Free Culture - Lawrence Lessig (hot novels to read .txt) 📗

your favorite songs with your best friend-- or your 20,000 best friends.

According to a number of estimates, a huge proportion of Americans have tasted file-sharing technology. A study by Ipsos-Insight in September 2002 estimated that 60 million Americans had downloaded music--28 percent of Americans older than 12.7 A survey by the NPD group quoted in The New York Times estimated that 43 million citizens used file-sharing networks to exchange content in May 2003.8 The vast majority of these are not kids. Whatever the actual figure, a massive quantity of content is being "taken" on these networks. The ease and inexpensiveness of file-sharing networks have inspired millions to enjoy music in a way that they hadn't before.

Some of this enjoying involves copyright infringement. Some of it does not. And even among the part that is technically copyright infringement, calculating the actual harm to copyright owners is more complicated than one might think. So consider--a bit more carefully than the polarized voices around this debate usually do--the kinds of sharing that file sharing enables, and the kinds of harm it entails.

File sharers share different kinds of content. We can divide these different kinds into four types.

A. There are some who use sharing networks as substitutes for purchasing content. Thus, when a new Madonna CD is released, rather than buying the CD, these users simply take it. We might quibble about whether everyone who takes it would actually have bought it if sharing didn't make it available for free. Most probably wouldn't have, but clearly there are some who would. The latter are the target of category A: users who download instead of purchasing.

B. There are some who use sharing networks to sample music before purchasing it. Thus, a friend sends another friend an MP3 of an artist he's not heard of. The other friend then buys CDs by that artist. This is a kind of targeted advertising, quite likely to succeed. If the friend recommending the album gains nothing from a bad recommendation, then one could expect that the recommendations will actually be quite good. The net effect of this sharing could increase the quantity of music purchased.

C. There are many who use sharing networks to get access to copyrighted content that is no longer sold or that they would not have purchased because the transaction costs off the Net are too high. This use of sharing networks is among the most rewarding for many. Songs that were part of your childhood but have long vanished from the marketplace magically appear again on the network. (One friend told me that when she discovered Napster, she spent a solid weekend "recalling" old songs. She was astonished at the range and mix of content that was available.) For content not sold, this is still technically a violation of copyright, though because the copyright owner is not selling the content anymore, the economic harm is zero--the same harm that occurs when I sell my collection of 1960s 45-rpm records to a local collector.

D. Finally, there are many who use sharing networks to get access to content that is not copyrighted or that the copyright owner wants to give away.

How do these different types of sharing balance out? Let's start with some simple but important points. From the perspective of the law, only type D sharing is clearly legal. From the perspective of economics, only type A sharing is clearly harmful.9 Type B sharing is illegal but plainly beneficial. Type C sharing is illegal, yet good for society (since more exposure to music is good) and harmless to the artist (since the work is not otherwise available). So how sharing matters on balance is a hard question to answer--and certainly much more difficult than the current rhetoric around the issue suggests.

Whether on balance sharing is harmful depends importantly on how harmful type A sharing is. Just as Edison complained about Hollywood, composers complained about piano rolls, recording artists complained about radio, and broadcasters complained about cable TV, the music industry complains that type A sharing is a kind of "theft" that is "devastating" the industry.

While the numbers do suggest that sharing is harmful, how harmful is harder to reckon. It has long been the recording industry's practice to blame technology for any drop in sales. The history of cassette recording is a good example. As a study by Cap Gemini Ernst & Young put it, "Rather than exploiting this new, popular technology, the labels fought it."10 The labels claimed that every album taped was an album unsold, and when record sales fell by 11.4 percent in 1981, the industry claimed that its point was proved. Technology was the problem, and banning or regulating technology was the answer.

Yet soon thereafter, and before Congress was given an opportunity to enact regulation, MTV was launched, and the industry had a record turnaround. "In the end," Cap Gemini concludes, "the 'crisis' . . . was not the fault of the tapers--who did not [stop after MTV came into being]--but had to a large extent resulted from stagnation in musical innovation at the major labels."11

But just because the industry was wrong before does not mean it is wrong today. To evaluate the real threat that p2p sharing presents to the industry in particular, and society in general--or at least the society that inherits the tradition that gave us the film industry, the record industry, the radio industry, cable TV, and the VCR--the question is not simply whether type A sharing is harmful. The question is also how harmful type A sharing is, and how beneficial the other types of sharing are.

We start to answer this question by focusing on the net harm, from the standpoint of the industry as a whole, that sharing networks cause. The "net harm" to the industry as a whole is the amount by which type A sharing exceeds type B. If the record companies sold more records through sampling than they lost through substitution, then sharing networks would actually benefit music companies on balance. They would therefore have little static reason to resist them.

Could that be true? Could the industry as a whole be gaining because of file sharing? Odd as that might sound, the data about CD sales actually suggest it might be close.

In 2002, the RIAA reported that CD sales had fallen by 8.9 percent, from 882 million to 803 million units; revenues fell 6.7 percent. 12 This confirms a trend over the past few years. The RIAA blames Internet piracy for the trend, though there are many other causes that could account for this drop. SoundScan, for example, reports a more than 20 percent drop in the number of CDs released since 1999. That no doubt accounts for some of the decrease in sales. Rising prices could account for at least some of the loss. "From 1999 to 2001, the average price of a CD rose 7.2 percent, from $13.04 to $14.19."13 Competition from other forms of media could also account for some of the decline. As Jane Black of BusinessWeek notes, "The soundtrack to the film High Fidelity has a list price of $18.98. You could get the whole movie [on DVD] for $19.99."14

But let's assume the RIAA is right, and all of the decline in CD sales is because of Internet sharing. Here's the rub: In the same period that the RIAA estimates that 803 million CDs were sold, the RIAA estimates that 2.1 billion CDs were downloaded for free. Thus, although 2.6 times the total number of CDs sold were downloaded for free, sales revenue fell by just 6.7 percent.

There are too many different things happening at the same time to explain these numbers definitively, but one conclusion is unavoidable: The recording industry constantly asks, "What's the difference between downloading a song and stealing a CD?"--but their own numbers reveal the difference. If I steal a CD, then there is one less CD to sell. Every taking is a lost sale. But on the basis of the numbers the RIAA provides, it is absolutely clear that the same is not true of downloads. If every download were a lost sale--if every use of Kazaa "rob[bed] the author of [his] profit"--then the industry would have suffered a 100 percent drop in sales last year, not a 7 percent drop. If 2.6 times the number of CDs sold were downloaded for free, and yet sales revenue dropped by just 6.7 percent, then there is a huge difference between "downloading a song and stealing a CD."

These are the harms--alleged and perhaps exaggerated but, let's assume, real. What of the benefits? File sharing may impose costs on the recording industry. What value does it produce in addition to these costs?

One benefit is type C sharing--making available content that is technically still under copyright but is no longer commercially available. This is not a small category of content. There are millions of tracks that are no longer commercially available.15 And while it's conceivable that some of this content is not available because the artist producing the content doesn't want it to be made available, the vast majority of it is unavailable solely because the publisher or the distributor has decided it no longer makes economic sense to the company to make it available.

In real space--long before the Internet--the market had a simple response to this problem: used book and record stores. There are thousands of used book and used record stores in America today.16 These stores buy content from owners, then sell the content they buy. And under American copyright law, when they buy and sell this content, even if the content is still under copyright, the copyright owner doesn't get a dime. Used book and record stores are commercial entities; their owners make money from the content they sell; but as with cable companies before statutory licensing, they don't have to pay the copyright owner for the content they sell.

Type C sharing, then, is very much like used book stores or used record stores. It is different, of course, because the person making the content available isn't making money from making the content available. It is also different, of course, because in real space, when I sell a record, I don't have it anymore, while in cyberspace, when someone shares my 1949 recording of Bernstein's "Two Love Songs," I still have it. That difference would matter economically if the owner of the 1949 copyright were selling the record in competition to my sharing. But we're talking about the class of content that is not currently commercially available. The Internet is making it available, through cooperative sharing, without competing with the market.

It may well be, all things considered, that it would be better if the copyright owner got something from this trade. But just because it may well be better, it doesn't follow that it would be good to ban used book stores. Or put differently, if you think that type C sharing should be stopped, do you think that libraries and used book stores should be shut as well?

Finally, and perhaps most importantly, file-sharing networks enable type D sharing to occur--the sharing of content that copyright owners want to have shared or for which there is no continuing copyright. This sharing clearly benefits authors and society. Science fiction author Cory Doctorow, for example, released his first novel, Down and Out in the Magic Kingdom, both free on-line and in bookstores on the same day. His (and his publisher's) thinking was that the on-line distribution would be a great advertisement for the "real" book. People would read part on-line, and then decide whether they liked the book or not. If they liked it, they would be more likely to buy it. Doctorow's content is type D content. If sharing networks enable his work to