Copyright Trolls and the Common Law

I.     Introduction

Copyright trolls have been treated as a kind of IP boogeyman—an apocryphal malevolent occupying the realm of legal nightmares. But Matthew Sag’s important new work, Copyright Trolling, An Empirical Study,1 helps bring copyright trolls out of the dark. In a way that only sharp data can, Sag’s research serves as a wakeup call to anyone who thinks that litigation trolling remains a rare and inconsequential burden on the U.S. copyright regime and federal courts. It also strikes me as further support for resolving the copyright troll problem judicially rather than legislatively.

Sag’s research draws from his database of all federal district court copyright lawsuits filed between January 1, 2001 and March 31, 2014.2 Within that time period, Sag focused on a form of trolling that has come to dominate the federal copyright docket—the Multi-Defendant John Doe (“MDJD”) lawsuit—and discovered a seismic shift in the nature of copyright lawsuits. Though almost unheard of in 2001 and rare before 2010, 43% of copyright lawsuits filed in 2013 were against John Does; most of those were related to pornography.3 Moreover, MDJD lawsuits constituted the majority of copyright cases in “19 of the 92 federal district courts” and “in the Third, Fourth, Sixth, Seventh, Tenth, Eleventh, and D.C. Circuits in 2013.”4

MDJD lawsuits target John Does who shared a copyrighted work—overwhelmingly pornographic films—in a BitTorrent swarm. By filing a single action in federal court against hundreds, even thousands, of John Does, the copyright owner is able to compel Internet service providers to reveal the name and physical address associated with the Internet Protocol address involved in the BitTorrent swarm. The copyright owner then often sends these individuals a letter accusing them of copyright infringement and indicating the owner’s desire to pursue statutory damages of up to $150,000. The letter includes, or is followed with, an offer to settle the dispute for somewhere between $1,000 and $5,000, with a frequently used $4,000 figure “calculated to be just below the cost of a bare-bones defense.”5

Sag works nicely through the economic asymmetries for plaintiffs and defendants in MDJD litigation.6 It is enough to say that considering the costs of defending an infringement claim—not to mention the time, uncertainty, and potential embarrassment of being publicly named for downloading porn—defendants are wise to settle, guilt or innocence notwithstanding. As one district court judge observed, pornography owners who file MDJD lawsuits have “discovered the nexus of antiquated copyright laws, paralyzing social stigma, and unaffordable defense costs.”7

Moreover, unlike other copyright troll targets, who might be able to claim at least colorably de minimis or fair uses, Sag notes that John Does are unlikely to find haven in traditional copyright defenses.8 There is nothing de minimis or fair about the unauthorized downloading of a pornographic film. However, “[t]he possibility of hacking, open Wi-Fi networks, Internet accounts accessed by multiple users, and mistakes by ISPs open the door to numerous highly fact-specific ‘it wasn’t me’ defense[s].”9 But, as noted, the defendants’ incentives push toward settlement.

The MDJD plaintiff’s success hinges on the effective use of statutory damages and quirks of litigation practice. As Sag explains, “[t]he economic viability of MDJD litigation depends on suing as many defendants as possible in a single action to keep costs low and leveraging the threat of statutory damages in order to maximize the flow of settlement dollars.”10 Statutory damages are low-hanging fruit—a “pot of gold at the end of the litigation rainbow”11—and, more significantly, a cudgel for extracting settlements.12 Permissive joinder standards, on the other hand, help keep costs down and maximize profit margins. After elucidating the MDJD problem and laying the normative foundation for reform, Sag proposes reigning in the availability of statutory damages and tightening lax joinder standards.13

Indirectly, Sag also sheds light on the question of who should redress copyright trolling: Congress or courts? By placing copyright trolls within the tradition of opportunistic plaintiffs and helping quantify the MDJD form, Sag reminds us that copyright trolls have a transient nature. As I discuss below, amorphous trolling forms are best addressed through ad hoc determinations rather than per se classifications. This understanding urges a judicial approach over a legislative one.

II.      Defining the Copyright Troll

The scope of the copyright troll problem remains unclear because trolls do not fit neatly into any definition. Trolls take a variety of forms, and focusing on MDJDs, of course, is an imperfect way to approximate the copyright trolling problem.14 Sag acknowledges this with a Section titled “When Do Multi-Defendant John Doe Lawsuits Amount to Copyright Trolling?”15 The short answer: when Multi-Defendant John Doe lawsuits amount to copyright trolling.

More generally, Sag cautions that we identify trolls in practice rather than craft a categorical rule of trolling in the abstract.16 In practice, Sag argues the MDJD copyright troll does at least one of three things systematically: “asserts rights it does not have, makes poorly substantiated claims, or seeks disproportionate remedies.”17 Sag claims that these cluster features fit within traditional understandings of litigation trolling generally—the troll is an opportunistic plaintiff that seeks to exploit litigation for profit.18

But determining the proxies for identifying when a copyright owner is using litigation as an independent revenue stream proves too much. Attempting to use one definition ends up sweeping in many plaintiffs who systematically seek statutory damages that are disproportionate to the injury, much like trolls, even when turning litigation into an independent revenue stream is not among the plaintiff’s goals. Consider the record companies that sought statutory damages against peer-to-peer sharers of copyrighted music.19 Sony BMG and other music studio plaintiffs, for example, enforced copyrights against file-sharer Joel Tenenbaum.20 With a strong legal claim for infringement, Sony BMG elected for statutory damages, and the court awarded $675,000 for thirty infringing downloads—$22,500 per song.21 This award was clearly disproportionate to the actual injury Tenenbaum caused.22 There was never any indication that Sony BMG sought the disproportionate award as an alternative revenue stream. In fact, Tenenbaum claimed that the record companies “intend to teach the public a lesson” and thus were “trying to make an example of [him].”23 But, for Sag, is the lack of a litigation-revenue purpose enough to overcome systematically seeking a disproportionate reward? Does it spare Sony BMG from the copyright troll label? If so, it is unclear what work the presence of a systematic cluster feature actually does.24 But, if not, the defining characteristics seem overbroad.

On the other hand, at the same time that not all of MDJD cases represent copyright trolling, some instances of copyright trolling are unrepresented in Sag’s data because they “do[] not fit particularly well with the predominant form of copyright trolling in federal courts over the last few years.”25 Indeed, the spark for the copyright trolling explosion, Righthaven, is absent because Righthaven sued named individuals, and did so individually.26 But Righthaven’s business practices still qualified as copyright trolling under both Sag’s definition and the status-based definitions proffered by Shyamkrishna Balganesh27 and myself.28 To be sure, identifying non-MDJD trolls in practice is infeasible, as it would require individually reading the dockets of every copyright case filed—and that still would leave out trolling behavior that resolved with settlement before litigation commenced.

Yet, the imperfect proxy of MDJD data still provides important new insights into the emerging troll problem.Even if exact quantification is elusive, it is increasingly clear that copyright trolls are real. Before 2010, there were occasional cases of copyright trolling, but no systematic business model. Righthaven changed that by showing that a content business built on litigation revenue could be profitable. Its practice was to partner with newspapers—perhaps the most desperate mass owners of copyrighted content29—to identify partner news content and photos that had been reproduced in whole or part online, to then purchase rights to that work from the newspaper, and then file no-warning lawsuits seeking statutory damages. In sixteen months, Righthaven filed 276 lawsuits and reportedly recovered $352,500 in 141 settlements.30 Righthaven’s collapse in 2013, after numerous cases were dismissed because it had failed to properly acquire the rights it sought to enforce, was not an indictment of copyright trolling per se, but rather a roadmap of pitfalls that future trolls needed to avoid.

Next came the so-called copyright porn trolls, which represent the overwhelming majority of the opportunistic plaintiffs in Sag’s research. The most notable of these has been Malibu Media, which Sag discovered filed 31.79% of all copyright lawsuits between January 1, 2013 and June 30, 2014.31 Prenda Law, though, garnered much more negative attention early on. A law firm affiliated with numerous shell companies,32 Prenda was accused of seeding a pornographic film that it owned onto BitTorrent and then filing MDJD lawsuits against those who downloaded it.33 Mere months after a Prenda lawyer boasted that he had made a “few million dollars” suing downloaders,34 Prenda imploded. The firm and the lawyers who ran it were sanctioned for lying to the court and for forging the signature on a copyright assignment,35 and subsequently were hit with defendants’ attorneys fees in other cases that “smacked of bullying pretense.”36

But where hubris has not taken over, copyright trolling generally, and porn trolling specifically, has proven profitable. And it has adapted to remain so. Sag notes, for example, that as judges have become more skeptical about permitting joinder, some plaintiffs have eschewed the MDJD form “and are now suing individual IP addresses and demanding significantly higher settlement amounts.”37 Copyright trolls likely will continue to evolve, regardless of the type of content, so long as a business built around litigation remains profitable.

Two key questions follow. First, should we care? That is, are trolls actually bad for copyright law, for the legal system generally? And, second, if trolls are a problem, who should address them? Part III discusses copyright trolls’ societal costs, and Part IV argues for a judge-focused approach to abating trollish behavior.

III.     Trolls and Their Tolls

In short, copyright trolls are a problem. To be sure, trolls are not all bad—for example, they create economies of scale, increase compensation to authors, and outsource costly and time-consuming efforts to monitor for and enforce against infringement.38 But their imposed costs outweigh the benefits.

To begin, copyright trolls encourage undesirable litigation. This happens in two different ways. In MDJD litigation, it occurs because mass joinder and a template for extracting settlements via cookie-cutter demand letters lower the costs of litigation too far.39 In other cases, trolls disrupt the enforcement equilibrium by encouraging litigation where the author otherwise would not be motivated to sue.40 The Righthaven defendants, for example, included an unemployed woman who wrote a barely-read blog from the perspective of a cat and republished there a newspaper article about birds killed in a fire, a use that posed little to no market threat to the original author, the Las Vegas Review-Journal.41 Additionally, where trolls acquire copyrighted works from another author, they contradict Demsetzian notions of commercial value serving as a proxy for social value by compensating authors for the work’s litigation value.42 And, finally, overlapping more closely with concerns about litigation trolling generally, Sag argues that the asymmetries in MDJD litigation prejudice defendants, whether innocent or guilty, and pervert the judicial system into a clearinghouse for “a judicially sanctioned hunting license.”43

Because copyright trolling generally is about monetizing litigation, there is something particularly unseemly about doing so via a law designed to promote social progress.44 That concern is particularly acute considering the potential for copyright trolls to chill speech and deter innovation.45

IV.     Searching for a Solution: Congress or Courts?

Attempts to combat trolls through statutory amendment will be undercut by the difficulty of defining a troll. As copyright trolls have garnered increasing attention, observers have suggested numerous corrective actions. Proposals fit into two categories for targeting trolls: general and specific. The primary general reform is to limit the availability of statutory damages.46 This would remove from copyright law the troll’s primary attraction. But there are two major limitations. First, Congress has deemed statutory damages necessary because of the challenge of proving actual damages in copyright cases; this is especially true in file-sharing cases.47 Second, and more practically, Congress has shown no interest in reforming statutory damages and I am skeptical that courts will reverse course and, as Sag argues, “recognize that there are logical and constitutional limits to the level of damages that deterrence justifies.”48

On the other hand, specific reforms offer solutions to various iterations of copyright trolling, but likely would prove ineffective with other variations, known and unknown. For example, Dan Booth has proposed a “one satisfaction” rule that would enable defendants in an MDJD action, whom the plaintiff has sued jointly and severally, to use the rules of joinder against the plaintiff.49 If courts adopted Booth’s argument that where defendants are joined there could be only a single infringing action and thus, under the Copyright Act, only one statutory damage award per infringed work,50 the troll’s advantage in filing MDJD lawsuits would disappear. But the one satisfaction rule would be impotent against individually filed lawsuits.

The biggest shortcoming of both the general and the specific approaches, though, is that their costs and benefits are dependent on a clear understanding of attachment. Quite simply: Who, exactly, is a copyright troll? Answering this question has proven surprisingly elusive. Commentators have used numerous definitions for copyright trolls.51 In previous work, I attempted to define the troll as a copyright owner who:

(1) acquires a copyright—either through purchase or act of authorship—for the primary purpose of pursuing past, present, or future infringement actions; (2) compensates authors or creates works with an eye to the litigation value of a work, not the commercial value; (3) lacks a good faith licensing program; and (4) uses the prospect of statutory damages and litigation expenses to extract quick settlements of often weak claims.52

But neither judges nor scholars have agreed on the common markers of trolls. Often, the propriety of the label seems to amount to “I know it when I see it.”53

Sag, who eschews the status-based definitions,54 argues in his Article that “[i]t makes more sense to define trolling in practice than to attempt to identify trolls in the abstract.”55 Upon reflection, I think that he is correct. That is not to say that defining the troll’s essential characteristics is futile. But sharper definition is likely to be over-inclusive at times and under-inclusive at other times. Trolling is circumstantial—“troll is as troll does,” Mark Lemley noted in the patent context56—and, as Sag argues, “[w]e should, in short, identify instances of trolling rather than looking for trolls per se.”57

With that in mind, a prerequisite question must be answered before we can determine how to best address copyright trolls. That is, who should decide: Congress or courts?

What Sag describes, and what copyright troll observers have seen in just a few years, is that copyright trolls are, fundamentally, opportunistic plaintiffs. Whether we call them trolls or sharks or another colorful creature, opportunistic plaintiffs are not unique to copyright law; they are not even unique to IP. They appear across legal fields, wherever a seam in the law can be exploited for profit. And trolls’ thus-far-brief history in copyright law demonstrates that what is vogue today might be passé tomorrow, replaced by a new opportunistic business model.

Congress does not do well with such transience. To the extent that copyright law is statutorily revised to plug the leak from which trolls are filling their buckets, another is likely to spring. What is needed, then, is not a legislative solution, but rather a recognition that trolls exist and are bad for the copyright system, and that judges are best-suited to address copyright trolling as it adapts to new circumstances. The amorphous nature of trolls and the likely need for ad hoc determinations rather than per se classifications urge a judicial approach over a legislative one.

Though copyright law is increasingly a statutory regime,58 it has deep common law roots59 and several doctrines, including the fair use doctrine,60 remain robust and flexible for judicially addressing the copyright troll problem; courts also could further develop a copyright misuse doctrine for addressing such litigation opportunism. Additionally, statutory fee-shifting discretion, which is authorized in any copyright action, frivolous or not, provides judges with a powerful tool.61 Moreover, the Federal Rules of Civil Procedure further empower judges to curtail some trolling practices. Sag argues that judges should be more liberal in exercising discretion to reject joinder motions, and they should, while remaining mindful of judicial economy considerations.62 But judges also have the authority, at least in some instances of copyright trolling, to impose sanctions.63

Each of these measures has shortcomings, and none standing alone could quash copyright trolls. But together they could seriously help address the problem, and could do so without the costs of statutory reform that would be both under- and over-inclusive.

  1. [1]. Matthew Sag, Copyright Trolling, An Empirical Study, 100 Iowa L. Rev. 1105 (2015).

  2. [2]. The raw data is available at Mathew Sag, Data Sets, Matthew Sag: Copyright L., Fair Use & Tech., (last visited Apr. 3, 2015).

  3. [3]. See Sag, supra note 1, at 1127 (“Cat videos do not feature prominently in MDJD lawsuits. Pornography is another story.”).

  4. [4]. Id. at 1117.

  5. [5]. Ingenuity 13 LLC v. John Doe, No. 2:12-cv-8333-ODW (JCx), 2013 WL 1898633, at *1 (C.D. Cal. May 6, 2013).

  6. [6]. See Sag, supra note 1, at 1134–35.

  7. [7]. Ingenuity 13 LLC, 2013 WL 1898633, at *1.

  8. [8]. Sag, supra note 1, at 1113.

  9. [9]. Id. at 1121 (quoting Third Degree Films, Inc. v. Does 1–72, No. 12-cv-14106, 2013 WL 1164024, at *6 n.7 (E.D. Mich. Mar. 18, 2013)) (internal quotation marks omitted).

  10. [10]. Id. at 1109.

  11. [11]. Id. at 1121.

  12. [12]. See Brad A. Greenberg, Copyright Trolls and Presumptively Fair Uses, 85 U. Colo. L. Rev. 53, 109 (2014) (noting that “trolls use copyright law as a stick for extracting profits via settlements”).

  13. [13]. See Sag, supra note 1, at 1135–45.

  14. [14]. It is both over-inclusive, in that not all MDJDs are examples of copyright trolling, and under-inclusive in that many instances of copyright trolling occur outside of MDJD litigation.

  15. [15]. Sag, supra note 1, at 1114.

  16. [16]. Id. at 1113–14.

  17. [17]. Id. at 1114.

  18. [18]. Id. at 1113–14.

  19. [19]. Sag asserts that “[t]he [Recording Industry Association of America] and its members [we]re not copyright trolls because the industry’s end-user litigation strategy was to send a message, not create an independent revenue stream[,]” id. at 1114, but it is unclear how an observer would know to disregard one of the cluster features that indicate trolling.

  20. [20]. Sony BMG Music Entm’t v. Tenenbaum, 719 F.3d 67 (1st Cir. 2013).

  21. [21]. Id. at 68.

  22. [22]. Id.

  23. [23]. Seth Rolbein, Joel Tenenbaum’s Big Day in Court, BU Today (Aug. 6, 2009),’s-big-day-in-court.

  24. [24]. Similarly, can the stated desire of creating a litigation revenue stream overcome the absence of any cluster feature? Cf. Sag, supra note 1, at 1115 nn.40–43 and accompanying text (discussing Voltage Pictures’ lawsuits against downloaders of The Hurt Locker).

  25. [25]. Id. at 1112.

  26. [26]. Sag discusses Righthaven early in his Article, but notes that “it is just one instance[,]” id. at 1112, and argues that focusing on Righthaven undesirably constrains copyright trolling discussions to the their disruption of accepted unlicensed uses, id. at 1113.

  27. [27]. See Shyamkrishna Balganesh, The Uneasy Case Against Copyright Trolls, 86 S. Cal. L. Rev. 723, 732 (2013).

  28. [28]. See Greenberg, supra note 12, at 59.

  29. [29]. See Brad A. Greenberg, Tollbooths and Newsstands on the Information Superhighway, 20 Mich. Telecomm. & Tech. L. Rev. 171, 172–75 (2013); Brad A. Greenberg, Comment, A Public Press? Evaluating the Viability of Government Subsidies for the Newspaper Industry, 19 UCLA Ent. L. Rev. 189, 192–94 (2012).

  30. [30]. Greenberg, supra note 12, at 67 (citing Welcome to Righthaven Lawsuits, Righthaven Lawsuits, (last visited Apr. 3, 2015)).

  31. [31]. Sag, supra note 1, at 1147.

  32. [32]. See Ingenuity 13 LLC v. John Doe, No. 2:12-cv-8333-ODW (JCx), 2013 WL 1898633, at *1–2 (C.D. Cal. May 6, 2013).

  33. [33]. See Greenberg, supra note 12, at 61 n.35.

  34. [34]. Kashmir Hill, How Porn Copyright Lawyer John Steele Has Made A “Few Million Dollars” Pursuing (Sometimes Innocent) Porn Pirates”, Forbes (Oct. 15, 2012, 2:09 PM),

  35. [35]. Ingenuity 13 LLC, 2013 WL 1898633, at *3–4.

  36. [36]. Joe Mullin, Appeals Court: Shell Game Over, Prenda Law Must Pay Sanctions in Full, Ars Technica (Aug. 1, 2014, 2:16 PM),

  37. [37]. Sag, supra note 1, at 1142. Of the 817 copyright lawsuits that Malibu Media filed in the first six months of 2014, each was against a sole individual defendant. Id. at 1142, 1147.

  38. [38]. This quick summary gives rather short shrift. For further discussion, see Greenberg, supra note 12, at 71–72.

  39. [39]. See Sag, supra note 1, at 1135 (“We expect landholders and copyright owners alike to enforce their rights when their benefits outweigh their costs. Litigation invokes significant public resources and has potentially significant public costs. Attaching a positive cost to litigation through filing fees or other procedural mechanisms is an important screening mechanism that deters marginal complaints and trivial claims.”).

  40. [40]. Balganesh, supra note 27, at 757.

  41. [41]. See Steve Green, 8 More Websites Sued Over R-J Copyrights; 34 Total, Las Vegas Sun (June 5, 2010, 1:53 PM),; Eric E. Johnson, Purr-loined Story Gets Cat Blog Sued, Blog L. Blog (June 8, 2010, 9:18 AM),; see also Greenberg, supra note 12, at 66 (noting that two other defendants were “a United States Senate candidate whom the Review-Journal had in fact endorsed, and a news source who republished on his own website the Review-Journal article quoting him and the research he had provided free to the newspaper”) (internal citations omitted).

  42. [42]. See generally Harold Demsetz, Toward a Theory of Property Rights, 57 Am. Econ. Rev. 347 (1967).

  43. [43]. Sag, supra note 1, at 1135.

  44. [44]. Id. at 1145; see also Greenberg, supra note 12, at 57 n.15 (“Even more than the copyright infringers whom trolls enforce against, trolls disturb the copyright system by exploiting incentives without contributing to the creative works market—they do not produce, distribute, consume, or use the copyrighted work—and do so in a manner that discourages other creators.”).

  45. [45]. Greenberg, supra note 12, at 81–91.

  46. [46]. The Copyright Act permits statutory damages of $750 to $30,000 per work infringed, or up to $150,000 if the infringement was willful. 17 U.S.C. § 504(c) (2012).

  47. [47]. Reducing the statutory damages range from $250 to $3,000 per infringed work for online file-sharing, as Sag suggests, could discourage MDJD copyright trolling. Sag, supra note 1, at 1139. But it would create an additional burden for non-trolls and also would be impotent against forms of trolling unrelated to file-sharing.

  48. [48]. Id. at 1137. The definitive work in this area is Pamela Samuelson & Tara Wheatland, Statutory Damages in Copyright Law: A Remedy in Need of Reform, 51 Wm. & Mary L. Rev. 439 (2009). Yet, in the six years since, courts have repeatedly rejected such arguments. See, e.g., Sony BMG Music Entm’t v. Tenenbaum, 719 F.3d 67, 70–72 (1st Cir. 2013); Capitol Records, Inc. v. Thomas-Rasset, 692 F.3d 899, 907–10 (8th Cir. 2012); Virtual Studios, Inc. v. Beaulieu Grp., LLC, 987 F. Supp. 2d 769, 777–78 (E.D. Tenn. 2013). But see Capitol Records, Inc. v. MP3tunes, LLC, No. 07cv9931, 2014 WL 4851719, at *23 (S.D.N.Y. Sept. 29, 2014) (finding damages of $23,220 per work in a copyright infringement to violate due process).

  49. [49]. See generally Dan Booth, The “One Satisfaction” Rule: A New Approach to Curbing Copyright Trolls, Landslide (Jan.–Feb. 2015), at 22.

  50. [50]. Id. at 25–26.

  51. [51]. See, e.g., Minden Pictures, Inc. v. John Wiley & Sons, Inc., No. C-12-4601 EMC, 2014 WL 1724478, at *8 (N.D. Cal. Apr. 29, 2014) (defining a copyright troll as “an entity which exists solely for the purpose of acquiring rights to pursue litigation”); Malibu Media, LLC v. John Does 1, 6, 13, 14, 950 F. Supp. 2d 779, 780 (E.D. Pa. 2013) (defining a copyright troll as “a non-producer who merely has acquired the right to bring lawsuits against alleged infringers”); Jason R. LaFond, Personal Jurisdiction and Joinder in Mass Copyright Troll Litigation, 71 Md. L. Rev. 51, 51 (2012) (“A copyright troll, roughly defined, is a person or entity that acquires a (usually narrow) license from an original copyright holder for the sole purpose of suing and obtaining settlements from alleged infringers.”); Pamela Samuelson, Is Copyright Reform Possible?, 126 Harv. L. Rev. 740, 759 (2013) (book review) (defining copyright trolls as “rightsholders who threaten or bring infringement lawsuits in order to induce users to pay to settle weak claims”). James DeBriyn’s definition of a copyright troll as “a plaintiff who seeks damages for infringement upon a copyright it owns, not to be made whole, but rather as a primary or supplemental revenue stream,” Shedding Light on Copyright Trolls: An Analysis of Mass Copyright Litigation in the Age of Statutory Damages, 19 UCLA Ent. L. Rev. 79, 86 (2012), has been the most heavily adopted by courts. See, e.g., PHE, Inc. v. Does 1–122, No. 13-cv-786, 2014 WL 1856755, at *3 (N.D. Ill. May 7, 2014); Malibu Media, LLC v. Brenneman, No. 3:13-cv-00332-PPS-CAN, 2013 WL 6560387, at *2 (N.D. Ind. Dec. 13, 2013); Third Degree Films v. Does 1–47, 286 F.R.D. 188 (D. Mass. 2012).

  52. [52]. Greenberg, supra note 12, at 59; see also Balganesh, supra note 27, at 732 (offering a similar categorization of the copyright troll).

  53. [53]. Cf. Jacobellis v. Ohio, 378 U.S. 184, 197 (1964) (Stewart, J., concurring) (saying of obscenity: “I know it when I see it, and the motion picture involved in this case is not that”).

  54. [54]. See Sag, supra note 1, at 1112–14.

  55. [55]. Id. at 1113.

  56. [56]. Mark A. Lemley, Are Universities Patent Trolls?, 18 Fordham Intell. Prop. Media & Ent. L.J. 611, 612 (2008).

  57. [57]. Sag, supra note 1, at 1113.

  58. [58]. Indeed, it is increasingly regulatory in nature. See generally Joseph P. Liu, Regulatory Copyright, 83 N.C. L. Rev. 87 (2004). In fact, other areas of copyright law would benefit from greater regulatory involvement. See Brad A. Greenberg, Against Neutrality: Embracing Technological Discrimination, 100 Minn. L. Rev. (forthcoming 2015).

  59. [59]. See generally Intellectual Property and the Common Law (Shyamkrishna Balganesh ed., 2013).

  60. [60]. For the cases in which it is relevant, see generally Greenberg, supra note 12.

  61. [61]. 17 U.S.C. § 505 (2012).

  62. [62]. Sag suggests three other procedural remedies that courts could allow: (1) use of a special master to oversee “the discovery process and ensure that the names and addresses of individuals are not given to the plaintiff until the initial round of ‘it wasn’t me defenses’ have been raised and investigated”; (2) offer defendants the option of an ‘it wasn’t me’ mini-trial; and (3) permit defendants to proceed anonymously. Sag, supra note 1, at 1144.

  63. [63]. See, e.g., Ingenuity 13 LLC v. John Doe, No. 2:12-cv-8333-ODW (JCx), 2013 WL 1898633, at *4–6 (C.D. Cal. May 6, 2013); Righthaven LLC v. Democratic Underground, LLC, 2:10-cv-01356, Dk. #138 (D. Nev. July 15, 2011).


Intellectual Property Fellow, Kernochan Center for Law, Media and the Arts, Columbia Law School; Visiting Fellow, Information Society Project, Yale Law School. 

Thanks to Tejas Narechania for helpful feedback and to the Iowa Law Review staff for thoughtful edits.