Frischmann on Carrier's Innovation in the 21st Century

Geoffrey Manne —  31 March 2009

This post is from Brett Frischmann (Loyola/ Cornell (Visiting))

I enjoyed reading Mike’s book very much. It provides an excellent primer on antitrust, IP, and innovation.  He synthesizes the legal and economic foundations, contours, and controversies in an accessible fashion. I applaud him for doing this because frankly, it is tough to do given that the fields are quite technical and specialized.  The book really is appropriate for a general audience.  That said, I agree with some of the previous commentators that at times Mike oversimplifies some of the very heated debates he summarizes; given the breadth and complexity of issues, I cannot imagine how he wouldn’t do so.  Still, I think readers should recognize that the debates Mike wades into are incredibly contentious and considerably nuanced.  Nonetheless, the primer is excellent, and the rest of the book is quite provocative.  Ambitiously, Mike makes 10 specific proposals to improve the copyright, patent, and antitrust laws.  I’ll focus my comments on his discussion of dual use technologies and the Sony rule.

In his first chapter focused on proposals for copyright law, Mike discusses dual use technologies (e.g., telephone, cameras, radio, photocopier, VCR, computer, Internet, P2P file-sharing software, etc.).  He explains that dual use technologies are often a form of disruptive innovation that creates new markets, opportunities and even more innovation along with new risks to copyright owners’ rights to control reproduction and distribution.  Copyright law has struggled with technological change in general and dual use technologies in particular throughout its history.  Mike explains how secondary liability theories of contributory liability and vicarious liability can be employed by copyright owners to hold dual use technology vendors accountable for copyright infringement of technology users, and he also explains how the US Supreme Court’s decision in Sony Corporation of America v. Universal City Studios (“Sony”) erected a doctrinal shield to protect technology companies from contributory liability claims where their technologies are “widely used for legitimate, unobjectionable purposes, or even if merely capable of substantial noninfringing uses.”  He then talks about P2P file sharing technologies, P2P litigation, and different interpretations of and challenges to the Sony rule.

The basic issue:  What should be the secondary liability regime for dual-use technologies such as P2P file sharing?   Mike’s proposal is essentially to preserve the Sony shield in its broad form.  He prefers the more protective version—where the defendant is off the hook if the technology is merely capable of substantial noninfringing uses—to the less protective version— where the defendant is off the hook if the technology is widely used for legitimate, unobjectionable purposes.  He offers a few reasons:  most important, the broad, bright line rule is easier to apply by courts and practitioners than other rules that take into account the primacy of certain uses, subjective intent of technology providers, and potential technological design options (e.g., whether a technology provider employed adequate technological precautions to limit infringement).

Generally, I am sympathetic to his approach (I wrote a brief essay on the topic here.)  My concern with his analysis is that he did not engage arguments made by Doug Lichtman and others concerning the potential benefits of crafting a rule that forces technology providers to implement cheap, easy technological fixes to deter or disable infringement or perhaps better enable copyright owners to detect infringers.  Mike touches on the arguments lightly in his discussion of Judge Posner’s decision in Aimster, but I would have liked to see a bit more.  The argument for a more nuanced rule that places some responsibility on technology providers is stronger in the context of dual use technologies that enable widespread copying and distribution—e.g., P2P file sharing technologies; the threat to copyright owners is arguably much greater and technological precautions implemented by technology providers may (or may not) be relatively cheap.

I enjoyed Mike’s discussion of asymmetries—innovation asymmetry, error-costs asymmetry, and litigation asymmetry.  He claims that innovation asymmetry occurs in dual use cases because courts tend to “systematically overemphasize the infringing uses and underappreciate the noninfringing uses.”  (p.128)  The reasons for this asymmetry are that the former are more readily observed and quantified while the latter are “less tangible, less obvious at the onset of a technology, and not advanced by an army of motivated advocates.”  (p.129)  The noninfringing uses are difficult to quantify and value.  As Mike puts it, “how do we put a dollar figure on the benefits of enhanced communication and interaction?”  Moreover, the noninfringing uses tend to develop over time in ways that are difficult to predict upfront. 

I agree with Mike’s observations about the innovation asymmetry and think that he is correct to emphasize how it leads to a systematic bias in how courts (and commentators) evaluate technologies and develop the rules to regulate technologies.  Of course, the asymmetry is not unique to the creativity-innovation dichotomy; it also exists when courts analyze different uses of a work protected by copyright.  That is, the quantification and valuation problems are more general that dual use technologies.  (I must note that I am still working my way through part of the book, and frankly, I hope to see this argument made elsewhere because I think the innovation asymmetry he highlights is pervasive.)  In fact, as I have argued elsewhere; see also here and here, this type of asymmetry provides a relatively strong argument in favor of the broad version of the Sony rule for some types of general purpose—multi-use—technologies.

As for error-costs asymmetry, his discussion is very brief, and I have my doubts about the utility of the error cost framework, because it does not seem to account for accuracy benefits very well (i.e., the framework focuses on the costs of false positive and false negatives but does not deal directly with the benefits of positive positives and positive negatives – some preliminary thoughts on this are here). Moreover, I think Mike is too quick to say that in the case of a type II error (where a court mistakenly upholds a technology), “Congress can always step in to compensate copyright holders”— (really, Congress can? It is that easy?), and in the case of a type I error (where a court mistakenly holds a technology provider liable), a technology will be abandoned and “consumers will never know what they are missing”—is this always the case? The claims seem a bit strong.

Mike’s discussion of litigation asymmetry is also important. He notes how the high litigation costs alone can stifle technological innovation and create substantial holdup problems. He gives a number of high profile examples to make his point.  Mike claims that the litigation asymmetry, along with the other asymmetries, “exert[s] a strong, though often hidden, pull in the evaluation of infringing and noninfringing uses.”  This is one of those claims that are quite difficult to prove empirically, but nonetheless, at least in my view, ring true.  In the end, Mike recommends a return to Sony, at least for P2P file sharing technologies.  While one might say it is too late to return to Sony, given Napster, Aimster, and especially Grokster, I think it is fair to say that Sony is alive and well, perhaps even in its broad form. 

In closing, I will note that Mike’s discussion of statutory damages in chapter 7 is probably the least controversial in the book.  Maybe I am wrong, but I suspect that most people would agree with Mike that “applying statutory damages to secondary infringers has startling, unjustifiable consequences [described well by Mike], which are not needed to carry out Congress’s purposes and which pose great peril for innovation.”  (p. 160)

Geoffrey Manne