Bryan Caplan  

Discrimination, Liberty, and the Sorites Paradox

PRINT
Bill Dickens Responds on Pover... Optimizing Your Family Size in...
The Sorites Paradox works in two directions.

Top-down: 1,000,000 grains of sand is a heap of sand; a heap of sand minus one grain is still a heap; therefore one or grain of sand (or zero!) is a heap of sand.

Bottom-up:
One grain of sand is not a heap of sand; a non-heap of sand plus one grain is still not a heap; therefore there are no heaps of sand.

Will Wilkinson's defense of discrimination laws appeals to the absurdity of the Bottom-up version of the Sorites paradox:
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.
In my previous critique of Will's argument, I said that "he was wrong to change him mind."  After further reflection, though, I've concluded that he moved from one wrong position to another wrong position.

Will's initial argument, as he now happily admits, was invalid.  The mere fact that one person's negative opinion doesn't impair your liberty does not show that many people's negative opinions do not impair your liberty. 

But what if we switch from the Bottom-up Sorites argument against discrimination laws to a Top-down version?  Like so:

1. If everyone else on earth treated you badly without violating your person or property, it would not impair your liberty. 
2. If N people treating you badly without violating your person or property would not impair your liberty, then (N-1) people treating you badly without violating your person or property would not impair your liberty.
3. Conclusion: Regardless of how many people treat you badly without violating your person or property, it does not impair your liberty.

This is clearly a valid argument for the conclusion that Will once held, then abandoned.  But is it sound?  Most people, probably including Will, would call it absurd.  But is it?  Consider the following claims:

1. If everyone else on earth was gay, it would not impair your liberty.
2. If everyone on earth refused to be your friend, it would not impair your liberty.
3. If everyone on earth refused to buy a ticket to your poetry recital, it would not impair your liberty - even if poetry recital were your only marketable skill.
4. If everyone else on earth believed in astrology, it would not impair your liberty.
5. If everyone else on earth refused to work for you, it would not impair your liberty.
6. If everyone on earth refused to dine at your restaurant, it would not impair your liberty.
7. If every woman on earth refused to date you, it would not impair your liberty.

You don't have to be a dogmatic libertarian to accept many or even all of these claims. 

What makes these claims broadly plausible?  The moralized theory of liberty that most of us share: If you don't rightfully own something, you can't legitimately complain that the way the rightful owners use it "impairs your liberty."  Since you don't own other people, you can't legitimately complain that their sexual preferences impair your liberty.  Since your don't own other people's money, you can't legitimately complain that how they spend their money impairs your liberty.  Etc.

The key difference between libertarians and normal people isn't acceptance of this sort of claim.  The key difference, rather, is whether you accept such claims for employers.  E.g.:

1. If every employer on earth refuses to hire you, it does not impair your liberty.
2. If every employer on earth is mean to you, it does not impair your liberty.
3. If every employer on earth refuses to offer you healthcare, it does not impair your liberty.
Etc.

My question: Why on earth should we regard employers so differently?

Because you need their money to live?  You could say the same about customers in a world of self-employment.  If customers utterly disdain your wares, you're in dire straits.  But have the world's customers impaired your liberty?

Because you'll be miserable if employers don't give you a decent job?  You could say the same if everyone refused to be your friend.  You might even be suicidal.  But has the human race impaired your liberty?

It is easy to dismiss libertarians for "refusing to see" the coercive power of social pressure.  The reality, though, is that almost everyone usually "refuses to see" the coercive power of social pressure.  The simplest explanation is that people "refuse to see" social pressure as coercive as long as people have a moral right to exercise it.  You don't have a moral right to a friend or a date or an audience, so you remain free despite your deprivation. 

The libertarian critique of discrimination law isn't some weird novelty.  Once you stop treating employers as uniquely morally suspect, the libertarian critique immediately follows from common-sense morality.  The Sorites paradox, properly deployed, merely cements the libertarian case.



COMMENTS (18 to date)
Matt C writes:

I am only guessing, but I think a liberaltarian might indeed see cause for redress in some or all of 1-7.

Again, I'm interested in Will's response. I don't normally follow him, so I hope you'll link if he replies.

Philip writes:

It is amazing, truly remarkable, that Brian cannot see that: if large numbers of people do not own any means of production, and all of the people who do own means of production refuse to hire them, then their liberty is worth very little. Even Nozick was far less locked into an axiomatic way of thinking about allowable policies than this, which is why he made admissions about how his theory depends on some fundamental respect for rights of acquisition that are quite clearly false in the real world.

Of course, if anyone suggests to Bryan that a philosophical exercise like this one is incapable of determining wise policy in the real world, which for most of us is about particular effects on real people in our particular history-laden world, he will just shrug it off as evidence that the rest of us are less logically rigorous than he is.

Perry Metzger writes:

The Sorites Paradox when it is true (not in the instances where it is sophistry) is better known as mathematical induction.

Caleb writes:

@ Matt C

I'm interested. What do you have in mind? Because no matter as hard as I think, I can't come up with a single classically liberal legal remedy for any of 1-7. That's not to say I can't think of scenarios where each condition is a condition precedent to a legally actionable occurrence. But each of those scenarios requires a fact not in evidence.

Dan Hill writes:

@ Matt C - what @Caleb said, or to summarize, you are wrong. In fact, you couldn't be more wrong. I haven't even heard extreme nanny-staters proposing remedies to any of 1-7.

Doug writes:

I find that the vast majority of people who have the knee jerk reaction against employers as a special form of oppressor have very rarely ever held a management position.

Ask anyone who's every run a fast food restaurant about their worst employee stories. It becomes very clear that most forms of cited "employer oppression" are manifestly needed given the state of the (particularly unskilled) workforce.

Without these countermeasures, due to the fat left tail of potential damage a bad/malicious employee can cause, enterprise profitability and hence wages would be much lower. Observe how employees at McDonalds are subjected to more "workplace oppression" than employees at Google. The former's typical workforce has a much higher concentration of "bad apples."

When the poor get angry at onerous and humiliating employer demands it's astounding that all their anger gets directed at the firm. They reserve no hate for their bad behaving compatriots. Who while small in proportion make up a very large concentration of the damage that necessitates those demands.

How does this translate to workplace discrimination. Most economic labor discrimination is driven by real statistical differences between groups. If race X is being discriminated against by firm A, why get made at firm A? Why not get mad at that disproportionate members of bad apples in your own race?

Matt C writes:

As I understand them, liberaltarian ideals embrace positive liberty as well as negative liberty. Being fired from your job should be considered a harm in much the same way as getting punched in the nose is.

Not so long ago the liberaltarians were clear that your employer setting a policy you didn't like, so you had to go along with it or quit your job, was unacceptable and justified government intervention. If we are looking at magnitude of harm, certainly not having any friends, or not having any prospect of finding a mate, is a greater harm than having to quit your job.

It may be that there is some layer of unfairness needed before the liberaltarian thinks intervention is justified. Perhaps the examples could be modified slightly to make them feel more unfair. Or maybe you're right and liberaltarians would regard these as absurd and outside consideration for compensation--but this is far from obvious to me.

I'd be interested in hearing from people who actually regard themselves as liberaltarians.

R. Jones writes:

This is a wonderful reductio ad absurdum of "liberty with a capital L", libertarianism.
IF nobody wants to protect me from harms --- bodily, emotional, financial, social, etc...then they aren't taking away my liberty, right? They always have that right to abstain? Then what happens when they all abstain? How is my "liberty" not violated?

Tracy W writes:

Philip: if large numbers of people do not own any means of production, and all of the people who do own means of production refuse to hire them, then their liberty is worth very little.

Human capital is a means of production. So there are not large numbers of people who do not own any means of production. Yes, people who are say in comas don't benefit much from liberty, but no political system can do anything more about that than to fund research into treating their medical problems.

If you define "means of production" so as to exclude human capital, then isn't it equally the case that if all the people who own human capital refuse to work for those who own the means of production, then the liberty of those who own the means of production is worth very little?

Anon. writes:

Not only is human capital a means of production, it comprises the VAST majority of total capital.

The idea that employers have a disproportionate amount of power in a service economy, in which they control only a tiny part of total capital, is completely and utterly absurd.

@Philip

The question isn't about the worth of liberty, but of liberty itself. There are many things that restrict liberty but increase its worth; that doesn't mean they are morally right.

Philip writes:

Tracy W:
I take your point, but don't draw the same implication as you seem to. First, it suggests to me that society has a quite profound obligation to ensure you do have human capital. For Brian, perhaps this just entails that the community ensure that children are not very badly mistreated by their parents; for me it seems likely that human capital formation can be crippled in far less dramatic ways. It also seems worth pointing out that human capital by itself is incapable of feeding anybody.

As for your second question, it doesn't seem exactly parallel. If it is people who own the means of production, presumably they will not join the general strike you contemplate, and so their liberty will not be as completely impaired. If we imagine computer overlords who are themselves incapable of supplying real-world labor to own all of the means of production, it is true that their freedom of property will amount to little.

But that was my point. As R. Jones writes, these sort of considerations lead most of us to conclude that there is something wrong (and indeed rather morally bankrupt) about imagining that this style of reasoning is capable of informing our real-world policy choices, which we conceive of first and foremost as trying to help people rather than just preserve their liberty. Probably even Bryan agrees with part of this thought--last I checked, he did not place himself among the anarcho-capitalists. That being the case, the exercise in this post pretty clearly proves too much to be helpful for, e.g., arguing about the particulars of appropriate discrimination policy.

Ken B writes:

What if everyone refuses to allow me on their property?

Consider if I own no property, or a small isolated parcel with no food or water?

--------
@Perry Metzger: Greetings and LTNS. I believe we had some squabbles in the early 90s back in the days of 96 baud ...

Andrew writes:

@Ken B

What if everyone refuses to allow me on their property?

Then where could you possibly be? By being some where, you have to be on 'someones' property.

@Bryan

The key difference between libertarians and normal people

Thanks for this... I spit up a little soda on my keyboard...

Ken B writes:

@Andrew: I could have been born and in a leased apartment, the lease now elapsed, but in any case you will note I anticipated your objection in my next sentence. The point remains.

Lawrence D'Anna writes:

You seem to be claiming that the ony legitmate claim that a person can make on anything external to themselves is ownership. Is that really your position?

Costard writes:
It is amazing, truly remarkable, that Brian cannot see that: if large numbers of people do not own any means of production, and all of the people who do own means of production refuse to hire them, then their liberty is worth very little.

Precisely the opposite. This is when their liberty is worth the most, because a free use of hand and mind is the only means at their disposal. But as you say the hypothetical is absurd. Liberty is a social right, and you've begun by assuming an anti-society.

@Ken B: Newborn infants trespass on doorsteps every day. Yet the nuns and the firemen never seem to sue them.

Yet war always exacts its toll.

Society may be imperfect but social breakdown is calamitous. The worst social policy is the one that destroys what it is intended to perfect, by pitting one constituent against another and making enemies out of free associates.

I would say that the examples being given here against liberty, could only arise in a society in which liberty didn't exist. Unanimous opinion is a feature of the totalitarian state. Rampant racism is not a hypothetical but a historical reality, and yet the oppression that resulted from European antisemitism was legal, not economic; for many centuries it was states, not markets, that punished Jews.

In fact the complaint usually leveled against free societies and markets is that they fail to exclude specific participants and practices and substances. If you wish to find an exemplar of "positive" rights, look no further than pre-revolutionary France, in which people had neither liberty nor equality nor food.

Ken B writes:

@Costard and Lawrence D'Anna:

Costard's point about nuns is irrelevant as we are not discussing likely outcomes but *principles*. And Lawrence's question is about exactly the point I am driving at. Trying to reduce everything to just property rights leads to absurd cases, like mine. It is no answer to say, as Costard does, 'oh that won't happen' when you are arguing 'it's fine if it DOES happen.'

Kendall writes:

It seems to me the statement a heap minus one grain is still a heap is not always true. Even if the boundry is not clearly defined there is a point where a pile of sand becomes a heap. This principle may apply in some of the other examples.

Comments for this entry have been closed
Return to top