Power holders in the United States tend to be white, male, straight, and (profess to be) Christian. Membership in these groups also seems to confer a degree of advantage relative to people outside of these groups. Yet, as been noted in the previous essays, some claim that the people in these groups are the “real victims” today. In this essay I will look at how a version of the fallacy of anecdotal evidence can be used to “argue” about who is “the real victim.”
The fallacy of anecdotal evidence is committed when a person draws a conclusion about a population based on an anecdote (a story) about one or a small number of cases. The fallacy is also committed when someone rejects reasonable statistical data supporting a claim in favor of a single example or small number of examples that go against the claim. The fallacy is considered by some to be a variation of the hasty generalization fallacy (drawing a conclusion from a sample that is too small to adequately support that conclusion). The main difference between hasty generalization and anecdotal evidence is that the fallacy anecdotal evidence involves using a story (anecdote) as the sample.
Here is the form of the anecdotal evidence variation often used to “argue” that an advantaged group is not advantaged:
Premise 1: It is claimed that statistical evidence shows that Group A is advantaged relative to Group B
Premise 2: A member of Group A was disadvantaged relative to a member of Group B.
Conclusion: Group A is not advantaged relative to Group B (or Group B is not disadvantaged relative to Group A).
To illustrate:
Premise 1: It is claimed that statistical evidence shows that white Americas are advantaged relative to black Americans.
Premise 2: Chad, a white American, was unable to get into his first choice of colleges because affirmative action allowed Anthony, a black American, to displace him.
Conclusion: White Americans are not advantaged relative to black Americans.
The problem with the logic is that an anecdote does not suffice to establish a general claim—an adequate sample would be needed to make a strong generalization.
But one must also be on guard against another sort of fallacy:
Premise 1: It is claimed that statistical evidence shows that Group A is advantaged relative to Group B.
Premise 2: Member M of Group A is disadvantaged relative to Member N of Group B.
Conclusion: The disadvantage of M is morally acceptable, or M is not really disadvantaged.
To illustrate:
Premise 1: It is claimed that statistical evidence shows that men are advantaged relative to women.
Premise 2: Andy was disadvantaged relative to his boss Sally when she used her position to sexually harass him.
Conclusion: The disadvantage of Andy is morally acceptable, or Andy was not really disadvantaged.
While individual cases do not disprove a body of statistical evidence they should not be dismissed. As in the illustration given above, while men generally have the workplace advantage over women, this does not entail that individual men are never at a disadvantage relative to individual women. It also does not entail that, for example, men cannot be the victims of sexual harassment by women. As another illustration, while white men dominate academics, business, and politics, this does not entail that there are not injustices against specific white men in such things as admission, hiring and promotions. These sorts of situations can lead to considerable moral debates about harm.
One excellent example is the debate over affirmative action. The oversimplified justification is that groups that have been historically disadvantaged are given a degree of preference in the selection process. For example, a minority woman might be given preference over a white woman in the case of college admission. The usual moral counter is that the white woman is wronged by this: if she is better qualified, then she should be admitted—even if this entails that the college population will remain almost entirely white. The usual counter is that the white woman is most likely better qualified because she has enjoyed the advantages conferred from being white. For example, her ancestors might have built wealth by owning the ancestors of the black woman who was admitted over her and this inherited wealth meant that her family has been attending college for generations, she was able to attend excellent schools and her family could pay for tutoring and test preparation. This can be countered by other arguments, such as how the woman did not own slaves herself, so it is unfair for her to not be admitted on the “merit” arising from her advantages. One can, of course, consider scenarios such as cases in which the black woman is from a wealthy family and the white woman is from a poor family. Such cases can, of course, be considered in terms of economic class—one could argue that this should also be a factor. This obviously all leads to the moral issue of whether it is acceptable to inflict some harms on specific members of advantaged groups to address systematic disadvantages, which goes way beyond the scope of this work. Fortunately, I do not need to settle this issue here. This is because even if such anecdotes are examples of morally wrong actions, they do not disprove the general statistical claims about relative advantage and disadvantage between groups. For example, even if a few white students are wronged by affirmative action when they cannot attend their first pick of schools, these anecdotes do not disprove the statistical evidence of the relative advantage conferred by being white in America. After all, the claim of advantage is not that each white person is always advantaged over everyone else on an individual by individual basis. Rather it is about the overall advantages that appear in statistics such as wealth and treatment by the police. As such, using anecdotes to “refute” statistical data is, as always, a fallacy. But what about cases in which members of an advantaged group do suffer a statistically meaningful disadvantage in one or more areas?
While falling victim to the fallacy of anecdotal evidence is bad logic, it is not an error to consider that members of an advantaged group might face a significant disadvantage (or harm) because of their membership in that advantaged group. As would be expected, any example used here will be controversial—I will use the Fathers’ Rights movement as the example. The central claim behind this movement is that fathers are systematically disadvantaged relative to mothers. While there are liberal and conservative versions, the general claim is that fathers and mothers should have parity in the legal system on this matter. Critics, as would be expected, claim that men tend to already enjoy a relative advantage here. But if the Fathers’ Rights movement is correct about fathers being systematically disadvantaged relative to mothers, then this would not be mere anecdotal reasoning. That is, it would not just be a few cases in which individual fathers were disadvantaged relative to a few individual mothers—it would be systematic injustice. But would this area of relative disadvantage disprove the claim of general advantage? Let us look at the reasoning:
Premise 1: It is claimed that statistical evidence shows that Group A is advantaged relative to Group B.
Premise 2: But Group A is disadvantaged relative to Group B in specific area C.
Conclusion: Group A is not advantaged relative to Group B.
As presented, this would be an error in reasoning—Group A being disadvantaged in one area would not prove that the group is not advantaged relative to Group B when all areas are considered. To use an analogy, the fact that Team B outscored Team A in the fifth inning of a baseball game does not entail that B is leading. It must be noted that a similar argument with multiple premises like Premise 2 could show that Group A is not advantaged relative to Group B—after all, establishing adequate statistical evidence would obviously be adequate. There are, of course, questions about how to determine relative advantage and these can be debated in good faith. One obvious point of dispute would be the matter of weighting. For example, if fathers are disadvantaged relative to mothers, how would this count relative to the pay gap between men and women? And so on for all areas of comparison. This does show the need to consider each area as well as a need for assessing value—but this is not unique to the situation at hand and one could, as is often done, assign crude dollar values to do the math.
In closing, while individual wrongs and wrongs done to members of advantaged groups as members of that group can occur, they do not automatically disprove the statistical data.
Leave a Reply