Let me suggest an analogy. Suppose I were to start claiming that I'd been awarded a Purple Heart. (The truth is that I have never even served in the military.) I wouldn't be taking anyone else's Purple Heart away. The award is felt to be valuable, but it doesn't have a clear price, and probably no one could prove that I had gained money by my false claim. Nonetheless most people would be likely to agree that by such a lie I would be harming soldiers who had rightfully earned the award. This example may seem too heavy, morally speaking, as I'm willing to admit, but I think it does prove at least this much: Whether it is right to take a thing from a person does not depend on whether it is abstract, and does not depend on whether the original owner is thereby deprived of his possession. A lighter example: sneaking into a half-empty movie theater through the exit. True, it's not a felony, but it is wrong. If they catch you, they do call your parents.
Since this is America, let me try once more to talk about the theft of art in terms of money. As Kant and Yglesias have observed, it won't quite do to define the property being stolen as merely a set of words (or bits). A good enough definition, perhaps: What you're stealing is the particular copy that might have been sold to you. Yglesias notes that it isn't correct to assume that every act of piracy "represents a lost sale." But it doesn't represent nothing, either. You pirated the work because you wanted it, so it did have some value to you.
Attacking from another angle, Yglesias suggests that illegal piracy reduces "deadwight loss," an economic term of art for the value lost to the market when prices aren't set merely by the intersection of supply and demand curves. If the government subsidizes corn, the resulting increase in corn-syrup-sweetened colas drunk is a deadweight loss. If the government grants Farrar Straus a monopoly on the sale of Elizabeth Bishop's poems, fans of hers who can't afford the monopoly price will have to forgo buying a collection—and the pleasure they won't have is another deadweight loss. This sounds like math, but if I've understood correctly (and I might not have, not being an economist), it's little more than a restatement of copyright's definition. As Yglesias himself concedes, "the whole point of copyright is that the owner of the rights ... has a monopoly on sales." Uh, yeah. That is the idea. To say you approve of copyright infringement because it reduces deadweight loss, therefore, is a little like saying you approve of tax evasion because it reduces the market-distorting appropriation of citizens' money by the government.
According to the Constitution, the purpose of copyright is "to promote the progress of Science and useful Arts." For that purpose I'm willing to not only eat Yglesias' lunch but cash his paycheck, too. Jeffrey Rosen of the New Republic recently suggested that "There is certainly a price below which authors and journalists won’t produce good work in the first place." Sure, answered Yglesias, and "the price is almost certainly negative." In other words, Yglesias thinks that writing, music, film, and television can prosper as hobbies—that in the copyright-free future, people might even pay for the privilege of writing and creating. Corporate bean counters of Slate! Did you read this? You have on your hands an economics experiment begging to happen. Mail me a fraction of Yglesias' paycheck, increasing the fraction from week to week. If he's wrong, we'll eventually cleave the heart of the mystery that is the "price below which authors and journalists won't produce good work." If he's right, though, he'll never blink. He might even start paying you.
For my part, I promise to spend whatever you send with the purest market efficiency.