Law in the Internet Society
Open for comments and edits.

-- DanielHarris - 05 Dec 2008

The recording industry never sues innocents, said the recording industry lawyer. Ah, but it sends takedowns to networked laser printers, said his fellow panelist.

The recording industry lawyer was, at that instant, arguing for the innocuousness of a strict-liability “making available” right under the 17 USC 106(3) exclusive right of distribution. You don’t need the processing power of a laser printer to poke holes in that argument: consider the case, not so far from one Eben brought up in class, of a computer-illiterate family member (younger child or great-grandmother perhaps) given an old laptop computer cast off by a recent college graduate.

The graduate might have left her carefully-assembled library of tens of thousands of songs on the laptop for her family member to enjoy. She might even have forgotten to turn off her peer-to-peer downloading software, set to automatically run in the background and share those sound recordings when the computer starts up. Even a proven so-called innocent infringement can leave our hapless victim liable for $200 per work, or $2 million for a not-extraordinary library of 10,000 songs.

The RIAA lawyer wants us to rely on what might be called prosecutorial discretion on the part of record labels. Although this discretion might prevent such a case from going before a jury, it has not yet stopped the sending of pre-litigation threats and settlement offers. As unlikely as the benevolence of label lawyers is, it presents what is likely the most complete solution under the current copyright system.

We cannot fix this troubling situation without completely overhauling copyright as a property right. Applying copyright, a strict liability property right against the world, to intangible property leads to this result: unlike tangible copyrighted works, one may reproduce, distribute, and perhaps even publicly display a bitstream without knowledge. Although a bookstore owner might unknowingly distribute, for instance, a lawfully-made copy of a tangible book, the doctrine of first sale in 17 USC 109 shields her from liability. Because sharing a bitstream in any way necessarily involves the making of a copy and a 106(1) violation, the interaction of 109 with 106(3) is not sufficient to protect an inadvertent digital infringer.

Like any good copyright visionary, I am unencumbered by practical considerations: treaty obligations, interest group influence, and so on. Although the correct solution might be to rewrite the copyright system to take account of the nature of bitstreams, the imposition of a culpability requirement or the redefinition of “copy” or “tangible medium” might be too hard to swallow. Instead, I propose a solution that would impose a knowledge requirement on bitstream copyright infringement using existing law: amend (or interpret) 106(3) to refer only to tangible copies.

Assume, for simplicity’s sake, a simple peer-to-peer system where one user (”D”) downloads each whole file from just one other user (”U”). It is clear that D is liable for infringement under 106(1) for reproducing the work. Even without 106(3), which current law would apply to U given evidence that the transfer actually occurred, U is liable for contributory infringement under 106(1). However, for the secondary liability to attach, U must contribute with knowledge of the infringing activity. Our innocent infringers are off the $200-per-work hook, but intentional peer-to-peer uploaders remain liable.

When the record labels say they want to prevent the “theft” or “piracy” of music, much of the public is sympathetic. Despite music’s long and distinguished history as culture, in our society music has been commoditized, can be owned, bought and sold. To someone ignorant of recording industry economics, copyright infringement of music can easily be seen as taking food from the mouths of the proverbial starving artists.

However, the main argument the recording industry would raise against my proposed reform is one which has so far failed to get much public sympathy. We need a “making available” right under 106(3), says the argument, because proving infringement under the law is too hard. This argument, however, applies just as easily to cases of tangible copyright infringement on an equivalent small scale. When my classmates in Hong Kong made photocopies of reserve course books, they were mechanically reproducing tangible works and creating new tangible copies. This seems just as impractical to police as does networked peer-to-peer sharing, and it might even be more difficult. Yet, nobody argues that leaving a book on a desk constitutes an infringing act of “making available.” Although some might scoff at the idea that someone would bother to photocopy each page of a book, the majority of my classmates did this--with the aid of student workers, paid by the university, at the campus copy center. I have seen graduate students at Columbia using one of the rare scanners in the Columbia libraries to do functionally the same.

This illustrates a sort of threshold, in the tangible world, between small-scale infringement for personal use and large-scale activities attracting attention of copyright holders and law enforcement. When one user burns a copy of Windows XP and gives it to a friend, Microsoft is clueless (and probably appreciates the market share). When a business outfits an entire office with pirated copies of Windows, or when factories in China press copies for sale on the street, Microsoft takes notice. The assertion of a “making available” right and failure of courts to uphold due process lets the sinking recording industry lash out at the small-scale infringers among its dwindling clientele in a way that was impossible before our world of intangible works. No longer being able to use the lower burdens of 106(3) as a strict-liability cudgel against small infringers would merely restore the natural balance in copyright enforcement.

I added a comment box into your page--I hope you don't mind.

I enjoyed reading your essay, and it gives a pretty simple straightforward description of the issues and why they're important. Regarding the knowledge requirement for 106(3)--wouldn't this in practice end the applicability of 106(3) to p2p filesharing, given that sharing is enabled by default and runs in the background? It seems like anyone could claim that they didn't know or intend to share. It's not like other knowledge requirements, where prosecutors can at least show that the defendants "reasonably should have known." Or is it?

Not that it would be a bad thing to effectively kill 106(3) for p2p, but I was just wondering what your thoughts were on actual practicality from the prosecutor's standpoint.


-- StevenHwang - 10 Dec 2008

I think that's exactly what he's saying, Steven.

-- KateVershov - 17 Dec 2008


Of course, I'd be delighted to effectively kill 106(3) for these small-time peer-to-peer cases. It would not definitively kill it, though--evidence of knowledge is still there and discoverable. Who installed the software (peoples' Kazaa usernames, as suits have shown, aren't always creative), who owned the computer when, when did the files get there, what is the user's sophistication, is the program "quiet?" Answering these questions requires effort and resources beyond the *AA's current shotgun approach to due process, but not prohibitive effort.

-- DanielHarris - 17 Dec 2008

Daniel, I'm not convinced that eliminating "making available" or even 106(3) entirely would have any effect on the RIAA's litigation strategy.

As you point out, knowledge is still discoverable. The RIAA could seek discovery from hosts or seeders on allegations of contributory infringement. Rule 11 only requires that factual allegations "will likely have evidentiary support after a reasonable opportunity for further investigation or discovery." While the RIAA has sent take-down notices to printers in the past, a fair proportion of the people they sue do probably share copyrighted content knowingly. That may not have been true 8 years ago, but fewer people today are truly clueless about the technical and legal details.

So long as we're modifying the copyright law anyway, why not simply lower the statutory damage minimum to $1 per work for non-commercial infringement?

-- AndreiVoinigescu - 17 Dec 2008

Maybe I'm missing something, but I thought that the "making available" charge was completely made up. Wouldn't just enforcing the law as it is be enough to curb the RIAA abuses? If you're litigating on a strategy that is bunk, and has been ruled to be so by judges, and doing it repeatedly, isn't that a Rule 11 violation? Also, on a case by case basis, all you need are a few complaints against printers (if the judge who receives the complaint sanctions the attorney on that particular case) and the quality of the representation of the RIAA will go way down. No one wants to be fined, censured and made unpopular amongst the printing and copying machinery community.

-- HamiltonFalk - 17 Dec 2008

Tony -- the pro-consumer view on making available might be sufficient:

Unfortunately, there is still substantial disagreement on the point and no appellate ruling. I'm not sure if this is a formal rule of law, but when you manage to fool a handful of district court judges you're probably out of Rule 11 territory.

Andrei -- as much as I'd like to lower it to $1, the content industry and its allies in legislatures would have a point in calling that an efficient breach too far. On the other hand, the new RIAA strategy (if it's to be believed) effectively gives up all statutory damages in favor of a privately administered pseudo-injunctive scheme.

-- DanielHarris - 19 Dec 2008



Webs Webs

r11 - 19 Dec 2008 - 14:24:48 - DanielHarris
This site is powered by the TWiki collaboration platform.
All material on this collaboration platform is the property of the contributing authors.
All material marked as authored by Eben Moglen is available under the license terms CC-BY-SA version 4.
Syndicate this site RSSATOM