7. See Neil Weinstock Netanel, “Locating Copyright Within the First Amendment Skein,” Stanford Law Review 54 (2001): 15 (citing Houghton Mifflin Co. v. Noram Publ’g Co., 28 F. Supp. 676 (S.D.N.Y. 1939); Houghton Mifflin Co. v. Stackpole Sons, Inc., 104 F.2d 306 (2nd Cir. 1939) (upholding the validity of the U.S. copyright in Mein Kampf ); Anthony O. Miller, “Court Halted Dime Edition of ‘Mein Kampf’: Cranston Tells How Hitler Sued Him and Won,” Los Angeles Times, February 14, 1988, § 1, 4 (giving Cranston’s version of the case’s underlying facts)).
8. The Corley court was uncertain about this point. (“Preliminarily, we note that the Supreme Court has never held that fair use is constitutionally required, although some isolated statements in its opinions might arguably be enlisted for such a requirement.”). Universal City Studios v. Corley, 273 F.3d 429, 458 (2d Cir. 2001). In my view, both logic and those “isolated statements” suggest that fair use is required. As I point out later, when the Supreme Court revisited the matter in the case ofEldred v. Ashcroft, 537 U.S. 186 (2003), it stressed that it was precisely the internal limitations such as fair use that made copyright law normally immune to First Amendment scrutiny. The Court added “when . . . Congress has not altered the traditional contours of copyright protection, further First Amendment scrutiny is unnecessary.” Ibid. at 221 (citing Harper & Row, 471 U.S. at 560). Yet that is exactly what the DMCA does: alters “the traditional contours of copyright protection” by handing out the exclusive right at the same time as it confers a legal power to remove the privilege of fair use.
12. One empirical study seems to challenge this assumption, though at modest levels. Rafael Rob and Joel Waldfogel, “Piracy on the Silver Screen,” Journal of Industrial Economics 55 (2007): 379–395. Rob and Waldfogel surveyed college students—traditionally a population that engages in high levels of downloading since they have “free” and extremely high speed Internet connections, lots of leisure time, and low disposable income. Even among this group, the authors found that total levels of downloading were low—2.1 percent of paid consumption. The authors also assumed that all unpaid downloading or DVD burning was equal to piracy—an assumption that is clearly false. The Sony case makes that clear. In fact, Rob and Waldfogel found a positive relationship between second time unpaid viewings and future paid viewings; watching the movie a second time on a downloaded or privately made copy burned from the airwaves actually was associated with more paid purchases. The authors were skeptical of any causal link, however. Ibid., 389.
13. Admittedly, section 1201 only affects works protected under the copyright act, so arguably the legal protection of the digital fence would expire with the copyright term. But even if the courts interpreted the statute this way, two problems would remain. First, since the DMCA prohibited the trafficking in tools which allowed the breaking of the encryption, the law would have effectively forbidden the production of wire cutters for gaining access to identically encrypted public domain works—remember Judge Kaplan’s discussion of the irrelevance of Mr. Johansen’s motives. Second, it would be trivially easy to add a trivial amount of new copyrighted material to the work that had fallen into the public domain. Access to the public domain work would then be prohibited for another period of life plus seventy years. And so on. The Copyright Office holds hearings on the question of whether there are any “classes of work” that need exemption from the DMCA’s provisions. So far, those exemptions have been highly restrictive in application.