On Fish's "When Identity Politics Is Rational"

Stanley Fish's essay on identity politics, he asks "Is it so irrational and retrograde to base one’s vote on the gender or race of [sic] religion or ethnicity of a candidate?" Here is Fish's example of when identity politics is rational:

Yet every African American – conservative or liberal, rich or poor, barely educated or highly educated – meets with obstacles to his or success and mobility that are all the more frustrating because they are structural (built into the culture’s ways of perceiving) rather than official. To the non- African American these obstacles will be more or less invisible, especially in a country where access to opportunity is guaranteed by law. It makes sense, therefore, that an African American voter could come to the conclusion that an African American candidate would be likely to fight for changes that could remove barriers a white candidate might not even see.

Here's how I see this argument breaking down, in simplified parts:

-- (Premise 1) Identity group members share unique experiences
-- (Sub-conc from P1) Group members share a unique viewpoint on the issues
-- (Premise 2) Non-group members are less likely to have had he same experiences as in-group members
-- (Sub-conc from P2) Non-group members are less likely to share the group's viewpoint on the issues, or even less likely to recognize the issues as issues at all
-- (Conclusion) Group members are more likely and more able to fight on the issues important to members of the group

Unfortunately for Fish, this example makes identity politics "rational" only in the most trivial sense. (I'm picking up here on a comment on Fish's essay, that Fish himself points out in a follow-up to his essay. Unlike blogger Nathan Nahm, though, I will not try to push Fish into the ethical realm. I will merely try to push him into a more substantive look at the "rationality" he has purported to establish.) It seems to me that the only rationality Fish has proven with his example is that peolpe have reasons for practicing identity politics (or more accurately, reasons for practicing the subset of 'interest' identity politics that Fish defines). But this standard is truly trivial. People have reasons to support any number of bad arguments. People have reasons for believing we never landed on the moon, reasons for killing other people, reasons for voting Republican. Plenty of arguments are supported by reasons, but are nonetheless incoherent arguments.

I don't mean to insult Prof. Fish. He is obviously well aware of the distinction between reasons and good/relevant reasons. He draws this distinction himself when criticizing 'tribal' interest politics ("'Because she is a woman as I am' is of course a reason, but it is not a reason of the relevant kind, a reason that cites goals and programs, and argues for them"). So what is the difference in the reasons given for 'tribal' indentity politics and 'interest' indentity politics? It appears Fish tries to establish that the reasons grounding 'interest' identity politics are tied to the issues -- they are just like any other interest. But if we look more closely at the argument, we see that the connection between the issues and 'interest' identity politics is tenuous at best.

I'll concede that the sample argument is valid -- the main conclusion certainly follows from the premises and inferences. If that's all Fish was going for, then kudos to him. But I'll reiterate, I find this accomplishment fairly trivial. I could argue: "All pigs have wings; all things with wings can fly; therefore, pigs can fly." I just provided a valid argument. So apparenlty, by Fish's standard in this article, people who think pigs can fly are completely rational. The question everyone actually cares about is whether the argument is sound. Fish seems like he might not actually be interested in this question. He doesn't hold against 'interest' identity politics that "the calculation [to favor a candidate based on identity] may or may not pan out (successful candidates both disappoint and surprise), but it is a calculation of the right kind." But I'd don't think we'd be as sympathetic to the man left with a dead pig after throwing it from his roof as Fish is being towards practictioners of identity politics. And I don't mean that in a normativity-laden way. I'm not asking Fish to step into the moral/ethical/social realm to answer this question. I'm asking him to critique the premises, inferences, and soundness of the argument with the same rigor he defended its validity. If we know the premises to be false, or the inferences along the way to be invalid, then we shouldn't be surprised when the pig can't fly. (More true to the analogy -- we have no good reason to throw the pig off the roof and believe it will fly.)

So, here are the two main problems I see in the argumentative structure of 'interest' identity politics. They are, in some ways, a re-focused reiteration of the problems with identity politics that Fish sought to show do not make identity politics irrational in the first place. My point is to show that he hasn't explained these problems away. Rather, they are the very problems standing in the way of the substantive ethical-netural/value-netural rationality of 'interest' identity politics.

1) The premises are false. A coherent notion of identity politics relies upon a coherent notion of identity. Fish dismisses the disparate ideas and experiences within various groups as irrelevant to the question of identity politics. At the same time, he ignores the possibility of strikingly similar ideas and experiences among widely disparate identity groups. But the sample rational argument from an interest identity voter is premised on the fact that identity groups share something exclusive and unique. For a counterexample to this premise, look no further than Toni Morrison's oft-quoted article on Bill Clinton as the first black president. She says, "Clinton displays almost every trope of blackness: single-parent household, born poor, working-class, saxophone-playing, McDonald's-and-junk-food-loving boy from Arkansas." This might seem like a small and insignificant step away from Fish's point. What's the difference between African Americans practicing 'interest' identity politics with another African American and practicing 'interest' identity politics with someone with African American-like experiences? Is it really any different, any more rational? Absolutely. That difference is at the core of this debate. Fish might be right that some "identity" voting is rational. But it is not along the hard and fast lines of identity that Fish thinks define indentity voting ("when you vote for or against someone because of his or her skin color, ethnicity, religion, gender, sexual orientation"). Even if an identity group votes on the premise that the politician's experience gives that politician a unique understanding of the issues important to the group, that experience (and thus that understanding) demonstrably does not break down along distinctions of race, gender, religion, sexual orientation, etc. To work under such an assumption would be no better than working under the assumption that pigs have wings.

2) The sub-conclusions are not justified. This second point is empirically testable. One could conduct a survey of any given identity group, scoring the issues most important to that group. Then, one could look at how well the proposed and enacted legislation of a politician of that identity group correlates to the needs and ideas of the group, compared to other non-group politicians. I obviously don't have such statistics. I would bet the expected identity match up would sometimes come out on top. Sometimes there would be a tie. Sometimes the non-identity politician might come out on top. The high likelihodd that there would be such a diversity of outcomes is enough to show that this is not a "calculation of the right kind." This point is also linked directly to my points in (1). A man who has spent years in the Hanoi Hilton might have a better understanding of the deep importance of control over one's own body than a wealthy girl from a liberal family whose freedom of choice was never challenged. And no, I'm not saying McCain is stronger on women's rights than Hillary -- it's just a hypothetical -- I have enormous respect for Hillary's great escape from her conservative family. But a more important indicator than man, woman, P-O-W, or P-O-GOP, is a critical look at a candidate's background and voting record.

As a final point, with a little twist of irony, it seems to me that the 'tribal' indentity politics, which Fish seems to dismiss offhand, actually might be the most rational type of identity politics. Fish cursorily brushes past the idea that "it is possible to argue that the election of a black or female president, no matter what his or positions happen to be, will be more than a symbolic correction of the errors that have marred the country’s history, and an important international statement as well." Like Fish, I'll avoid making a firm value judgment about this kind of argument. But as much as we strive to be race-blind, color-bling, sexuality-blind, etc, these distinctions are obviously very much at play in our discourse and in our lives. Perception of the group by other groups, and self-perception of the group from within will undoubtedly be altered by the election of a marginalized figure to the highest office in the land. It then becomes a balancing act, a value judgment, whether any merits derived from the election would outweigh / complement various policy proposals. While I don't think such an argument about women and blacks can tip the scales when both a woman and an African American are in the race (OK, so maybe just a little value judgment!), the rudimentary 'tribal' version may be the most sound and coherent way in which identity politics can function.