• Freedom
  • Innovation
  • Growth

About that "graduated response doesn't work" paper

So this morning I see news of a release of a paper from a law professor in Australia who finds that graduated response enforcement, such as “three strikes” policies, doesn’t work.

And that’s as far as almost anyone goes these days in our information overload society. You see a subject line in an email that says graduated response doesn’t work according to a new study, and you tuck that little detail away in your mind. “I remember reading somewhere that graduated response doesn’t work” the little voice in your head will say 18 months from now when the topic comes up in a discussion somewhere.

And then Mike Masnick over at TechDirt and Tim Lee over at the Washington Post will gleefully blog that, according to this new study, graduated response doesn’t work. The echo chamber does it’s work. And that’s that.

UPDATE: It's already happened. See here and here and here and here.

So I decided to actually go to the trouble of reading the study. And not only that, but also reading a few other things the author has written, look into some of the author’s affiliations and assumptions, etc.  It takes time, of course. But I do these things as a service to you.

I find that the author, Rebecca Giblin, likes to write about copyright. In fact, she likes to write about copyright and about networks, which is a common intersection of interests these days. For some reason a lot of activists and professors have decided to focus on the two primary means of production in the Internet economy, the content generators and the network operators. It seems to really bother them that private interests (capitalists) are permitted to control the means of production for the Internet economy, which is a topic that intrigues me and will be the subject of a future blog entry. Of that you can be assured . . .

I find further that Professor Gilblin submitted comments arguing for vastly expanded copyright exceptions and limitations. She also wrote a paper arguing that New Zealand’s proposed graduated response system was a bad idea and should not be implemented. I don’t suppose THAT reveals any particular bias on Professor Giblin’s part on the topic, does it?

[Note: I want to drive that point home. A few months before writing a paper supposedly evaluating graduated response efforts, she wrote a paper opposing their implementation.]

So the first thing Professor Giblin does in her paper is thank Pamela Samuelson at Berkeley for being her mentor and for generously supporting her work (p. 1). Hmmmmm. If you don’t already know, Pamela Samuelson is not exactly neutral and open-minded when it comes to either copyright or privately controlled networks. So that’s a clue.

She thanks Gwen Hinze of the Electronic Frontier Foundation, a name that pops out at least for me, from having seen Gwen at work any number of times at the World Intellectual Property Organization (WIPO) criticizing copyright and intellectual property in general. Gwen is an anti-IP crusader, make no mistake about it.

Another thanks (twice listed for some reason) is issued to Heesob Nam, who is with InfoJustice, an aggressive opponent of intellectual property protection.

You get the idea. These are not researchers—they are anti-copyright idealogues. One gets the idea that if one took the time to research each of the names Professor Giblin generously thanks, one would find something similar.

Aw, for the heck of it, let’s do one more. Philippe Laurent is another name I recognize from my international work on IP. He’s a crusader for open source software. Now without going off on a tangent, there’s nothing wrong with open source software. But I can tell you that many of the open source crusaders like Philippe are not simply proponents of open source—they are critics of proprietary software, and openly push open source as morally superior to proprietary. It’s not a matter of open source being an alternative model for development and licensing—it’s open source as the morally superior alternative to the offensive and immoral ownership of content through IP protection.

So I think by now I have a feel for who Professor Giblin travels with. It’s not in itself an indictment of her paper, but it provides helpful clues.

The tone of the paper itself drips with contempt against the content industry. She refers disparagingly to the “war against online copyright infringement” (p. 1). Here’s a great one: “There has never been universal agreement about what copyright law’s aims are, or even about what they should be” (p. 3). Really? We don’t even know why we have copyright? And on precisely what policy topic have we ever had "universal agreement" in the entire history of civilization?

Then the paper has a nice little section expressing skepticism that reducing infringement is even a legitimate policy goal (p. 3-4). She implies that things like “more creation” or “higher quality of creative output” would be more useful policy goals, though she finally resigns herself to having to use reduction in infringement as one of her criteria for judging whether efforts to reduce infringement work. Wow. Thanks for throwing us THAT bone, Professor Giblin.

And we’re only on page 4. From there she artificially and illogically creates a framework which sets reducing infringement and promoting knowledge as mutually exclusive or at least as opposing goals, which of course they are not. She argues that those who take a natural rights view of IP of course logically conclude reducing infringement, while (convenient to her purpose) those who take a utilitarian view should logically conclude promoting knowledge rather than reducing infringement. Again, working from the assumption that the two goals are in conflict, which they of course are not.

How on earth can we rely on someone with such a philosophically skewed and illogical perspective on copyright to fairly evaluate anti-infringement policies?

Turns out we can’t, as the rest of the paper demonstrates. In France, for instance, she views the Hadopi regime as a failure based on the fact that the new Socialist government of Hollande made changes to it, though she grants that there is evidence that the Hadopi regime reduced P2P infringement (p. 9). [Later in the paper she’ll get around to disparaging that evidence.] In South Korea, it is apparently sufficient to rate their regime a failure because “the National Human Rights Commission of South Korea called for it to be repealed” (p. 18).

In the U.K., the good professor grants that their graduated response law hasn’t even come into operation yet and cannot be evaluated. But that doesn’t prevent her from dedicating portions of 6 pages to describing its onerous (in her view) impact (p. 19-25).

There follows an extended discussion of the new regime in the United States, though as the author acknowledges, it has only been in place for a few months and so there is very little information available for evaluating its impact.

Next, the author goes country-by-country and 1) grudgingly admits that, in countries where data is available, it tends to show reductions in infringement, and 2) discounts any data or analysis supplied by the content industries. There’s always a “but”:

  • Yes, the data by the university researchers or government agency show a reduction in infringement, BUT they might just be getting their illegal content elsewhere, or
  • Yes, the data show reduction in infringement, BUT the data is supplied by the content industry so it can’t be trusted, or
  • Yes, the data show a reduction in infringement and increased consumption of legitimate content, but that might be due to reasons other than graduated response, or
  • The system hasn’t been in place yet long enough to make a determination, BUT they must already know it’s not working by virtue of the fact that they won’t release the data.

Astonishingly, on page 46, she asserts that because the content industries have such vested interests and powerful PR capabilities, if the graduated response systems were working well, surely it would have been widely published. But throughout the paper, every time the content industry asserted any analysis or data, she discounts it. The anti-copyright bias here is blatant.

This is getting long, so I’ll simply point out that in the final section of the paper the author attempts to discredit studies and data that indicate increases of legitimate content access that have been a result of the graduated response regimes. There’s fairly clear evidence, for instance, that iTunes purchases have increased with the introduction of graduated response mechanisms. It’s our present author’s clear intention, however, to time and time again raise doubts and questions about that evidence. She has a point to make rather than an evaluation to complete.

Perhaps the piece de resistance is this spectacular example on page 51:

It is certainly the case that music revenues have increased and music and film piracy have decreased within Korea. However, there’s no evidence that either outcome has been caused by Korea’s graduated response law. There are other possible explanations, including the impact of other anti-piracy policies. . . The increased availability of legitimate options might also explain the piracy decrease.

This passes for an academic evaluation?

Finally, can you guess her conclusion? After all of this dissembling and biased treatment, downplaying arguments and evidence that undercut her goal and always suggesting amorphous, unknown “other possible explanations,” Professor Gilbin concludes that there is a fairly overwhelming case that graduated response policies are a disastrous failure and should be repealed.

Professor Gilbin is an anti-copyright hack. She travels with anti-IP activists, she ignores or disparages data that disagree with her prevailing bias, and she draws predetermined conclusions.  But that won’t stop her from becoming a player in the anti-copyright echo chamber.

Look, take my word for it--I'm in this business. "A new study" means nothing unless the arguments in it are sound. You have to read the paper, not just listen to the echo chamber.

blog comments powered by Disqus