All of this writing and data analysis is a lot of work! So after more than five years of posting, I've finally launched a Patreon to help pay the bills.


Why do Trump voters say they support Trump?

Dylan Matthews, in a new piece for Vox, is back to rejecting "the idea that [Trump] voters are motivated by economic struggles". This time around, he has a new argument:
Taking Trump voters' concerns seriously means listening to what they're actually saying...describing these people as motivated by racial resentment...[is] supported by extensive amounts of social scientific research and indeed by the statements of Trump's supporters themselves. 
This, of course, is supposed to cleverly turn on its head the notion that we're being empathetic when we acknowledge that economic anxiety may play a role in motivating Trump voters. Which might seem like a nuance that avoids completely demonizing them - but in fact, Matthews argues, such explanations are "insulting" because they don't take "the stated concerns of Trump voters...seriously in the slightest." In fact, if we take Trump voters seriously, we should dismiss these personal theories of motivation and take them at their word when they say they're motivated by racism.

Fair enough. But if Matthews wants to lean on deference to other perspectives, why does he then invoke "social scientific research" in addition to "the statements of Trump's supporters themselves"?

These two things are not the same! Almost exclusively, however, Matthews relies on a collection of studies that statistically correlate support for Trump with various indicators of racism, and on his inference about the "message this research sends". It's true that those indicators are usually self-reported racism - people making statements like "if black people would only try harder they could be just as well off as white people", for example. But from this evidence, researchers are concluding that one thing motivates the other, and pundits are concluding that one exclusively motivates the other. These may be defensible conclusions about the evidence, but they are not, as Matthews suggests, the conclusions of the respondents themselves.

That distinction is crucial, because Matthews is not making an argument about rigor, but about deference. He wants to cast people who see an economic role in support for Trump as ideologues who are even ignoring the explanations that Trump voters give for their own politics - that way, he gets to turn around the empathy argument so often wielded against liberals who are accused of ignoring economic hardship.

But to pull off that move, you would have to rely on what Trump voters are explicitly saying about their own motivations. And that point is fairly clear: when asked "why they support their candidate", 76% of Trump voters credit his "views on the economy", while only 28% credit his "views on race relations". The single tweet Matthews offers as evidence that "it's not about economics" doesn't somehow overthrow these results.

Personally, I think it's pretty ridiculous to defer to the theories Trump voters give about their own motivations. For example, when only 28% say they're motivated by his views on race, I suspect that the long-understood problem of social desirability bias is coming into play and that a lot of them are being dishonest. I also suspect that when 78% credit Trump's "views on terrorism" for their support, what they actually mean are his views on Muslims and other minorities, meaning that the role of racism is actually much higher than 28% suggests. Additionally, I also suspect that Trump supporters are unlikely to even realize when they're motivated by implicit racism, even though that almost certainly plays a huge role in their politics. These are all good reasons why we should not "take the stated concerns of Trump voters...seriously" in the sense of allowing them to discredit a more sophisticated analysis.

But of course, when we pursue a more sophisticated analysis, we often get more sophisticated conclusions. Thus for example, in the same Gallup study that Matthews thinks "confirms" his analysis, the author explicitly affirms the role of economic factors in Trumpism: the evidence, he writes, indicates "support for the idea that Trump supporters are less prosperous than others" and he goes on to speculate that "material circumstances caused by economic shocks...are the true underlying causes."

Acknowledging a role for economics in the Trump candidacy does not, of course, preclude a role for racism as well - and as Jeff Spross points out, that was never the left argument to begin with. Regardless, Trump voters are obviously not telling us to rule out economic motivations, and even if they did, it would be ridiculous to defer to their analysis.


Perhaps Trump is a good opportunity to rethink presidential democracy

Most liberal-left discussion of Donald Trump's candidacy has focused, at least implicitly, on keeping him (and similar candidates) out of the White House. Predictably, that goal usually dictates the standard two-pronged agenda of American activism. On one hand, we are enjoined to remedy the various socioeconomic pathologies (economic anxiety, white ethno-nationalism, and so on) that created Trump's constituency in the first place; on the other hand, we have to fight him for the presidency itself.

Perhaps we shouldn't be surprised that this framing of the liberal-left agenda is so popular in an election year - it neatly advances the theory that we need to elect a progressive president, one who can both keep Trump out of office and fight the socioeconomic conditions that led to his candidacy. Every one of his electoral opponents, of course, is necessarily making this sort of argument, and the media will predictably let the way politicians talk about politics dictate their coverage.

There is, however, an alternative that neither the Clinton nor Stein campaigns are likely to ever bring up: we get rid of the presidency altogether.

Yet that's precisely the solution that came to mind this morning when I re-read Juan Linz's The Perils of Presidentialism, available in full here. In that classic essay, Linz weighs the advantages and disadvantages of presidential against parliamentary democracy, and concludes that "the odds that presidentialism will help preserve democracy are far less favorable."

What is truly striking is how closely Linz's nightmare scenarios for democracy correspond with the dangers posed by candidates like Trump. He notes, for example, the
risk that [the president] will tend to conflate his supporters with "the people" as a whole...[this may] bring on a refusal to acknowledge the limits of the mandate that even a majority - to say nothing of a mere plurality - can claim as democratic justification for the enactment of its agenda. The doleful potential for displays of cold indifference, disrespect, or even downright hostility toward the opposition is not to be scanted...
This risk is fostered by the very nature of the presidency - thus, we even see it realized in candidates like Clinton, whose partisans routinely invoke the rhetoric of democratic consensus while marginalizing the opposition's agenda as (to quote Linz) "the selfish design of narrow interests". That move has some particularly ominous precedents when the Clinton campaign starts leaning heavily on nationalistic paranoia about meddling foreign agents.

If that dynamic is detectable in Clinton's campaign, however, it's ramped up to eleven in Trump's, which is built entirely on contrasting the will of "real Americans" with the sinister machinations of all kinds of out-groups (Muslims, terrorists, immigrants, Jews, and so on). And dangerously, the mandate of "Real America", in the eyes of the Trump right, is their license for a draconian eliminationist politics, which includes everything from religious tests at the border to assaulting political protesters.

These essentially fascist currents in American politics are in no small part driven by the battle over the presidency, which many people see as their only vehicle for exercising political agency. As Linz continues,
...a president bids fair to become the focus for whatever exaggerated expectations his supporters may harbor. They are prone to think that he has more power than he really has or should have and may sometimes be politically mobilized against any adversaries who bar his way. The interaction between a popular president and the crowd acclaiming him can generate fear among his opponents and a tense political climate.
Again, this cuts both ways. The 2016 election has clearly become the locus of intense anxiety for millions of Americans, who see the whole of their political fates bound up in a single decision. It may be Clinton's politics that causes the right to fear gun confiscation, Sharia law, high taxes, and so on - but it is the expansive power of the presidency that makes them fear that she might actually be able to do all of this, and that incites them to express intense, boundless fear and paranoia as radical enthusiasm for Trump. Meanwhile, that same enormous power seduces Trump supporters into seeing him as the silver bullet for all of their problems, and cultivates intense frustration and disappointment when he encounters obstacles to victory.

These incentives aren't unique to Trump supporters - partisans for Clinton are driven by many of the same anxieties. Suffice to say that this is not how you have to arrange your democracy. In a parliamentary democracy, political struggle is diffused into a larger number of electoral decisions, which has the paradoxical effect of lowering the stakes and diluting the toxin of social conflict.

Obviously, while a presidential system may amplify democratic discord and express it in fundamentally dangerous ways, the presidency isn't the exclusive source of conflict in the United States. The familiar problems of poverty, inequality, racism, sexism, and so on are all contributing to polarization and the rise of extremism in the United States. But this isn't a counterpoint to Linz's argument - in fact, it simply affirms it. Repeatedly, the author points to the United States as the major exception to the historical rule of instability and dysfunction in presidential democracies. But there is, Linz argues, a reason why the US has been able to make it work:
the uniquely diffuse character of American political parties - which, ironically, exasperates many American political scientists and leads them to call for responsible, ideologically disciplined parties - has something to do with it... In countries where the preponderance of voters is centrist, agrees on the exclusion of extremists, and expects both rightist and leftist candidates to differ only within a larger, moderate consensus, the divisiveness latent in presidential competition is not a serious problem. 
In other words, the rise of Trump likely bodes the end of the presidential system, one way or another. Either we replace it with a more democratic parliamentary system, or we watch political polarization turn the presidency into an increasingly undemocratic institution incapable of mediating social conflict. For obvious reasons, if we prefer the former outcome, we probably ought not rely on presidential candidates to fight for it.


How do gender *and* age shape the electoral college?

Yesterday, Nate Silver posted a widely-disseminated article aspiring to show how the 2016 election would play out if only men or women voted. Predictably, the election swings deep blue when only women vote, and bright red when only men vote. Out of curiosity, I used Reuter's new States of the Nation tool to take a quick look at this year's age gap, and while the results were similarly predictable, this approach encounters three significant problems. First, the tool only allows one to drill down to the 18-30 year old bracket, which excludes Millennials ages 31-34. Second, a lot of the sample sizes are too small to produce reliable data, which means that even states like South Carolina (with 9 electoral votes) can't be called. And third, 538 used a completely different methodology to produce its maps:
Here’s a quick way to estimate it. In the polls I cited above, Clinton is doing 10 points better among women than among the electorate overall. So we’ll add 10 points to her current polls-only margin in every state to forecast her performance if women were the only ones who could vote.
This is a pretty fast-and-loose approach that I don't have much confidence in, but it at least makes it easy to make some apples-to-apples comparisons. So relying on the 538 approach, here's what the election looks like if only Millennials voted:

In a youth election, Trump is held to single digit electoral votes, and Clinton wins every state except two: Wyoming, and Nebraska's third district. Compare this to what happens if no Millennials vote:

Here, Clinton still manages a win, but Arizona, Iowa, North Carolina, and Maine's second district all swing into Trump's column. More likely, this could threaten Democratic hopes of capturing the Senate by allowing Republicans to keep control of seats in Indiana and Wisconsin - both states Democrats hope to flip.

The conclusions here are pretty straightforward:
  • In the electoral college, at least, the gender gap is bigger than the age gap: there is a 270 vote swing from men to women, compared to a 213 vote swing from olds to millennials.
  • These gaps create significantly different outcomes, however, because they're centered at different points: women are more reactionary than millennials, and men are more reactionary than olds.
These points raise an interesting intersectional question: how do age and gender combine to influence electoral outcomes? To answer this, the first thing we need to do is break down the vote margins for every combination of old and young men and women, which gives us a table like this:

Already, it should be completely obvious that age ends up playing a far more important role in predicting preference than gender. Trump has a slight advantage among old men, and because of the sheer number of olds voting, it's easy to look at the electoral college and conclude that this pattern holds for all age groups. But in fact, not only does this pattern not hold among millennials - it actually reverses, so that Clinton has an 8.4 point advantage among millennial men compared to millennial women. Even among millennial women, however, Clinton's advantage is astronomically larger than it is among old women, who actually do .2 points worse than the national average.

If we look at how these differences would play out in the electoral college, it's clear where Trump's support comes from and where Clinton's support comes from. Among older women, Clinton would actually lose an electoral vote in Maine's second district. And among young men, Clinton wins every electoral vote except one, in Nebraska's third district. It's youth that most powerfully predicts support for Clinton, with gender proving unreliable at best.


Trump's tape vindicates the left-feminist critique of liberalism

Donald Trump's [latest] inflammatory remarks, just publicized from a 2005 recording, have sparked an extraordinary and much-deserved political backlash - even within his own party. A lot of the outrage has simply been directed at the "extremely lewd" character of his remarks, as when he calls a woman "a bitch" and ridicules her "phony tits"; some critics have been most offended by the "vile degradation" of "hitting on married women"; others have criticized his "boasts about sexual assaults".

One aspect of Trump's comments that has been largely overlooked, however, is his explicit admission of the way that power licenses his odious behavior:
When you're a star, they let you do it. You can do anything...grab them by the pussy. You can do anything.
The few responses to this have been telling. When Joe Biden decries such behavior as "an abuse of power", he isn't actually criticizing the power that lets people like Trump "do anything" - he's simply condemning its "abuse". And within capitalism, Kash CEO Kaz Nejatian insists, this is the approach we must take:

This is a decisively liberal feminism: one that takes hierarchies of power for granted. Within liberal feminism, the opportunity to abuse power must be defended, and even the opportunity to abuse women; if your "freedom" to acquire ungovernable economic power puts women at risk, so be it. Liberal feminism allows us to do things like "condemn" patriarchy, and shame it - but any collective action that actually mobilizes the arm of the state is prohibited, as this would be an imposition on "freedom". Contrast this to this second approach, proposed by Matt Karp:

Unlike Biden and Nejatian, Karp is actually prioritizing the danger posed by Trump. For obvious reasons, the VP and CEO are both first and foremost interested in preserving hierarchy, and only within that framework are they willing to consider the safety of women. But when Trump says that it is power that allows him to "do anything", Karp takes this seriously, and concludes that people should not have the kind of power Trump has if it puts women at risk. If this is a problem for capitalism or an imposition on the freedom of men to "abuse power", so be it.

Obviously, as Karp's own formulation suggests, the "extinction of the billionaire class" alone can't solve the problem of patriarchy; things like shaming have a role to play, too. But a feminism that exclusively relies on the latter because it prioritizes the preservation of billionaires over the preservation of women hardly deserves the name feminism; it's just capitalism with a "feminist" brand.


Harold Myerson thinks we have a white racist millennial Clinton-defector problem. He's wrong.

I anticipated this in my previous article, but since Harold Myerson is making the argument directly, I'd like to shoot it down directly:
Hillary Clinton is still having trouble winning the allegiance of the apt description of the millennials holding out for the third-party candidates: They’re all white...a hard core of young, white Bernie-or-Busters may yet believe that voting for Stein, or even Johnson, is an expression of their disdain for the system.
As noted, Clinton is not actually having trouble winning the allegiance of millennials - she's winning about 53% of them, compared with a minority of 38.9% among olds. Meanwhile, a plurality 39.3% of olds are voting for Trump, compared to just 24.7% of millennials. Clinton does not have a millennial problem - she has an old people problem, and a millennial solution. Analytically breaking down these age groups into conveniently gerrymandered sub-demographics (white millennials! third-party defector millennials! white third party defector millennials!) does nothing to contest the broader, obvious age trend; it just makes one's analysis increasingly narrow, and increasingly irrelevant.

Faced with these brute numbers, Myerson now only has one possible move: to compare today's young voters with those from 2012. But if we do that, the second premise of his argument collapses: Myerson wants to blame "white skin privilege" for the failure to support Clinton, but if the last election is our baseline, it's young people of color who are running from the Democratic party. He can, that is to say, only salvage his critique of millennials by abandoning his critique of white people, and vice versa.

A second problem with Myerson's race critique is that he repeatedly tries to make it into a specific critique of leftists. He opens with an anecdote about Stein and Sanders supporters; he quotes an organizer who singles out Sanders supporters; and he closes by once again brooding about the notorious "Bernie-or-Busters".

But to do this, he has to play fast-and-loose with the numbers, and in a way that strikes me as pretty deliberate. Consider, for example, the crux of his argument, which the article even highlights in a pull-quote:
Presumably, this 2% discrepancy demonstrates some kind of white privilege among leftist voters. That's why it's interesting that he omits a directly relevant fact: the same poll reports that Stein is also at 4% among Latinx and Asian Americans. An even more interesting point is that he includes those same numbers when he reports on Gary Johnson, who has the backing of "15 percent of whites...but just 8 percent of Latinos, 6 percent of Asian Americans, and 4 percent of African Americans." Comparatively, it seems clear that white supremacy is far and away the province of young libertarian voters, and that there is no detectable third-party voting trend that's unique to young white leftists; it also looks a lot like Myerson erased Stein's support among Asians and Latinxs precisely to obscure this point.

Finally, it's worth putting the millennial attrition issue into perspective. When Myerson notes that Stein and Johnson have 4% and 11% of the millennial vote, one's tempted to conclude that this amounts to 15% of the youth vote for Clinton. But in fact, as YouGov reports, only 35% of third party voters under 30 say that they prefer Clinton to Trump - meaning that she's really only losing about 5% of the youth vote to third parties. And since there's no significant evidence that this 5% is disproportionately white, it's hard to escape the impression that Myerson is accusing young leftists of racism for no good reason, while ignoring support for Trump in how own generation that's larger by several orders of magnitude.


The blame-the-kids two-step

In just the past few weeks, The Hill, The Observer, Salon, Paste, Fox News, Newsweek, New York Magazine, Bloomberg, and Vox have run articles referencing "Clinton's millennial problem" - and that's leaving out the endless parade of television and social media pundits saying the exact same thing. From, this, one might suspect that there is some kind of specific and known challenge that Clinton faces among millennials - and usually, the conclusion is that it is up to millennials to fix it.

But look closer, and you'll actually notice that the anti-millennial grievance has switched between two lines of criticism, each equally baseless in their own way.

Argument 1

Here, in argument (1), millennial support for Clinton is inadequate compared to other age groups. Specifically, olds like to point to the comparatively high youth support for third-party candidates in order to suggest that support for Clinton comparatively low. This of course is demonstrably incorrect on multiple grounds. Voters under 30 give Clinton her second highest margin of all age groups, and millennials (when we bracket generations correctly) give her a margin greater than every other age group combined. And while millennials may vote for third parties in high numbers, most of that attrition is coming from Trump's camp, not Clinton's. For instance, among supporters of Gary Johnson, every other age group has a significantly higher attrition rate from Clinton's camp than millennials do.

Thus, argument 1 fails completely: millennials are voting in higher numbers for Clinton than anyone else, and defecting from her far less. 

Argument 2

That's why, particularly in the past week or so, we've seen a slightly different complaint:

In argument 2, millennial support for Clinton is inadequate compared to their support for Obama. The premise here is that we should expect age groups to give Democrats a certain level of support based on the last election, and that their failure can be measured by how much that support declines.

It should probably be enough to say that pointing to the last election really just defers the question: instead of asking "are millennials underperforming," we're now asking "did millennials overperform last time?" But instead of leaning too hard on that, I'd like to make a distinct point:

18-30 year old voters

Clearly, if failure is simply measured by decline, then young African Americans are failing liberals more than anyone, followed by Asian Americans and then Latino/as. Any shift among support among young white voters, meanwhile, is so small that it's barely outside the margin of error (as is any current difference between young whites and older whites). So you cannot, that is to say, talk about the decline in youth support for Democrats without laying almost exclusive blame on young voters of color. Somehow, I doubt that Clinton's (largely white) media surrogates will have the audacity to rally behind that complaint.

What if Clinton does not actually have a millennial problem

Often when a grievance has to rely on multiple, shifting lines of critique, we can take this as a good indication that the arguments are just being backfilled to support the complaint. This is particularly true when the criticism turns out to be unusually baseless and flimsy; when this happens, it often makes sense to start looking for motivation. One egregiously incorrect data mistake is understandable - two are suspicious, and beg for an explanation.

Here, I think the explanation is pretty simple. Clinton's lead over Trump has tightened over the past month, to the point of occasionally disappearing altogether. This is making Clinton supporters nervous, and naturally they are looking for someone to blame. And because olds are significantly overrepresented on the editorial boards and in the opinion columns of elite media, this anxiety will predictably express itself as an inclination towards blaming young people. The persistence of this grievance says almost nothing about the actual numbers - past or present - but it says a lot about the ageism in our media.


Unpacking "alt-"

I've been fascinated by the recent popularization of the prefix "alt-" as an ideological descriptor in American political discourse. Anyone who pays much attention to popular politics online knows that it's been around for quite some time (at least since 2008) in the formulation "alt-right", which generally just referred to white capitalist ethnonationalists. This was, as Shane Burley notes, a term of their own making, and one that they continue to embrace.

In just the past year, however, the etymology took a decisive turn. That's when the American right, faced with widespread opprobrium of the alt-right, made their standard move of constructing a political equivalence: there is also, they argued, an "alt-left". Look at the history of this term, and you'll only see it pop up sporadically over the past few years, for instance on an obscure and stagnant Reddit page. But just this year, you'll find an insurgency of articles like What About the 'Alt-Left'?:
The New York Sun wants no part of [the alt-right] and neither does the GOP. But the Democratic Party has its own fringe for which to answer before Mrs. Clinton has any standing to make a megillah of the “alt-right.” What about the alt-left?
The argument here, quite explicitly, makes two moves that define the entire genre. First, crucially, it de-centers the alt-right by placing it on the "fringe". This isn't necessarily implicit in the original formulation, and in fact the alt-right continues to see itself as "a silent majority or hidden mainstream" in America. While rejecting its claim to popularity, meanwhile, the liberal-left long made a similar argument: as Luis Miranda put it, "the GOP made Trump through years of divisive and ugly politics". There was, that is to say, a widespread consensus that the alt-right was merely just another variety of the American right, just like libertarianism, neoconservativism, paleoconservativism, and so on, and that all were implicated in the same reactionary politics. But this new definition of alt-right implicitly contests that, and its political function here is plainly to distance the "true" right from a damaging label. 

Thus, "alt-" went from meaning something like "white capitalist ethnonationalist" to something like "fringe" (a spatial metaphor synonymous with "not-center"). Having reframed the prefix to mean deviation from the center, the American right could then easily reorient it into their own attack. This is the second function of alt's redefinition: to create a slur that can be used to punch left. 

For the New York Sun, punching left simply means attacking Hillary Clinton, by creating the false equivalence that she, too, has her own extremists. There is, of course, no equivalence between the fascist base of Donald Trump's movement and any faction of Hillary Clinton's coalition; substantively, the charge doesn't make sense, because if there were white ethnonationalist capitalists supporting Clinton, we'd just call them alt-right Clinton supporters, not alt-left Clinton supporters. But by reconceptualizing "alt-" to mean "fringe", one can get around that substantive problem and apply it to whatever one considers the not-center-left. 

This etymology of alt- must be the starting point of any analysis of its role in modern political discourse. Historically, it is laden with the censure that all decent people extend to everything that is detestable about the American right, and historically it is inextricable from the broader evolution of right-wing politics. But rhetorically, it now functions to shield the "center-right" from that critique, and to place in its cross-hairs anything that deviates from the center. Uncoupled as it is from any substantive meaning, defined purely by the centrist metaphor of "fringe", we should not be surprise to see this attack drifting into the hands of centrist liberals - who themselves are always looking for new slurs to wield against the American left.


Shunning is mostly performative and the left has usually opposed it

Growing up in an Anabaptist community, I occasionally found myself in the middle of debates over the politics of shunning. For those who are unfamiliar, shunning is an old practice shaming and exclusion based on a few lines written by the Apostle Paul:
I wrote to you in my letter not to associate with sexually immoral people - not at all meaning the people of this world who are immoral, or the greedy and swindlers, or idolaters. In that case you would have to leave this world. But now I am writing to you that you must not associate with anyone who claims to be a brother or sister but is sexually immoral or greedy, an idolater or slanderer, a drunkard or swindler. Do not even eat with such people. (I Corinthians 5:9-11)
In practice, this could have some pretty dramatic consequences. While the basic rules of shunning are fairly narrow - don't eat with the person, don't do business with them, don't accept anything from them, etcetera - and sporadically observed, the ostracization that it legitimizes could effectively shut the target out of social life and turn them into a pariah. Justifying all of this, of course, was an elaborate apparatus of theology that few outside of the Amish church would find compelling.

In addition to theological justifications, however, the community also developed various pragmatic rationales that the modern liberal-left will find familiar. "By shunning [the offender] in all social relations," Hostetler writes, "the community gives him a status that minimizes the threat to other members of the community." The Dordrecht Confession of Faith, a central text of the Radical Reformation, advocates shunning so that the offender "may be made ashamed, be affected in his ways." In other words, shunning was supposed to have two practical consequences: 1) to engineer wokeness in the community, and 2) to shame the offender into rehabilitating.


Both of these rationales emerge time and time again in modern liberal-left advocacy for shaming and ostracization as tools of social engineering and personal discipline. And yet strangely enough, anyone at all familiar with the standard left critiques of shunning should have rejected both long ago.

To take the second point first, there is little reason to believe that shunning actually has any kind of rehabilitative effect on its target, and considerable reason to believe that it can actually amplify the problem. Delaney notes that "the effects on the shunned person can be devastating...[and] akin to psychological torture." Tanaka notes research on shunning that
indicates a severe distortion of the self image, for example, 'I am a type of person that everyone hates'...This long-term effect suggests a huge impact on one's identity...[it] has a strong impingement on emotional development, which as Kahn points out is the essence of cumulative trauma. 
Tanaka goes on to add that as a defense mechanism, the target of shunning may "develop a victim's identity...[that] may fix and solidify further their negative identity." This should be an all-to-familiar experience for anyone who has tried to shame an offender, only to watch them double-down and embrace the attack. The point here is not to argue that shunning is simply mean - it's to point out that it's often directly counterproductive in terms of its supposed goal. Instead of rehabilitating the offender, it can just as easily harden the offender and give him a powerful psychological / emotional stake in continuing his behavior. As Massaro observes,
Psychological accounts of shame suggest that the behavioral consequences of this emotion are unpredictable, and may include anger and a desire to retaliate against the one inflicting the shame. The shaming advocates' relative indifference to these concerns suggests that they likely are not particularly concerned with rehabilitating the offender.
Massaro adds that this unpredictability also comes into play regarding rationale (1) - that we should shun people as an exercise in social engineering:
Both the psychological and the anthropological works indicate that the general deterrence and expressive effects of shame measures are likely to be highly contextual and unpredictable...shame penalties often will have multiple potential meanings, depending on the communities to which these expressions are directed, and thus will have an uncertain impact on the targeted audience's behavior. 
Again, this just confirms experiences that everyone is already familiar with. Efforts to shun someone may effectively remove them from discourse and community and attach a social taboo against their behavior - but it is just as likely to do the exact opposite. Frequent readers will probably recognize in this line of criticism frequent skepticism about discourse gaming. The implicit theory behind rationale (1) is that instead of reasoning with people, and persuading them to avoid certain types of behavior, we tactically use all kinds of psychological tactics, like shaming, to manipulate them into behaving appropriately. Say what you will about the ethics of this approach, but as a matter of basic pragmatism there's no compelling reason to believe that it actually works.


The psychology and sociology on shunning and its efficacy are all fairly straightforward - and yet, particularly among the liberal-left, the tactic is still fairly popular as a way to mediate social conflict. A few theories on why we're actually still trying this:

  • Often, we just adopt shunning as a default measure when other efforts don't seem to be working: "Historians and criminologists have noted the extent to which shaming and shunning sanctions emerge from the public's frustration with conventional punishment options," Miller writes.  
  • As Posner writes in Laws and Social Norms, shunning is often just an exercise in self-interested performance. People participate in exercises like shaming "to show each other that they are cooperative types" and because doing so "serves as an opportunity for everyone to signal his reliability...the chief motive for shaming is to enhance reputation, not to do justice."
  • Often, I get the sense that liberals in particular are operating on an essentially capitalistic, marketplace-of-ideas model of socialization where we can essentially boycott and blacklist problematic people out of business. Here, behavior is commodified as a product that we can either patronize or shut out of the market by manipulating demand, which means that sociopolitical relationships can be simplified into a kind of consumer activism.
  • Historically, as suggested above, shunning was often a decisively religious procedure, couched in all kinds of metaphysical beliefs about purity and holiness. Instead of shunning people in order to achieve politically or socially productive ends, one is simply honoring a deontological commandment; there is, that is to say, no theory of social or personal harm at stake if we don't shun people, just some rule that it's what you should do.
Needless to say, none of these motives are particularly compelling, particularly from a left-liberal perspective interested in substantive sociopolitical progress. Without laying out the case for it here, my personal position on the topic is that the best way to deal with people engaged in deviant or problematic behavior is often to present them with arguments on why they should change it. Often, it can also help to establish the kind of good-faith relationship with them where they see you as a constructive critic rather than an adversary - this is really just psychology 101. This sort of relationship isn't always possible, of course, but it's exceedingly rare that anything resembling constructive interpersonal influence appears without it.

UPDATE: Readers from Twitter will have recognized that this piece was written amid a controversy over a Lebanese man who tweeted a violent threat to an American woman. Specifically, however, I wrote it in response to a particular argument, floated by Daniel Sieradski, that one should not follow this user for the specific reason that doing so "confers legitimacy on him".

It seems obvious to me that one can criticize that narrow argument without dictating broader conclusions about the controversy at hand. One can for instance say that a Twitter follow does not confer legitimacy, but that we should nevertheless shun out of solidarity with victims, in order to make them feel welcomed and safe and so on. That's a justification, grounded in personal support for and empathy with the oppressed, that is entirely distinct from Sieradski's elaborate scheme of using shunning as a mechanism for social-engineering legitimacy and illegitimacy norms.

In an effort to focus my argument on that narrow point and disentangle it from broader conclusions, this piece avoids reference to the Twitter controversy entirely. I also rely on the example of the Amish community, not simply because it gives background into my own experience with the issue, but also because it largely avoids the considerations of solidarity that emerge when shunning involves a victim. My focus here is simply on the social engineering argument, for reasons that any frequent reader will immediately recognize: I have a longstanding interest with liberal-left ventures in discourse-gaming / social-engineering and am generally skeptical of their efficacy, a point I've written about at length on multiple occasions.

Fortunately, the overwhelming majority of feedback I've received on this article has been positive, coming from readers (a majority of them women, incidentally) who recognize this narrow concern. Nevertheless, a few readers (almost all of them men, by the way) have read into this piece dispositive conclusions about the Twitter controversy and independent claims about solidarity. This reading has little to do with anything I've actually written here, for reasons given.


Poverty increases political polarization among race groups

A standard premise of identitarian demographic analysis is that privileged groups should vote like their oppressed counterparts. Instead, that is, of second-guessing their votes with our own analyses and rationalizations, privileged groups should defer to the judgment of the oppressed, whose positions are informed by the lived experience of oppression. An analytical corollary to all of this is that when privileged groups are not voting like their oppressed counterparts, we can understand this as an expression of privilege. When, for example, we see a difference in the way that white people are voting and the way that people are color are voting, we should suspect racism.

The racism index

If one accepts this approach, it's pretty straightforward to determine which of the two major candidates is supported by which race constituencies:

Overwhelmingly, voters of color prefer Clinton to Trump by a 47.4% margin. White voters, meanwhile, prefer Clinton to Trump by a -7.2% margin - which is to say, they actually prefer Trump. Graphed out, these margins just look something like this:

In theory, white voters, in solidarity with and deference to voters of color, should also support Clinton by a 47.4% margin. Instead, they oppose her by a 7.2% margin - a difference of 54.6%. For the sake of this analysis, I'm going to call that number the "racism index". Ideally, that index should fall to 0% as white people vote in unity with persons of color; at worst, it could hypothetically hit 200%, if both groups threw 100% of their support to opposing candidates.

The racism index, by income

Having established a racism index, it's pretty simple to conduct an intersectional analysis of race and income as manifest in the 2016 elections. All we have to do is apply the above methodology to particular income brackets. First, we chart out the preferences:

Then, we look at the margins:

And from here, we can easily assign each income bracket a racism index based on the difference between the margins. For instance, the racism index of the lowest income bracket is (47.2%+6.6%=53.8%), while that of the highest bracket is (36%+.8%=36.8%). Graphed out, our racism index will look something like this:

So what does this graph tell us? It's pretty straightforward: voting is less polarized between whites and people of color the higher everyone's income is. The only exception to this trend is among people making less than $25k a year, who tend to show more solidarity than people making more than $25-50k. Otherwise, the trend is direct.

Of course, income is not the same as wealth or precarity, and it's an even worse proxy for class; nevertheless, the numbers here seem to conform pretty closely to leftist intuitions about a relationship between racism and material conditions. Looking at a graph like this, I would be tempted to suspect that there's something about being poor that exacerbates racism, and something about prosperity and financial security that helps to foster political solidarity. If redistributing some of the wealth at the end of this graph could help bring down the racism in poorer income brackets, that strikes me as a win/win for anticapitalists and antiracists, even if it can't completely solve either problem.


Some intersectional analysis on the Trump and Clinton coalitions

The demographic divides that define the 2016 election have been fairly clear for months, though some pundits, of course, persist in getting them wrong. As I tweeted out earlier tonight, the general state of affairs looks something like this:

If you're white, and/or if you are a young boomer, you're more likely to vote for Trump than Clinton. Otherwise, you're more likely to vote for Clinton than Trump. Within those two coalitions, of course, the strength of support also varies, but any demographic analysis of this election has to begin with race and age.

This, of course, is a significant simplification. A more intersectional approach wouldn't just look at how identity, as defined by a single dimension, determines one's politics - it would look at how the intersection of multiple identities do so. And when we do this, the picture gets more complicated. 

The boomer race split

Consider, for example, how the numbers look at the intersection of age and race (here, black vs. white):

A few significant trends stand out. First, of course, race is by far the most important factor predicting one's vote - this is implicit in the first chart. This tendency is so strong, in fact, that it masks a second trend that only becomes visible here: a split between black and white middle-aged voters. Support for Trump isn't simply coming from young boomers. It's specifically coming from young white boomers, and also from ageing white gen-xers. Among black Americans, meanwhile, we see a complete reversal of this trend: support for Clinton is the strongest among black Americans between 50 and 59.

This fact of black American life gets completely erased by the usual simplified way that we talk about political demographics. It is numerically correct that Trump's support comes from boomers, but only because the overwhelming majority of boomers happen to be white. Such statistics tell us just as much about racial proportions among age groups as they tell us about political preferences. It's only when we dig into the data, and do an intersectional analysis of the trends, that we get a more accurate picture of what's going on.

In what sense is Trump the candidate of white men?

Or consider, meanwhile, the common characterization of Trump as the candidate of white men. This is obviously true in the narrow sense that Trump has the highest margin of support among white men - but let's put this fact in context:

Once again, we find that the most important divide at work is race. Race is what divides preference for Trump from preference from Clinton, and race proves far more important than gender even within the two coalitions. (This is why, for instance, support for Clinton is stronger among black men than Hispanic women.)

This point is worth attending to when, for example, we have multiple white women in media characterizing Trump as the candidate of white men. As we see above, this is mostly an exercise in last-place-avoidance - but it does little to distinguish white women otherwise. Their shared whiteness with white men is what ends up mattering; being a woman knocks off a few points off of their overall preference for Trump, but it does nothing to bring them into a coalition with people of color. Here, the intersectional high ground white women can claim over white men is about as legitimate as 40-49 year old whites scolding 60-69 year old whites for liking Trump just a little more than they do.


One takeaway from all of this is that the overwhelming majority of our demographic analysis is painfully simplistic. What I've done here is more sophisticated than what we usually see, and even this was an extremely limited effort. A more ambitious analysis would look at how multiple demographic dimensions - race, age, gender, income, and so on - all combine in unique points of intersection with unique political tendencies. Doing this with two genders, four races, five age groups, and six income brackets would leave me with 240 data points to analyze, and I don't do that kind of work for free. It is, nevertheless, the bare minimum of what anything resembling an adequate intersectional demography would have to involve.

A more important point, however, is that once you start digging into the way various forms of identity intersect, the usual simplifications that popular demographic analysis deals in end up obscuring as much as they reveal. It's fine to talk about the role that boomers and white men play in supporting Trump, but not if this erases the role of middle-aged black voters in opposing Trump, or if it masks the complicity of white women.

This kind of simplification isn't the only problem with what passes for identitarian demographic analysis, but it's one of the most pernicious and egregious - and if we're going to do this, we should at least get it right.