Metrics and levers, or how to cheat at meat.

Substantive representation, descriptive representation, and bigotry

It’s only a precision instrument until you lean on it

If you’ve been around the deli for any period of time, you know how important it is to ensure that what you’re buying is accurately measured. It is all too easy for the store to cheat by having someone put their thumb on the scale, skewing it in their favor. After the first time you take home what’s labeled as a pound of cold cuts and find that it only comes to 13 oz. on your own scale, you learn to be more cautious and less trusting.

Skewed measurement is also a fundamental problem in politics, where the public weighs in through voting. The core principle of democracy is that the legitimacy of the government is rooted in the consent of the people. Our leaders are not emperors but representatives of our will; they follow us. This is the ideal that we aspire to but all too often fall short of.

The fundamental measure of the success of a democracy is to what extent officials, particularly elected ones, act to further the goals and interests of their constituency. Distortions such as gerrymandering, voter suppression, and outright cheating leave “our” leaders free to follow an agenda hostile to those they act in the name of. They undermine the checks and balances intended to prevent them from going rogue.

When they take the power we grant them and abuse it for their own gain, this is the essence of corruption. What we instead want is a match between citizens and government on matters of policy: this is called substantive representation.

However, when you hear “representation” invoked today, it is often in a related but distinct meaning. Hashtag slogans like #representationmatters refer, instead, to the diversity of identities. That version of representation is invoked to justify voting for people on the basis of how closely they mirror the appearance of those who elected them: this is called descriptive representation.

There’s no conflict between the two, and in a fair system—one free from discrimination—we would expect that our leadership naturally exhibits demographics in line with those of the voters, just as their policies would match those of the voters. Any disparity would itself be an indicator of discrimination.

So, for example, the fact that women make up a quarter of Congress despite being half the population is a bad sign. It is indicative of systemic bigotry. But is it a symptom or the cause? How do we tease out the reason why?

In the end, there are really only two explanations for such ongoing disparity. Either the oppressed group continues to be oppressed or its members really are (as the bigots claim) inherently inferior. So if someone denies that the underrepresentation is caused by discrimination, they are necessarily bigoted. But, given that a lack of descriptive representation reveals bigotry, does increasing such representation counter it?

Well, to the extent that being a member of a group leads to representing its interests, yes, having a more descriptively representative leadership would entail having a more substantively representative one. So, for example, if women could be counted upon to support women’s issues, then having more female representatives would help women be more equally treated. But to what extent is this true?

The situation seems mixed at best. It is, unfortunately, not difficult to find examples of officials whose policies do not further the interests of the disadvantaged identity groups that they themselves are members of. For example, there are any number of women in politics who oppose women’s rights, especially reproductive rights, despite being women.

I could also offer examples such as Ben Carson, a Black man who works for an overtly racist president, denies that systemic racism still exists, and insists that it is the liberals who are actually racist. I could point to Ted Cruz, a Hispanic who supports xenophobic, racist anti-immigration policies against Hispanics. I could go on and on with examples of these identity group traitors, and I will mention a few more before I’m done, but I want to also look at the patterns.

The first women to gain entry into a boys’ club have to work extra hard to be seen as “one of the boys”. From this, you can expect that early examples of descriptive representation might fall short of substantive representation, particularly if they’re under the banner of a party opposing equality. If anything, they’d have to show themselves to be huge identity traitors in order to succeed.

Since the RNC has been the party of bigotry since Nixon, we can look to them for examples. The first Black Supreme Court Justice appointed by the RNC is Clarence Thomas and their first female VP candidate was Sarah Palin. Contrast this with Thurgood Marshall and Geraldine Ferraro, respectively, and you can see that descriptive representation can be entirely compatible with bigotry. Moreover, having a token on their side is a great way to silence criticism of their bigoted policies; the political version of “but I have a Black friend”.

Even if we limit our consideration to those who support equality, it is not clear that, say, a female advocate for women’s rights is a better choice than a male one. We might hope that a woman would care more about these issues, prioritizing them more highly, and that may well be the case. However, as was just pointed out, it is frequently not. Being a woman is neither necessary nor sufficient for supporting equal rights for women.

It might also be counterproductive. In terms of appearances and impact, when a woman speaks in favor of her own rights, it’s easy for bigots to dismiss it as mere partisanship. When a man does, that removes one counter. You have more credibility arguing for the sake of a group that you are not a member of, because it cannot be interpreted as mere self-interest. Bigots are more likely to take you seriously simply because you look like them as opposed to the people they’re bigoted against. And while it’s nice to preach to the choir, we make progress by persuading, or at least silencing, our opposition.

There’s also the problem of losing. Discrimination against a group makes it harder for members of that group to gain positions of political power so they can then fight against this discrimination. Consider that Hillary Clinton, a strong champion for women’s rights, faced rampant sexism in her presidential run precisely because she is a woman. Her husband, whose views closely mirror her own, got elected despite them because he has a penis. So, for example, the goal of equal rights for women may be better served by choosing a man, if the man can win and the woman can’t.

Even when a candidate from a disadvantaged group overcomes this disadvantage and wins, the reaction against that victory can cause harm. Blowback can be deadly. Our first Black president, merely by his existence, united the far right against him, leading to the election of Trump. The presence of a Black man in the White House enraged the racists, propelling the Tea Party into dominion over the RNC, and buoying Trump’s own foray into politics through anti-Obama Birtherism.

Trump has since made it his mission to undo all of the good that his predecessor managed to accomplish and drive us back into the days of Jim Crow, if not a post-apocalyptic Stone Age. Given all this, an argument could be made that, not only would Black people in America be better off today if we had never had a Black president, we all would.

So far, the examples I’ve given are of people who clearly do or don’t support policies that represent the interests of their own identity groups: Hillary Clinton vs. Ted Cruz. But there’s a third case that is uniquely terrible: cosplay representation.

Consider that the populist left is ostensibly in favor of equal rights, but cares only about the battle for economic equality while disparaging the rest of the war for equality as mere “identity politics“. You might think this would mean that they’d follow a color-blind approach by running the best candidates to support their policies, regardless of what they look like, but that would not be slimy or hypocritical enough.

Instead, through the so-called Justice Democrats, the left-populists have made a point of cynically playing identity politics by choosing candidates who are some combination of non-male, non-white, and non-straight. But they’re just infiltrators; definitely not substantive representation.

Just as these anti-Democrats are running against Democrats while pretending to be Democrats, they are anti-anti-bigots who are embracing the very same notions of identity politics that they disparage, using it as a tool to undermine anti-bigots while pretending to be anti-bigots themselves. And once these “minority” politicians get elected, they pursue Marxist policies that do not address bigotry. They are stealth identity traitors.

The disparity between how people look and what they do has has long been recognized, as shown by the saying, “skinfolk ain’t always kinfolk”. In other words, just because they share the color of your skin doesn’t mean they’re family: they do not have your interests at heart. Their representation is descriptive, not substantive. In the end, policy is more than skin deep, so we voters need to look deeper, too.

There’s another way that descriptive representation can go wrong, which is that it can lead to voters refusing to accept excellent candidates who do not happen to look like them. We’ve seen this lately in the case of white women who supported Elizabeth Warren attacking Kamala Harris because she’s Black. We’ve even seen a few Black people swearing they won’t vote for Joe Biden unless he chooses a VP who’s Black. This is both discriminatory and self-defeating, but it’s a natural outcome of a push for descriptive representation.

A key requirement for any official is competence. Here, the disadvantaged strangely have a leg up. Due to systemic discrimination against them, they have to “have to work twice as hard for half as much”. On that basis, we should expect the only Black woman in the Senate to be well above average, and she is. Likewise, the first woman nominated for the presidency by a major party was simply overqualified. So, all things being equal, the call for representation could lead to better candidates.

Even so, due to that discrimination, they can expect to be held to a higher standard, and treated harshly and unfairly for any real or perceived failings. Hillary is the poster child for that, with “but her email” and a host of other baseless attacks, and Harris’ “she’s a cop” is par for that course. And when one of them is legitimately incompetent, this is used to harm the political aspirations of everyone who looks like them for years.

Identity politics, when used as a slur, refers to supporting someone purely due to shared identity; for descriptive representation. This sort of tokenism does still exist and can work; witness the modest success of AOC and her skinfolk squad in leveraging their identity, both to gain votes and to defend against accusations of bigotry.

Of course, the elephant in the middle of the room when it comes to identity politics in the most negative sense is white men voting for white men. It is by far the largest-scale example, and the most harmful because they already had equality and then some, so what they sought was further supremacy. This demonstrates how the line between descriptive representation and bigotry can be blurry, or nonexistent.

Fundamentally, the problem with pushing for descriptive representation is that it doesn’t reliably yield substantive representation, and can even backfire in a variety of ways, as outlined above. The deep reason for the failure is that you can’t just fake it until you make it.

Equality of opportunity leads to equality of representation, but it doesn’t necessarily work the other way around. Pushing for appearances is voodoo politics, reversing causality. It is the tail trying to wag the dog, never addressing the root cause of inequality, only covering it up.

It also interferes with achieving the goal of equality because of Goodhart’s law, which can be phrased as “when you use a metric as a lever, it ceases to work as a metric”. In order for descriptive representation to be a useful measure of substantive representation, we can’t also use descriptive representation as the lever by which to achieve it. It would be like increasing what the scale reads by pushing on it directly with our thumbs instead of adding a few more slices of cold cuts. It’s cheating, but only cheating ourselves.

Not only does seeking appearance over substance fail to move us toward our goals, it obscures how far we’ve fallen short, and leaves us vulnerable to cynical manipulation, whether it’s by a Palin or an AOC. Moreover, it can be indistinguishable from the sort of bigoted discrimination that has kept the identity hierarchy in place for all these years. You cannot attain equality as a result without being dedicated to equality in your methods.

Does this mean that we should always or never vote on the basis of identity? It’s not that simple. We should vote for the best candidates, the ones who will succeed, both in gaining power and in exercising it for the good of all. Sometimes the best candidate also stretches the envelope of representation, and that has positive consequences in itself. Sometimes the candidate is best regardless of their identity groups, or even despite them. But if we focus on descriptive representation, we undermine our own efforts and help the bigots win.

3 Replies to “Metrics and levers, or how to cheat at meat.”

  1. When I initially commented I clicked the “Notify me when new comments are added” checkbox and now each time a comment is added I get four e-mails with the same comment. Is there any way you can remove people from that service? Many thanks!

    1. I’m getting spammed on that post. Although I’ve marked them as spam, you probably still get notified and I don’t know how to do anything about that. It’s died down lately, so hopefully the worst is over. If not, best you can do is cancel that notification.

Leave a Reply

Your email address will not be published. Required fields are marked *