When, if ever, is affirmative action morally justified? 

1 Introduction

Few topics can rouse intemperate discussion like affirmative action. In this essay, I shall offer reasoned, moral justifications for affirmative action, and I shall focus on affirmative action as it relates to race. In section two I shall offer some definitions, and examine Pojman’s taxonomy of affirmative action, adding a type of my own devising: moderate affirmative action. In section 4 I shall examine which types of affirmative action might be morally justified. In section four I shall consider some objections to my position and conclude in section 5. 

2 Definitions and Pojman’s taxonomy

Firstly, I define racism as unjust or wrongful discrimination against a person based on that person’s race. Codified racism is racism perpetuated in policy or action by social institutions such as corporations, governments or the academe. Affirmative action is actions or policies undertaken by governments, corporations, or other social institutions that seek to correct historical or ongoing racism in those, or other, institutions. However, this definition is quite broad, so I am going to use Pojman’s (1992) taxonomy of distinct types of affirmative action and then offer my own addition.

Strong Affirmative Action (SAA) – involves ‘positive steps to eliminate past injustice, such as reverse discrimination, hiring candidates on the basis of race… in order to reach equal or near equal results, proportionate representation in each area of society.’ (p. 183) Examples of this might be all racial minority shortlists for political party candidate selection, quota systems within corporations for racial minority representation, or preferential selection of racial minority candidates for academic places or employment. 

Weak Affirmative Action (WAA) – involves ‘the elimination of segregation… widespread advertisement to groups not represented in certain privileged positions, special scholarship for the disadvantaged classes… using underrepresentation or a history of past discrimination as a tie breaker when candidates are relatively equal.’ (p. 183)

In addition to Pojman’s categories, I’d like to add one of my own:

Moderate Affirmative Action (MAA) – includes all the steps of WAA plus encouraging the view in those who are to enact MAA that racism has a negative effect on the constellation of virtues and vices that make up the life of members of racial minorities. Decision-makers undertake to take account for this, in a morally relevant way, when reaching decisions where racial minorities are competing for a given good. So, in the case of applying for a job, if an applicant is of a racial minority who have been or are viewed as academically inferior, then the panel doing the hiring should consider that given this, the environment in which the applicant cultivated their academic virtues was hostile (to some degree or another) and that this is both praiseworthy and may explain underachievement in a candidate. 

The literature seems to be unanimous that WAA is morally acceptable, but it shall be my contention that MAA is justified in the context of ongoing societal racism (even in the absence of codified racism) because of this societal racism and the detrimental effect it has on the ability to cultivate relevant virtues for members of racial minorities, and that SAA is liable to create as much injustice as it seeks to correct. In the following section, I shall argue that only MAA can be properly sensitive to the correct targets of affirmative action.

3 Targets of affirmative action

In this section I shall discuss the proper targets of affirmative action. In 3.a. I shall discuss whether groups or individuals are the proper targets of affirmative action, argue for the view that only individuals are proper targets of affirmative action, and that only MAA is properly sensitive to this requirement. In section 3.b I shall address whether victims of historical codified racism, in the absence of ongoing codified racism, are proper targets for affirmative action, and argue that they are, and that MAA is the best form of affirmative action for tracking the ethical considerations raised by this question.

a: Individuals or groups?

One argument for affirmative action runs roughly as follows (paraphrased from (Sher, 1975)): 

1. Racial minority groups have suffered injustice and wrongdoing in the past due to racism

2. Injustice and wrongdoing deserve restitution

3. Therefore, members of those groups should get strong preferential treatment, in the form of affirmative action

At first glance this argument seems compelling, however, it is wrong-headed. Here, I follow Sher (1975) in arguing that we only think of persons (be they corporate persons or individuals) as the proper targets of injustice and restitution, since to be a target for restitution implies being a target for culpability. We do not permit that a group can be held to give restitution to another group; consider, we would find it abhorrent for the ethnic majority group, in it’s entirety, to have to give up a proportion of their assets as restitution to the minority groups, regardless of any given individual’s culpability. Simply put, to be the proper target for restitution is to be the proper target for culpability, and similarly so for justice and injustice. Groups, therefore, cannot be proper targets of these sorts of judgments and actions. So, we might reformulate this in the following way:

4. Individual members of racial minorities have suffered injustice and wrongdoing in the past due to racism

5. Injustice and wrongdoing deserve restitution

6. Therefore, these individuals deserve restitution in the form of affirmative action

This argument is significantly less problematic, since it properly targets individuals. Nonetheless, different forms of affirmative action have different moral statuses in relation to this argument. SAA programs have no way of targeting the proper individuals who have or have not historically suffered. Let us consider two members of racial minority X. We’ll call them Jill and Susan. Now, let’s imagine Jill suffered under historic codified racism, and is therefore older than Susan, but Susan did not. Under SAA programs, as members of X, both Jill and Susan are deemed to be equally deserving recipients of the benefits of affirmative action, yet surely this is an injustice against Jill, if this argument is right (which surely, it is, if we view historic racism as an injustice) Furthermore, WAA similarly presents both Jill and Susan as equally deserving of the benefits of affirmative action, solely by reference to their being X. MAA, however, has the tools to discriminate between Jill and Susan, since, if Jill and Susan are competing for some good, let’s say, access to social housing, the decision-makers can take into account how racism has effected the development of Jill’s virtues compared to Susan’s. Let us say that one criterion that the decision maker must consider is integration into the local community. Let us also say that minority X, and Jill, suffered under segregationist policies in the past. Now, by the lights of MAA and using knowledge off Jill’s (and Susan’s) age and previous addresses, the decision-maker can make a reasonable inference that Jill has suffered under the segregation regime and this will have significantly affected her environment in such a way as to discourage the cultivation of virtues relevant to integrating with the community, but that Susan has suffered only the social after-effects of such policies. Thus, MAA can properly discriminate between members of minority groups in the right sort of way and prevent injustices when members of the same racial minority are in competition for the same resource, and this is surely the intuitively just result.

B: Historical codified racism

Our discussion of Jill and Susan brings out an interesting point about those who suffered directly under codified racism and those who are their descendants. Consider argument (6), which suggests that individuals who suffered directly under racist institutions deserve restitution from said institutions. That, however, says nothing about Susan. As far as argument (4) - (6) is concerned, Susan deserves no special treatment. And, we can see why by way of a simple thought experiment. Let us imagine that an interventionist deity in Susan’s society put an end to racism in all its possible forms, including wiping every racist thought, value and norm from every mind overnight before Susan was born. Born into a society totally free of racism and never having suffered directly under racist policies, Susan does in fact seem underserving of the benefits of affirmative action, since racism in all its forms ended overnight.

However, I wish to argue that this is an obviously incomplete position. Racism could never disappear overnight. It shall be my argument that historical codified racism creates ongoing societal racism, and that affirmative action is the correct form of restitution:

7. Institutions (especially governments) historically advocated, perpetuated and/or benefitted from racist values and norms, taboos and stereotypes within wider society,

8. Those norms and values are still acted upon in society, sometimes explicitly and sometimes implicitly, and those actions harm their targets i.e., racial minorities, including by those minorities themselves

9. Institutions owe restitution for the damage done to individuals for the perpetuation of racist norms and values

10. Therefore, institutions should implement affirmative action by way of restitution.

Now, here it should be noted that institutions that are relatively recent may not be as strongly bound by this argument (or the previous one,) however, these institutions benefit from older institutions, such as their financiers, and in so doing, inherit some benefit from the states of affairs these older institutions helped create, and thereby should inherit some of the responsibility.

Again, different forms of affirmative action have different statuses with regards to these arguments. For affirmative action to be justifiable it must succeed in identifying those norms and values that have caused harm and be appropriately sensitive to them. In doing so, it must be able to show differing levels of restitution for differing levels of offence. So, let’s broaden our scope briefly and consider cases of just competition between racial minorities. If any form of affirmative action can give us the just and moral result here, then it should give us the right results in cases of competition between racial minority and racial majority persons. Consider a society with two racial minorities, minority Y and minority Z. Now, the institutionally-created norms and values spread about each of these differ. In the case of minority Y, the racist values advocated and perpetuated by institutions pertained to their lack of intellectual capacity, and they would always be bad at anything involving any amount of intellectual work. The norms that pertained to Y, therefore, were that they would not be accepted into positions that required intellect, but they were otherwise seen as members of the community. Minority Z, on the other hand, were viewed as a contaminant, the majority value-judgement about them that they were unclean or somehow unfit, and therefore the norm was that they were restricted to a small number of educated professions, particularly in this case professions α, β, γ, but there was no taboo against them attending university or to imply that they were intellectually deficient. Now, let’s also say that the institutional advocation and perpetuation of these values is some time into the past, let’s say 30 years, but these norms and values persist both explicitly in some sections of society and frequently implicitly elsewhere.

Next, consider Monty and Zach, who hail from minorities Y and Z respectively, and two scenarios in which they are competing for a scarce resource and see which formulation of affirmative action gives us the intuitively respectable outcome. In the first instance, Monty and Zach are competing for the same academic place at a university. Each is approximately equal in all the relevant criteria considered by the university, though Zach has a very slight advantage. With regards to the argument (7) – (10), who should get the place? Those adhering to WAA have nothing to say here, since they are both from racial minority groups and cannot be distinguished in this manner for the tiebreaker. Either candidate is equally good. But doesn’t that do a disservice to Monty, who to get to the same place must have cultivated virtues in an environment hostile to their cultivation? Those adhering to SAA will make a decision based on the relevant quota or percentage distribution at play and will simply ignore the fact that the norms and values that surrounded Monty will have strongly discouraged the cultivation of academic virtue, and so even if Monty does get the place in the end (due to filling the appropriate ‘racial slot,’) he will do so for unjust and demeaning reasons. The university would not have considered whether he possessed the virtues that would make him a good student. Again, however, MAA can distinguish in this case. The admissions supervisor(s) can consider how the norms and values of society at large have affected Monty’s development of academic virtues as part of the larger case in his favour. That may not guarantee that Monty gets the place, but it does mean that the admissions supervisor is discriminating in the right way; accounting for how Monty came to be where he is without blindly equating him with his racial characteristics.

Now, let’s imagine that both Monty and Zach are applying for the same promotion in the same company. This would take them into profession θ at a senior level. Now, the employer in question is convinced that both men are of equal intellect having worked with both for years. In fact, as far as the employer can tell, both are equally capable of doing the job. If the employer operates using only WAA, then he can essentially flip a coin or pick a favourite. However, that seems unjust to Zach, who may have fought against disapproving parents or, perhaps, the employer himself sees it as ‘not quite right’ that a member of Z wants to do θ. If the employer uses SAA, again, we repeat the ‘wrong reasons’ objection above; whomever gets the job, it seems they get it for the wrong reasons. MAA, however, should help the employer overcome his worries about Zach entering the profession, and give him reason to think that Zach has cultivated virtues in conditions hostile to their cultivation, and this is to tell in Zach’s favour. This is surely the right result and answers worries found in Pojman (2006, Section 7) about unjust discrimination between different racial groups.

c: Summary of my position

Drawing together the above arguments, I wish to clarify my overall position in answer to the titular question. Affirmative action is justified when it targets individuals (section 3.a.) who are members of racial minorities that have suffered historical, codified racism (section 3.b.), and when that racism has resulted in ongoing social disadvantage at the hands of the norms and values that, implicitly or explicitly, result in racist actions, and when that affirmative action is moderate, which is to say, properly sensitive to the way in which the particular forms of racism are likely to have affected a given individual’s cultivation of virtues relevant to the to the good that they are attempting to acquire. 

4 Objections and responses

a: Affirmative action is unnecessary

Given that rejecting racial segregation is included in WAA, I daresay that at least WAA is necessary. To deny that affirmative action is necessary is to deny that Governments should end racial segregation, of the form ‘separate but equal,’ and should not undertake to better the lot of its most vulnerable by any means whatsoever. Even if we set this worry aside, considering the arguments deployed in section 3a, particular (4)-(6) and (7)-(10), it seems like justice demands that action be taken to right the wrongs done to the individuals in question, and MAA can provide this. To deny affirmative action is necessary is to deny either that (i) historical racism has had no bearing on extant individuals or (ii) that ongoing societal racism is simply non-existent. As for (i), my preceding arguments have shown this to be untrue; historical racism has had a significant effect on the ability of members of ethnic minorities to cultivate certain sorts of virtues that are relevant to the workplace. I hope that (ii) stands as unworthy of rebuttal; the amount of research that demonstrates this is unenumerable and I could not possibly hope to cite even a cross-section of it, for some examples from across many disciplines see (Zamudio & Rios, 2006), (Samuel & Wane, 2005) and (Barbee, 1993). If either individuals affected by historical racism or ongoing societal racism exist, then the arguments for MAA stand, and if they stand for MAA, they stand for WAA also. 

b: MAA is ‘reverse discrimination’

Versions of this argument appear in (Pojman, 1992; 2006) and refutations in various forms in (Burns & Schapper, 2008; Shaw, 1988; Sher, 1975). The following argument is taken from Pojman (2006), adapted for my moderate case, and my defence inspired by the others.

Pojman’s ‘reverse discrimination’ argument goes roughly as follows:

1. ‘Respect for persons entails that we treat each person as and end in him or herself’

2. ‘What is wrong about racism against Blacks is that it fails to treat Black people as individuals, judging them instead by their skin color [sic] not their merit.’

3. What is equally wrong about Affirmative Action [SAA] is that if fails to treat White[s]… with dignity as individuals, judging them by their race.’

4. ‘Weak affirmative action discriminates weakly, strong affirmative action discriminates strongly’ (2006, Section 7 )

5. Therefore, any affirmative action stronger than WAA is unjustifiable (Pojman does not argue against WAA.) 

Now, Pojman’s obvious prefers a deontological framework, as he mentions it throughout the paper, and I do not. I could simply argue that we operate with fundamentally different ethical assumptions and leave it there, however, that would be unsatisfying. Therefore, let us grant Pojman his deontological framework. I would then argue, however, that (3) is false, specifically when considering MAA. MAA does not, in fact, in any way denigrate the personhood of the racial majority, because it simply looks to use information about history and society at present to inform decision-markers about the personhood of minority candidates. MAA uses extant information about the minority candidate’s race to enhance the mental picture of the circumstances in which this candidate became the person they are; to imagine more fully the minority candidate as an end in themselves. In short, under MAA, the race of the minority candidate is taken to be tangentially salient (in that it can be taken to give information about the candidate, (and the sort of person/end they have become) but it has nothing to add, nor nothing to detract from, the racial majority. Therefore, (3) is untrue under MAA and the argument is invalid. 

5 Concluding remarks

Herein I have advocated the moral justification of moderate affirmative action, in a context of historical institutional and ongoing societal racism, by reference to an environment of hostility to cultivating certain virtues in racial minorities, that is sensitive to the sort of discrimination that such groups suffer. This argument is designed to be sufficiently general that it can be, with some work, re-applied to any group that meets the criteria above. Of course, there remain many questions, like when should someone count as part of a racial minority? This questions, and other like them, must be the topic of future research. 

Bibliography

Barbee, E. L. (1993). Racism in U.S. Nursing. Medical Anthropology Quarterly, 7(4), 346-362.

Burns, P., & Schapper, J. (2008). The Ethical Case for Affirmative Action. Journal of Business Ethics, 369-379.

Pojman, L. P. (1992). The Moral Status of Affirmative Action. Public Affairs Quarterly, 181-206.

Pojman, L. P. (2006, 10 07). The Case Against Affirmative Action. Retrieved 02 14, 2018, from Sacremento State: http://www.csus.edu/indiv/g/gaskilld/business_computer_ethics/the%20case%20against%20affirmative%20action.htm

Samuel, E., & Wane, N. (2005). "Unsettlingg Relations": Racism and Sexism Experienced by Faculty of Color in a Predominantly White Canadian University. The Journal of Negro Education, 74(1), 76-87.

Shaw, B. (1988). Affirmative Action: An Etthical Evaluation. Journal of Business Ethics, 763-770.

Sher, G. (1975). Justifiying Reverse Discrimination in Employment. Philosophy & Pubblic Affairs, 159-170.

Zamudio, M. M., & Rios, F. (2006). From Traditional to Liberal Racism: Living Racism in the Everday. Sociologgical Perspectives, 49(4), 483-501.