Why are group-level differences important for making a judgment that there is structural inequality?
Word count: 4706
If structural inequalities exist, then our approach to evaluating social justice should be able to take such inequalities into account. In this paper, I argue that in order for judgments to be made about structural inequality, a group-conscious approach is required. I make this argument in the following way. First, I show that equality is morally valuable when discussing justice and injustice in society. This step is required because any claim about the injustice of structural inequality necessarily depends on equality having moral relevance when discussing justice generally. Second, I argue that structural inequalities exist and that the only way to identify and consequently make judgments about these requires a group-conscious approach. Third, I summarize Young’s group-conscious approach, explain how it depends on making group-level comparisons, and argue why doing so is important. Fourth, I examine five possible objections to the group-conscious approach and show why these are not persuasive or do not ultimately undermine the approach. Briefly, these objections raise concerns about making groups the focus of justice, the salience of cultural explanations over structural explanations, the skewing of significant individual differences in group classifications, the ambiguity of structural explanations, and the problem of averages in comparing groups. Lastly, I provide a concluding summary of my argument and findings.
Throughout this paper, I make continued reference to Iris Marion Young’s article ‘Equality of Whom? Social Groups and Judgments of Injustice’ (2001). Instead of advancing a fully original argument, I have attempted to develop Young’s group-conscious approach and raise relevant critical objections that have been overlooked or incompletely addressed.
Before continuing, a definitional clarification is required. By the ‘group-conscious approach’, I am referring to Young’s theoretical framework that uses group-level differences to identify structural inequalities. Structural inequality refers to the relative constraints faced by a group of similar individuals due to the cumulative effect of their asymmetric social positions and other systemic disadvantages (Young 2001: 15).
1. Is equality morally relevant?
Any normative argument about the importance of structural inequality requires an argument for the moral significance of equality. This is because an argument about structural inequality would depend on a specific understanding of inequality that states what kind of equality is important, why it is morally relevant, and to whom does it apply. However, I argue only for a minimal conception of equality here in order to maintain the focus on the argument about the group-conscious approach.
Why should we be morally concerned about equalities and inequalities? When faced with certain cases of social, political, economic, or legal inequality, our intuition is to say that there is something wrong, bad, or unjust about these circumstances. For instance, denying specific groups access to public facilities and voting booths on the basis of their ethnicity, as in the United States before civil rights legislation in the 1960s, would be discriminatory and a morally relevant social and legal inequality. Paying women less than men for the same work – while controlling for other salient factors – would be an instance of an economic inequality. In a society with extreme wealth polarization between the rich and the poor, it would seem unjust to have people dying in the streets from hunger and exposure due to no fault of their own while the rich live in lavish exuberance. These are intuitively clear cases of morally relevant inequalities, but we still require a reason to think equality is morally valuable.
There are two primary approaches to establishing the value of equality (Beitz 2001: 97; Gosepath 2011). The intrinsic view of equality would state that it is morally relevant because it is a good in and of itself. This view has been widely criticized in its pure form because it would lead to situations in which people would be made worse off in absolute and relative terms for the sake of equality. Alternatively, there is the instrumental view that states that equality can be morally relevant because it is a useful means to reach our fundamental moral aims. For example, Charles Beitz argues that we have derivative reasons to care about inequality, such as the alleviation of terrible deprivation, suffering, and humiliation (2001: 99-105). This approach to the value of equality is stronger since it permits a focus on more extreme wrongs or cases of injustice. This view has similarities to Harry Frankfurt’s sufficiency view, which argues that equality has derivative moral importance since it identifies and helps address cases of extreme deprivation such as pain, chronic hunger, illness, and other needs (Frankfurt 2001: 89, 94).
In less extreme cases of injustice, some forms of inequality exist only because they exist within a social order. For instance, equal access to voting stations is contingent on there being a social convention of voting that has socio-political importance in a society. If there were not an electoral system, then such an equality would refer to nothing and have no importance. In this way, the relevance of equality also has a basis in the normative structure of a social order, because these govern why and how individuals should be treated in that system. For instance, a liberal constitutional democracy guarantees individual rights and the equal treatment of individuals under the law. However, these would not exist without the system itself. Since we live within a social system governed by a legal and political order, we have a further reason to care about equality because it functions as a broadly applicable principle in the treatment of individuals.
Therefore, we have two compelling reasons to be morally concerned about certain inequalities. First, some inequalities highlight extreme cases of suffering and deprivation that generate a moral obligation to correct them. Second, the normative structure of a social order creates rules and principles that should be followed so that societal governance is conducted fairly, impartially, and with respect to individual dignity: all of these rely to some extent on a principle of equality.
2. The need for a new approach to deal with structural inequality
Why do we need a group-conscious approach? In society broadly, there is a tendency to distinguish between groups when discussing inequality and injustice. Our ordinary use of language suggests that certain groups suffer from injustices or inequalities in society. For instance, group-level inequalities are often referred to when discussing housing policy, education policy, incarceration and policing practices, and social mobility. In such cases, certain ethnic or social groups are seen as disadvantaged, discriminated against, or otherwise disparately impacted. Some of these inequalities can be referred to as structural inequalities, which consists in
the relative constraints some people encounter in their freedom and material well-being as the cumulative effect of the possibilities of their social positions, as compared with others who in their social positions have more options or easier access to benefits. (Young 2001: 15)
In other words, structural inequalities refer to disadvantages faced by groups of individuals in relevant areas of life that are generated by the dynamic of constrained social positions relative to other groups in a given society. For instance, individuals of a particular ethnic descent may face recursive disadvantages in pursuing better employment opportunities on the basis of implicit attitudes against people from that group. The cumulative effect of such discrimination over a period of time, coupled with other salient disadvantages such as locality, education opportunities, family stability, and health care access, will create constraining structures on that group. That is, the individuals of that group will occupy lower social positions on average when compared to other groups.
The relevance of structural inequality depends on it being an actual phenomenon – a conclusion for which there is strong evidence. For instance, a study on ethnic discrimination in the Swedish labor market shows that job applications received from individuals with Swedish-sounding names are 50% more likely to be invited for an interview than individuals with Middle Eastern-sounding names (Carlsson and Rooth 2007). Another study highlights the lack of healthcare provision for Aboriginals compared to non-Aboriginals in Australia (Henry, Houston, and Mooney 2004). A skeptic might disregard these findings as anomalies, but their purpose is to highlight that inequalities can be subtle, subliminal, and the result of complex social phenomena. Regardless, more obvious and wide-reaching examples exist as well. In the United States, individuals of African-American descent face a structural inequality rooted in their historical maltreatment from slavery, to Jim Crow, to contemporary institutional racism and public biases (Wax 2009: 15-6).
As structural inequalities exist, then there should also be a way to identify such inequalities. As Iris Marion Young (2001: 4) points out, it is often assumed that political theorists should evaluate equality and inequality through the comparison of individuals – call this methodological individualism. However, she argues that an assessment purely on the basis of individuals provides little foundation for making judgments of social justice (2001: 7). For example, Ronald Dworkin argues that his approach
aims to provide a description [… of the] equality of resources person by person, and the considerations of each person’s history that affect what he should have, in the name of equality, do not include his membership in any economic or social class. (1981: 340)
Here, Dworkin is arguing for methodological individualism in determining inequalities. In his evaluation of Rawls’ difference principle, he further states that the preferences, ambitions, and ideas of the good life in a group are too diverse to make groups meaningful units in this evaluation (1981: 343). However, preferences and ambitions are not what we are concerned with in cases of structural inequality, but rather, the constraints faced by a particular group through systemic disadvantages in a system. Thus Dworkin’s individualism is too restrictive in its approach to account for injustices in society, because not all relevant causes of unjust inequalities are derived from individual preferences, actions, or bad luck.
Take the example of an individual A born into a low-income family. She is unable to attend a higher educational institution because her family cannot afford it. In addition, she has difficulty securing a well-paid job due to an implicit negative bias against individuals from her ethnic background. She is therefore much more likely to be stuck in a low-income setting than an individual that is not from a low-income family that can afford higher education and does not suffer from an implicit attitude bias due to their ethnicity. Moreover, individual A does not face these disadvantages in isolation; a significant portion of people of her ethnic background also face these same challenges, much more so than other ethnic groups. As a result, a significant portion of this ethnic group suffers from disproportionate constraints that are not faced by other groups.
The inequality suffered by person A cannot be explained on the basis of methodological individualism: that is, in terms of her preferences, her actions, or bad luck. This is because her preferences did not lead to her unequal circumstance, nor does her status follow from actions she has taken or decisions she has made. An inclination to retroactively explain such differences in outcome with reference to personal preferences would be suspicious, because these explanations tend to regurgitate “ideologically suspect stereotypes about particular social groups” (Phillips 2004: 6). Additionally, it would be incorrect to describe her situation in terms of bad luck, since it wrongly equivocates her ethnic background as an unfortunate happenstance comparable to being born with a severe physical disability or suffering an accident. Since methodological individualism cannot explain the cause of this inequality in individual terms, it would not be able to appropriately identify its structural origin and consequently address it. This suggests that in order to identify and make judgments about structural inequalities, we require an approach that methodologically can account for group-level differences: viz., a group-conscious approach.
3. The group-conscious approach defined
Young’s group-conscious approach is an analytical framework that can be used to evaluate and make judgments regarding structural inequality on the basis of group-level differences. It involves combining an understanding of structural injustices with a methodological step that accommodates group-level differences (Young 2001: 15). A judgment of structural inequality according to the group-conscious approach would look like the following:
A group suffers from injustice iff that category of individuals is unequal to others in a relevant way and a plausible story can be told about how the relations, rules, expectations, and cumulative consequences of collective action specifically condition the lives of members of that group.
Group-level comparisons necessarily place individuals into categories in which they share a commonality. As Young argues, such categories are not arbitrary in the case of structural inequality, since only groups that are salient in terms of the conditions for making a judgment of structural inequality are considered (2001: 15). Such relevant groups consist of categorized individuals that occupy recognized social positions based on patterns of race, socio-economic class, age, gender, and so forth.
This group-conscious approach allows us to utilize group-level comparisons. This is important because group-level comparisons open the door to fulfilling two necessary analytical steps in determining structural inequality. First, group-level comparisons are necessary for discovering social structures (Young 2001: 6). The starting point for doing so is to identify constraints in the lives of individuals, followed by a comparison with similar individuals to reveal “a net of restricting and reinforcing relationships” (ibid. 10). The ‘net of relationships’ (i.e. structure) being referred to is a relational process, since individuals occupy different positions in a social space, in which their positions exist in determinate, dynamic relation to one another (ibid. 12-3). Since we can meaningfully identify differences in the constraints and advantages faced by groups, we can look at how particular groups face comparatively more disadvantages, constraints, and discrimination than do other groups. The totality of such disadvantages can be referred to as the social structure that determines the persistent constraints on such a group.
Secondly, group-level comparisons are necessary for making judgments about inequalities grounded in social structures (Young 2001: 6). Since we are able to assess structural advantages and disadvantages faced by different groups, this opens the way to making judgments of just and unjust equalities and inequalities. This is because the group-conscious approach widens the scope of relevant causes of inequality in our considerations of social justice. We can thus make judgments about structural inequalities provided the two conditions of plausible narrative and membership in a relevant, unequal group are fulfilled.
4. Objections and responses
In this section, I consider five possible objections to the group-conscious approach. Briefly, these relate to the problem of viewing groups as the focus of moral justice, ignoring relevant cultural explanations for inequality, high-performing individuals in disadvantaged groups, the problem with comparing groups on the basis of group averages, and the ambiguity of structural explanations. I provide responses to these objections and show why they are ultimately unpersuasive in undermining the robustness of the group-conscious approach.
a. Individuals, not groups, are objects of moral concern
The categorization of individuals into groups would appear to make groups objects of moral concern over individuals. Ronald Dworkin (1981: 343) and Larry Temkin (1993: 101) argue that the problem here is that groups should not be seen as entities of moral concern, as this moral concern is only locatable in individuals. Daryl Levinson argues that Young believes equality should be evaluated on the group level and not the individual level (2002: 1374). Furthermore, if groups have moral value in considerations of justice, then by sheer membership or non-membership in a group, an individual’s rights and claims can fluctuate.
In response, Levinson appears to be oversimplifying Young’s group-conscious approach. Young agrees with the central point of Dworkin’s and Temkin’s objection: the purpose of making assessments regarding inequality is a concern for individuals as irreducible moral equals (2001: 6). To this end, the group-conscious approach does not overlook individuals since individuals still remain the ultimate moral focus of justice in the calculation. Nor should the group-conscious approach be seen as overriding or displacing individual-based comparisons of equality and inequality, but simply as an analytical tool that permits the analysis of structural causes of inequality that escape individual-based assessments. Therefore, the group-conscious approach is not affected by the objection of group-centered moral concern because it is rooted in moral individualism.
b. Cultural explanations, not structural inequality, explain group inequalities
Another possible objection to the group-conscious approach is that it incorrectly interprets aggregated individual choices, preferences, and lives as being constrained by group-external instead of group-internal factors. In other words, what the group-conscious approach mistakenly presumes is that inequalities experienced by groups are necessarily caused by external social structures rather than internally by a group’s culture. The key distinction here is that a social group may have a cultural disposition within a broader societal system that results in an outward appearance of inequality, rather than a social order creating systemic disadvantages for, say, a minority group.
For instance, take one explanation of the comparatively low performance of African-American children in education in the United States. Fordham and Ogbu (1986) argue that the academic achievements of black students are hampered by a fear of being accused of “acting white.” Their oppositional culture hypothesis suggests that the academic achievements of black students are partially reducible to group-internal factors, i.e. group culture (Fordham and Ogbu 1986: 201). Brian Barry also explains some structural inequalities on the basis of cultural preferences, arguing that the fact that groups cluster around certain low-income occupations is not evidence of oppression (2001: 98, 108). Therefore, a group’s internal patterns of behavior can be the culprit in generating structural inequalities. But it is unclear how the group-conscious approach takes these into account, because it is unclear how these cultural causes can be distinguished from structural causes. Consequently, the group-conscious approach is incomplete because it would overlook salient group-internal causes of structural inequality that would suggest the resulting inequality is not unjust.
While the specific findings by Fordham and Ogbu are contested, there is broad research on the subject that notes the importance of cultural explanations in explaining racial inequalities. Nevertheless, Elizabeth Anderson argues that there is a tendency among scholars to avoid cultural explanations of inequality in favor of structural explanations because cultural ones may reinforce stigmatizations and effectively produce “blame the victim” narratives (2012: 108).
Two counterpoints can be made in response. First, particular group cultures can also be the outcome of structural inequality. For instance, in the U.S., black parents utilize feedback opportunities at schools less effectively than white parents due to a persistent distrust for the educational system as a result of its racist history (Anderson 2012: 112-3). This suggests that cultural explanations can ultimately be rooted in structural inequalities, especially when the dominant cultural group determines the system in which certain cultural qualities result in better social capitalization. Therefore, different cultural dispositions do not necessarily mean that the resulting inequality is justified.
Another counterargument to the cultural explanation objection is that particular groups develop preferences and patterns of behavior on the basis of the options provided to them. In responding to Brian Barry’s argument about women disproportionately making life choices that are incompatible with high-achievement occupations due to innately different career dispositions, Anne Phillips argues that this difference is due to the past and current social structures that reinforce such choices (2004: 12). Moreover, Phillips’ argument further reinforces the significance of group-level comparisons, because incongruity between equality of opportunity and equality of outcome are most obvious when comparing groups rather than individuals (ibid. 18). For this reason, in certain contexts, group-cultural explanations of dispositions seem less plausible than do structural explanations.
Secondly, even if a particular group’s cultural capital is less effective in a certain societal context, the organization of all public facets of society should be geared toward reducing inequalities that result from providing advantages to particular cultural groups (Anderson 2012: 121). This is especially pertinent if there is a plausible narrative about a group’s unequal position in society that points to the cumulative consequences of collective actions that disadvantage them. In short, as Young argues, if inequalities are socially caused, then communities bear the responsibility for collectively remedying them (2001: 16). So even if cultural dispositions are not generated by structural inequalities, there is still a good reason to believe that society should work to correct these in order to reduce inequalities in outcome between groups.
c. Overlooking individuals: the case of high earners and performers
Another objection to the group-conscious approach is that it categorizes high performing, achieving, or earning members into a disadvantaged group, suggesting that mere membership means they are necessarily victims of injustice. The consequence of this would be that salient individual differences are skewed in the group classification. Consider the following example. Take the case of an ethnic group A in society S. Group A is unjustly disadvantaged in a society in comparison to groups B and C. But 10% of group A does very well in that society and are not disadvantaged in terms of equality of outcome. In fact, they do much better than the average members of groups B and C. Are these well-off individuals still structurally disadvantaged merely due to their membership in group A?
The answer both yes and no. As Young states, it is important to note that structural constraints do not have to be insurmountable, as individuals can overcome constraints (Young 2001: 15). Additionally, membership in a group is not a sufficient condition for saying that an individual suffers from inequality. Rather, membership in a group that suffers from structural inequality is indicative of an increased risk of being disadvantaged. So in some sense, the high-achieving 10% may be constrained at the level of opportunity, but not constrained at the level of outcome.
d. The problem with averages in inter-group comparisons
Young discusses Douglas Rae’s critique that group-conscious assessments reduce individual circumstances to a statistical average when viewed in the group context (2001: 5). Consequently, similarities between individuals in a group are over-emphasized while differences between groups are exaggerated. As these group-level differences are observable in statistical variation, relevant distinctions between individuals in groups are glossed over. The result is that corrective action to rectify a structural inequality cannot rely on group-level comparisons alone, since individuals are ultimately the concern of justice and individual variation exists. This raises doubts as to whether or not the group-conscious approach is precise enough to be action-guiding after a structural inequality has been identified.
Recall the case with ethnic group A with the subgroup of well-off members that compose 10% of the total group. Group A is disadvantaged in the society due to structural inequality, fulfilling the conditions of being relevantly unequal compared to others with a plausible narrative of being historically disadvantaged. It is known that group A is unequal because its average welfare is lower than groups B and C. Now suppose the society collectively decides to rectify this problem through various policies, resulting in statistical equality between all three groups. However, the outcome of these policies is to make the well-off subgroup of A very, very well-off, effectively raising the group average to an equal footing with groups B and C.
In one sense, it is difficult to escape Douglas Rae’s critique of using group-based assessments. This is because inter-group comparisons necessarily look at these in terms of averages, the problem of which can never be fully eliminated. Nevertheless, two things can be said in response. The first is that the categorization of individuals into groups should be as precise as possible in terms of qualifying characteristics. The more precise it is, the more definite is the evidence that there is a state of inequality despite individual variations. This suggests that the objection to using group averages is contingent on the classification not being useful for addressing those least well off, but not that such classifications are necessarily problematic.
Secondly, the case of artificially raising a group average by merely elevating the most well-off would be an instance where the group classifications should be more narrow. If a clearly delineated subgroup varies widely within its own group, then the initial group classification should be made more specific in order to take into consideration only those individuals that are actually disadvantaged. It should be noted that the purpose is not to equalize groups here, but to provide individuals with a strong equality of opportunity (Young 2001: 17). These two points serve to show that the problem of using averages between groups in assessments is not necessarily a serious problem and that the group classification method should, as much as possible, represent only individuals that are actually structurally constrained.
e. The ambiguity of structural explanations
The final objection to the group-conscious approach is that it relies on structural explanations that are ambiguous. The problem specifically is that it is unclear what exactly constitutes a plausible causal narrative since the features of structural inequality can be complex, subliminal, and subtle.
Young advances two primary criteria to determine when a group suffers, which are that that category of individuals is unequal in a relevant way and that a plausible causal narrative can be provided for how the rules, relations, and cumulative consequences of group action specifically conditions the lives of that group (2001: 15). These conditions for determining if a group suffers an injustice are very demanding, in that they require an exhaustive and clear case of structural inequality that takes into consideration counter-explanations of group culture. However, ultimately the force of this objection is rather small, since it can be easily conceded that there is room for a certain amount of doubt regarding any narrative of structural inequality. Regardless, this does not mean those plausible narratives of structural inequality are impossible or too ambiguous to even be plausible.
5. Conclusion
In summary, I have argued that in order for judgments to be made about structural inequality, a group-conscious approach is required. Since equality has moral value in considerations of justice, inequalities should be identified and addressed. But conventional approaches cannot properly take into account structural inequalities, because these require group-level comparisons and assessments of cumulative social structures. The group-conscious approach offers a way to assess these in a way that makes a judgment of structural inequality contingent on there being a category of individuals that is unequal in a relevant way and due to relevant social reasons.
Additionally, this approach can address and overcome important objections that demonstrate its theoretical robustness. It maintains individuals as the ultimate focus of moral concern, avoiding objections from methodological and moral individualists. It can also serve to explain why the cultures of certain groups are less successful in society, since these may be the result of past and current structural inequality. Also, it does not necessarily skew salient individual differences, since the evaluation uses group-level comparisons as an indicator of inequality rather than as a sufficient condition of inequality. Lastly, while there will be a degree of ambiguity in structural explanations, the vast evidence for multiple disadvantages faced by certain groups suggests that this does not negate such explanations.
What I have not addressed in this paper is the way in which structural injustices can or ought to be corrected. I do not provide action-guiding prescriptions for policies, nor do I offer a justification for restitution or compensation in rectifying structural inequalities. Rather, the group-conscious approach offers a robust theoretical framework for diagnosing and making morally relevant judgments about social justice. Any corrective action to follow from this would depend on the extent, plausibility, and form of the structural injustice in question, the effectiveness of the specific proposed policy measures, and the distributive cost of doing so.