Introduction
The British reversionary story is not the only one that shows how reversion’s potential has been thwarted by poor design and systematic undermining by rightsholders. The United States has a long history of time-based reversion rights intended to benefit creators, first via a renewal right, and later via rights to unilaterally terminate copyright grants they had made decades before. In each case, however, these interventions failed to make any meaningful difference to the balance of power between creators and rightsholders, and did little to direct more money into creators’ pockets.
This chapter traces that history of failure, and argues it should come as no surprise that the current US termination right is generating few benefits for authors: its design ensured its ineffectiveness from the very beginning.
3.1 US Reversion Rights Have Been Consistently Undermined
3.1.1 1790: The Mirroring of the Two-Term Statute of Anne Model
The US mimicked the Statute of Anne when it enacted its first Federal copyright law in 1790, granting 14 years of copyright and then 14 more to the author if they were still alive at the end of the first.Footnote 1 But unlike the Statute of Anne, which explicitly returned the rights to authors, the 1790 Act allowed the renewal term to go to authors or their executors, administrators or assigns.Footnote 2
It was unclear from the get-go whether the reference to ‘assigns’ meant that the author could transfer their rights for both terms in their initial contract, and Bently and Ginsburg argue that the more likely intention was that they could not.Footnote 3 Unsurprisingly, however, publishers routinely claimed to take the full duration of both terms upfront, before the value of the work was known.
Various draft bills were proposed to clarify this point in favour of authors, but never actually passed into law.Footnote 4
3.1.2 1831: The Two-Term System Remains, but Is Undermined by Rightsholder Contracting Practices
The US retained the reversionary term in its 1831 law, after the British had abandoned its dual-term model. But despite the problem being well known, its revised law still failed to include any explicit prohibition on investors extracting rights over the reversionary term at the same time as the first.Footnote 5 Unsurprisingly, publishers took full advantage of that omission, and by the 1870s, publishing contracts ‘systematically’ provided for both terms to be extracted in full, with authors receiving no special payment for the second.Footnote 6 Although it was by no means clear that this was actually permitted under the law, the practical effect of this conduct was to deprive authors of the ability to re-license still valuable works once their copyrights were renewed.
In the lead-up to what became the 1909 US copyright law, it was proposed to extend copyright terms to the author’s lifetime plus 50 years.Footnote 7 However, disputes broke out about what would happen to existing copyrights. There was consensus that they should also be extended – the dispute was over whether it should be authors or publishers that benefited from the windfall.
An early draft proposed limiting publishers’ rights to the term they had actually contracted for, returning them to authors to enjoy the benefit of any additional years of protection after that.Footnote 8 But publishers objected fiercely, insisting instead that they get the benefit of the extension even though they had never bargained for (or paid for) those rights.Footnote 9
The Publishers Copyright League proposed that the new term of life plus 50 years incorporate a second renewal term that could be added to an initial fixed term and first renewal.Footnote 10 Sneakily though, they included language stating that this additional renewal term would only be enjoyed by the author or their heirs if the copyright had not been previously assigned.Footnote 11 Since publishers had been routinely extracting authors’ rights in full for decades by this point, this would have meant an enormous windfall for publishers – all the while professing their ‘willing[ness] that … [authors] should have the full benefit of this extension of those copyrights’.Footnote 12
Bently and Ginsburg describe this proposal as a ‘wolf in … sheep’s clothing’ that would have given publishers something they’d never before had – ‘explicit statutory recognition’ that authors could assign reversionary and renewal rights before they had even had a chance to vest.Footnote 13
3.1.3 1909: An Additional 14-Year Term, Undermined
3.1.3.1 The Poor Design of the 1909’s Two-Term System
Ultimately, Congress abandoned the life plus 50 years proposal, and in 1909 simply extended the renewal term from 14 years to 28.Footnote 14 Congressional lawmakers seemed to understand how beneficial reversionary rights could be to authors – at least when they were able to hold on to them. There was powerful real-world evidence of this in the experience of Samuel Clemens (better known by his pen name, Mark Twain). He had sold Innocents Abroad very cheaply for the first term, but had somehow managed to keep hold of the renewal rights.Footnote 15 That was just as well, because it ended up being his best-selling work during his lifetime. If the rights had not reverted for the renewal term, nearly all of the profits would have ended up in other people’s pockets.
Congress acknowledged this, and expressed its intention for authors to benefit from their statutory reversion terms.Footnote 16 And yet somehow, once again, they failed to make it explicit in the 1909 Act that publishers could not extract authors’ renewal rights at the same time as the initial term. Publishers continued to take advantage of the uncertainty by insisting they could,Footnote 17 and, since few authors had the means or power to challenge them, the question of whether advance transfers of rights were binding remained unresolved for another 30 years.
3.1.3.2 The Supreme Court Undermines Creator Rights
This uncertainty was finally resolved in 1943, when the Supreme Court in Fred Fisher Music Co v Witmark found that publishers could take both terms from the author in the first instance.Footnote 18 The case involved three composers of the music and lyrics to the song ‘When Irish Eyes Are Smiling’, who assigned copyright to music publishers, M Witmark & Sons in 1912 for the first copyright term. But five years later, one of the song’s lyricists, George Graff Jr, granted his rights in the renewal term to Witmark for ‘When Irish Eyes’ and 69 other songs for US$1,600, and a power of attorney for Witmark to complete the processes necessary to renew the copyright for its second term, as he was ‘in need of immediate funds’.Footnote 19 When the 28-year term was up, Witmark registered the renewal and transferred the second term to itself.Footnote 20 But Graff had a change of heart and sought to register the renewal himself, following which he assigned the renewal interest to Fisher Music Co.Footnote 21
The dispute between Fisher and Witmark eventually came before the Supreme Court mainly on this point of law: could copyright in the renewal term be assigned before the renewal right had accrued? The Supreme Court found that it could.Footnote 22 Although Congress had clearly indicated that authors were given two terms so they could negotiate the second term of exploitation from positions of greater bargaining power, the Supreme Court’s decision ensured that rightsholders could prevent them from doing so.
Extraordinarily, Justice Frankfurter phrased his rubbing out of authors’ rights in a way that suggested he was doing them a favour:Footnote 23 ‘We are asked to recognize that authors are congenitally irresponsible, that frequently they are so sorely pressed for funds that they are willing to sell their work for a mere pittance, and therefore assignments made by them should not be upheld.’Footnote 24 But, his Honour continued,
We cannot draw a principle of law from the familiar stories of garret poverty of some men of literary genius … We do not have such assured knowledge about authorship … or the psychology of gifted writers and composers, as to justify us as judges in importing into Congressional legislation a denial to authors of the freedom to dispose of their property possessed by others.Footnote 25
His Honour’s reasoning was that it may well be that authors ‘have habits making for intermittent want, [but] they may have no less a spirit of independence which would resent treatment of them as wards under guardianship of the law.’Footnote 26
The authors who had advocated so strongly for statutory clarification that their reversion rights couldn’t be prematurely extracted may well have been taken aback by the nature of this solicitude. By the time of this decision publishers had a track record of systematically extracting broad rights from authors for at least 70 years, with none but the most powerful and successful authors typically having any choice about whether to give them up.Footnote 27
Nonetheless, the Fred Fisher Court found that the thing authors would most resent (amidst their struggles to pay their bills and provide for their families) was whether the law took an overly paternalistic view of their copyright interests!Footnote 28
The Fred Fisher decision is perhaps unsurprising given the Supreme Court’s early twentieth-century deference to freedom of contract.Footnote 29 But it also demonstrates the importance of precise drafting that accurately gives effect to the legislative intent: the law could not have been undermined if authors’ rights had been properly protected in the first place.
The Supreme Court did subsequently provide one small consolation prize for destitute families of deceased authors – a ruling that, if the author assigned the right and then died, the 28-year renewal term would go to heirs free and clear: they would not be bound by the author’s own earlier grant.Footnote 30 But investors promptly made sure to take this away too. They started requiring creators not only to give up their own rights, but to obtain their family members’ agreement to give up their rights too.Footnote 31 Until that happened, they couldn’t get paid. A ‘considerable amount of trafficking’ grew up around these rights.Footnote 32 This anecdote from music publishing veteran Jay Morgenstern shows just how seriously investors took this. As he tells the story, a music studio ‘demanded all of the signatures and we were able to comply. [Then] the author died while his wife was expecting. Everybody at the studio went into cardiac arrest thinking that the new heir could renege on the license at reversionary time.’Footnote 33
Renewal rights obviously had potential value – if they didn’t, there would have been no reason for rightsholders to cling to them so tightly. Authors should have been able to share meaningfully in that by having the right to re-negotiate their deals during the renewal term, once everyone had a better idea of how much their work was worth. However, the design and interpretation of this regime ensured that it could be largely captured by investors instead. This undoubtedly harmed the public interest in access as well. As we’ve seen elsewhere, just because a publisher holds the rights doesn’t mean they will actually exploit them. If authors had been free to negotiate fresh deals, as Congress intended, it would have increased the chance of the rights ending up in the hands of an investor who actively wished to do so.
3.1.4 1976: The New Copyright Act’s ‘Inalienable’ Termination Right
The renewal structure’s failures influenced Congress to create an entirely new reversion scheme as part of the 1976 Copyright Act, which remains the law today. But as we show below, this too was undermined by poor design and fierce rightsholder lobbying. The result is a case study in what not to do if you want a reversion right that will meaningfully benefit creators.
3.1.4.1 A Radical Proposal: Automatic 20-Year Reversion Except in Cases of Continuing Royalties
In the lead-up to the revision of federal copyright law, the Copyright Register, Abraham Kaminstein, had been scathing about the renewal right’s failures. He pointed out that its primary purpose had been to protect authors and their heirs against ‘unprofitable or improvident disposition[s]’, and that because the law had been interpreted to allow authors to transfer their renewal rights in advance, it had ‘largely failed to accomplish its primary purpose.’Footnote 34
In response to these failures, the Copyright Office advanced an initial proposal for rights to revert 20 years after transfer unless the contract provided continuing royalties for use.Footnote 35
Publishers and other rightsholders lambasted the proposal, claiming that would not give them enough time to ‘recoup their investment[s]’ and would ‘make [copyrights] … less valuable.’Footnote 36 And in any event, they insisted, ‘authors, if they ever were, are no longer in a weak bargaining position and need no special protection of the sort contemplated here’.Footnote 37 Authors weren’t happy about it either, though their reasons were very different: they thought it would be too easy for publishers to evade the law by paying nominal royalties for the renewal period, and asked for stronger protections.Footnote 38
3.1.4.2 Automatic Reversion and a ‘Bestseller’ Clause
The Copyright Office took the authors’ point, and suggested two new alternatives for consideration. In a proposal evocative of the 1737 British bill that would have limited transfers to 10 years, Alternative A would have seen rights revert automatically to authors 25 years after transfer.Footnote 39 Alternative B would have given authors the right to apply to a court to reclaim their rights if, after 20 years, the profits earned by investors were ‘strikingly disproportionate’ to the author’s share.Footnote 40
Alternative B was discarded quickly, with authors and investors both concerned about its potential costs and uncertainty.Footnote 41 But it was Alternative A on which investors really unleashed their ire. Many of their arguments echoed Justice Frankfurter in suggesting the proposed protection was insulting to authors. In transcripts of the negotiations to thrash out the law, content industry representatives variously described it as making authors out to be ‘improvident’, ‘nincompoops’ and ‘fools’.Footnote 42
It turned out, however, that authors didn’t want copyright investors to protect them from the new law. In fact, they were ‘ardently’ in favour of reversion protections: to the point they would oppose any amendment of the copyright law in the event they weren’t included. Irwin Karp, representing the Authors’ League of America in those negotiations, explained that this vehement support was not because authors were improvident or foolish, but because the opposite was true. ‘[P]rudent, provident author[s]’ want to limit their grants of rights, because of that information problem we’ve talked about before: they know their value can’t be properly assessed at the time the initial bargain is made.Footnote 43 However, they rarely have the opportunity to do so. ‘The reason you need a reversion’, Karp explained, ‘is because the imbalance in bargaining positions makes it impossible for the author to obtain the sort of bargain that he would like to obtain’.Footnote 44 In the publishing industry, for example, ‘[t]he basic terms of a book contract are the same wherever you go. You can’t, no matter how strongly you bargain as an author, limit the lease of rights to publish a book.’Footnote 45
Copyright investors argued their businesses would be ruined if authors were limited to granting them just 25 years of rights. A film industry representative claimed that it might not be enough time to make a motion picture, ‘for various complicated reasons’; book publishers claimed it would be ‘devastating’;Footnote 46 and the music industry claimed it would ‘mean the death knell’.Footnote 47 But Karp, the authors’ representative, persuasively demolished those claims. Under the then-current law, as he pointed out, investors lost their rights if the author died before the 28th year was up. No evidence had been presented of suffering by content companies as a result, so ergo, a term around that length must be ample.Footnote 48
Rightsholders also solemnly exhorted the importance of freedom of contract. Horace Manges, speaking for the book industry, was particularly unsympathetic to those who came off second best in their deals. ‘[I]mprovident bargains are made by all classes of people, but the law does not give them another chance to change a disadvantageous contract.’Footnote 49
But Karp had yet a powerful response to this argument too, pointing out that, at that exact moment, book publishers were attempting to interfere with retailers’ freedom of contract by setting the prices at which books must be sold. In his acid observation, ‘Each of us will object to certain interferences and support other interferences, depending on whether we are the owner of the ox who was gored or whether we want to gore somebody else’s ox.’Footnote 50
3.1.4.3 The Eventual Compromise
Congress eventually passed the 1976 Copyright Act with two new termination rights – one applying to pre-1978 grants and the other to those that came after.Footnote 51 To terminate rights transfers, applicants must serve a notice on the grantee or their successor setting out the effective termination date, work title, author, and other key information such as details of the grant in question.Footnote 52 Those notices had to be issued not less than 2 years and not more than 10 years before the termination date,Footnote 53 and had to be validly served on the rightsholder as well as filed with the Copyright Office.Footnote 54 The base fee for recording a notice is US$125 (at the time of writing), and increases with the number of grants included in the notice.Footnote 55
Lawmakers were explicit that they were necessary because of the power dynamics typical of creator-investor relationships: ‘A provision of this sort is needed because of the unequal bargaining position of authors, resulting in part from the impossibility of determining a work’s value until it has been exploited.’Footnote 56 As we show in Section 3.2, however, the operation of this creator-focused scheme has been anything but, with the law proving to be unclear, complex, susceptible to undermining by rightsholders, and generating little benefit for creators.
3.2 How the Current US Termination System (Actually) Works
Despite Congress’s good intentions, the termination legislation they ultimately enacted gored the authors’ ox deep. The provisions were a far cry from the Copyright Office’s original author protective proposal. Rights cannot be reclaimed until 35 years after transfer. Terminations must be initiated by authors in accordance with complex technical rules. And numerous carveouts introduce substantial new uncertainties that have been exploited by investors to even further reduce the practical benefit of the new rights for their intended beneficiaries.
In this section, we explain how the scheme works in practice, with a focus on the main controversies:
(a) whether sound recordings are exempt from termination;
(b) the complexities of the termination procedure;
(c) whether rights are in fact ‘inalienable’; and
(d) the way it disproportionately benefits the most commercially successful creators.Footnote 57
Consistent with the history we’ve chronicled so far, there is a huge gap between this legislation’s stated intention of protecting authors, and its actual real-world impact.
3.2.1 Are Sound Recordings Exempt from Termination?
One of the most controversial aspects of both 1976 Act termination rights concerns scope. Congress learned from the decades of uncertainty around prior reversion regimes, and finally made termination rights inalienable – their authors cannot contract them away.Footnote 58 However, there’s a big exception. US law features the legal fiction that works ‘for hire’ are authored by the employer – even when that is a corporation, incapable of authoring anything.Footnote 59 Although termination rights are inalienable, the law provides that in cases involving works ‘for hire’ their actual creator never had authorship rights in the first place, meaning that there’s nothing for them to terminate.Footnote 60
3.2.1.1 The Work-for-Hire Exception
The extent to which this carveout would affect creators was always going to depend on how works ‘for hire’ was defined. In the negotiations to craft the new Copyright Act, rightsholders succeeded in obtaining a broad definition that included not just all works created by employees in the scope of their employment, but also works within nine other defined classes – so long as the parties expressly agreed on that designation in writing (Table 3.1).Footnote 61
| Type of exclusion | Description |
|---|---|
| In their entirety | Works prepared in the scope of employment |
| By express agreement, having been specially ordered or commissioned for use: | As a contribution to a collective work As part of a motion picture or other audiovisual work As a translation As a supplementary workFootnote 62 As a compilation As an instructional textFootnote 63 As a test As answer material for a test As an atlas |
As you can see, the additional carveouts focus largely on circumstances in which recognition of individual authorship would undermine the feasibility of making the work in the first place (such as documents prepared for the benefit of a corporate employer), or where the existence of multiple authors might make individual ownership unworkable (as can be the case for compilations and collective works).
When rightsholders lobbied for those carveouts from the termination right, they focused particularly on the potential problems termination would cause for the creation of encyclopedias and textbooks, which tend to have many different authors updating them regularly over time.Footnote 64 That is a valid point: after 35 years it would indeed be very difficult to untangle any individual’s contributions, and allowing any one individual to terminate could risk a much larger investment.
In practice, however, these carveouts have not been limited to circumstances such as that. Rightsholders have taken advantage of drafting ambiguities to assert that they apply well beyond the scope they were obviously intended to cover. As foreshadowed above, this has most notably occurred in the context of sound recordings – the category of works most likely to have enduring value decades after they first hit the market. As we’ll show, their attempts to broaden the reach of these carveouts has done much to subvert the law’s intended benefits.
3.2.1.2 Record Companies Try to Include Sound Recordings as Works-for-Hire: And Succeed (Temporarily)
In 1999, more than a decade before the first § 203 terminations could be initiated, a congressional staffer quietly inserted four words into the Satellite Home Viewer Improvement Act that had the effect of adding sound recordings to the Copyright Act’s definition of works for hire. The 11th-hour change occurred without any debates,Footnote 65 and artist representatives didn’t find out about it until after it had been signed into law.Footnote 66 Uproar ensued. The resulting investigation discovered that the amendment had been inserted at the request of the Recording Industry Association of America (RIAA) – the industry lobby group representing the interests of the world’s most powerful record labels – rather than any elected member of Congress.Footnote 67
While Marybeth Peters, the then Register of Copyrights, denounced the ‘middle of the night’ amendment for having been made ‘without any input from performers’,Footnote 68 the RIAA played innocent, describing it as a mere ‘technical’ change.Footnote 69 US law permits technical amendments to statutes to bypass some of the usual lawmaking processes, but that is limited to fixing obvious mistakes like typos – not changing the substance of what has been deliberately decided by Congress.Footnote 70 But Hilary Rosen, the RIAA’s then-president and CEO, falsely claimed that ‘in everybody’s view this was a technical issue’, because ‘record companies have long registered recordings with the Copyright Office as works-for-hire’.Footnote 71 As one legal writer retorted, ‘Rosen’s idea of “everybody” … must not include the artists, their attorneys, Intellectual Property professors and authors, top officials at the Copyright Office, or the courts for that matter, because all believe differently.’Footnote 72 As if there wasn’t already enough stench around this surreptitious manoeuvre, the congressional staffer who had clandestinely slipped in the amendment, Mitch Glazier, was hired by the RIAA just a few months later.Footnote 73 He subsequently became, and is at the time of writing, its chair and CEO.Footnote 74
We’ve told this story in detail because it demonstrates the lengths powerful rightsholders will go to avoid meaningful reversion rights – that is, those that might actually help level the playing field for creators. If they didn’t have such potential, the music industry would have had no cause to go to such extremes to undermine them.
The RIAA’s sneaky change didn’t remain on the statute books for long. Recording artists were outraged when the theft came to light, and began a fierce and coordinated campaign to have it rolled back, finally succeeding after almost a year of gruelling and expensive efforts.Footnote 75
In practice though, the artists’ victory has been far less effective than they would have hoped. Although the labels’ brazen attempt to steal musicians’ reversion rights over sound recordings was ultimately defeated, they still consistently resist efforts to terminate sound recording transfers on the grounds that they fall within the carveouts. Label licensing agreements routinely designate sound recordings as works-for-hire.Footnote 76 Most commonly, they argue that’s because they’re part of an audiovisual work, because they’re specifically ordered or commissioned for use as a contribution to a collective work, or because they’re compilations.Footnote 77 The latter arguments ignore the fact that sound recordings typically have none of the features that justified compilations and collective works from being made excludable in the first place, and they’re made even though labels argue albums are not compilations with equal fervour when that outcome would increase the sum payable to them by way of statutory damages for infringement!Footnote 78 Some leading industry contract templates even designate artists as their employees exclusively for the purpose of the copyright ownership clause.Footnote 79 Recording artists are obviously not label employees (and certainly aren’t treated as employees when it comes to minimum pay or worker protections), but if successful, such machinations would deny artists their right to terminate even sound recordings which are obviously outside the enumerated categories.
While some federal courts have explicitly recognised that sound recordings aren’t one of the specified carveouts, there is a lack of specific higher court authority that could provide certainty to artists as to whether the labels’ claims have merit.Footnote 80 The dearth of litigation is unsurprising. Litigation is expensive, particularly in the US where even successful claimants usually have to bear their own costs.Footnote 81 Termination cases have sometimes been instituted by stars with deep pockets, including Sir Paul McCartney,Footnote 82 Boston country music star Dwight Yoakam,Footnote 83 or as class action lawsuits like those headed by singer John Waite and New York Dolls frontman David Johansen.Footnote 84 But labels typically ensured those cases are settled, usually with non-disclosure agreements, preventing not only judicial precedent but also the kind of knowledge transfer that might benefit artists more broadly.Footnote 85
The upshot of all this is that, almost half a century after the 1976 law was enacted, many potential beneficiaries remain in limbo about whether they have any rights at all. The label arguments may well fail if and when they’re finally tested in court, but until then the practical effect of the legislation is as if the RIAA’s legislative heist had in fact succeeded: few artists can use their statutory termination rights to get a better deal for their sound recording copyrights.
This situation is especially aggravating given the changing economies around music distribution. As we explained in Chapter 1, in the 1970s royalty rates as low as 4 per cent were not uncommon, while deals inked today are likely to pay at least six times that amount.Footnote 86 If heritage artists were able to use their reversion rights the way Congress intended, they could renegotiate those contracts to let them share more fairly in the fruits of their labour. As matters currently stand, however, record labels simply add that windfall to their profits instead. Though it’s artists’ strong moral claims that justify beyond-incentive copyrights, the termination law’s failure ensures those rewards fatten the wallets of executives and shareholders instead.
3.2.2 Termination Rights Are Difficult to Exercise
The termination law’s second major failing arises from its complexity. As the proposed rights worked their way through the legislative process, the substantive and procedural processes for reclaiming rights became ever more complicated, sometimes ambiguous, and often impossible for laypeople to understand. Indeed, former Register of Copyrights Maria Pallante describes the termination provisions as ‘almost incomprehensible on their face, particularly for the authors, widows, widowers, children and other heirs who need to navigate them’.Footnote 87 As composer Dr Elizabeth Vercoe explains, this is a real deterrent to use: ‘the process is complex and few of the composers are sufficiently persistent to jump over the hurdles required to file for terminations.’Footnote 88 Copyright law academic Pamela Samuelson agrees, describing the process as ‘so cumbersome and complicated that most authors will not realistically have a meaningful opportunity to terminate’.Footnote 89
3.2.2.1 Creators Can Easily Misinterpret Termination’s Procedural Requirements
The legislation’s detailed prescriptions ensure that those who aren’t entirely deterred from exercising their rights have plenty of opportunity to get something wrong. Consider the example of artist Daniel DeCarlo, who worked for Archie Comics for over 40 years, during which time he created popular characters such as Josie and the Pussycats, Cheryl Blossom and Sabrina the Teenage Witch. Months after losing his job due to a breach of contract lawsuit he filed against Archie Comics,Footnote 90 he sought to terminate the transfers to some of his work, but the US District Court for the Southern District of New York declared his notices invalidFootnote 91 – DeCarlo had understood the termination clock began 35 years after initial publication of the works in 1967, but the Court confirmed it began from the dates of the agreements (1988 and 1996).Footnote 92
DeCarlo is not the only one confused. Our study of copyright termination notice records unearthed at least 24 titles from nine § 203 notices with effective dates of termination before 1 January 2013. It’s not clear if this was due to errors when those notices were recorded by the Copyright Office, or whether the creators themselves entered the wrong dates.Footnote 93 In any case, it should have been apparent that pre-2013 dates would be legally invalid, since that was the first year any § 203 termination could take effect.Footnote 94
3.2.2.2 Compliance Can Be Difficult Even When Interpreting Termination Procedures Correctly
Even where applicants interpret the requirements correctly, compliance can still be difficult. For example, since decades must pass before a termination notice can be issued, it can be difficult to determine precisely when it needs to be filed, or the date on which termination can have legal effect. There are any number of good reasons why, decades after a copyright agreement was executed, it might not be possible for its author to lay hands on it – and that can prevent them from being able to supply the key details needed to comply with the detailed substantive and procedural requirements of the notice process. The challenges are magnified when the transfers are evidenced over multiple documents or involve multiple parties.
Innocent errors won’t automatically invalidate notices so long as they don’t ‘materially affect the adequacy of the information required to serve the purposes of [the termination provisions]’.Footnote 95 However, while courts have sometimes let minor defects slide,Footnote 96 the overall approach has been to require strict compliance with termination formalities.Footnote 97 Legal representation can help creators get that right,Footnote 98 but are very expensive – another disincentive to creators enforcing their rights.Footnote 99
The termination system’s complexity reduces its effectiveness as a mechanism to reward creators and promote widespread access to knowledge and culture. The best practice principles we set out in Chapter 6 prompt lawmakers to reflect on what is actually necessary to best achieve copyright’s access and reward goals.
3.2.3 An Inalienable Right: So far as That Goes
The creators who campaigned for reform to the existing US termination law lost on many fronts, but secured one huge victory: Congress, after decades of dodging this question, finally made termination rights inalienable. The legislation expressly permits the termination of transfers ‘notwithstanding any agreement to the contrary, including an agreement to make a will or any future grant’.Footnote 100 This is a huge improvement on the previous law, which allowed rightsholders to exercise their outsized bargaining power to snap up the benefit of the renewal term at the time the initial contract was entered into.Footnote 101
But the efficacy of this new inalienability ‘shield’ in upholding the interests of authors and heirs has been undermined, as with other parts of the 1976 termination law, by its poor design. Keenly aware of the money and power at stake, rightsholders have developed ingenious arguments to test, and exploit, weaknesses in this shield. Controversies have thus arisen about precisely how inalienable termination rights are, of which we now examine the three most prominent:
(a) creators’ testamentary freedom being curtailed;
(b) the risk of statutory heirs losing their US termination rights; and
(c) the risk of creators losing their US termination rights through the application of foreign law.
3.2.3.1 Creators’ Testamentary Freedom Being Curtailed
When Congress finally got serious about protecting authors against being obliged to give up their termination rights, it got very serious. While creators may leave their copyrights to any person they choose (and those testamentary transfers are not themselves terminableFootnote 102), they can’t bequeath the rights to terminate any previous transfers – those beneficiaries are designated via the statute.Footnote 103 This bifurcated system means that the owner of the copyright might not be the person who comes to own the termination rights.
That was the situation that emerged with the estate of Nobel Laureate John Steinbeck: he bequeathed the entirety of his copyright interests to his third wife, Elaine, but the Copyright Act split the associated termination interests between his wife and two children from a previous marriage.Footnote 104 When Elaine died, the termination rights were controlled exclusively by the two children, even though she had explicitly excluded them from her will.Footnote 105 A similar situation arose with Ray Charles’ estate. He had left his copyrights to his charitable foundation, but since the statute gave the termination rights to his children, they were able to exercise them in a way that diverted the proceeds away from people with disabilities.Footnote 106
It’s obviously important to prevent creators from being pressured to bequeath their reversionary interests in ways that benefit the current licensees, but facts such as these are a reminder that it’s also important to protect creators’ wishes when it comes to the future stewardship of their works.Footnote 107 While a detailed consideration of the pros and cons of the current US approach is outside the scope of this book, we flag it to note that any policy discussion around the inalienability of reversion rights will need to grapple with the issue – a matter we return to in Chapter 6.Footnote 108
3.2.3.2 Statutory Heirs Losing Their US Termination Rights
The statutory bifurcation between who an author (A) can leave their copyrights to (their estate, E), and who the termination rights will pass to automatically (the statutory successors, S) creates potential situations where E, in tandem with a grantee, can pre-emptively renegotiate a new deal for the rights and eliminate S’s termination rights entirely in respect of that copyright. At first glance, this scenario appears to fly in the face of Congress’s intention to protect the author and their statutory heirs by giving those heirs the right to terminate grants. The question that remains unanswered is whether these agreements can be ruled unenforceable as ‘agreements to the contrary’. Courts have not settled the matter to date, leaving the potential for statutory heirs to be disinherited of these valuable termination rights.Footnote 109
Steinbeck’s two children found this out after they sought to terminate a 1938 grant Steinbeck made to The Viking Press, who eventually assigned the rights to publisher Penguin.Footnote 110 Penguin refused, arguing that a 1994 agreement with Steinbeck’s widow Elaine (who inherited the copyright upon Steinbeck’s passing) effectively ‘re-grant[ed] publishing rights to [those] … works’.Footnote 111 The Second Circuit agreed. The 1994 agreement was clearly intended to contractually terminate the 1938 agreement, which meant there was nothing for Steinbeck’s children to use the statutory termination right under § 304(d) of the Act on.Footnote 112 And the 1994 agreement was not an ‘agreement to the contrary’ under § 304(c)(5) of the Act.Footnote 113 Even if it had removed a termination right, that right would not have been exercisable when the 1994 agreement was executed (because § 304(d) related to works that benefited from the 20-year copyright term extension in 1998) – so any rights the 1994 agreement purported to bypass had not come into existence when it was executed.Footnote 114 The upshot of this decision was that creators/heirs had one chance to use the termination right, but that Elaine had effectively used up this chance through the 1994 agreement.Footnote 115
The Ninth Circuit reached the same conclusion in a case involving the estate of Winnie the Pooh creator AA Milne.Footnote 116 Milne had initially granted US and Canadian rights to producer Stephen Slesinger in 1930, and Slesinger granted these rights to Disney in 1961.Footnote 117 By this point Milne was deceased, having bequeathed the interests in the Pooh works to a trust benefiting first his widow, then his son Christopher and granddaughter Clare.Footnote 118 Christopher favourably renegotiated these agreements in 1983 (the Pooh trust being estimated to have made a net gain in the hundreds of millions of dollarsFootnote 119) on the proviso that he would not exercise his § 304 termination rights under the original agreement.Footnote 120 Clare sought to terminate the 1930 agreement in 2002 (renegotiating the grant of the rights she expected to regain with Disney on the same dayFootnote 121), but would only have had the right to do so if the 1983 deal was an invalid ‘agreement to the contrary’.Footnote 122 The Ninth Circuit held that it wasn’t, since the 1983 agreement had effectively replaced the one from 1930.Footnote 123 In any event, it reasoned, there was no need for termination here because the 1983 agreement had achieved substantially the same goal, giving the authors/heirs a second opportunity to profit from the work.Footnote 124
Both Milne and Steinbeck suggest a creator’s estate can disinherit statutory heirs of the termination right (to the extent they are different) by executing a subsequent agreement that replaces the old grant and leaves nothing to terminate under the statute.Footnote 125 The subsequent agreement may well be lucrative for the copyright heirs (e.g. Milne’s 1983 agreement) or lead to greater access to knowledge and culture. But as we showed in Chapter 1, such success is simply not guaranteed. In the meantime, enforcing such agreements enables rightsholders to deprive statutory beneficiaries (under the Act it is the creator’s family) of their right to terminate, and all the intended benefits that could flow from being able to negotiate from a more informed bargaining position as to the value of the work.
There is some uncertainty as to whether the Milne and Steinbeck approaches will be followed.Footnote 126 Amidst this uncertainty, statutory termination heirs and testamentary beneficiaries (if different parties) should ensure they are as aligned as possible on the exploitation of the works and termination. Otherwise, these kinds of disputes are likely to keep arising until either Congress or the Supreme Court brings clarity to this area of the law.
3.2.3.3 Can US Termination Rights Become Alienable by the Operation of Foreign Law?
Rightsholders have also tried to undermine attempted terminations of US copyrights under the 1976 system by securing creator agreement to, and then relying on, choice-of-law/choice-of-forum clauses to have exploitation contracts adjudicated under foreign domestic law and secure rulings preventing the effect of those terminations, whether in foreign or US courts.Footnote 127 This strategy directly challenges termination’s inalienability, especially given rights grants covering all or most of the world are prominent in leading industry templates (and most creators are likely to have difficulty negotiating substantive departures from these standard form agreements)Footnote 128 – rightsholders could simply insist that any disputes are governed by laws and fora without such ‘inalienable’ creator protections.
3.2.3.3.1 Undermining US Reversionary Systems Using Foreign Law before 1976.
A 1963 judgment of the High Court of England and Wales, Campbell Connelly & Co Ltd v Noble,Footnote 129 had already shown that the two-term renewal system under the 1909 Act was susceptible to this sort of undermining. There, a worldwide assignment of copyright, governed by English law, was held to assign not just the first term but the renewal term of US copyright as well.Footnote 130
But was the 1976 Act, with its express provision for inalienability, better protected against such incursions under foreign law? The answer thus far remains uncertain. Another High Court decision from 2016 appeared to fling wide the door to wholesale undermining of the 1976 termination rights by the application of foreign law to contracts assigning US copyrights, although (as we explain in Section 3.2.3.3.2) this outcome is by no means certain.
3.2.3.3.2 Undermining the 1976 Termination Rights: Breach of Contract in a Foreign Jurisdiction.
Gloucester Place Music Ltd v Le Bon and Ors is arguably the most prominent twenty-first-century example of rightsholders neutralising the US termination rights.Footnote 131 The case involved the spat between members of the world-famous pop band Duran Duran and music publisher Gloucester Place Music.Footnote 132 The band members tried to statutorily terminate the grants of US copyrights made as part of a 1980 worldwide copyright assignment to Gloucester (previously known as Tritec Music).Footnote 133 But Gloucester brought legal action in the High Court of England and Wales claiming the filing of the notices was a breach of contract.Footnote 134
The contracts specified that English law applied, and since neither party disputed the choice of law question, the Court did not have the opportunity to determine it.Footnote 135 In these circumstances Justice Arnold decided, ‘not without hesitation’, that the band members exercising their termination rights was a breach of the relevant contracts under UK law.Footnote 136 The clause assigning worldwide rights was ‘wide and general’, and in context required the worldwide copyrights to stay with Gloucester for the whole copyright term.Footnote 137 Further, the band members had agreed they would not transfer the rights to any other person, which Justice Arnold interpreted to mean ‘any person other than [Gloucester]’.Footnote 138 Any exercise of the US statutory termination rights would thus deprive Gloucester of the ownership promised by the contract.Footnote 139 Duran Duran was granted leave to appeal the decision in 2017, but the dispute was subsequently settled.Footnote 140
3.2.3.3.3 Undermining the 1976 Termination Rights: A Potential Closure of the Loophole.
Gloucester Place illustrates the potential for rightsholders to bypass the clear congressional intent for termination to be inalienable by simply enforcing a clause that prescribed the law and forum for any dispute. But the extent of this potential is not clear. In 2019 the US Second Circuit refused to uphold a similar claim by Bixio Music against renowned Italian composer Ennio Morricone,Footnote 141 even though the parties had agreed Italian law applied.Footnote 142 One of the Second Circuit’s reasons was that the contract provided worldwide rights were granted ‘for the maximum total duration permitted by the laws in force in each country in the world’,Footnote 143 and for US rights this was 35 years ‘plus such additional period as the assignor allows until the exercise of the option to terminate’.Footnote 144
Arnold (who adjudicated Gloucester Place) and Ginsburg (who was an advocate in Morricone) subsequently published a joint paper arguing that Morricone suggests that the termination right limits the scope of US copyright in a way that restricts what can actually be assigned by contract: ‘[I]t suggests that a U.S. copyright should be conceptualized as being inherently contingent upon the author’s exercise (or not) of her right of termination.’Footnote 145 In their view, this means any assignment of US copyright, even in a contract governed by English law, would not preclude the exercise of statutory termination rights.Footnote 146 We’re yet to see whether this will be accepted by future foreign courts, but the outcome in Morricone and the public argument raised by these two distinguished experts makes it certain that choice of law won’t simply be conceded the next time the issue comes up. In the meantime, creators are left sitting with the uncomfortable possibility that their termination rights are perhaps not as inalienable as Congress intended them to be.
3.2.3.4 Alienability under the 1976 Termination System: Balance Askew
The controversies we examined in Section 2.3 show the 1976 system’s approach to alienability is askew. On the one hand, it may be too restrictive on creators by forcing the bifurcation of their termination rights (which go to statutory heirs) and copyrights (which go to their estate). This could lead to situations where the statutory heirs reclaim copyright in ways the creators did not envision or desire. Whether this imposition on testamentary freedom is desirable is a discussion beyond the scope of this book, but the debate highlights the importance of crafting reversion systems with a clear vision of the obligations those inheriting reversion rights should have in relation to the exercise of those rights (which we integrate into our best practice analysis in Chapter 6).
On the other hand, the 1976 system has been shown to be vulnerable to rightsholder undermining through exploitation contracts. They have successfully disenfranchised statutory heirs by renegotiating rights grants with the estates, and in Gloucester Place they neutralised the termination right entirely by bringing a successful breach of contract action in the UK (albeit in a decision of a lower court, and where choice of law was not disputed). Generally speaking, reversion systems are instituted because contractual practices are undesirable; subverting creator/heir access to these systems maintains the undesirable status quo. Examples like these inform our recommendations in Chapter 6 for strict limits on contracting out, whether in domestic or foreign settings, when crafting domestic reversion mechanisms.
3.2.4 The 1976 Termination System Privileges Famous, Successful Creators (and Is of Limited Utility to the Rest)
3.2.4.1 A 35-Year Termination Time Limit Works Mostly for Successful Creators
Even if the previous three issues were rectified with the termination system, the 35-year period for most post-1977 copyright grants means most works are unlikely to be worth terminating.Footnote 147 This is yet another example of a poorly designed system relative to its creator-protective goals (and more broadly copyright’s access and reward goals). It is most likely to benefit the highly successful creators whose works are likely to retain value after 35 years. But those superstars are the artists (and works) that generally need termination the least. They have already been, and continue to be, handsomely rewarded from their creative labour,Footnote 148 and their works are more likely to be in demand (and therefore available to the public as knowledge and culture).Footnote 149 But for the vast majority of creators it is not likely to be worth filing to terminate after 35 years. Without termination intervening in the creative lifecycle, it’s likely that knowledge and culture will continue to be locked away from the public once their initial commercial shelf-life is exhausted.
3.2.4.2 The Formal § 203 Termination Process Has Been Used Sparingly, and by Successful Creators
Our recent world-leading empirical study into copyright termination notices, referred to above, adds further weight to this argument.Footnote 150 The study (the full methodology, results and limitations of which are set out in our published paper) was a whole-of-universe study of all termination notice records on the online US Copyright Office Catalog. Of course, a termination notice being filed doesn’t mean it occurs, because terminations can be challenged (as we showed above with sound recordings and works-made-for-hire).
Additionally, notices are not valid just because they are filed with the Copyright Office – errors could still invalidate them. The actual notices themselves are also not generally available for analysis online, meaning most researchers are reliant on the data uploaded to the Catalog (which contains flaws and inconsistencies).Footnote 151 And perhaps most importantly, the filing of a termination notice is only part of the termination system’s impact. The other part is the indirect impact of the prospect of termination on rightsholder behaviour, for which we have located only anecdotal evidence.Footnote 152 Further research using other data sources and methods (e.g. interviews with creators and rightsholders) could usefully begin to fill these gaps in our knowledge as to termination’s impact.
Despite their unavoidable limitations, the Catalog records still allow us to draw some preliminary conclusions: how little the termination system has been used and how it may skew towards famous creators.
Comparing our data against complementary Copyright Office registration data compiled by leading US copyright scholars Robert Brauneis and Dotan Oliar, we found that at most, the number of works subject to termination notices in our dataset was approximately 1.6 per cent of the number of copyright registrations in the years that would have corresponded to those notices (1978–85).Footnote 153 We then looked at the top 10 list of creators by the number of works they had subject to termination notices, and found several prominent artists: funk legend George Clinton accounted for 2.49 per cent of all works, pop duo Hall and Oates made up 1.02 per cent and country songwriter Harlan Perry Howard accounted for 1.58 per cent.
These figures were even more telling for the relatively few literary works that were subject to termination (n = 840): that list was dominated by heavyweights from romance (Debbie Macomber, Nora Roberts, Linda Howard, Jayne Krentz), horror and science fiction (Stephen King) and children’s fiction (Ann M Martin). Sweet Valley High author Francine Pascal was the most active author, with notices covering 305 of her titles – 36.3 per cent of all literary works with notices issued under § 203.
Our preliminary findings suggest a system where few creators see value in terminating their rights formally, while the top 10 lists feature industry stalwarts with ‘bestseller’ works – it’s great for them to re-exploit their works, but not for most other creators. This is consistent with our argument that a 35-year termFootnote 154 is a poorly designed, excessively lengthy time period. It also suggests lawmakers may have too readily jettisoned the shorter 20- and 25-year models proposed in the copyright revision hearings in the face of rightsholder lobbying, despite strident advocacy for creators by Karp and his ilk.Footnote 155
However, these findings inform our best practice analysis in Chapter 6 about crafting time limits that better help creators recapture and disseminate their works while they retain commercial (and cultural) value, particularly in light of the innovative new distribution models we outlined in Chapter 1.
3.3 Conclusion
In this chapter we laid out in stark detail just how ineffectively statutory reversion rights have been developed and operated in the US. Rightsholders fought hard to reduce the value of termination rights to authors, by blowing out the time before they can be exercised, insisting on onerous and expensive formalities,Footnote 156 and carving out as many works as possible. Once enacted, they embarked on a systematic campaign to reduce the value of the rights still further by insisting on the broadest possible definitions of ‘work for hire’, settling lawsuits with NDAs to avoid precedents and prevent the kind of knowledge spread that could help artists successfully navigate the legislation’s ambiguities, and attempting to subvert Congress’s intention to make termination rights inalienable by entering into new deals with heirs.
Congress itself is also to blame here, repeatedly passing poorly designed laws: failing to specify key inclusions/exclusions, mandating complex formalities and enshrining a lengthy pre-termination waiting period that renders it near-irrelevant for non-superstar creators. And the role of the courts can’t be overstated, as the Supreme Court and various circuits of the Court of Appeals handed down rulings that opened the door for rightsholders to undermine termination’s intent through their contracting practices.
Nevertheless, the lessons we have learned from the sinking ship that is the US termination system will usefully inform our best-practice reversion principles in Chapter 6. And hope can be drawn from the fact the record industry attempted to steal back the termination rights of recording artists via their clandestine, dead-of-night manoeuvre, showing that they believe that meaningful reversion rights have real potential to shift money and power from labels to artists. While the US system is far from ideal, that should give hope to creators that there is a better way.
Open access