Bryan Caplan  

Eubulides, Wilkinson, and Discrimination

The Auto Boom: Healthier Econo... U.S. Federal Budget Cuts in th...
Eubulides of Miletus is best-known for the Sorites paradox:
The paradox goes as follows: consider a heap of sand from which grains are individually removed. One might construct the argument, using premises, as follows:
1,000,000 grains of sand is a heap of sand (Premise 1)
A heap of sand minus one grain is still a heap. (Premise 2)

Repeated applications of Premise 2 (each time starting with one fewer grains), eventually forces one to accept the conclusion that a heap may be composed of just one grain of sand...

Will Wilkinson uses the Sorites paradox to construct a novel defense of anti-discrimination law:
I used to think that if negative rights to non-interference were strictly observed, liberty was guaranteed, but I don't now. Here's how I had thought about the matter. One racist acting in a private capacity on his or her racist beliefs can't violate anyone's legitimate, negative rights. (No one is entitled to another's good opinion!) Two racists acting as private citizens on their racist beliefs can't violate anyone's rights. Therefore, I inferred, thousands or millions of racists acting non-coercively on their racist beliefs can't coercively violate anyone's rights. I now think this is quite wrongheaded.
My immediate reaction: Why draw the line at actions motivated by racism?  It seems like on Will's logic, the same applies to actions motivated by Christianity, lookism, environmentalism, patriotism, or any other widely-held position.  Indeed, why draw the line at actions?  Since people care about others' beliefs, it seems like beliefs as well as actions are suspect.

Will continues:

Eventually I realised that actions that are individually non-coercive can add up to stable patterns of behaviour that are systematically or structurally coercive, depriving some individuals of their rightful liberty. In fact, rights-violating structures or patterns of behaviour are excellent examples of Hayekian spontaneous orders--of phenomena that are the product of human action, but not of human design. This shift has led me to see racism and sexism themselves as threats to liberty.  Racism and sexism have come to matter more to me in that I have come to see them in terms of the political value that matters most to me: liberty.

This seems like a strangely unpluralist position, especially for someone like Will.  Why not just say that private racism and sexism are expressions of liberty, but many expressions of liberty have undesirable consequences?  Why not say that the legality of racism and sexism exposes a tension between, say, meritocracy and liberty?

Will goes on:

And so I have become much more sympathetic to policies that would limit individual liberty in order to suppress patterns or norms of behaviour that might pose an even greater threat to freedom. So I've become fairly friendly toward federal anti-discrimination law, affirmative action, Title 9, the works.

While I admire Will's invocation of the Sorites paradox, I still think he was wrong to change him mind.  If state action against workplace racism and sexism is warranted, why not state action against racism and sexism in dating and marriage?  Why not state action against every group that workplaces shun: the ugly, the socially awkward, guys with facial tattoos?  Indeed, from Will's standpoint, it seems like every artist who produces art that nobody wants to buy can justifiably accuse the world of "coercion."

The obvious reply for Will, of course, is that it's a matter of degree; we have to balance the liberty to negatively judge others against the liberty to not be habitually negatively judged.  For racism and sexism, the latter far outweighs the former.

This answer will no doubt appeal to defenders of the status quo, but it fails on multiple levels.  Most obviously: When prejudices are ubiquitous juggernauts, democracies don't make them illegal!  If virtually everyone held a prejudice, who would vote to suppress it?  Virtually no one.  That's why the ugly, the socially awkward, and guys with facial tattoos now quietly suffer: even nice, respectable people inwardly scoff when they complain.  If Will should be standing up for anyone, he should be standing up for these silent minorities. 

Think about it this way:  When discrimination is legal, to whom do its victims turn for protection?  The people who oppose that form of discrimination.  Under democracy, the continued legality of a form of discrimination shows that people who oppose that discrimination are rare and marginalized.  The continued illegality of a form of discrimination, in contrast, shows that people who oppose that discrimination are common and influential.*

The upshot: Will's framework may give him a good reason to support a long list of eccentric discrimination laws that our democracy will laugh out of court.  But his framework gives him little reason to support the discrimination laws our democracy has actually adopted.

* If opposition to workplace racism and sexism is so common and influential, why does it persist after fifty years of illegality?  Largely because the war on "racism" and "sexism" deliberately overlooks a long list of reasons why races and sexes objectively differ in job performance.

P.S. If you're sure that standard racism and sexism are the most injurious of all prejudices,  ponder the vast number of people against whom discrimination is not merely permissible but mandatory.  To quote Proudhon, "That is government; that is its justice; that is its morality."

Comments and Sharing

COMMENTS (21 to date)
Tom writes:

The sand issue isn't a paradox, it's a case of using poorly defined words mixed with well defined words. If instead of removing a grain of sand you removed a "handfull" would you still have a heap left? The answer is ambiguous as it depends on how big a handful is and how big a heap is.

WW makes a similar mistake. Is a million people a lot of people or a few? It depends on the overall population. 10 private racists could create coercion in a population of 11 but a million might not in a poppulation of hundreds of millions.

To support actual programs like the ones he mentions he has to believe that the proportion of racists/sexists to population fits into a range where their actions become coercive and yet there is still enough popular support to get those laws passed and enforced.

Since he doesn't bother to define the range that causes non coercive individual actions he ends up blindly supporting solutions without knowing if they are nessecary.

MichaelM writes:

The unspoken answer to the objection about democracies and discriminatory majorities is that an educated, liberal (in the old sense) minority should be given power separated and protected from majoritarian, popular feelings.

Educated liberals like Will Wilkinson ;)

The gentry have been trying to run this country since we threw out the British. This kind of argument is nothing new.

D writes:

Is there strong evidence that, after controlling for IQ and criminal background, blacks are heavily discriminated against in the marketplace?

IOW, how much is racism and how much is a fairly accurate assessment of the person's likely economic contribution (remember: IQ is the #1 predictor of job performance in the industrial/organizational psychology literature)?

Hugh writes:

Take Title 9 that WIlkinson likes so much:

It seems to be saying that girls should not be turned away if they want to practice sports at their university; who could object to that?

But we now find that boys are being turned away from practicing the sports they love, just to keep the statistics in line. That makes no sense to me (a man), and I am rapidly learning to discount everything Wilkinson says. Maybe working at the Economist has poisoned his mind.

John T. Kennedy writes:

It's the Fallacy of the Beard.

andy writes:

Eventually I realised that actions that are individually non-coercive can add up to stable patterns of behaviour that are systematically or structurally coercive

I wonder what definition of 'coercive' he uses. The logic behind this seems to me more as if he was claiming that if you added a grain of sand to a heap of sand, ultimately you would end up with a steel-reinforced sky-scraper.

david writes:

Cry 'pecuniary externalities!'

Where an individual's actions have a unilateral ability to affect another's material welfare: surely then it matters whether we assert that this is legitimate.

david writes:

For those interested in high economics: the traditional consequentialist defense of libertarianism - that intuitive notion that implementing the harm principle and free exchange leads to some Pareto-optimal social good - depends on all individuals being price-takers. Nobody can exert pecuniary externalities.

That occurs quite easily by having an infinite number of individuals. If we works backwards, as Wilkinson does via invoking the continuum paradox, we find that the result collapses. Wilkinson collapses it in a particularly emotive way.

There's some stuff in there about starting-line equality and such as well... heh.

ryan yin writes:


Pecuniary externalities don't preclude Pareto optimality. Wilkinson is talking about something different.


I wonder if this argument is a bit too clever. Your last pre-footnote sentence is undeniably correct if you were to say "But his framework gives him less reason to support the discrimination laws our democracy has actually adopted." But have you demonstrated "little"? What if I can be largely harmed by a significant chunk of the population that doesn't itself have control of the anti-discrimination laws? Or what if re-framing the argument makes people feel bad about something they were doing (or makes them feel bad long enough for it to be set in the law, at which point I don't need to convince the majority anymore, just a judge)?

Sieben writes:

I also don't understand how widespread racism could ever be coercive.

If racism really were widespread, prohibiting it would be mass thought policing...

stephen writes:

"Eventually I realised that actions that are individually non-coercive can add up to stable patterns of behaviour that are systematically or structurally coercive, depriving some individuals of their rightful liberty"

What happened, I am guessing, was that he wanted to affiliate more with a different group, so he switched his beliefs to fit in. I am willing to bet good money there was no new observed event.

Why should he be allowed to choose the beliefs and groups that he prefers, cause, you know, his own argument?

Chris Brennan writes:

Wilkinson, like many people, spends more time trying to win arguments by trying to define words in the particular ways he prefers than he does in trying to use words to effectively communicate the underlying substance of his ideas.

Jason Malloy writes:

The evidence suggests that the returns to education are comparable across ethnicity, especially when controlling for ability measures.

This is fairly easy to find out, so it comes across as dishonest when people pretend like the issue that motivates them is discrimination instead of inequality.

The reason for this dishonesty is that discrimination is far more morally persuasive, and evidently unjust, than inequality.

Which, of course, is another instance of the phenomenon from your Bastiat post, where false, but morally persuasive premises are systematically encouraged to push through certain policies.

Matt C writes:

I hope Will responds to your arguments. Who knows, maybe he supports affirmative action for ugly people.

I don't understand how you weigh the collective right of women and blacks (or ugly people or hillbillies) to be free from discrimination, against the right of other people to choose their associations, and come up with the right answer.

In general I don't see how collective rights are supposed to work within a libertarian framework.

Admittedly, Will does not call himself a libertarian any more.

I might find liberaltarianism more convincing or at least more interesting if I felt like I could clearly predict the liberaltarian position on an issue without reading something from Will or the guys at BHL first.

Some Dude writes:

Morality is a social construct.

Caring about racism and sexism is a necessity to get along in liberal social circles, especially as a white male. It's a signal that you fit in.

Beauty-ism and height-ism may have an even greater impact on those effected. But talking about them will get you odd looks, and may even cause others to question the seriousness with which you hold to liberal orthodoxies.

If you oppose some harmful discrimination against a group, and there is no group studies department at Harvard about that group, does anybody care?

david writes:

Pecuniary externalities don't preclude Pareto optimality...

Yes, they do. In their presence, there are almost always agents who would prefer to trade at the margin, but do not, because doing so would affect the market price and therefore their existing gains from previous trades. The archetypical example is the textbook price-setting monopolist.

But that's not really the point. The point is that where price-setting power exists, it is possible to (adversely) affect another individual's material welfare unilaterally, rather than (positively) through bilateral voluntary trade alone. It is rare for single individuals to exert such power, but prices have to come from somewhere, and so individuals in the aggregate can do so.

blink writes:

Will is right to challenge knee jerk reactions on these issues; there are no easy solutions. However, your point about democracy seems to be a knock-down argument here: Protecting the oppressed almost always means going *against* the majority so it is naive to look to democratic institutions for protection. Instead, Will's thinking should lead to individual action: Stand up, speak out, and challenge the wrongful treatment of others when you see it.

Steve Sailer writes:

What century is this anyway? Will Wilkinson is girding his loins to fight a battle his side won before he was born. The jousting tournament at Ye Olde Renaissance Faire is about as relevant as Will's crusade against racism and sexism.

t3 writes:

Will changed his mind about the size of the effect of racism and sexism on blacks and women. But for an analytical guy, he doesn't seem to want to make a case on this key ground: his new weighting of the balance between liberties lost and gained. (Maybe he does it somewhere else, but I bet not.)

Next, Will will discover he was wrong about another apostasy:

"I have come to understand that the combined effects of individual decisions on emissions can create a stable pattern of overuse. Each additional molecule of C02 may have a neglible impact. But together, they could destroy our planet, which would threaten liberty. So I've become fairly friendly to carbon taxes, government discouragement of coal plants, and things like that. It's a feeling I have, like being in love or having to go to the bathroom or being welcomed to Ezra Klein's place for homemade sushi."

D writes:

"But for an analytical guy..."

He also claims to be an empiricist when it comes to the social sciences. In that case one would have expected he would have a response(or to have actually pre-empted it)to someone like Jason Malloy.

ryan yin writes:


Don't focus on price-setting power -- the reason we say pecuniary externalities aren't market failures has nothing to do with infintesmally small agents (pollution doesn't stop being an externality just because every single person is "small"). It's that strictly speaking, pecuniaries are just a distribution thing. Market power is more than just a pecuniary externality. Though I'm probably quibbling with the logic rather than the conclusion.

Comments for this entry have been closed
Return to top