I’ll be back on Tuesday with, I expect, something new to say.

 Print

Share/Save

[Before you tell me that giving up meat is socially beneficial because it holds the price of meat down, remember that low prices are good for buyers only to exactly the same extent that they're bad for sellers. Changing a price does no net good. The rigorous proof of this is part of the theory of pecuniary externalities, on which the Wikipedia entry is uncharacteristically useless.]

Observing Lent or Passover has much in common with things like running around a track: You push yourself to do something hard, you feel good about it, and you leave the world pretty much the way you found it. What a shame that you didn’t push yourself to do something useful instead. I bet you could have learned to feel almost as good about that.

The two tragedies have very different micro foundations. Athletic events are wasteful essentially because they’re arms races; the winner succeeds only by preventing others from succeeding. (I posted about this here and here.) On Passover, by contrast, we can all be equally successful at sticking to the same ridiculous diet. So athletic competitions and religious observance are tragic for fundamentally different reasons, even though they’re tragic in the same way.

 Print

Share/Save

Is it okay to steal? Certainly not, and I’ve already told you why: The time and effort you spend stealing things is time and effort you could spend producing things instead. Theft leaves the world poorer than it could have been.

Is it okay to counterfeit? Certainly not, because counterfeiting is stealing. The time and effort you spend producing a phony dollar bill entitles you to a Hostess cupcake or a bus ride or a Blockbuster video rental without adding anything to the world’s stock of food, transportation, or entertainment. The cupcake you eat is made of flour and sugar that someone else could have eaten.

With that as background, Eli asked me this:

Is it okay for me to counterfeit if the central bank is not being sufficiently expansionary?

I stalled for a minute by asking if Eli was going to keep the seignorage (that’s economics talk for asking whether he was going to use his counterfeit money to buy stuff for himself instead of others). Eli rightly pointed out that this was a purely distributional question and hence irrelevant to the economic efficiency standard I’d been touting. (He later blogged his further thoughts on the matter.)

So the question then becomes: When is it okay to break the law in furtherance of improved public policy, while possibly pocketing a little profit along the way? When is it not just okay, but admirable?

Often, I’d say. Surely it is admirable to offer harbor, aid and comfort to Jews in Hitler’s Germany, escaped slaves in Roger Taney’s Maryland or “undocumented aliens” in Barack Obama’s America. These are easy calls. Why, then, do I instinctively balk at counterfeiting?

Partly, I think, because Eli didn’t tell me what he means by an “insufficiently expansionary” monetary policy. I understand what’s wrong with putting people in concentration camps, or enslaving them, or preventing them from crossing borders in pursuit of a better life. But I’m not sure I understand what’s wrong with a 3% monetary growth rate. The answer must depend on all sorts of auxiliary assumptions about the stickiness of prices, wages, and so forth, and I suspect the welfare consequences of counterfeiting might depend on those assumptions. And I can’t imagine ever being confident that I had those assumptions right.

That might explain my gut reaction, but of course it doesn’t answer the question, which I think I will continue to mull over. Meanwhile, there’s an interesting asymmetry here. Suppose you think it’s morally imperative to correct the Fed’s monetary policy by counterfeiting when monetary policy is too tight and (symmetrically) burning currency when monetary policy is too loose. Then you’ll get richer doing the right thing in tight-money times but you’ll get poorer doing the right thing in loose-money times. If you trade off personal gain against moral imperatives, you might consistently correct the Fed in one direction only. Is that better or worse than never correcting them at all?

 Print

Share/Save

In a post last week, I asked:

How can it be okay to remain childless but not okay to have children and treat them badly—given that the children themselves would presumably prefer being treated badly to not being born at all?

There are a lot of comments there worth reading, including one from Ryan Yin that rephrased the problem so clearly that I want to highlight it here.

First, though, some answers that I think don’t work. Cos asked:

Why is it so puzzling that we value the preferences of people who exist over the hypothetical preferences of people who might’ve existed but don’t?

Sierra Black drew the same distinction between people who exist and those who don’t, though she phrased it in terms of rights:

Children aren’t possessions or art projects, they’re people. Before you have a child, you’re a person with a certain right to control what happens to your body. After the child is born, that child has some of the same rights you do, including the right not to be mistreated.

Ryan Yin (after agreeing with Sierra!) put his finger on exactly why I find her answer (and others like it) so unsatifying— though I’m not sure he intended his observation to be used in exactly this way:

Suppose you were thinking of having a child. Suppose that once the child is born, the child would wish he’d never been born (and prefer that to killing himself), and further that this is entirely predictable. By the logic given above, you aren’t allowed to care about this fact until after the child is born (at which point there’s nothing to be done). This seems like a very odd way of valuing the preferences of others.

Exactly! If (as Cos says) it’s perfectly reasonable not to value a person’s preferences until after s/he’s born, or (as Sierra says) people don’t acquire rights until after they’re born, then there can be no objection to creating a miserable life. I might be dooming my child to a life of misery, but that’s okay because s/he hasn’t been born yet and therefore has no rights and no preferences I choose to care about.

Surely this can’t be right. And if it’s not, then the Cos/Sierra story can’t be exactly right.

 Print

Share/Save

I’m not sure how old I was at the time, but I must have been very young because I already knew the refrain by heart when my father first told me about foreign languages. In other countries, people use different words than we do. We say “cat”, but in Spain they say “gato” and in Russia they say “koschka”.

Well, I could easily see where this was going. Before my father could finish his sentence, I jumped in to announce that yes, those people might use other words, but we’re right and they’re wrong—right, Daddy?. I’m not sure whether he recognized his own refrain, but he looked quite taken aback as he gently explained that, well, no, there is no right and wrong, and one word is as good as another as long as the people around you understand what you’re saying.

In that moment I decided that the world made absolutely no sense and there was no point in even trying to understand it. Yesterday we were right and everyone else was wrong. Today everyone’s equally right. Was there no pattern here?

Somehow I outgrew my disillusionment. Today I dare to hope the world does make sense, or at the very least that it’s worth trying to find some sense in the world. I believe that racism is evil, that foreign languages are benign, and that, the mental limitations of four-year-olds notwithstanding, it’s not too hard to find a moral framework that can reconcile that paradox. Other paradoxes seem much harder. Here’s one that I keep coming back to: How can it be okay to remain childless but not okay to have children and treat them badly—given that the children themselves would presumably prefer being treated badly to not being born at all? (I am assuming here, for the sake of argument, that if you couldn’t treat your children badly, you’d choose not to have them. Obviously this doesn’t apply to everyone, but it does apply to some people. Those are the people I’m calling “you” in this question.)

I don’t know the answer to that question. I do persist in thinking about it. Feel free to address it in comments (or to share your own favorite moral paradox), though it’s not the focus of this post. Mostly I just felt like telling you this story.

 Print

Share/Save

Over the next dozen years, Grothendieck and his disciples—first class mathematicians in their own right—turned the vision to reality, overcoming vast technical obstacles that might have led almost anyone else to change course—but Grothendieck was a force of nature. The completed vision was published in a series of twelve volumes totaling over 6200 pages, the Seminaire de Geometrie Algebrique, known to mathematicians simply as SGA. These in turn rested on a series of 8 foundational volumes, also sprung mostly full blown from the mind of Grothendieck, the Elements de Geometrie Algebrique, or EGA. Every one of these many thousands of pages is dense with content; I happen to have idly picked up a volume a few days ago, read two random paragraphs, and spent two days digesting them.

There is a very real sense in which all subsequent work in algebraic geometry—and much subsequent work in several related fields—consists of commentary on EGA and SGA. These are the indispensable texts. And they’re hard to get ahold of. SGA was published by the German publisher Springer-Verlag, but it has long been out of print—and as the population of algebraic geometers has mushroomed (due largely to the fertility of Grothendieck’s ideas) there haven’t been nearly enough copies to go around. There are scans available on the web, but they are hard to read (they were produced on an electric typewriter, with the mathematical symbols handwritten in), not searchable, and littered with typos.

Enter Bas Edixhoven, a Dutch mathematics professor who conceived the rather marvelous idea of organizing teams of volunteers to re-type all of SGA using modern typesetting software, correcting typos and inserting references to more recent literature where appropriate. Edixhoven secured the blessings of all of Grothendieck’s collaborators, but was unable to contact the great man himself, who has spent most of the past two decades in complete seclusion in an unknown location. (More on this below.) Confident that Grothendieck would not oppose the project, Edixhoven launched the project. After eight years, the first two of the twelve volumes of SGA—SGA1 and SGA2—are now complete and available on line.

Last month, the project crashed to a halt when a letter surfaced from the 82 year old Grothendieck, opposing any further dissemination in any form of any of his work. He asks that all such work be “removed from commerce” (which would presumably include the World Wide Web) and from all libraries. He does not threaten legal action (though as far as anyone can tell, he does seem to be the copyright holder), but he appeals to his readers’ “sense of shame”.

The Edixhoven project is now halted, with the hard work of many volunteers presumably set aside.

Why did Grothendieck send this letter? One can only guess. But in the many years since Grothendieck retired from mathematics and withdrew from society, he has periodically (though not for several years now) released long, eccentric and beautiful manuscripts containing thousands of pages of mathematics, autobigraphy, philosophy and theology, and these give some clues. The Grothendieck Revolution enabled a generation of mathematicians—including Grothendieck’s students—to solve a vast array of formerly unapproachable problems, and to formulate and solve new problems that would have been inconceiveable prior to SGA. In his long autobiography Recoltes et Semailles, Grothendieck expresses bitter disappointment about the direction of much of this research. The thrust of his objection—and I am a little hesitant to summarize the subtleties of his position in a single phrase, but here goes—is that everyone’s picking the fruit and nobody’s tending the garden. Nobody, that is, is doing the sort of deep foundational work that marked Grothendieck’s own career. That ongoing disappointment, presumably, has something to do with the letter that arrived last month.

Well, now what? Legally, it appears that Grothendieck does own the copyrights. Ethically, should those copyrights be respected? (No, I do not plan to consult Randy Cohen on this matter.) Should an embittered recluse be allowed to impede the progress of a vast scientific enterprise that he almost singlehandedly created? Does it matter that he waited eight years to object to this project? What are the issues here? Do weigh in.

Edited to add: Several commenters have asked whether Springer still has an interest in the copyright. My understanding is that the copyright has reverted from Springer to Grothendieck under the applicable European laws.

 Print

Share/Save

More striking even than Cohen’s characteristic “ethics by pronouncement”, refusing to acknowledge, let alone address, the underlying issues, is that he doesn’t even seem to notice that these questions have something in common. He treats them as two separate reader inquiries, from two separate and non-overlapping universes. Thus it’s okay for the doctor to turn away a patient because “You cannot be forced to practice medicine” and because the patient can always find another doctor. One might wonder, then, in the case of the employer, why it’s not true/relevant/dispositive that “You cannot be forced to provide employment” and/or that the candidate can always find another job.

That’s not to say that the situations are identical. In the cases before Judge Cohen, the employer has partners who don’t share his politics; maybe there’s a relevant obligation to those partners. That thought, however, seems not to have popped into Cohen’s head during the five minutes he devoted to thinking about this column.

As I said the last time I blogged about this bozo, it’s not his conclusions I’m objecting to. It’s his apparent belief that “No you may not” is a substitute for logical analysis based on clearly stated principles that are at least stable enough to be maintained for the length of a newspaper column.

It is possible to do this stuff right. I claim to have done so in, for example, Chapter 18 of The Big Questions. More to the point, Tim Harford does it on a regular basis in his “Dear Economist” column, as my reader Jon Shea noted in comments here on this blog last week:

Contrast Cohen’s “The Ethicist” with Tom Harford’s “Dear Economist” in the Financial Times. Harford frequently sources both classic and new research papers. He also uses named economic theories to help explain his answer. As a result “Dear Economist” doesn’t feel like just an Anne Landers style advice column. When I read Harford I don’t feel like I’m getting a stranger’s opinion, but instead I feel like a trained economist is applying his skills and knowledge in a way I might not think to.

Hear, hear.

 Print

Share/Save

You’re redesigning a website and you want to include a photo of a generic customer. The client does not want the generic customer to be African-American, partly because he has never had an African-American customer and thinks it unlikely that he ever will. Is this okay?

My objection is not to Cohen’s answer (which is “no”) but to the way it’s dispensed, as if from an oracle, with no attempt at a derivation from clearly stated principles.

Here’s the best he has to offer:

Race may be a factor in selecting this photograph only if race is germane to the product or service the franchise provides. For instance, if the company sold hair-care products used almost exclusively by African-Americans, then you could rightly indicate as much through the photo you post on the Web site.

Well, okay. But why? Cohen doesn’t tell us.

Nor does he test his policy against the hard cases. Race, he says, may be a factor if it’s germane to the product or service. What if this is a product or service that African-Americans rarely purchase? Does that make race germane? Does it matter why they never purchase it? What if all—or most—African-Americans had a genetic aversion to this product? Is a race-correlated aversion morally equivalent to a race-correlated hair type? What if all—or most—African-Americans had a culturally induced aversion to this product? Is that morally equivalent to a genetic aversion? Why or why not?

I don’t pretend to know the answers to these questions, but that’s partly why I don’t call myself “The Ethicist”.

 Print

Share/Save

Bryan’s example is about discrimination, a subject that has come up before on this blog, but I want to emphasize that the argument Bryan quotes is quite separate from the arguments we got into in that earlier thread, and, for the sake of clarity, I hope we manage to keep them separate.

Bryan (paraphrasing me!) starts with the rather strong intuition that it’s okay for tenants and workers to discriminate. If you don’t want to live in an Albanian-owned building or an work for an Albanian employer, that’s your right (no matter how strongly we might strongly disapprove of your attitude). By analogy, then, it might seem that landlords and employers should have the same right to discriminate.

Now clearly the situtation is not that simple; landlords and employers are not the same as tenants and employees. But the question is: Are they not the same in any way that is morally relevant? The most frequently cited difference (in my experience) is that landlords and employers tend to have more market power than tenants and workers. Putting aside the question of whether that’s true, it can’t possibly be a full justification for treating landords and employers differently, and here’s why: There are plenty of instances where we don’t think that market power takes away your right to discriminate. Extremely attractive people have a lot of power in the dating market, but I think it’s safe to say that almost nobody thinks the most beautiful among us should be forced to date Albanians, or to prove that they choose their partners according to some objective criterion other than national origin.

So if you think it’s okay for tenants to discriminate but not landlords, you’ve got to face the question: What is the ethically relevant distinction here? It’s clearly not market power, so what, if anything, is it?

I do not deny that there might be a good answer to that question, but I must admit I can’t imagine what it would be.

 Print

Share/Save

A similar issue comes up in Chapter 20 of The Big Questions, where Mary the landlord won’t rent to, say, Albanians. Ought we force her to?

In The Big Questions, I make two separate (but closely related) arguments on Mary’s behalf. I was about to write a blog post offering the same arguments on behalf of Sony when I realized that only one of them applies. So I am forced to conclude that I should be a little less sympathetic to Sony than I am to Mary.

My first argument is that Mary never had any moral obligation to rent to anyone in the first place—and if she has no general obligation to rent to anyone, then she can have no specific obligation to rent to Albanians. Likewise, Sony has no moral obligation to provide anyone with video games—and if there is no moral obligation to provide me with a video game then there is no obligation to provide one to Alexander Stern. Fine so far.

But my second argument is that Mary, appearances to the contrary, is actually doing some good for Albanian apartment seekers. By renting rooms to non-Albanians, she takes a little pressure off the housing market, driving down rents and making it easier for Albanians to find apartments elsewhere. Sure, she could be doing even more for them, but she’s already doing more for them than I am, since I don’t rent apartments to anyone at all. How can she be at fault for doing small amounts of good when I’m given a free pass to do no good at all?

Now this second argument is actually a little slippery. When I say Mary is doing the Albanians a small amount of good, you’re entitled to respond “Compared to what?”. Compared to taking her building off the market altogether, she’s surely doing them some good. But what about compared to selling her building to a non-bigot?

Answer: Compared to selling to the non-bigot, Mary is doing no particular good, but she’s doing no particular harm either (except, perhaps, to Albanians with idiosyncratic reasons for preferring that building to all others). No matter who owns the building, it’s going to take, say, ten renters off the housing market and have the same effect on rents elsewhere. So depending on the comparison you want to make, Mary might or might not be doing small amounts of good for Albanians, but at least she’s doing no harm.

Now let’s try applying that second argument to Sony: Alexander Stern is having trouble playing Sony’s video games. But without Sony those video games wouldn’t be there in the first place. According to the argument, those games force down the price of other games, including the ones that Stern can play.

But in this case, the argument is probably wrong. Here’s why: Mary, by running a single apartment building, can’t drive all the Albanian-friendly landlords out of the marketplace. All she can do is drive rents down, which is good for all renters including Albanians. But Sony, by dominating a segment of the video game market, might well drive out some of its blind-friendly competitors—and that’s bad, not good, for Alexander Stern.

The fundamental difference is that an apartment is what economists call a private good—it can be occupied by only a small number of people at a time. A video game, by contrast, has at least some of the characteristics of a public good—once you’ve developed it, there’s no limit to how many can play. This gives Sony the ability to cause damage that Mary can’t.

So although Mary is a contemptible bigot, whereas Sony seems to be making reasonable economic decisions, I think I am forced to conclude that the case against Sony is stronger than the case against Mary. Not strong enough for me to support it, though.

 Print

Share/Save