Author: David Newhoff

David is an author, communications professional, and copyright advocate. After more than 20 years providing creative services and consulting in corporate communications, he shifted his attention to law and policy, beginning with advocacy of copyright and the value of creative professionals to America’s economy, core principles, and culture.

Decision in Seuss/Trek Mashup Case is One for the Books

Last week, the Ninth Circuit Court of Appeals issued a fair use decision in Dr. Seuss Enterprises (DSE) v. ComicMix. After oral arguments were presented in April, I wrote about this case as an example from which creators could learn what not to do when they propose to make substantial use of protected works—especially very famous works. The lawsuit involves an unpublished book called Oh, the Places You’ll Boldly Go (Boldly), a Dr. Seuss/Star Trek mashup that principally uses material from Seuss’s Oh, the Places You’ll Go! (Go!).

The appellate panel held unequivocally that ComicMix’s fair use defense failed on all four factors, and I imagine this opinion will be cited for years to come because the court so clearly inked fresh lines around several fair use principles that many avid copyright critics have endeavored to blur or erase through litigation, academic writing, or public editorial. To summarize, the Ninth Circuit opinion written by Judge McKeown held that Boldly was not a parody under factor one; that because the works copied are expressive, fair use did not favor ComicMix under factor two; that ComicMix’s use of the original works was both quantitatively and qualitatively substantial under factor three; and that Boldly represented significant potential harm to DSE’s market under factor four.

Understanding Parody

In that April post, I suggested a guideline for creators that if one truly intends to parody a work, especially a work that is instantly recognizable, that one need not engage in extensive, verbatim copying to achieve that result. In the case of Boldly, I maintain that the amount of detailed copying, which is weighed under fair use factor three, militates against ComicMix’s claim that the book is a parody at all, which is considered under factor one. As the opinion reminds us, the four factors of the fair use test are interdependent:

[The first] factor has taken on a heightened significance because it influences the lens through which we consider two other fair use factors. The third factor—the amount and substantiality of use—‘will harken back’ to the first factor. And the fourth factor, relating to market harm, is influenced by whether the commercial use was transformative.

Considering the factors in order, the court first held that:

Boldly is not a parody. ComicMix does not seriously contend that Boldly critiques or comments on Go!. Rather it claims Boldly is a parody because it situated the ‘violent, sexual, sophisticated adult entertainment’ of Star Trek ‘in the context of [Dr. Seuss]’ to create a ‘funny’ book. We considered and rejected this very claim in an appeal involving another well-known book by Dr. Seuss—The Cat in the Hat (Cat).

For discussion of that prior Cat case, see Stephen Carlisle’s post, but this clarification by the court in ComicMix cannot be overstated:  copying existing works into novel contexts is not necessarily parody just because it is meant to be funny. And mashup, merely because it changes context, is not inherently parody. Moreover, when one is copying famous works, the user will more likely be seen as attempting to ride the coattails of the prior author’s notoriety, unless a clear element of commentary on the original work, like parody, is present in the new work. But citing the gold standard case on transformative use for the purpose parody, Campbell v. Acuff-Rose, the McKeown opinion reiterates, “Lacking ‘critical bearing on the substance or style of’ Go!, Boldly cannot be characterized as a parody.”

It is reasonable to assume that, if the court had found Boldly to be parodic, its third factor analysis would likely be more liberal regarding the amount of the original works used; but I would argue that the amount used is a fundamental reason why no parody can be said to exist in this particular case. After all, what was actually taken? Not ideas, which are not protectable, and certainly not Seuss’s meter or rhyming schemes, which are also not protectable. The primary expressions that were copied—and stylistically copied in minute detail—were a number of visual works belonging to DSE.

So, even if we were to strip away the verses, ignored intended literary expressions in either book, and exclude any knowledge of Star Trek, an ordinary observer, comparing ComicMix’s illustrations to Seuss’s originals, would not necessarily know that Image B is a parody of Image A.

To the contrary, the observer could very easily assume that Image B is a derivative work prepared by DSE because it painstakingly matches so many elements in Dr. Seuss’s original drawings. And according to the opinion, it was apparently ComicMix’s intention to prepare an unlicensed derivative work that they hoped DSE might acquire, hence at least one reason illustrator Ty Templeton spent so many hours copying Seuss’s drawings in such intricate detail.

Keeping our focus on the fact that what was copied was a collection of visual works, note this response from the opinion: “We also reject as ‘completely unconvincing’ ComicMix’s ‘post-hoc characterization of the work’ as criticizing the theme of banal narcissism in Go!.” It is not uncommon for defendants to concoct a “parody” narrative in response to an infringement claim, and/or to misconstrue, at the time of use, the meaning of “parody” as expression that intends to be funny through the language of a prior work (e.g. none of Randy Rainbow’s anti-Trump songs is a parody of the songs themselves*).

It is the work that has been copied that must be the target of the parody, or other commentary. So, how can a visual work that is meticulously faithful to the original hope to convey parody through the act of too much replication? Something about the subsequent work must cue the observer that a lampoon of the prior work is present. Otherwise, the new work will be considered a derivative work, which only the copyright owner has the right to prepare.

As an example, imagine a drawing of Cindy Lou Who that has borrowed just enough Seussian elements that we recognize her, but she is depicted sneering and giving the Grinch the finger in response to his bullshitting her about the Christmas tree. That image by itself would plausibly be a comment on the original work in the same spirit that “Pretty Woman” was held to parody “Oh, Pretty Woman.” And the parodist would not have to mimic Seuss in every scribble and hue to achieve this result.

The third factor asks whether “the amount and substantiality of the portion used in relation to the copyrighted work as a whole” favors a finding of fair use. And a component of this analysis is whether the defendant used only as much of the original work as was necessary to achieve the desired result. In this case, because the desired result (an alleged parody) was held do disfavor fair use, then, logically, it should not matter very much whether ComicMix quantitatively copied one page of Go! or, as the opinion notes, 60% of the book along with “significant illustrations from Grinch and two stories in Sneeches.”

While that is a substantial quantitative taking, I would maintain that it was the qualitative copying whereby the defendant “replicated, as much and as closely as possible from Go!, the exact composition, the particular arrangements of visual components, and the swatches of well-known illustrations” that should have doomed ComicMix under the third factor test. Again, it is not necessary to mimic the originals so precisely in order to achieve the desired result of parody—if that purpose had been achieved in any sense.

Importantly, this opinion also reminds us that the third prong does not ask how much work was copied relative to the complete works produced by the original author. I’ve seen this one before, often in music infringements, where a defendant argues that he only copied a song or two owned by an artist with a large catalog. Here, the court rejected as a legal fallacy ComicMix’s argument that it only copied portions of five books out of the sixty produced by Dr. Seuss. Reiterating the law for users everywhere, the opinion responds that, “Under ComicMix’s theory, the more prolific the creator, the greater license a copyist would have to copy and imitate the original works. Nothing supports that argument.”

Potential Market Harm

The court’s discussion of the fourth factor also contains several important reiterations of fair use doctrine from which any creator can learn a great deal and reminds us that “potential” is the most important word in this part of the statute. This is because “potential” implies three underlying considerations:

1) what would be the effect of the specific use at issue on the market or value of the original work?

2) what would be the effect if the same kind of use were widespread among multiple parties?

3) although the copyright owner has not exploited the work in the manner at issue, does the new use constitute a derivative work, which the author is not obligated to produce or license?

In consideration of all the above, the court found that Boldly was not a fair use under factor four. As I have tried to emphasize, the amount of verbatim copying done by ComicMix produced an unlicensed derivative work, and the court noted that this placed Boldly in direct conflict with other derivatives of Go!, which DSE has licensed for the market. But perhaps the most important statement by the court regarding factor four is that it soundly rejected ComicMix’s assertion that the plaintiff bears the burden to prove potential market harm.

Not much about fair use doctrine lends itself to absolute statements, but the Supreme Court and our circuit have unequivocally placed the burden of proof on the proponent of the affirmative defense of fair use. ComicMix tries to plow new ground in contending that fair use is not an affirmative defense and that the burden shifts to Seuss to prove potential market harm.

Not only is that a judicial smackdown of a largely academic theory that fair use is a “right” rather than an affirmative defense, but the court further stings ComicMix’s erroneous citation of the very case that so many copyright skeptics consider a standard bearer of this invalid doctrine. The boondoggle litigation colloquially known as the “Dancing Baby” case, in which the EFF, on behalf of Stephanie Lenz, sued Universal Music Group for wrongful takedown of her YouTube video, is about as close as litigation has come to upending the affirmative defense principle of fair use. But here, the Ninth Circuit, which ruled in Lenz, states:

In an effort to distinguish controlling precedent, ComicMix argues that in Lenz v. Universal Music Corp., we deviated from our precedent construing fair use as an affirmative defense. This view misreads Lenz, which involved fair use in a different corner of copyright law, the safe harbor for Internet service providers under the Digital Millennium Copyright Act (DMCA)….In no way did we deviate from our characterization of fair use as an affirmative defense under §107. To the contrary, in addition to clarifying that, unlike copyright misuse and laches, fair use is not an excuse to copyright infringement, we reiterated that ‘the burden of proving fair use is always on the putative infringer.’

Oh, The Places We Don’t Need to Go

As I say, I predict ComicMix will prove to be an important and instructive case going forward. It should not be misread as “the big guy defeating the little guy” because, in fact, if ComicMix’s legal arguments were supportable, such a ruling would make independent creators of fresh material more vulnerable to substantial takings by large entities. It also should not be understood as a referendum on mashups in general, as the court explicitly stated that mashups can be fair use, just not this one.

In the end, I personally have sympathy for the ComicMix guys, recognizing that they put a lot of work into a project based on some very dubious copyright theory. But especially because the trio of creators who decided create Boldly are all professionals with skills and talent, the underlying lesson remains that it is almost always better to invent than to copy—let alone to copy legendary works as extensively and precisely as was done in this case.

*I cite Randy Rainbow as an example typically misunderstood as parody, and make no comment on the legality of his videos re. copyright.

Inconvenient Friction: DMCA Review 2020

Tomorrow (December 18, 2020), the Senate Judiciary Committee will present draft legislation with proposed amendments to the Digital Millennium Copyright Act of 1998. Whatever is in the draft will probably set someone’s hair on fire—or perhaps everyone’s hair on fire who has an interest in digital-age copyright enforcement. But any initial shouting will then hunker down for the drudgery of legislative haggling, which could take years. As Chairman Tillis noted in his opening remarks in the final 2020 DMCA hearing held on 12/15, it will most likely take all of his second term to get any new legislation “across the finish line.”

Tuesday’s hearing, titled The Role of Private Agreements and Existing Technology in Curbing Online Piracy, was another two-panel session with testimony by witnesses from both copyright owner and platform owner industries. Heavyweights YouTube and Facebook were in attendance, and it was noted that Twitter declined to participate. The copyright owners were most prominently represented by CreativeFuture and Copyright Alliance. All the familiar themes were reprised by both sides, but I would draw readers’ attention to the testimony of Noah Becker, music creator, and co-founder of the digital rights management company AdRev, because it presents the following information:

AdRev, which works in conjunction with YouTube’s Content ID system to identify protected works, reviews thousands of disputed takedowns every week and finds that 30% of disputes are valid, while 70% are invalid. Why do I highlight that stat? Because there is so much generalized noise about wrongful takedown allegedly silencing speech and stifling “new” creative expression despite the fact that these generalizations are not well supported by data.

The observation by AdRev that most disputes are invalid (e.g. erroneous claims of fair use*) concurs with anecdotal evidence (e.g. most people don’t understand fair use); but the fact that 30% of disputes are found to be valid, and therefore left online, also demonstrates that a more robust system of technological enforcement can benefit both owners and users of copyrighted works. As Becker stated in his testimony:

[Filtering systems] allow bona fide fair uses to thrive on the platform. YouTube’s amazing success attests to the fact that the system, works. Content ID can match millions of fingerprints against billions of hours of video while providing copyright owners and users a set of tools to easily communicate with each other about rights claims and related issues.

YouTube’s own testimony states that fewer than 1% of the takedowns it receives are disputed and that among the disputed, 55% of these favor the user of the material. This indicates that fewer than .5% of all takedowns on the platform are potentially wrongful.

Echoing the foundation of the class-action suit against YouTube by composer Maria Schneider, Becker notes that although systems like Content ID work quite well, small and independent rightsholders are not granted access to that technology by YouTube, and they generally cannot afford outside services like AdRev. Nevertheless, opponents of more robust online enforcement (i.e. advocates for the status quo of Section 512) will continue to ask Congress to ignore independent, professional creators for the sake of “new creators,” which is odd because many “new creators” want to be professionals as well.

The “New Creators” Argument Against Enforcement

In Tuesday’s hearing Joshua Lamel of Re:Create Coalition alluded once again to a somewhat ambiguous class of internet users known as “new creators,” a term so expansive that it encompasses new artists who would like to become professionals, pure hobbyists, small-business product makers, social media influencers, and media creators who presently earn revenue from online-only platforms (e.g. YouTubers receiving a share of ad revenues). While no party, whether amateur, professional, or would-be professional, should be ignored in the DMCA conversation, Congress should note that organizations like Re:Create tend to paint a very broad tapestry with the hope that nobody considers how much of this “new” creative activity is not in conflict with copyright enforcement.

On the contrary, many “new creators” are either immediately or eventually just as dependent on copyright as the “old creators.” For evidence, the IP Subcommittee need look no further than the small businesses operating in every state that find their creative products counterfeited on sites like Amazon, Etsy, and eBay. Further, with regard to generalizations about the class of “new creators,” Congress should be mindful of at least the following five considerations:

1) there does not appear to be any clear evidence proving that chronic, wrongful takedown is as rampant as alleged by various parties, let alone that the class of “new creators” has been substantially harmed by same;

2) per AdRev’s evidence, broadening the application of STM should support “new creators” by extending technological capabilities to more notice senders and respondents;

3) “New creators” like YouTubers are equally concerned with protection of their works (as YouTube’s Copyright Match system demonstrates);

4) The online marketplace for “new creators” remains too young and mercurial for policy to be predicated on over-eager predictions about economic or cultural benefits; and

5) The major platforms are an oligopoly where a single site like YouTube can amend its terms at will, leading to events like the “adpocalypse” that abruptly cut off ad revenue for many “new creators.”

With regard to points 4 and 5, Congress should view these negotiations in the context of copyright’s history and remember that at various moments in amending U.S. copyright law, one industry or another has advanced an interest in preventing some class of creators from enforcing its rights. At the time of the 1909 Act, the newspapers fought to avoid licensing photographs, and the player-piano companies didn’t want to license musical compositions; and a few years later, at the time motion pictures were added to copyright in 1912, the movie producers generally hoped to avoid licensing adaptations of novels. This is not a new story.

Today, the major internet companies would prefer to abrogate copyright enforcement altogether as an inconvenient friction gumming up the free flow of money through their advertising platforms. One big difference in this era is that YouTube, Facebook, et al pretend that they are not the parties exploiting creative works while they profit to the tune of billions of dollars from the exploitation. That ruse has persisted long enough. And at no time in copyright’s legislative history has protecting the individual creators’ interests from the users’ interests failed to fulfill copyright’s purpose to promote progress.

*AdRev also notes improper claims of valid licenses and/or that the material used is in the public domain.

Intense Accusations in Schneider Class Action

When one files an allegation of fraud in court, one must present evidence. No, I am not referring to that freak show of litigations about the 2020 election that have been tossed out by over fifty federal courts, including the Supreme Court. I’m talking about YouTube. As discussed in this July post, composer Maria Schneider leads a class action lawsuit against YouTube, alleging that the platform knowingly facilitates copyright infringement of the musical works belonging to the putative class.

At the heart of the Schneider complaint is the fact that YouTube makes its copyright management system Content ID available only to institutional rightsholders with huge catalogs, while leaving independent copyright owners to fend for themselves—i.e. one DMCA notice at a time with no automated detection of infringing files. As noted in that earlier post, Schneider’s complaint presents several arguments whereby YouTube could be found to have voided its DMCA safe harbor, including allegations of “red flag” knowledge of infringement. This suit is also occurring at the same time that Congress has been reviewing the DMCA; and the chronic allegation that platforms like YouTube play fast-and-loose with the knowledge standard will very likely be addressed in any proposed legislative changes.

In September, YouTube filed a counterclaim seeking dismissal on the grounds that Schneider’s co-plaintiff Pirate Monitor LTD used anonymized agents to “fabricate infringement claims by having its own agents upload material from its copyrighted works to YouTube.” The platform alleges that Pirate Monitor engaged in this practice in order “to obtain access to [its] copyright management systems.” In addition to its allegation of fraud, YouTube avers that Pirate Monitor’s actions are an example that justifies its policy of limiting access to Content ID because, it states, some parties cannot be trusted with such tools.

That is one lulu of an accusation, to which Pirate Monitor responded in a brief filed in November that YouTube provides no evidence in its counterclaim of any relationship between the anonymized uploaders of the files in question and Pirate Monitor. The response argues that YouTube’s counterclaims “rest solely on conclusory allegations,” that its assertions of fraud do not meet the evidentiary standard under the rules of civil procedure, and that YouTube lacks the standing to request injunctive relief for the alleged harm caused by the alleged fraud. More broadly, the response accuses YouTube of trying to change the subject, “to spin this case away from being about YouTube’s facilitation and inducement of copyright infringement, and make it instead about who can and cannot be ‘trusted’ to use Content ID.”

I am not about to assert any comments of a factual nature without facts in hand. We’ll see what evidence is, or is not, presented. But I will reiterate that the soul of Maria Schneider’s complaint expresses precisely what hundreds, or thousands, of independent creators contend with on a piracy-rich landscape like YouTube. As noted in the first post, this suit reprises several matters of unfinished business left in limbo after Viacom and YouTube settled their copyright infringement suit in 2014.

At issue then, as it remains today, is evidence suggesting (or even proving) that YouTube’s operators have knowingly attracted and retained highly lucrative market-share by walking the razor’s edge of its responsibilities under the DMCA. In that spirit, YouTube’s preliminary statement in its counterclaim is a cue for a spit-take if nothing else. “Since its founding in 2005,” the brief begins, “YouTube has gone far above and beyond its legal obligations to assist copyright holders in protecting their rights.”

That self-congratulatory assertion hardly squares with the record. When Google acquired YouTube for $1.65 billion, the platform had achieved category-killer status based on a policy of copyright infringement. And although YouTube developed Content ID as a semi-compromise with the corporate rightsholders, it has done nothing for the countless small and independent creators whose works are used and monetized without permission. That story doesn’t even satisfy the industry’s bombast about supporting entrepreneurial creators, let alone anyone’s definition of “going above and beyond.”