“That man who has nothing to lose:” Black Americans and Superfluousness

Long before white Americans felt like their society had abandoned them, Black Americans knew the feeling. Just like whites do today, some Black Americans responded to earlier superfluousness by “clinging to guns and religion” to use Barack Obama’s famous analysis. (cf. Kinsley gaffe) Here’s James Baldwin, describing the Nation of Islam:

“I’ve come,” said Elijah, “to give you something which can never be taken away from you.” How solemn the table became then, and how great a light rose in the dark faces! This is the message that has spread through streets and tenements and prisons, through the narcotics wards, and past the filth and sadism of mental hospitals to a people from whom everything has been taken away, including, most crucially, their sense of their own worth. People cannot live without this sense; they will do anything whatever to regain it. This is why the most dangerous creation of any society is that man who has nothing to lose. You do not need ten such men—one will do. And Elijah, I should imagine, has had nothing to lose since the day he saw his father’s blood rush out—rush down, and splash, so the legend has it, down through the leaves of a tree, on him. But neither did the other men around the table have anything to lose. –James Baldwin, “Down At The CrossLetter from a Region in my Mind

Baldwin was no fan of Elijah Muhammed’s movement, but he tries to understand it and he seems to sympathize. And what he calls out in these lines is an overriding sense of loss–one that can justify any effort or sacrifice to overcome. That loss of worth, which Baldwin wants to depict as something much deeper than “self-esteem,” is tied not to airy questions of recognition but to material harms and embodied injuries: frisks and kicks by “legitimate” authority that go unanswered, murdered fathers that go unmourned by white society.

Baldwin even interprets the turn to Islam as a turn away from Christianity’s whiteness–from the forgiveness that it seems constantly to demand from white supremacy’s victims.

Today I know more Black Muslims than Arab ones, but mostly they’re imprisoned. And that, too, is a form of superfluousness: of the 2.3 million people incarcerated in the US, 40% are Black, even though Blacks make up only 13% of the population.

PPI 2016 race

It’s hard not to see imprisonment of African-Americans as primarily a reaction to their enforced superfluousness. Labor market prejudices create a circumstance where unemployment rates for African-Americans are roughly double the rates for whites, at every education level. My students believe (and I agree) that the War on Drugs is largely a war on Black participation in the black labor market. (Recall the Ice Cream truck war of the summer which helps explain how police and the courts can make black markets less or more violent.) It’s an attempt to foreclose available forums for Black entrepreneurship. It’s notable that as marijuana legalization proceeds state-by-state it begins in white places, and the new profits and businesses are primarily white-owned.

To be rendered superfluous is a particularly odd phenomenon, and at least in the formulation I’ve lately been thinking about, it seems uniquely tied to relative deprivations of status and respect. I want to believe that respect and recognition matter less than life or health. We all have things we could lose, and honor is the least of these.

But that’s simply not how people act. Flourishing matters to most people more than survival, and flourishing requires a community of esteem. It requires reputation and character assessments, it requires that the agents of the state give you equal protection and don’t target you as a unique threat.

Some parts of Black America responded to superfluousness by clinging to god and guns, but for the most part African-Americans responded by becoming the center of cultural attention. There’s little argument that Black culture simply is American culture at this point, as almost all of our distinctively American institutions and cultural traditions are shot-through with Blackness even if it is unacknowledged. Baldwin charged that even myths of Black laziness or violence serve an important function in White America: they are our “fixed star” and moving out of their place of subordination would shake our “heaven and earth… to their foundations.”

Yet whites don’t seem to be willing to create their own economies of esteem in this way. Something–perhaps supremacy itself–has rendered them too lazy, too dependent on long-lost tradition and long-gone cultural victories. Consider Katherine Cramer’s formulation: rural Wisconsinites resent that the urban centers have deprived them of “power, money, and respect.” But they’re simply wrong: they’ve seen no real deprivations of power or money compared to Black Americans. It’s certainly true that they receive less respect than they used to get, less deference and less cultural attention. But this is a downfall from absolute supremacy and literal enslavement–enforced by rule of law. We have a long way to go before we aren’t still reaping what WEB Du Bois called “psychic wages” of white supremacy. (I prefer to call them psychic dividends, since it’s important that they are not a reward for work, but rather like a racial trust fund.)

How much should that matter? In my heart, I want to believe that people with a surplus of money, power, and respect should share. And I want to believe that it’s not a finite resource, that more souls can and will produce more of each if they’re not forced into superflousness. I think of Malcolm X, whose industriousness and self-invention could not be halted or stultified by racism or even Elijah Muhammed’s theocratic nonsense, only assassinated. But that’s where my analysis is probably wrong: when people have the money and power, they’re well-placed to demand the respect or punish its absence.

Any Cook Can Govern: Populism and Progressivism

I have lots of feels and lots of arguments about these two pieces by Peter Levine on an alt-left populism: pluralist populism and separating populism from anti-intellectualism.” (This post on identity politics is also relevant.)

Peter even goes so far as to call himself a populist, which is a surprising move to restore the term’s sense in a year when we’ve watched a wave of populist elections sweep through the industrialized world on the back of nationalism, Islamophobia, and anti-elitism. Though the left frequently makes populist appeals as well–especially when we’re criticizing agency capture by industry or the undue influence of the very rich–it’s not always obvious to me that progressive political goals are compatible with populism’s mass movements and drive towards uniformity. Progressives tend to be pluralist and cosmopolitan in their egalitarianism.

Is Populism Inherently Nationalist?

In these pieces, Peter argues that populism can be pluralist and intellectual, and he uses great examples. (He gives the JCI Scholars Program a shout out, as well as our mutual friend Laura Grattan.) But many political theorists argue that populism is intrinsically nationalist and reactionary, usually anti-elitist and anti-immigrant, as well as racist and anti-Semitic. The counter-examples, like the late Nineteenth Century Populist Party run by farmers in the Midwest and South, seem never to actually achieve their goals or become all that… popular. 

For these critics, populist impulses tend towards the violent elimination of difference. Put another way, populist movements tend to become mass movements. Populists appeal to a mythical common good that renders class and geographical interests uniform, and usually identifies an evil or corrupting Other as the people’s enemy. For populism’s critics, the kind of anti-racist and grassroots intellectualism Peter has been describing is something else if it’s possible at all: class solidarity that re-organizes antagonisms without suppressing internal disagreements.

Is Populism Inherently Anti-Intellectual?

Famously, the Progressives of the early 20th Century were quite hostile to the Populists that had gone before. Populist hostility towards elites often swept up intellectuals as well, and the Populists–being farmers–had targeted urban dwellers, financiers, and Jews as their enemies. There is a tendency to lump the rich and the knowledgeable together, so efforts to raise the status of regular working people sometimes try to lower the status of scholars, teachers, and upper-middle class professionals. That’s a worrisome tendency.

But Peter quotes the JCI Scholars Program website, a group I helped found, on our motivations for working with traditionally excluded groups: we do 

as collaboration between teachers and students, and to make classes free-ranging discussions and workshops more than lectures.

But is that populist? Even at the JCI Scholars Program, we’re working with the talented tenth: at most 150 students, in a prison that has between 1400 and 1800 prisoners. The main question among prison educators is the extent to which we are engaged in a truly populist project, and the extent to which we are cultivating what Antonio Gramsci called “organic intellectuals.”

My co-founder, Daniel Levine, likes to invoke CLR James’ Every Cook Can Govern on this question. Pulling from Greek sortition, James praised the capacity of ordinary folks to take up the tasks of governance. Of course, this required a much simpler state, and much shorter periods of governance before passing the responsibility on to another. But perhaps our state has become so complex precisely as a result of–and perhaps as justification for–elite domination.

The deeper problem is that sortition required institutional safeguards as well as agreement. And it’s probably relevant that it didn’t survive, suggesting it wasn’t sustainable. As I’ve written elsewhere, Greek sortition depended on a number of institutional factors to function:

The three norms of isonomy are mutually reinforcing: equal participation requires that the office-holder act with the understanding that she might be replaced by any other member of the community. She cannot abuse her office without being held to account at the end of her term. For the same reason she must regularly give reciprocally recognizable justifications for her actions, without which her decisions might be reversed by the next office-holder, or even punished when her office no longer protects her from prosecution. The ideal result of such a regime is a strong preference for deliberation, consensus, and mutual respect, alongside a cautious honesty and transparency with regard to potentially controversial decisions.

So is it really “every cook can govern” or is it “any cook can govern?” This isn’t quite an embrace of full analytic egalitarianism: everyone is not equally capable of governance. And I don’t mean by this to substitute mere equal opportunity rhetoric for substantic equality, as we see in the classical liberalism of Pixar’s Ratatouille:

What I mean is that lots of people have an odd mixture of impatience and arrogance that makes them convinced that they already possess the requisite knowledge. This is dangerous when it renders them–or us–unable to change our views in light of new information, or incurious at the promise of new evidence.

Consider the man who claimed he could figure out the gist of important matters without doing much reading, that he was more accurate than guys who have studied it all the time,” and “[doesn’t] have to be told the same thing in the same words every single day….” We often think of analytic egalitarianism as more epistemically humble, because it requires humility among those to whom we currently defer, the experts. But there’s plenty of arrogance among non-elite groups as well. It strikes me that many people find the combination of arrogance and anti-intellectualism appealing, and that this is the thing to fear in populism.

Populism cannot fall into demagoguery. If facts are relevant to a decision, they must be given proper weight, even if facts cannot be a substitute for values. But though I doubt that populism is inherently anti-intellectual, the problem is that too often our society mistakes credentials for knowledge, which means that anti-elitism requires the motivation of a certain kind of anti-intellectualism.

Probably the best assumption to start with is the universal claim that all humans are epistemically capable. This what we call a “defeasible” claim, to be held until proven otherwise, as we do with juries. But we do exclude people from juries, and we don’t entrust juries to carry out their own investigations any longer. A provision of analytic egalitarianism is that a corrupted society will corrupt its citizens’ epistemic capacities. To paraphrase Rousseau: we’re born [epistemically] free, sure: but everywhere in [epistemic] chains, as well.

With the right support from experts, probably anyone can govern, where governing” means selecting from a menu of options supplied by those experts—it’s just that the experts who control the framing and flow of information have many opportunities to manipulate those who depend on them (or think themselves superior to them without actually doing the work) that can only be overcome by becoming an expert oneself. My own view is that many more people are capable of that expertise than our society will currently admit.

That because we attach governance responsibility to meritocratic credentialism. But it’s not always clear that we’re valuing knowledge over ignorance, rather than valuing exclusion itself. The nature of the competitive managerial class is that it sets up zero-sum competitions so that winners can capture the lion’s share of the benefits from their education and knowledge. Just because it’s unlikely that every cook can govern equally well doesn’t mean that we must restrict governance to the winner of the most Spelling Bees.

A lot of citizens can govern, and it’s a waste of those talents to relegate those capable of informed participation from doing so: we’d be better off if many more of us were able to take on those responsibilities. There ought to be many more opportunities to exercise one’s civic capacities, rather than such a limited number that our capacities atrophy from disuse. And meritocracy doesn’t just award these opportunities to knowledge-elites, it also tends to reduce their number. It creates both privileges and ignorance.

So that’s the version of populism I can support: one that celebrates the even distribution of insight and institutionalizes a fear of the even distribution of ignorance and arrogance.  A populism that is pluralist and cosmopolitan. But I have to admit that this doesn’t really sound much like populism; I usually call it “democracy.”

Your Enemy is Your Best Teacher

My good friend Sarah Shugars is subtweeting me:

Resistance is a way of life, it’s a form a citizenship. It’s a commitment to speaking out and, importantly, creating space for others to speak out. It’s a bold declaration that all people are created equal and its an unequivocal call that we will not, cannot, rest until that equality is manifest is our society. Resistance comes in every word you say, every action you take.

I find this approach to civic life unhelpful, just as Shugars intended (her subtitle: “an unhelpful guide.”) I think we–we scholars who tackle the civic arena–ought to be able to give advice, and not simply advocate a life of unspecified restless action. Too often we study the politics of governments but we need to practice a different politics: of relationships and of institutions. But I don’t yet know what advice to give. I am still a little bit heeding the instructions: don’t just do something, sit there.

The activists have only changed the world. The point is to understand it.

The Dalai Lama has said that in the practice of tolerance, your enemy is your best teacher.” I suspect he meant something about love and difficulty, about how we must learn to tolerate those we truly disagree with and not those to whom we have grown accustomed. But here’s what I want to know: why can’t we actually learn from our opponents?

Republicans responded to the elections of both President Clinton and President Obama with radically obstructionist tactics. Remember? Special prosecutors, filibusters, impeachments, Benghazi, endless attempts to repeal the Affordable Care Act, budget showdowns, government shutdowns, fiscal cliffs, refusing to hold hearings or approve nominees at every level.

They played constitutional hardball, and won. Maybe Democrats should try that. 

I imagine someone will say that it works better for the party that claims to oppose government than the party that supports it. But in public, they’d simply say they were opposed to the policy X, not ALL government. We’re about to see that they believed that, too: the federal government is not about to suddenly shrink so small it can be drowned in a bath tub. They’ll cut taxes and spend profligately like they’ve done since Ronald Reagan “proved” that deficits don’t matter. Republicans haven’t been afraid to contradict themselves, which meant lots of bad soundbites for Jon Stewart to satirize, but kept spelling electoral success. A lot of what made the GOP so effective over the past eight years was running out the clock with nonsense and using pre-commitment strategies to win various games of policy chicken.

I imagine someone will say that the Democrats don’t have a single house of Congress in which to stage these obstructions. That’s how the GOP felt in 2009. Yet 2010 is widely recognized as their crowning achievement. They founded the Tea Party to oppose themselves, and gerrymandered themselves a longterm majority.

So why not try that? What do we have to lose, control of government?

There is, unfortunately, an objection I can imagine someone could raise: the problem with chicken is that someone has to be willing to swerve. If both sides play constitutional hardball then we will fall into Juan Linz’s nightmare of constitutional crisis. The likely result of conflicts between a popular president and an unpopular congress will be a weaker legislature and a stronger executive branch. So there is something to lose, and that is the lesson: winning policy victories does not tame the prince, when the prince has a clear mandate from the people. Knowing that this is a possible fate makes thinking through such scenarios important, and it makes informed guides to strategy and action valuable as well.

So as I said on November 9th, it is a good time to ask me to make future commitments of help and support. Tell me what you plan and what you need and how I can help.

Race, Income, and Elections: The White (Male?) Working Class

In my last post before the election, I quibbled with Peter Levine’s strategic argument that Trump’s supporters might be momentarily richer than average, but only because they were older, maler, and whiter. I worried that it was a kind of mistake, even if it’s perhaps an analytic effort designed to enhance our ability and willingness to achieve strategic ends. Since the surprising election results last week, many more people are returning to questions about the working class, and specifically the “white working class.” I have been trying to think about this for a while, and I’m still coming up short.

So I’m returning to that argument, especially given his follow-up, “To beat Trump, invest in organizing:”

Meanwhile, we also need organizations in red states and red counties, in rural areas and exurbs. The point of organizing there is not to show empathy to Trump voters or to honor their concerns. The point is to win. Particularly in 2018, anti-Trump votes will be very poorly distributed–far too concentrated in the great cities to win the House and Senate back. Every extra vote in a white non-urban county will matter, and that requires organizations to change minds, to empower the disenfranchised, and to offer real benefits. By the way, although I think the Democratic Party is a necessary component of the opposition, it is not sufficient. Electing or reelecting responsible and caring Republicans in red districts is also essential.

This is very much of a piece with what he wrote before the election, when he tried to distinguish the strategic argument from the analytic one (emphasis mine):

If you have no organizations behind you, you’ll typically feel powerless. If that’s how you feel, you are unlikely to want to participate in a difficult conversation, make sacrifices and tradeoffs, acknowledge any unfair advantages, or negotiate. Again, to use Trump voters as an example: they are overwhelmingly White, and it would be appropriate for them to acknowledge White privilege when issues of racial injustice arise. But I think they are very unlikely to acknowledge their own privilege, let alone agree to concessions, as long as their overwhelming experience is one of powerlessness. And I think they are powerless if they are unorganized and represented only by unaccountable celebrities. This implies, by the way, that one of the most important tasks confronting us today is organizing the White working class.

Just who is the white working class? Levine defines it as white people without college degrees working blue collar jobs and living in counties with increasing white male mortality. This is remarkably precise, and perhaps it does indicate a set of shared interests that must be organized. And yet…

Depending on the way of measuring, as few as 1/5 of workers or as many as 2/3 of workers are in the working class. When economists use the term, it gets precisely defined, but then glossed by others in a confusing way. Are all jobs that don’t require a college degree blue collar? Does the working class include the service sector, or manufacturing only? Are we talking about all non-agricultural low skill manual labor? What about medium- and high-skill manual labor? That variation from 23% of workers to 66% of workers provides a massive opportunity for equivocation, and suggests that the term is well beyond anything that tracks a specific group. It reminds me of the way we talk about the “middle-class,” with precise definitions on offer in such variety that almost anyone can find one that they fit. With that much equivocation, the term lends itself to rhetorical manipulation or propaganda. 

But the most important question is this: why “white?” Why assume that–whatever the working class is–its interests are racialized? I wonder to what extent the “White Working Class” formulation merely reproduces the actual racism of a group of people who share thereby share a sense a pride and solidarity even if their material interests are not always aligned. Something that can be a social identity, a racial identity, and a class identity all at once.

I don’t fully know where I stand as I watch the dizzying play of free association. Just because a concept is ideological doesn’t mean it doesn’t call out a group of people who are interpellated by it and then cathect it themselves. Just the opposite, I think: we imbue arbitrary conceptual constellations with meaning in order to make our world liveable and loveable.

And there’s a triptych of books out recently trying to figure this out: J.D. Vance’s Hillbilly Elegy, Kathy Cramer’s The Politics of Resentment, and Arlie Hochschild’s Strangers in their own Land.

Each picks up a different group, but in the first and last case I don’t know if they would fit the standard Labor Department’s classification of the blue collar worker. The Politics of Resentment does, though: using Wisconsin rural and exurban voters whose work is primarily physical and does not require a college degree, but are well-enough paid that they’d qualify as middle-class, yet have not seen their standards of living increase. This group in particular found Barack Obama’s campaign initially promising yet soured when he made comments about them clinging to God and guns. After years of field work, Kathy Cramer diagnoses three causes of their resentment:

That feeling is primarily composed of three things. First, people felt that they were not getting their fair share of decision-making power. For example, people would say: All the decisions are made in Madison and Milwaukee and nobody’s listening to us. Nobody’s paying attention, nobody’s coming out here and asking us what we think. Decisions are made in the cities, and we have to abide by them.

Second, people would complain that they weren’t getting their fair share of stuff, that they weren’t getting their fair share of public resources. That often came up in perceptions of taxation. People had this sense that all the money is sucked in by Madison, but never spent on places like theirs.

And third, people felt that they weren’t getting respect. They would say: The real kicker is that people in the city don’t understand us. They don’t understand what rural life is like, what’s important to us and what challenges that we’re facing. They think we’re a bunch of redneck racists.

So it’s all three of these things — the power, the money, the respect. People are feeling like they’re not getting their fair share of any of that.

This makes some sense: everyone wants power, money, and respect! These are the three hands: the invisible hand of the market, the intangible hand of esteem, and the iron hand of state coercion. And yet everyone always feels as if they are getting the back of these hands to some extent.

I still worry that this analysis makes a mistake about the relationship between race and class in support of a strategic point. To what extent is there a group of people called the “white working class” that is particularly injured by the ways that the political economy is organized? Remember that my position is that we ought to be wary about racialized organizations of white people. Indeed, I followed Arendt in worrying that the primary form of organization towards which such a group would be prone would be a racist or imperialistic one. That is, it’s precisely the effort to organize the disorganized that creates a populist nationalism. But this is not a popular view!

Yet this is a mainstay of American demographics: we cannot understand our country easily through income or education unless we add the racial coefficient. Once we’ve demographically controlled for race, only class remains:

[This conclusion] requires us to divvy up the working class into whites and non-whites. Having done this, we then find that non-white working class members strongly support Hillary Clinton, and white working class members strongly support Donald Trump. Thus something called “social class” predicts candidate support!

I still think this is right. That is, we can’t fruitfully rebut claims about race by dividing the working class into racial groups. It’s certainly true that these groups live and work in different places and may well have different cultures, although I do think that the Black Jeopardy SNL sketch captured something impressive about their shared outlooks:

Yet that’s quite clearly not enough to justify the overall claim that they belong to the same class. To do that, one has to ignore race. For instance, Levine concludes that:

[I]f class means social status, and status involves occupation and education, then Trump voters tend to be downscale Whites in downscale White areas.

Here’s what best predicted Trump support before the election: occupation, education, mortality level, gender, and age. Trump’s voters tend to be blue collar, to lack a college degree, and to live in those counties driving the white lifespan downward. Yet none of these explains Trump support as well as race. And indeed, after the election we have reason (although we must be careful of exit polls) to believe that whites with college degrees joined less educated whites to support Trump!

It’s clearly true that many of Trump’s voters are “downscale” compared to other whites. But they are equally “upscale” compared to non-whites. So the extent to which we can accept the class analysis of their support is the extent to which we can both assume whiteness (to capture the group that is not African-American and Latino and supports Clinton) and erase that whiteness. That’s the extent to which there is a group of people with a distinct class interest we can call “the white working class,” preserving and erasing whiteness at once. So what worries me is that we seem–for strategic reasons I mostly share–to want to deny that white nationalism is about race.

After all, by analyzing away their whiteness we miss one of their primary concerns: the relative loss of prestige associated with older white males.

I feel like I am being uncharitable to Levine and other class-first analysts of this election cycle, so let me ask some questions that might help me understand the position:

  1.  Are there non-strategic reasons to analyze the white working class separately from the non-white working class? For instance: in presidential elections, poor white voters vote in much higher numbers than poor Black and Latino voters. They seem more like white people than like poor people. Why does cutting the working class at the racial joint seem so smooth to you (and not to me)? Can you help me see my error?
  2. Is there a sense in which the white working class really is worse off than their incomes suggest? We often talk about this group’s recent surprising decline in lifespan, and I would add that they are uniquely threatened by the new college credentialism craze. But the white working class lifespan is still higher than for non-whites of the same class (or even much higher classes!) Perhaps the loss of status is measurably worse the the perpetual lack of it is: certainly non-whites have more optimism than whites in this income group.
  3. Put simply: is the racial division of the working class warranted? Can we cut the pie into racial pieces without eliminating our ability to rebut racial resentment explanations?

One last thing: since I wrote my last post, I have read this very interesting Monkey Cage post on research by Wayne, Valentino, and Oceno supporting the claim that sexism (and anger) is a better explanation of Trump support than racism (and fear.) Here’s a Vox gloss of the study. Though the Monkey Cage post is older, I had already been thinking about this because of the work of the philosopher Kate Manne (here, here, and here, for instance).

I think it’s difficult to see whether sexism causes voters to support Trump or Trump support causes voters to become more sexist. That is: how many of these misogynistic voters would vote for the Democrat if the Republicans had selected a woman candidate? It’s notable that sexist voters strongly preferred Romney to Obama in 2012, though less so than they prefer Trump. So it could be that when a man runs against a woman, sexists support the man, or it could just be that sexists tend to vote Republican. The fact that sexists are more likely to support Trump than Romney suggests the former. But there’s a third possibility: when a woman runs, her opponents become more sexist. They reach for reasons to oppose her, as we all do when we have a position to defend, and in a misogynistic culture the reasons closest to hand will tend to be gendered, misogynistic ones.

Explainer Journalism Needs Better Explanations

Corey Robin got some nice jabs in at the current class of younger non-academic pundits a while back:

A lot of these pundits and reporters are younger, part of the Vox generation of journalism. Unlike the older generation of journalists, whose calling card was that they know how to pick up a phone and track down a lead, the signature of this younger crew is that they know their way around J-STOR.

He went on to diagnose the the current generation as ahistorical positivists who read too much economics and not enough history. I have a different explanation: bad social science journalism, a breed of the general problem of bad science journalism.

Some people–perhaps even some journalists–seem to think of explanation as a fundamentally neutral recounting of facts, not something that can have formal conditions or persuasive interests. But explanations are arguments, too. And sometimes explanations fail because the premises don’t have the right relationship to the conclusions: though no explanatory argument is valid in the logical sense, some are much stronger than others.

Very often our chosen explanations are meant to convince us of something quite apart from a causal account of how an event occurred. We don’t read the newspaper: we get our news from TED talks, op-eds, comedian newscasters, and  “explainer” journalism. And in that space, explanations often follows the form, “Y is a surprising reason for X. But Y also implies that we ought to care more about Z.” In that case, the “explainer” quite often cares much more about Y and Z. X is merely a reason to climb onto their favorite hobby horse.

Much of what gets billed in headlines as an explanation of “hows” and “whys” actually fails some of the basic rules we teach in inductive logic about inference to the best explanation. As a fan of the Vox-ification of media, I need to be reminded of this myself sometimes.

Inference to the best explanation has simple rules: all things being equal, it counts in favor of an explanation that there is some new piece of evidence that could falsify it; all things being equal, it counts in favor an explanation that it can explain more than just the single set of facts we see before us. It’s also better to avoid multiplying metaphysical entities beyond need, and to preserve as many of our pre-existing beliefs as possible. What’s more, explanations should not jump to conclusions or overclaim beyond what the evidence can support.

Regressions are not Explanations

I’m all for data journalism, but I think there is a particularly dangerous version of it that equates a regression with an explanation. One kind of explanation is a causal one, and we can often test causality using concomitant variation. We can ask: if a sick person takes the drug, do they get well? Then we can add some math and ask: are they more likely to get well than a person who didn’t take the drug? That’s statistics in a very small nutshell.

But there’s a big problem in applying these case-control methods to demographic information: we often can’t vary the big social factors while holding all else equal. So we approximate. We try to use the sheer morass of human variation to uncover independent variables that predict changes. This has led to a lot of really cool math to produce real cool studies that show that some variable predicts another variable with high confidence… while only predicting a very small part of the variation. Gee whiz!

Consider a pie that has been carefully segmented into different flavors: shoefly, sweet potato, apple, cherry, blueberry, and many others the bakers have left as a surprise. What a lot of social science regressions do is try to carefully slice out one of the sections of that pie to figure out its specific flavor. As a part of their analysis, they prove that they can say with near-certitude (high confidence interval!) that they’ve gotten nearly every crumb of the Key Lime section. What does that tell us about the rest of the pie?

Well, anyone who likes pie can tell you this: gobbling the piece completely without any crumbs doesn’t mean you’ve eaten the whole pie. I talk a lot more about low R2 R-squared values in the famous study of American oligarchy here, but this is the upshot: if there’s a lot of variability in outcomes, as you would expect in a big population of wonderfully weird, defiantly diverse, and polymorphously-perverse human beings like us Americans, and you have a very strong predictor of a very small part of that variability, you do not have an explanation of the whole thing. You do not understand the whole of America! You have that very accurate predictor of the small piece, you haven’t necessarily gotten any insight into the whole population thereby. The part–which has been carefully defined by its difference from the whole–bears no necessary or predictable relation to whole.

Donald Trump: A Case Study of Stymied Explanation

Take Donald Trump’s nomination: there’s a lot to be said for his extraordinary candidacy. A man without any political experience has been nominated by one of the United States’ two major parties, on a platform that has very little clarity and against at least one dynasty candidate within his party and several other establishment favorites. How could this happen?

I don’t think I can definitively answer that question, but I think there are two major candidate explanation types:

  1. One kind of explanation comes from the wellspring resentments, of which we have an increasing supply. Whites are angry at their loss of prestige. Workers are angry at the support both parties give to millionaires and bankers. As we saw with the Brexit vote, these voters seem to be willing to spite themselves and destroy their own prospects if it’ll frustrate elites. And rage over growing inequality and bank bailouts might need to find some outlet, even a destructive one.
  2. Another kind of explanation is that as we grow wealthier, we can afford more irrationality. It’s the rich who forgo vaccines. Safe and moderate and establishment-vetted candidates are sort of like vaccinations against bad outcomes. So as we grow richer, we become more tempted to forgo them.

I believe that most of the efforts to explain Trump fall into these categories: we’re sometimes told that we can blame Trump on the current president’s competence, for instance. Or perhaps we can blame elites who failed to respond to the financial crisis with sufficient punitiveness, as Andrew Sullivan alleged.

In journalism, though, it is more common to emphasize and sub-divide the first sort of explanation: Trump is either explained by racial resentment or by economic anxiety. (Note that both are Vox links!)

White Supremacy

If your first explanation of the Trump phenomenon is “white racism,” then it’s quite likely that we’d get along well. That’s the standard answer in most of my social network. But it basically can’t be a complete explanation: the United States has been a white supremacist country since its founding. Racism is literally written into our Constitution, and certainly structures all of our major institutions. Yet Trump’s nomination represents an unprecedented event in the US, unheralded by our history. The best precedents seem to be European nationalist parties like the UK Independence Party and the French National Front, where parliamentary politics makes such a minority view easier to advance at the national level. So we need to give an account of what has changed here to make his success possible.

A story that makes a bit more sense is to call this revanchist white supremacy. Some people really are enraged that we have a Black man as president. What’s more, some people are better able to articulate their other grievances when the President is a Black man, just as when one of the candidates is a woman the sheer efficacy of misogynistic tropes in our culture makes it difficult to avoid them. (I think in particular the trust gap with Hillary Clinton is alarming; she is one of the most honest candidates ever, and seems to have a lot of difficulty dissembling even when an easy lie would benefit her. Yet out culture’s misogynistic mistrust of women pins her to some mythical deceitfulness while her husband–a perjurer!–gets a pass.)

Possibly it makes sense to say that Trump’s candidacy is better understood not as anti-Black racism, the sort that structures our country, but as anti-immigrant racism and also of Islamophobia. But it was only a decade ago that George W. Bush was insisting that we not blame Islam for terrorism. And the hatred that Trump and his supporters show for Latino immigrants is particularly notable because it’s the one thing in which establishment Republican politicians struggled to join him: they know they can’t afford to alienate Latinos. It’s the one policy plank which he has held clearly and unambiguously from the start, and it’s the most obvious contrast with his rivals.

But could it be that anti-immigrant nationalism–a kind of racism–explains Trump’s ascendancy? That, too, seems unlikely. Immigration has been on the decline for more than a decade.

PH_2015-09-28_immigration-through-2065-09If immigration peaked in 2005, then any explanation of current discomforts would require us to believe that it has taken people more than a decade to realize it was a problem. And let me be clear: there’s almost no evidence that it has been a problem in aggregate.

But it has been a problem for the segment of the population who are also voting for Trump. The anti-imigration story is usually associated with the effect that low-skill immigrants have on the labor market participation of low-skill American workers. So anti-immigrant attitudes can understood as partly an economic concern, and these could partly motivate Trump’s supporters.

The Trumpenproletariat

Since I’ve been writing about superfluousness lately, you will not be surprised to hear that I worry that there is a superfluousness explanation available here. Low-skill immigrants put particular pressure on the low skill workers who are only marginally attached to the workforce. So it is reasonable for them to fear that they will be undercut by undocumented workers who can undercut their wage requirements. But there’s an inconvenient fact that this explanation must negotiate: Trump’s supporters have a much higher average household income than most Americans, and certainly higher than any of the Democratic candidates. The median Republican primary voter who picked Trump reported a household income of $72,000 per year. (The median American household income is $56,000.)

With such a high income, it seems unlikely that his support is primarily drawn from within the lumpen proletariat, that counter-revolutionary group that Marx described in The 18th Brumaire of Louis Napoleon:

Alongside decayed roués with dubious means of subsistence and of dubious origin, alongside ruined and adventurous offshoots of the bourgeoisie, were vagabonds, discharged soldiers, discharged jailbirds, escaped galley slaves, swindlers, mountebanks, lazzaroni, pickpockets, tricksters, gamblers, maquereaux [pimps], brothel keepers, porters, literati, organ grinders, ragpickers, knife grinders, tinkers, beggars—in short, the whole indefinite, disintegrated mass, thrown hither and thither, which the French call la bohème.

That’s not Trump’s supporters. If income is any guide to one’s role in the economy, then his voters are significantly less “superfluous” than either Clinton’s or Sanders’ supporters. But this is comparing apples and oranges. How can we understand this income gap?

  1. Trump’s supporters are, first, voters. Poor people don’t vote in the same numbers as the middle-class and rich, so we should expect (and find) that voters have higher median household incomes than non-voters.
  2. Trump’s supporters are Republicans. And Republican voters just are richer than Democratic voters. Trump’s supporters are still the poorest of the bunch: Cruz’s voters were about the same, while both Kasich and Rubio had median household incomes more than $10,000 higher. (Kasich’s supporters had a median of $91k; Rubio’s supporters had a median of $88k.) But we have to be careful here: Republicans are richer because they are whiter and older than Democrats.
  3. Trump’s supporters tend not to have college degrees. Thus even if they are currently employed, they’re experiencing a decline in their prospects under our new credential economy.
  4. Trump’s supporters come from poor places where the lifespan is decreasing. I still think this is the single best explanation of this pie-slicing sort: lifespans decreased for a certain population in the first time in modern history, and at roughly the same time, those groups chose a surprising candidate.

That’s why I associate Trump with the superfluous ones. Yet it’s important to recognize that his supporters are experiencing relative and not absolute impoverishment: they are worse off than they were, but not worse of than the Muslims, immigrants, and African-Americans they seem to despise. That loss of status may be a better indicator of how surpluses turn into superfluousness than any other; immigrants aren’t at all useless, they’re too busy being exploited! Thus racism still is a very relevant part of the story.

Against Epistocracy

It’s notable how uniform the establishment reaction to Trump has been: ordinarily measured newspapers and even conservative magazines have lined up against him in large numbers. Pundits and wonks see him as a threat to democracy and the rule of law, which is highly inflated rhetoric infrequently applied to major party presidential candidates. So how could he have won his party’s nomination, and seem to have such a good chance of winning the presidency? (Though perhaps not SO good a chance?)

It may seem that Trump’s apocalyptic rhetoric has forced the chattering class to respond in kind. But in fact, elites have been speaking in apocalyptic terms for a while now. My readers are mostly academics, so they will be able to immediately recall the long list of threats to democracy: wealth inequality, the failure of campaign finance reform, the growth of long-term unemployment, the coming entitlements crisis, mass incarceration, police brutality, de-industrialization, racism, multiculturalism, presidential overreach, climate change, epistocracy, automation, etc. Perhaps one or a few of these threats did fundamentally break democracy?

Or perhaps it’s the establishment itself that has become too uniform, that has begun to substitute both its factual and moral judgment for an honest consultation of the will of the people. As a member of that credentialied elite, I’m generally sympathetic to this. But the major problem with the rule of credentialed elites is three-fold:

  1. Even those who tend to get the right answer may still be wrong, but overconfident errors tend to be more costly.
  2. Experts have a tendency to speak beyond their legitimate expertise.
  3. Experts can self-deal in ways that are difficult for non-experts to detect, and indeed they have managed to claw back most of the shared gains from their expertise.

An additional problem might be that the rule of expertise is only considered legitimate insofar as it doesn’t lead to undemocratic control of policy. By creating invidious comparisons among citizens, elite rule creates a class of superfluous men and women who must live with the constant reference to our democratic culture while recognizing that they are excluded from it. They are thus living their lives at the point of conflict between principle and practice, the contradiction between rhetoric and lived experience, much like African-Americans in the US, who must constantly hear our lip service about racial equality in its practical absence.

We live in a time of self-government limited by many forces, but the most relevant force most Americans experience is the way that their projects and desires are hamstrung by wonks and nerds. It’s not usually the very wealthy who tell us what to do. It’s usually an upper-middle-class professional: a lawyer, a doctor, perhaps an engineer or a psychologist.

In almost every case, it was ultimately a college-certified teacher or professor who put and end to a person’s dreams and alienated them from credentialed elites; if that didn’t happen–as it is unlikely to have permanently done for any of my readers–then you’re more likely to find yourself within the top thirty or forty percent of incomes and influence. Teachers put up walls to success in our economy.

The White Working Class

My friend Peter Levine argues that the data supports the supposition that low social class explains Trump support. One odd thing about this finding is that it requires us to divvy up the working class into whites and non-whites. Having done this, we then find that non-white working class members strongly support Hillary Clinton, and white working class members strongly support Donald Trump. Thus something called “social class” predicts candidate support!

I have ample evidence that Peter is smarter than me. But I just don’t see how he has come to this conclusion. It seems to me that he is saying that white supremacy cannot be an explanation, because poor white people support Trump and poor Black people don’t. Yet Peter is well-aware that one of the key components of white supremacy is the weaponization of the poor, the idea that preserving the social privileges of whiteness can allow elites to prevent working class consolidation.

So I have to ask: in this conception of social class, is it the “social” or the “class” doing the work? I understand the strategic point Levine is making, that the best response to superfluousness is organization, and this even applies to poor whites tempted by white supremacy. I am sympathetic to that point even though I’ve argued that:

we should not hope for 100% political participation, at least insofar as that requires that white supremacists and chauvinists find viable politicians who will court them openly.

Probably we can and will come to rapprochement on the strategic issue, since there are alternate, non-supremacist ways for those groups to organize. But it seems to me that his enthusiasm for participation has led him into an analytic error. Give that he is much smarter than me, I’m sure I’ll turn out to be wrong about this… but I must wait to hear the correction before I’m convinced.

The Null Hypothesis: Incompetence and Fear

In explanations, it’s important to give the null hypothesis its due. Sometimes no particular thing had a strong effect on an outcome: sometimes the best explanation is chance, bad luck, or incompetence. And I think all three play a role in Trump’s rise. In particular, the Republican Party failed to seize on a “better” candidate during the crucial early states because too many of them were pledged to Jeb Bush. Long after Bush ceased to be viable, they couldn’t coordinate around an alternative for fear of offending his family. And with so many candidates, and no room for backroom deals, elections become chaotic; they fail to choose the person that most people would prefer (the Condorcet winner).

Yet I have come to believe that the best guide to his rapid ascendance is not some statistic but what he says. To me he looks overwrought, bombastic, and silly. Yet he regularly draws larger crowds than my lectures, so maybe he’s on to something. Why is this rhetoric so effective?

One thing that unifies much of the the US is the sense that there are major threats left unaddressed. We see constant reports of terrorist attacks, police shootings, and mass shootings that target both civilians and law enforcement. We live in a time of unprecedentedly low crime, and yet the news seems to be full of criminal atrocities. No institution feels legitimate any longer; we have one of the most upstanding presidents in modern history, and yet he is regularly de-legitimized in the rhetoric of both the left and the right. (And I too can think of dozens of reasons but the criticisms have grown faster than the reasons.)

So my suspicion is that the best explanation of Trump is simply that he is able to mobilize our fears–well “our” fears–not mine, but the ones that trouble our culture. It remains to be seen whether “our” fears are truly irrational; I think that they are. But we should think of Trump as running to be the Commander-in-Chief for the ongoing Global War on Terror.

If this is right, describing Trump as racist is irrelevant. His primary appeal is as a strongman, a defender against terror; racism is irrelevant except insofar as it helps to identify the source of the threats against which he’ll defend us. His strongman appeal is bolstered by its disconnection to any actual strength. It almost seems to be helped by its association with with his remarkably fragile ego.

Now, here’s the thing: citing terrorism as the main explanation for such a complicated rise is indeed a “surprising explanation” of the sort I warned against above. I am coming very close to reiterating a debunked theory of Trump, that his supporters are more “authoritarian” than other voters. The counter-narrative is that Trump’s supporters are… nationalist populists, proud to be American and cynical about elites. But in this excellent recent piece, James Kwak captured part of the problem with this sort of analysis:

Racism isn’t a virus that falls out of the sky. It’s the product of historical contexts. I can’t prove that today’s heightened racism results from the Great Recession, although it seems perfectly plausible to me. But by the same token, saying “It’s racism!” doesn’t preclude the role of economic factors in making that racism attractive.

Similarly, mistaken feelings of poverty and vulnerability (among a relatively stable and wealthy group) isn’t a virus that falls out of the sky: it doesn’t preclude the possibility that nationalism and the hatred of the currently non-white and likely forthcoming non-male President-as-elite-representative cause those feelings of nationalism–racism by another name–and and misogyny–which is what hatred of elites looks like when it’s at home.

So I suspect that it helps to spell out a useful understanding of Trump’s appeal, not at the margins but for the mass: they are actually quite safe, and yet they don’t feel safe: they feel vulnerable, fragile, and in need of protection. This makes him a great candidate to run against a hawkish Secretary of State–because she is a woman and cannot so easily play “strong man” in our subconscious.