games such as the Prisoner’s Dilemma can evolve into stable cooperation under certain conditions, such as the anticipation of future interaction (Axelrod 1984). In relationships where intimacy and identification are present, the importance of self-interest as a motive for behavior is considerably lessened (Austin 1977). As was noted earlier, Rawls’s (1971) theory of justice, for example, suggests that one principle that underlies the moral judgments made in human society is the principle that social and economic arrangements are established to maximally benefit those who are less advantaged with respect to desirable social goods. Thus, one can see that equity theory as stated by Walster et al. (1978) probably overstates the need to specify a type of utilitarian-based logic for defining justice.

Implicit in this theoretical statement and the experimental literature on which it is based are the assumptions of cognitive dissonance theory (Festinger 1959). Thus, an important way to resolve dissonance is to adjust one’s beliefs to conform to reality. However, while the theory’s allowance for changes in beliefs is an important feature, there are limits to what people will believe. Elster comments as follows on the concept of ”belief adjustment”:

Dissonance reduction can also take the form of belief adjustment. Workers who take jobs in unsafe industries alter their estimated probabilities of accidents. As a result, when safety equipment becomes available, they may choose not to purchase it. Here, as in other cases, misformation of private beliefs (or preferences) creates a case for government intervention. . . . Belief-oriented dissonance reduction is a form of wishful thinking [but]. . . acting on beliefs formed in this way can be disastrous and is likely to force a change in beliefs. When action is not called for, the wishful beliefs can be more stable. The ”just-world” theory, for instance, suggests that people adjust their beliefs about guilt and responsibility so as to preserve their belief that the world is fundamentally just. The best-known example is the “blame the victim” syndrome. . . . (Elster 1989, pp. 22-23)

Presumably, many beliefs, including beliefs about inequality and justice, achieve a high degree of stability relatively early in adulthood and may not be subject to alteration (Alwin 1994). Expectations may not be altered easily if they are rooted in firmly held beliefs about the legitimacy of inequality.


In the early 1970s, Deutsch (1975) began to question the proportionality, or equity, principle implicit in much theorizing about justice. He, among others, argued that equity is only one of several principles used to evaluate the justice of outcomes in social life. Principles ofjustice are used as a basis of ”judging individual persons and in judging the basic structure of societies” (Cohen 1986, p. 1). At least five competing principles—equality, equality of opportunity, equity, rights or entitlement, and need—are applied routinely in most realms of social existence. Other principles, such as chance, are really devices for ensuring that a principle of justice—namely, equality of opportunity—is procedurally adhered to (see the discussion of lotteries in Elster 1989, pp. 62-122). While these principles may more or less exhaust the possible criteria for evaluating justice, perhaps the equity principle has been seen as most relevant in many cases. For example, in the work of Homans ([1961] 1974), the proportionality principle is given a prominent place in the discussion of distributive justice, and it is clearly present in many discussions of equity theory (see Adams 1965; Berkowitz and Walster 1976; Cook and Hegtvedt 1983).

However, with time, theoretical work and empirical research have cast the problem of equity into the broader framework of distributive justice, in which justice may be evaluated by one principle or a combination of many different principles. Contemporary interest in the study of distributive justice can be traced to the early articulation of relative deprivation theory, particularly in the work of Stouffer, Merton, and Homans (see Williams 1975). In the last several decades, issues of distributive justice have moved from a primary focus on the psychological and behavioral consequences of patterns of social reward distributions in small group settings (see reviews by Adams 1965; Berkowitz and Walster 1976; Crosby 1976; Cohen and Greenberg 1982; Hegtvedt and Markovsky 1995) to a more recent focus among sociologists on studying principles ofjustice evaluation (Alves and Rossi 1978; Jasso and Rossi 1977;Jasso 1978, 1980; Jasso and Wegener 1997).

Justice Evaluation and Social Comparison

Current theories of distributive justice implicitly or explicitly specify social comparison as the basis for justice evaluation and the reference standard persons use in evaluating the fairness of social rewards. The following discussion summarizes the author’s theory of justice evaluation and social comparison (Alwin 1987). As in most contemporary theories, it is assumed that in evaluating the fairness of a particular reward allocation, persons compare themselves with others (see Berger et al. 1972; Pettigrew 1967; Williams 1975; Gartrell 1982). For example, Homans’s rule of distributive justice often is stated as follows (see Walster et al. 1978):

Justice = [P's Reward / P's Inputs] – [O's Reward / O's Inputs] = 0 (1)

Here P and O are two persons in a local exchange in which P is assumed to evaluate his or her comparison ratio or rewards to inputs against that of another individual, O. If P perceives that O’s comparison ratio is equal to his or her own, then he or she will perceive a state ofjustice. If this is not the case, some degree of injustice is presumed to exist, and P will perceive that he or she is underrewarded or overrewarded relative to O.

The term ”inputs” is used in equation (1) to represent the general reward-relevant characteristics of individuals that are involved in making assessments of the fairness or justice of rewards. These may be contributions, investments, resources, or global status characteristics (see Cook 1975; Cook and Yamagishi 1983). This formulation ignores the concept of costs and their effects on rewards and inputs. In other words, inputs and rewards are thought of as positive quantities. For a somewhat different formulation involving negative inputs and rewards, see Walster et al. (1978).

Several investigators (e.g., Blau 1971; Berger et al. 1972; Jasso 1978) have pointed out that in viewing this identity (equation [1]) objectively, it is impossible to determine which parties are over- or underrewarded when perfect justice does not prevail. Moreover, formulations of justice evaluation as local comparisons cannot cope with the possibility that from the perspective of a more general referential standard, both P and O may be unjustly rewarded even though their comparison ratios may be equal. This view in no way denies that people make local comparisons of such ratios. The point is that when people make local comparisons, referential standards existing outside the local situation typically are invoked to evaluate fairness.

Berger et al. (1972, p. 122) criticize this formulation, arguing that distributive justice issues arise only in the presence of a stable frame of reference and that justice evaluations are inherently made on the basis of reference to generalized individuals rather than specific others. Using this observation as a basis for reconceptualizing the classical exchange-based conception of justice, Berger et al. (1972) formulate a ”theory of status value” that formalizes the process by which persons evaluate the fairness of rewards. They formulate the process in terms of referential standards: frames of reference that contain existing information regarding the characteristics and rewards of generalized others. The referential structure formalized by Berger et al. (1972) consists of information about the relationship between levels of characteristics possessed by general classes of persons and the associated levels of social reward. According to this theory, through social exchange persons develop normative expectations about the reward levels typically associated with general classes of individuals, and when persons perceive their reward-relevant characteristics to be similar to those of a particular general class of individuals, they come to expect that their reward levels also are similar. As a consequence of these beliefs about ”what is,” normative expectations are formed about reward levels that persons can legitimately claim (Berger et al. 1972, p. 139). This conclusion is consistent with Homans’s ([1961] 1974) and Heider’s (1958) observations that the ”ought” is determined in the long run by the ”is” and with recent sociological theorizing that argues that social inequalities are often the major basis of their own legitimation (Sennett and Cobb 1972; Della Fave 1980; Stolte 1983).

The Berger et al. (1972) status-value formulation, however, is limited in its consideration of the process by which a person selects a referential comparison standard among the many possibilities. Referential comparisons may be based on relatively small groups of persons such as one’s coworkers or large classes of persons such as occupational categories. This may appear to be a flaw in the status-value theory; however, it also may be viewed as an asset in that it allows flexibility in specifying the role of various types of referential standards in justice evaluation processes.

Referential Comparisons. While near consensus exists in the social psychological literature that persons use referential comparisons to evaluate how satisfactory their outcomes are, Gartrell (1982) observes that little is known about the origin and visibility of comparative frames of reference. In his own research, for example, Gartrell finds that information on wage rates is often invisible, and such information for persons in other jobs frequently originates from concrete, personal references rather than from knowledge of rates for broad social categories. Moreover, the awareness of wage comparisons frequently appears to relate to somewhat idiosyncratic factors, such as informal social contacts (see Walster et al. 1978). This is consistent with other studies of relative deprivation and status comparison, in which persons are found to rely heavily on information from their own social circles (see Runciman 1966; Rainwater 1974; Coleman and Rainwater 1978). Thus, it may be difficult to specify the origin or basis of a person’s referential comparison with objective accuracy. Some experimental research (Major and Forcey 1985) suggests that subjects are most interested in same-sex and same-job wage comparisons.

Moreover, the individual’s subjective judgment regarding the ”fairness” of a given reward outcome may be a more relevant concept. Jasso’s (1978, 1980) theory of distributive justice introduces the term ”just reward” to refer to the reward level individuals expect on the basis of referential comparisons conceived in general terms. Jasso formulates the status-value model of Berger et al. (1972) for justice evaluation as follows (1978, p. 1402):


That is, for a person to determine the justice or fairness of his or her reward, the actual level of reward is simply compared with the reward expected on the basis of existential (or other) criteria.

This formulation is more general than the one given by Berger et al. (1972) because it permits a wide range of reference group comparisons and because both existential and nonexistential criteria may be used in the calculus of the ”just reward” (see Jasso 1978, 1980). As Blau (1971, pp. 58-59) points out in his criticism of Homans ([1961] 1974), ”not all existing practices reflect justice; some are unjust by prevailing moral standards, and the fact that they are expected to continue to exist does not make them just.” Thus, there are both existential and nonexistential standards of justice, and either or both may be combined to determine the ”just reward.”

This ”comparison difference” formulation of distributive justice makes the nature of over- or underreward clear, unlike the formulation used by most equity theorists (e.g., Walster et al. 1978). Moreover, in contrast with the classical formulation given in equation (1), both quantities in the equation are expressed in the same units: units of reward. Further, this formulation (equation [2]) satisfies the notion that the individual’s subjective judgment regarding the ”expected” or ”deserved” reward may be the most relevant concept. Finally, the distinction between kinds and degrees of injustice can be measured on a scale that starts at zero, where ”perfect justice” occurs. Southwood (1978, p. 1157) has proposed a model very similar to Jasso’s model involving what he calls ”subtractive interaction” that is intended to estimate the effects of departures of actual reward from expected (or just) reward. This model simply involves estimating the effects of the quantity x1 – x2, where x1 represents the actual reward and x2 represents the expected reward.

Reformulating the Justice Evaluation Model.

Using Jasso’s concept of the just reward, it is possible to reformulate the classical exchange-based conception in a way that permits the measurement of the direction and magnitude of departures from justice. First, it is necessary to recast Homan’s ([1961] 1974) ”rule of distributive justice” given in equation (1) as an equivalence of the ratio of P’s and O’s rewards to the ratio of their inputs as follows (see Patchen 1961; Adams 1965; Homans [1961] 1974, 1976):


Here the comparison ratios are different from those in the classical statement (Walster et al. 1978). They now involve terms in common units: reward units on the one hand and units of input on the other hand. The present formulation of the classical model has two desirable properties: It is intuitively simpler to have the numerator and denominator of such ratios in the same units, and this formulation fits with the psychological mechanism often assumed in justice evaluation: Persons expect their inputs (contributions, investments, resources, or general status characteristics) to be in constant proportion to the rewards they associate with a standard of comparison.

If the Berger et al. (1972) theory of status value is correct is stating that a person, P, uses referential structures that specify levels of reward-relevant characteristics that are similar, indeed equivalent, to his or her own (see also Pettigrew 1967; Williams 1975), it is possible to equate P’s and O’s inputs on the right-hand side of equation (3), setting the second term on the right at unity, as follows:


If one generalizes the concept of ”O’s reward” to be the same as Jasso’s ”just reward,” the justice evaluation process devolves to a comparison of P’s actual and expected/just rewards:


An examination of this expression indicates that the classical formulation restated in this way permits the distinction between kinds and degrees of injustice measured on a scale that starts at zero, where perfect justice occurs. These units may conveniently be thought of as justice units because when the comparison exeeds zero, overreward occurs, and when it is less than zero, P is said to be underrewarded.

Note the convergence of the reformulation of the classical model given here with that proposed by Jasso (1978), which was derived empirically from the analysis of vignette data. The natural logarithm of equation (5), which is derived from classical exchange and status value theories, equals the formulation for justice evaluaton proposed by Jasso. I have derived theoretically a principle of justice evaluation that is equivalent to Jasso’s (1978, 1980) empirically derived ”Universal Law of Justice Evaluation.” This formulation can be used as a basis for defining departures from justice.

Jasso argues, however, that the simple ratio of actual to just rewards does not capture the justice evaluation phenomenon precisely: It does not account for the fact that positive departures from justice (overreward) are not equivalent to negative departures (underreward), and ”this appears to violate the human experience that deficiency is felt to be more unjust than a comparable excess” (Jasso 1978, p. 1403). In other words, the injustice created by an actual reward above thejust reward k is not equivalent to the injustice created by an underreward of the same magnitude. Jasso (1978, p. 1415) resolves the problem by proposing the natural logarithm of the comparison ratio (i.e., actual reward/just reward). Such a formulation assumes that an overreward of k times is equal in the magnitude of injustice to an underreward of 1/k times. Using satisfaction with material well-being, Alwin (1987) empirically examined Jasso’s hypothesis that the effects of underreward are more potent than the effects of overreward. While support was found for the importance of the sense of injustice in the prediction of material satisfaction, the hypothesis that the extent of satisfaction depends on measured departures from justice was not supported.


As was mentioned above, distributive justice issues arise in response to two sets of questions: (1) the realm of the ideal, that is, who should get what, and how, and (2) the real, or who gets what, and how. As was noted, these issues may be phrased with regard to the individual, the micro-justice level, or groups of individuals or the whole society (i.e., macro-justice). Regardless of the justice principles one espouses, behavior and sentiment are conditioned by how the system works or is perceived to work. One of the important aspirations of social science is the understanding of who gets what and how, and it was the major focus of several important works on social stratification and mobility in the 1960s and 1970s (e.g. Blau, and Duncan 1967; Hauser and Featherman 1977; Featherman and Hauser 1978; Jencks et al. 1972, 1979; Levy 1988; Moynihan 1968; Sewell and Hauser 1975; Wright and Perrone 1977). As a result of that work, a great deal more is known about the ways in which race, gender, and class affect inequalities and the manner in which families and educational institutions operate to promote or deny access to opportunities for socioeconomic advancement. However, the study of social stratification and mobility is an arcane field of inquiry, and there is not always agreement among social scientists about the main factors that create socioeconomic inequalities (see Herrnstein and Murray 1994 and the debate it provoked, e.g., Fischer et al. 1996).

Sociologists often assume that there is a causal linkage between the structural conditions of society (e.g., the standard of living and economic inequality) and public beliefs and sentiments regarding the acceptability of those conditions, but little is known about the nature of that linkage. There is a growing body of empirical data regarding trends in income distribution and income inequality, much of which indicates growing economic inequality and hardship for segments of the population (e.g., Duncan and Rodgers 1991; Levy 1988; Levy and Murnane 1992; Thurow 1987), but much less is known about subjective interpretations of economic conditions. Thus, while social scientists attempt to understand the inner workings of the stratification system, there is much that remains to be understood, and it is important to realize that a person’s beliefs about sources of inequality affect evaluations ofjustice as much as if not more than objective conditions do (see Kluegel and Smith 1981, 1986; Robinson and Bell 1978; Kluegel et al. 1995).


This article has argued that one can distinguish three components of the distributive justice process with regard to any primary good: (1) the principles for the allocation of goods, (2) the system that governs the application of those allocative principles, and (3) the resulting distribution. Justice sentiments derive from comparisons of what is received with what one believes should be received, that is, a comparison of the real with the ideal in a particular context.

According to Jasso and Wegener, empirical justice analysis has four major objectives:

(i) to obtain numerical approximations of the quantities and relations identified by justice theory; (ii) to gauge the extent of interindividual and intergroup variation in the quantities and relations; (iii) to explain their etiology, including the effects of social structure and of the observer’s position in the stratification structure; and (iv) to assess their behavioral and social consequences (1997, p. 393).

Three fundamental quantities pertain to justice: the actual condition, the just condition, and the justice evaluation. While justice evaluations involve the comparison of the other two quantities, it is not clear that individuals actually quantify justice in the way theoretical formulations suggest, and it remains to be seen whether most people ”calculate” more than a general ”sense of justice” from this comparison.

To summarize current research on social justice, one would have to focus on a wide range of distributional issues with respect to the primary social goods of distributed in society, including basic freedoms, political rights, power, authority, status, income and wealth, education and employment opportunities, housing, health care, and the pursuit of happiness. Below, this article briefly mentions five areas in which a consideration of justice theory is relevant to sociological understanding: (1) income inequality and the welfare state, (2) discrimination and affirmative action, (3) gender, work, and comparable worth, (4) divorce, child custody, and child support, and (5) inter-generational relations.

Income Inequality and the Welfare State.

One of the central preoccupations of sociologists who study distributive justice has been the economic realm, with an explicit focus on wages or earnings (e.g. Alves and Rossi 1978; Gartrell 1982; Gartrell and Paille 1997; Jasso 1978, 1999; Jasso and Rossi 1977; Patchen 1961; Randall and Mueller 1995; Robinson and Bell 1978). Research in the United States shows convincingly that individualistic attributions for poverty and wealth predominate (Kluegel and Smith 1986). Such attributions are found across the spectrum of socioeconomic positions, and among lower socioeconomic status (SES) groups these beliefs are held concurrently with structural explanations. At the same time, Americans show considerable antagonism toward any form of systemwide redistribution of income beyond current welfare assistance to children, the disabled, and the indigent and current forms of social security for senior members of the population.

Whereas Americans hold equality as the standard of justice in the political realm, inequality is the standard in the economic realm. A recent multinational comparison of a U.S. sample with comparable data from other Western nations showed that in assessing the deservingness of their own earnings, ”what ought to be” is strongly linked to ”what is” (Alwin et al. 1996). Thus, existential considerations play a strong role in the development of judgments about levels of deserved income, although there was considerable variation in the magnitude of those linkages. The weakest effects, as predicted on the basis of perceived system legitimacy, were in the postcommunist economies of eastern and central Europe and the former republics of the Soviet Union. The strongest effects were in the Western capitalist democracies. Indeed, the linkage between job desserts and job income was so strong in those countries (Germany, Great Britain, the Netherlands, the United States, and Japan) that it seemed hardly possible that any factors other than current income levels could contribute to variation in perceptions of justice. By contrast, the level of perceived family need played a much stronger role in evaluations of the justice of earnings in the eastern European countries than it did in the West (see Alwin et al. 1996, pp. 123-128).

Discrimination and Affirmative Action. Discrimination by dominant groups against ethnic minorities and women has been a significant concern of those interested in equality and justice. This set of issues can be addressed easily within the framework of social justice as an instance of the lack of congruence between justice principles such as equality of opportunity and the actual workings of society. Although considerable progress has been made in establishing constitutional prohibitions against discrimination on the grounds of race, ethnic origin, and sex in employment practices, education, and public accommodations, reality has lagged behind those ideals. Substantial inequalities among racial groups persist despite substantial opinion that racial discrimination is no longer a problem in American society (Blauner 1989). Considerable research has focused on the consequences of discrimination.

The concept of affirmative action has been used in the United States and other countries to refer to social policies that go beyond prohibitions against discriminatory practices that deprive minorities of their rights and aim social policy toward remedying the effects of past discrimination. Affirmative action represents an effort to restore equity to social groups, rather than to individuals, by targeting women and minorities for educational opportunities, jobs, promotion, government contracts, and other arenas where past discrimination has been documented. Affirmative action policies have been controversial because they appear to represent a form of reverse discrimination inasmuch as they violate the principle of of equal opportunity by giving preferential treatment on the basis of race and national origin. These policies have faced a number of legal challenges that are likely to continue as long as they are perceived to be unjust by some members of society. Regardless of how one views these policies, they have placed increased numbers of women and minorities in good jobs and selective educational institutions, but they may have increased tensions over these matters.

Gender, Work, and Comparable Worth. An important application of the social justice framework has been the examination of equity in the job rewards of men and women. In the United States, as in other countries, there are legislative guarantees to the right of equal pay for equal work, such as Title VII of the Civil Rights Act of 1964. The concept of comparable worth was designed to go beyond this mandate to describe a situation of equal pay for work of equal value, that is, work that requires comparable levels of effort, skills, and responsibility. Rather than being a remedy for past discrimination, as in the case of affirmative action, this approach is aimed at developing policies that will guarantee equal treatment for men and women. While this idea seems to be catching on, there are still substantial gaps in the workplace authority and earnings of men and women throughout the world (see Wright et al., 1995; Jasso and Wegener, 1999), and researchers face a ”paradox of the contented female worker” (Crosby 1982). Generally speaking, women have jobs with lower pay, less autonomy, and less authority than jobs held by men, but their level of earnings satisfaction is no different from that of men. Mueller and Wallace (1996) suggest that when levels of perceived justice are taken into account, substantial amounts of the difference between men and women in their satisfaction with earnings can be explained. What remains to be understood are the factors that contribute to differences between men and women in perceptions of what is just.

Divorce, Child Custody, and Child Support.

Marital disruption is a conspicuous feature of the contemporary family that raises issues of justice not only for the people involved but for the public as well. Apart from justice matters arising from the division of property and assets in a divorce, when children are involved, the settlement becomes more complicated. Child custody disputes and child support issues present unique problems for justice analysis because in the first case the primary goods are indivisible and in the second case costs (or negative rewards) are being allocated. Procedures for resolving both sets of issues are governed by state and federal statutes or guidelines as well as by local customs and have varied considerably across time and culture. Historically, courts have employed justice principles involving parental entitlements to various degrees, but in recent times, the principle has evolved that the best interests of the child ought to be the sole or major consideration in custody decisions. As Elster (1989, p. 126) puts it, ”although the child may to some extent and for some purposes be considered a consumption good for the parents, he is also and predominantly a person in his own right” who has an interest in the allocation. Elster argues, however, against the principle of the best interests of the child and suggests that in contemporary society, when joint custody is not feasible, three options present themselves: a presumption in favor of the mother, a presumption in favor of the primary caretaker (usually the mother), and tossing a coin.

Child support payments are vastly more determinate, and those judgments are considerably less Solomonic. Indeed, courts generally have allowed the participants, on the advice of counsel, to negotiate the nature of the awards. In child support situations, decision making occurs in two steps. There is a preliminary determination of an amount to be divided based on the needs of the child and then a decision on how that amount should be divided based on the relative resources of the parents. As Schaeffer (1980, p. 158) puts it, ”beliefs abut child support awards differ in important ways from other beliefs about justice . . . [because] allocating child support obligations involves allocating not rewards, but responsibilities expressed as contributions.” Schaeffer (1980) analyzed beliefs about the fairness of child support judgments by using a factorial survey involving vignettes. She found that child support awards are allocated according to a ”proportional contribution-variable need” system in which parents’ contributions are proportional to their resources. She concluded that preferences for child support awards embody a ”modified version of ‘from each according to their abilities, to each according to their needs”’ (1980, p. 172).

Intergenerational Relations. Relationships among generations are one of the most important structural features in all societies. Some authors have argued that the prevailing social contract between generations in Western society regarding expectations, obligations, and well-being is changing (Bengtson and Achenbaum 1993). This is due to the changing demography of age but also to shifting cultural understandings of age differences and issues of equity. Bengtson (1993, p. 4) argues that ”we have reached a cultural watershed concerning the implicit understanding of rights and obligations between age groups and generations in human societies. Never before have so many elders lived so long; never before have so relatively fewer members of younger age groups lined up behind them in the succession of generations.” He argues further that as a consequence, ”we are faced with new and historically unique dilemmas of family life and social policy agendas regarding the expectable life course and the succession of generations.”

These observations raise a number of questions regarding the nature of intergenerational conflict and the linkage between age and economic expectations, economic performance, and evaluations of material well-being. The concept of justice has been applied to the study of intergene-rational relations (see Norris 1987), but little work has explicitly linked social psychological theory regarding equity or justice evaluation to this range of issues. Moreover, researchers often have settled for conclusions based on conjecture or weak and inappropriate kinds of evidence. There is a wide range of intergenerational issues to which the justice framework can be applied, including not only the material well-being of older age groups but also the sense of obligation that adult children have about the support of their elderly parents as well as public concern about the future health care and social security systems that will support the elderly in the future.

Next post:

Previous post: