So Grokster is out, and really the Court did the only thing that it could logically have done. The Ninth Circuit's interpretation of Sony let a tech provider make a mockery of copyright in reality, while sheltering under merely possible or plausible noninfringing uses of that tech as a general matter. The Court makes it clear--we aren't putting P2P on trial here--just one specific application of it, under one set of circumstances. And, as I read the opinion, the tech might still be used widely for infringment, but still be sheltered from liability if there is no evidence that the provider intended to encourage such uses. So Grokster will need to ... get a new business model (where have I heard that before?).

But what about Justice Breyer's concurrence? It is eloquent, yet, I think, ultimately wrong. Here's why:

--Justice Breyer wants a bright-line test, but he does not grapple with the fact that a consequence of such an interpretation of Sony will be that increasingly sophisticated tech will shelter just outside that bright line, undermining copyright. He does not grapple with the fact that suits against individual users of P2P networks cannot be brought in sufficient numbers to make compliance with copyright a reality.

--He is concerned that an approach to liability that focusses on actual evidence of intent will be too complex and uncertain to shelter innovation adequately. But the burden of proof of coming up with the evidence will be on the plaintiff (content) side. And the circumstances in which there will be such evidence will be rather rare. The tech industry is unlikely to be staggered by such a marginal change in the law.

--He would ask the content industry to affirmatively prove that P2P undermines markets to the extent that creators are reducing production.

Why should creators *not* have to prove this, especiallly in a case of secondary liability? For one thing, the result would be manipulable--knowing that there would be such an inquiry would give creators a reason to reduce production, or to appear to do so. Even in the absence of such manipulation, the economic impact of unauthorized copying on actual production could vary widely over time (as broadband becomes widespread worldwide). Other factors (many) no doubt play a role in production levels as well, and it would be very difficult to disentangle these. A liability rule based on such a particular economic inquiry might well change every five years. The significant thing is the big picture potential for an impact on incentives in the long run.

Dangers of manipulation aside, Breyer's approach shortchanges consumers. Consumers should want creators and producers of content to have an incentive to pursue, not just big, safe revenue streams, but new and risky niches as well. To do this, the copyright regime needs to shore up the whole market, not just enough of it to keep creators and producers from going under. Each consumer should be a potential revenue stream. Every free rider should be treated as significant, not just crippling masses of them.

This doesn't mean we have to shoot them, or impose secondary liability for any and every direct violation. But we need not require specific proof that individual property rights encourage productivity as a general matter and that widespread violations are significant.