Home > ILN IP Insider > Entitled To Copyright Erasure?: A Fair Use Search For A Derived Yet Transformational Work

Entitled To Copyright Erasure?: A Fair Use Search For A Derived Yet Transformational Work

The title is the first thing the reader sees or hears …—and getting it right is the single most important … decision you’ll make. The title forms the basis of the reader’s judgment ….” Tucker Max

For those of you who may read my past ILN posts, you will not be surprised that I subscribe to the Tucker Max approach:  Make the title attention-grabbing, memorable and searchable, informative, easy and not embarrassing to say, and short.  OK, maybe not the last prong so much, but I nonetheless have one regular reader who has noted that he thinks that I have “a real knack for catchy titles!” I will not debate the merits of that opinion right now.

Instead, I will just note that I generally let the title stand on its own, and do not deconstruct or explain it.  But here, I am actually going to explain expressly a bit about the title, and what I meant it to convey.  There is a lot packed into the possibilities of this title and each issue may not be as apparent as it is in some of my earlier titles, such as those with obvious puns (like  “…this is my life”: Corporate Biography, Moral Rights & Being Slow To Berne” or Maybe Axanar Could Klingon To Its Fair Use Defense In A Parallel Copyright Universe)  or specific revelations of focus (like “…as best as your interests don’t conflict with mine”: Lawyers Fighting Over Intellectual Property)… This title –“Entitled To Copyright Erasure?” — could have several meanings.  It could signal the notion that copyrights should not exist, and that the ability to copyright anything should be erased.  Alternatively, it could suggest that adopting certain rules could eliminate the practical ability to protect certain compositions from creative reuse even if there is no express desire to create a fully open-source world.  But it could also encapsulate the question actually examined here, which is:  “Can ‘erasure poetry’ be copyrighted independent of the work on which it is based and by someone other than the creator of the work from which it was drawn?  Naturally, our title here also plays on the multiple possible meanings of “entitle,” from “to name,” to “to allow,” to “to deserve,” and so on.

  1. What is Erasure/Found/Blackout Poetry?

Of course, to understand that question, one must understand what erasure poetry is:

Erasure is a form of found poetry or found object art created by erasing words from an existing text in prose or verse and framing the result on the page as a poem.[1] The results can be allowed to stand in situ or they can be arranged into lines and/or stanzas.

Here is a nonce example using text from the November 2003 version of the English Wikipedia’s Main Page:


and free

we started

and are



you can

right now

Writers/artists have adopted this form both to achieve a range of cognitive or symbolic effects and to focus on the social or political meanings of erasure. Erasure is a way to give an existing piece of writing a new set of meanings, questions, or suggestions. It lessens the trace of authorship but also draws attention to the original text. As with any allusion, interpretive questions include:

  • What is the attitude to the original work? Does a gesture celebrate, denigrate, subvert, or efface the source completely?
  • What is the process for selecting erasures? One can erase intuitively by focusing on musical and thematic elements or systematically by following a specific process regardless of the outcome.

[Erasure (artform), Wikipedia]

If one wants a source beyond Wikipedia, the Academy of American Poets also explains erasure poetry and provides examples to review.

This is something I discovered myself only in late January 2021 by reading about it in the New York Times.  A few weeks later, inspired to tie that article together with a thought-provoking Wall Street Journal article on the state of Catholics in the United States or an image-laden report of the coup in Myanmar found in the next day’s New York Times, I tried my own hand at erasure, or found, poetry.

Of course, I was a little late to the party.  Evidently, according to those in the know, “Globally” as early as “2016,…Blackout poetry,” another of its monikers, was “revolutionizing the world of poetry but, at the same time, it is blurring the line between creationism and plagiarism even more.”  In fact, when I looked into its history, which some have called brief, I learned that by 2009, it was already ever-present:

The poetry of erasure is taking place all around us. Underneath the pavement, behind newspaper headlines, on paste-layered billboards and graffiti-laden walls, our communal landscape is continuously peeling away and papering over itself. Its very surface is a living thing in flux between the dueling processes of decay and renewal, driven in the name of progress to adapt to the shifting contextual demands of culture or be replaced, removed, re-imagined. While this process can, at times, be artificially postponed, nothing escapes its effects forever. This world demands of its denizens a constant and vigilant revision of form.

[Travis Macdonald, A Brief History of Erasure Poetics (2009) (emphasis added)]

In fact, according to The History of Blackout Poetry, “the lineage of blackout poetry or redacted poetry actually traces back to the 18th century,” and connects to a neighbor of Benjamin Franklin; “Through invention and gradual, but steady, modeling and remodeling of the art form, found poetry evolved and grew. Thanks to these innovative artists, the groundwork for the various forms of found poetry we know about today, was laid in the early days.”

And it seems to be getting more popular, as there is even in 2021 a Master Class on the topic of blackout poetry.  Likewise, in 2019, Mass Poetry sponsored an erasure/blackout poetry contest to mark National Poetry Month.  It has also become the stuff of museum exhibits, such as the one that notes that “Destruction is also creation. And this [exhibit] is all about a form of destruction that is both visual art and lyric: erasure poetry…Travis Macdonald[, a commentator quoted above,] argues that time and the elements were the first erasure poets since they have effaced sacred scriptures, stone engravings, the works of Sappho, and many, many other texts.  Nonetheless, there can be beauty and expression in the act of erasure.”  There was even for a time, The Found Poetry Review, which has numerous poems to be examined.  That it came into and out of existence without me becoming aware of the existence of found poetry gives me pause.  But my interest and curiosity, tardy but in no way tepid, remains.

Since early February 2021, I have been trying to compose at least one found poem per day, and collecting them in a document entitled Poetica ArchaeologicaPoems Uncovered, Discovered, and Revealed Both acknowledging and apologizing to Horace and the Ars Poetica., here is what I had to said there about what we face today:

I also greatly appreciate the efforts of the reporters and columnists who wrote the articles and columns that captured my attention and allowed the discoveries here.  I believe, and trust they believe, the use to which I have put portions of their texts to be fair and transformative rather than merely derivative.  Whether the transformation has a value is for others to determine.

And, as for the title, it echoed to me.  Ars Poetica is, of course, the famous poem by the Roman poet Horace in which he advises readers and other writers on the art of writing poetry and drama.  I went with Poetica Archaeologica as the title of this volume, which loosely translated to “archeological poetry” or “poetry through the archeological,” though the closest Latin phrase for archeology is actually scientia antiquitatis.  To the English-speaking eye and ear my phrase captures better my intent, as Poetica Archaeologica sounds, at least to me, like poems you dig up or dig out, or uncover from, what is before you—in other words, it sounds like found poetry (the name which I prefer to erasure poetry, as the positive is what I find in other’s broader words, not in what I erase from their fine works).

I was acknowledging those who created before me, but also claiming that my use of their material was fair and transformational.  That was not an unproblematic claim, as I do not tend to mask or move away from the source material’s subject, as some blackout poets do to enhance claims to fair use.  Mine, instead, are a way of reacting to current events, news, and views.

B. How Would The Law Look At Found/Blackout/Erasure Poetry?

There are certainly arguments that I was not any freer from a charge of infringement than a rapper or DJ using snippets of others’ music.  But as we have seen, not every such suit for such infringement is successful and the answer may depend on a variety of factors, including whether the suit is one brought in the United States or elsewhere, as Leah Hurvitz notes. Beyond that, in found poetry, I may be plucking single words, the equivalents of single notes, or a phrase, the equivalents of a chord, separated by many other notes and chords.  How is that possibly unfair to an earlier author who merely picked his or her words from a longer list, whether a remembered lexicon or an actual dictionary?  Why is my decision to limit the words available to me to choose from any different than an artist’s choice to limit a palette to only a few colors?  When a sculptor chisels away wood or marble to project his or her intended creation it is not usually considered less his or her artwork because the material may have been from a block or other form already earlier shaped by another.  It would seem that what could be true for a block of wood could be true for a block of words.   As you can see, I tend to agree with those who say “Erasure is an expression of self, just like all forms of art. It is not plagiarism to draw from the muse of other authors because erasure alters the meaning of the text.”

But that remains somewhat undetermined legally.  It does not appear that there has yet to be a US case involving claims that a found or erasure poem was infringing or was a mere derivative work of the earlier source material.   Cases may be on the way soon, however, because some published authors of erasure poetry books (there may be a market for my stuff, yet who knows) are lauded as revealers of “the art of every day words” at the same time that doctoral dissertations on the ontological meaning of erasure poetry are leading to advanced degrees. And there are people focused on the issue.  For instance, the Center for Media & Social Impact has issued a Code of Best Practices in Fair Use In Poetry that notes in part that “erasure” and “use of ‘found” material” from “ephemeral journalism” as “a point of reference” demands that fair use concerns be considered.  As CMSI notes, such use should only occur without license if “quotations are re-presented in poetic forms that add value through significant imaginative or intellectual transformation.” As one commentator notes, some of the inherent value in the erasure/blackout poetry process is its added depth and complexity:

Working within another author’s framework adds another layer of depth and a challenge but it is not a necessity to the content of the poem created. The words that makeup the poem are just words that obviously could be written elsewhere, the visual effect and the challenge posed is what sets blackout poetry apart from other styles and makes it a kind of erasure literature.

[“On Erasure and Copyright Law,”  Art is 99 Percent Robbery (2016)]

As Samuelson has noted, there is an appreciable loss to the public where “lawful reuses” of an early author’s text are avoided for fear of an infringement charge—“Whenever an author forgoes the opportunity to reuse portions of another author’s work out of fear that the use might be challenged as infringing, there is a loss not only to that author but also to the public. The public cannot benefit from the insights that the second author’s reuse of a first author’s work would have enabled. There is, moreover, some loss to freedom of expression and to access to information when lawful reuses are forgone.”  To avoid such losses, she prescribes limitations and exceptions to copyright enforcement, which others like Renee Hobbs have supported with specific reference to blackout poetry as fostering public interest in learning, the creative process, information dissemination, and self-governance.

Though there are not decided cases specific to erasure poetry, there are copyright infringement tests that deal with copying less than every consecutive word of work or passage within a work, and these may not be as liberal or liberating as Samuelson would suggest.  For instance, there are two existing theories to consider.  There is the “comprehensive non-literal similarity test,” under which infringement occurs when the non-copyright holder’s work is such that “’the fundamental essence or structure of one [copyrighted] work is duplicated in another.’”  Castle Rock Entertainment v. Carol Publishing Group, 150 F.3d 132, 140 (2d Cir. 1998) (quoting and citing 4 Nimmer On Copyright §13.03[A][1] at 13-29; §13.03[A][2] at 13-45 (hereafter “Nimmer”)).  There would be strong arguments to make that all fundamentally similar works would be infringing under the “comprehensive non-literal similarity test,” as an erasure poet’s effort to “restate” or “rephrase” a news report in poetic form might nonetheless usurp the article’s fundamental essence.  Of course, to the extent that structure beyond subject matter theme and word order matters, this test might be inconclusive.  Another test, the “fragmented literal similarity” test, is one covering situations “in which small bits of specific expression are copied but the overall structure is not,” as Adam Steele notes.  This test has “won wide scale judicial acceptance” along with the earlier noted test, as ways to articulate “sometimes alternative, sometimes complementary methods of demonstrating the existence of a probative and substantial similarity sufficient to establish infringement.”  Nimmer, supra, at §13.03[A][3] at 13-51.

It will come down often to how much of the earlier work is copied, which is a good result from the perspective of the erasure poet, whose entire work will, by definition, be made of material taken from the original work:

[N]o easy rule of thumb can be stated as to the quantum of fragmented literal similarity permitted without crossing the line of substantial similarity. The question in each case is whether the similarity relates to matter that constitutes a substantial portion of plaintiff’s work—not whether such material constitutes a substantial portion of defendant’s work.

[Nimmer at § 13.03[A][2][a].

Still, some courts will take note of how much of the infringing work is made of matter from the earlier work. See Craft v. Kobler, 667 F. Supp. 120, 128 (S.D.N.Y. 1987) (Supreme Court precedent makes clear that mentioning in a non-exclusive list of factors an approach using the copyrighted work as the denominator  “does not mean that [that]word count fraction. . .is the only relevant approach to the issue of the substantiality of the appropriation”); see also Harper & Row v. Nation’s Enterprise, 105 S.Ct. 2218, 2233-34 (1985) (finding infringement where 13% of the infringing work was compromised of quoted passage from copyrighted work).  Ultimately, it may come down to the common sense of the ordinary observer.  Nimmer, at §13.03[E][1] at 13-78 (courts have generally applied “the ordinary observer or audience test” to determine issue of “substantial similarity”); see also Nimmer, at §13.03[A][1] at 13-29 (test necessarily involves “line drawing” and that line be drawn “somewhere between the one extreme of no similarity and the other of complete and literal similarity” thereby “marking off the boundaries of ‘substantial similarity”).  Hence, substantial similarity can be found in analysis, without reliance on expert testimony, of the intrinsic comparison of expressions used in light of whether an ordinary reasonable person would understand them to convey the same “total concept and feel.”  Sid & Marty Kroft TV Productions v. McDonald’s Corp., 562 F.2d 1157, 1161-1164 (9th Cir. 1977).

But, the question remains–does fair use allow for a derived transformational work free from copyright claims of the author of the original work?  Traditionally, a “derivate work” is defined under Section 101 of Title 17 as “a work based upon one or more preexisting works, such as an abridgment, condensation, or any other form in which a work may be recast, transformed, or adapted. A work consisting of editorial revisions, annotations, elaborations, or other modifications which, as a whole, represent an original work of authorship, is a ‘derivative work.’”  Such derivative works can be claimed by the creator of the original work in many cases, as the Copyright Office makes clear.

The question of whether a derivative work always infringes the original usually comes down to fair use, which either means a ratio/feel/structure analysis like that noted above, or a conclusion that the new work is transformative.  According to the Copyright Office, “’transformative’ uses are more likely to be considered fair.  Transformative uses are those that add something new, with a further purpose or different character, and do not substitute for the original use of the work.”  Transforming prose into poetry is certainly somewhat transformative in all cases, and as Rachel Koupf has noted, can illustrate new creation quite vividly:

Procedures like erasure poetry are highly constrained yet highly unpredictable, opening up innumerable possibilities even while limiting the procedures available to deletion and maybe reformatting. While erasure demands that one not stray from the source material, it does not impose a determined structure upon one’s treatment of that material: the eraser must still determine which pieces to cut, a process which probably entails considerable reading, interpreting, and experimenting, often all at once. Along the same lines, the procedures that govern found poetry require that one rely upon preexisting texts to build a poem, yet what those texts are and how one combines, arranges, and revises them (again perhaps not copying with complete accuracy) are undetermined by procedure alone. At both ends of a spectrum marking constraint, in both looser and tighter procedures, there is interplay between personal preference and impersonal text-generation. Procedures shape writing into an orderly yet open system.

But the question of whether remaining within the original text, no matter how distilled, is sufficiently transformative to preclude infringement is likely limited to a case-by-case analysis. See Warner Bros. Entertainment, Inc. v. RDR Books, 575 F.Supp.2d 513 (S.D. N.Y. 2008) (Harry Potter encyclopedia found to be “slightly transformative” but this was not enough to justify a fair use defense in light of the extensive verbatim use of text from the Harry Potter books).  In the end, the best one can likely do is suggest a fair use checklist, explain how to use it, and suggest one emulate Austin Kleon’s optimistic views as to the impact of fair use rules, rules that say are “legal constraints” that “can actually be turned into artistic constraints. Rather than limiting my creativity, these constraints make the poems better.”

Because courts have not addressed erasure, found, blackout poetry (or seemingly its cousin golden shovel poetry, which is about to become a new pastime of mine), it has not gotten as much attention from traditional legal commentators as other forms of what Professor Fisher has somewhat unfortunately dubbed “appropriation art.” (I say “unfortunately” because it seems to bring an unnecessarily negative connotation to such creations).  A number of traditional legal comment fact have gotten together to assess whether the so-called Obama Hope poster was or was not a fair and transformative use of a photograph by an AP reporterSee William W. Fisher III, Frank Cost, Shepard Fairey, Meir Feder, Edwin Fountain, Geoffrey Stewart & Marita Sturken, REFLECTIONS ON THE HOPE POSTER CASE, 25 Harv. J. L. & Tech. 243 (2012), and this provides a useful lens for assessment of found poetry.  Though the entirety of the lengthy article is worth reading, the last thirteen pages by Professor Fisher lay out specific approaches that would support arguments and practical reforms as to “Why Copyright Law Should Not Proscribe Appropriation Art,” and, my criticism of that single word aside, the analysis in those pages is one I would endorse,  Id. at 313-326, and in a sense have already endorsed going back to earlier posts suggesting expanded protections for pastiche and seeking to continue to analyze fair use flexibility. Simply stated, Fisher suggests “giving artists more freedom to make creative uses of copyrighted materials,” and bases it in part on some shifts on how the law of “originality” and the concepts of scènes à faire should apply to found art and what is “necessary” to create “appropriation art” or something similarly inspired and differently named. Id. at 318-321. He also documents the evolving nature of transformative use cases toward standards tying transformation toward protecting use of past works that is “socially beneficial” or  has a “purpose … different from that of the [earlier] work.”  Id. at 322.  Those are supported by the already existing case law he cites and practitioners, though he would push further toward what he considers a “more straightforward” approach of providing a “safe harbor” for any use that is “creative,” which he defines as anything that “either constitute[s] or facilitate[s] creative engagement with intellectual products.”   Id. at 323.  I’ll embrace that approach for my own found poetry, but advise clients of the possibilities in the cited case law he mentions.

Deciding in my drafting to so advise my clients may have been prescient.  Although I see much merit in the Fisher philosophy of fair use, the United States Court of Appeals for the Second Circuit rejected an artistic intent or purpose test for fair use on March 26, 2021, in The Andy Warhol Foundation v. Goldsmith, i.e. after I had first written a draft of the preceding paragraph.  In the recently-decided case, which the Warhol Foundation had won below, the appellate court rejected the claim that Warhol’s uses of certain photographs of Prince by Linda Goldsmith (with photographs and Warhol works depicted in Slip Op. at 7-9) were transformative, stating that:

[W]hether a work is transformative cannot turn merely on the stated or perceived intent of the artist or the meaning or impression that a critic — or for that matter, a judge — draws from the work. Were it otherwise, the law may well “recognize[e] any alteration as transformative.”

* * *

Although we do not hold that the primary work must be “barely recognizable” within the secondary work… the secondary work’s transformative purpose and character must, at a bare minimum, comprise something more than the imposition of another artist’s style on the primary work such that the secondary work remains both recognizably deriving from, and retaining the essential elements of, its source material.

[Slip op. at 26, 28 (quoting Nimmer, § 13.05(B)(6) at 26)]

Because the appeal court concluded that “any reasonable viewer with access to a range of such photographs including the Goldsmith Photograph would have no difficulty identifying the latter as the source material for Warhol’s Prince Series,” the Second Circuit rejected the claim of fair use. Slip op. at 55.  In doing so, the Second Circuit rejected what seemed like a fairly persuasive amicus argument (illustrated with photos at pages 9-26)  that “recognizable similarity in expression is not, in itself, substantial similarity,” that only limited aspects of the Goldsmith work was protectable, and differences in aesthetic have been determinative in previous cases.

So it seems, for now, we are left with the weighing ratios of content and recognizability from work to work without any bright-line rule or even directional percentages, and wondering why protectable art requires an undefined “something more” than the imposition of an artist’s own style onto materials used in the creative process. Slip op. at 19 (noting that adding “’new aesthetics’ by placing the work ‘in a different context,’ …or by any other means is the sine qua non of transformativeness,” but not every second work that adds that “to its source material is necessarily transformative.”). Indeed this debate continued almost immediately, as both the United States Supreme Court and the Second Circuit addressed fair use cases in recent days, and gave the second artist’s creative “purpose” a much greater continuing role in determining transformativeness than Warhol would suggest was appropriate. First, the Second Circuit itself in Marano v. Metropolitan Museum of Art (decided April 2, 2021), held in a case about a photograph of “Eddie Van Halen playing his ‘Frankenstein’ guitar” that “whether the use is ‘transformative’…constitutes the ‘heart of the fair use inquiry,’” and that the purpose the defendant sought to serve was a primary factor in determining fair use.  Summary order at 2-5 (allowing “the Met’s ‘copying the entirety of [the Photo]’”).  Then, on April 5th the US Supreme Court decided in Google v. Oracle that “fair use” is an “equitable rule of reason” requiring “judicial balancing” of “the sometimes conflicting aims of copyright law”  so that copyright does not “stifle the very creativity which the law las was meant to foster.” (Op. at 13-15.)  Neither Marano nor Google cite Warhol, though each address at some length, including in a Supreme Court dissent, the role “purpose” plays in assessing whether a use is “transformative.” See, e.g., Google, Slip op. at 24-28, Dissent at 15-17.  So the debate goes on.

C. Isn’t It More Interesting To See How Found Poetry Views The Law & Its Institutions

As unsatisfying as that might be from a bright-line legal rules sense, it need not overlying curtail the creation of vital, meaningful erasure, or blackout, poetry.  For example, using public domain documents like the Declaration of Independence or other legal documents largely avoids these issues.  It also may portend more moving, revealing works, which may have been what motivated the Mass Poetry contest noted above.  That contest specifically sought “erasures of political documents ‘including the US Constitution, Bill of Rights, Letter From a Birmingham Jail, and the speech ‘Our books and our pens are the most powerful weapons’ Malala Yousafzai delivered to the UN,’” noting that “Originals of those documents, in various sizes, are available for download through the submission website.”  Indeed, erasure poetry uses power and precision to find hidden meanings, including in some works like Tracy K. Smith’s short poem Declaration and Kenneth Goldsmith’s book-length work Sports.  In fact, Smith has been lauded using found poetry to reveal history in all its “heartbreaking beauty.”  Declaration has been noted to be “the textbook example of how an erasure poem is about both what is on the page and what is missing with how it makes its statement…”

As the Library of Congress has noted for teachers, while “[r]aw material for found poems can be selected from newspaper articles, speeches, diaries, advertisements, letters, food menus, brochures, short stories, manuscripts of plays, shopping lists, and even other poems,” using primary historical sources can deepen the learning experience by having students concentrate on:

analyzing and interpreting historical primary source content, in synthesizing the information, and in making personal connections with history, ask them to articulate their understanding. …[by] the writing of “found” poetry. Using rich primary source texts, students select words that allow them to retell the historical content in poetic form… Careful observation and analysis of an image will provide historic details and to historical understanding and can spark writing ideas. Careful observation and analysis of an image will provide historic details and supportive information, and may even offer rich language for the found poem. … Notations about objective and subjective observations will be invaluable when they begin to retell history through their own poems.

[Library of Congress, PRIMARY SOURCE SET: Found Poetry]

A Delphic position for the Copyright Office—in promoting erasure poetry, is that office implying an expert agency view that erasure poetry constitutes a fair use?  Or, in emphasizing the use of “historical primary source content” as the basis for the poems, is that office suggesting one avoids infringement liability only by using sources in the public domain?  Our mystery continues.

Having gone through the exercise of writing this piece and applying the checklist to my own poems, I am comfortable that my erasure scribblings on current news events over the last month and a half amount to fair use, and probably only fair poetry.  But it can remain an enjoyable pastime for me or others, as we celebrate in National Poetry Month.  It also led to my discovery of Tracy K. Smith, which made it more than worthwhile.  Her words—wherever she found them–will not be erased from my memory.