Sunday, April 30, 2006

Hypothesis on Plagiarism and Copyrights [corrected]

Hypothesis: Incidents of plagiarism increase as the length of copyright increases.

Reasoning: Plagiarism is an academic crime. It is prevented by attribution. Copyright infringement is a legal crime. It is prevented by licensing or by building off works from the public domain.

Copyrighted works eventually become part of our general culture. Copyright is a bargain between the needs of authors to be compensated for their creations and the public good of having their creations available for free use. The full argument is made in Lawrence Lessig's Free Culture.

Currently, a copyright lasts until 70 years after the author's death or 95 years from publication or 120 years from creation if the author is a corporation. Congress can extend (and has extended) the terms of copyrights each time they are in danger of expiring. By the time any works created today enter the public domain, most will be culturally irrelevant. The majority of works written after 1923 may be assumed to be copyright protected in the US.

The length of the copyright term encourages attempts to “cheat the system,” by removing attributions and hoping that nobody will notice. Building new works out of older works is part of how our culture expands and evolves. Disney makes movies out of public domain novels and fairy tales, updating them and bringing them new cultural relevance. They also lobby Congress to make sure that these new creations are never cycled back into the public domain.

As one example of the benefits that come to a work from its being in the public domain, see Dracula 1897, a LiveJournal-based "real-time" edition of Bram Stoker's Dracula, enhanced for a medium that Bram Stoker never knew would exist. (hat tip: Lawrence)

Technorati tags: ,

In the case of an institutional author, it's 95 years from publication or 120 years from creation, whichever comes first. 17 U.S.C.A. ยง 302(c).

Reading this always makes me vaguely wish I were an institutional author, though of course 1) I have no possible use for 120 years of copyright protection, and 2) I can at this age hope to do better anyway.
Though I agree with parts of your hypothesis, especially that lengthier copyrights encourage infringement, there is an element of this case that I disagree with.

Plagiarism is an ethical crime that is separate from copyright infringement. It is possible to infringe a copyright withoug plagiarizing and to plagiarize without infringing anyone's copyright.

The age of copyright has no effect on the ethics of attribution. If I were to use quotes from Plato in my recent essay, I would be fully expected to cite my source. To not do so would be a lie.

Just because in the above example I didn't infringe on Plato's copyright doesn't mean that I did not lie, mislead my readers and misrepresent my work. Copyright infringement is a crime against the work's creator, plagiarism is a crime against the creator and the reader.

I fail to see how the ever-lengthening nature of copyright can affect the morals of attribution. Academic honesty is always expected, whether or not the word is protected by copyright.

These are just my thoughts though, the world may disagree...
Thanks for your comment. It gives me a chance to clarify something that may not have been made very clear.

Plagiarism is an ethical crime that is separate from copyright infringement. It is possible to infringe a copyright without plagiarizing and to plagiarize without infringing anyone's copyright.

I agree. But, my hypothesis is proposing that one is more likely to see a direct attribution such as "This work derived from ..." if the original work were in the public domain. Ideally, one might also expect to see fewer attempts to make it harder for the material's original source to be found (changing small details). In reality, original sources can sometimes simply be lost to the author.

I'm not sure that I believe Kaavya Viswanathan/Alloy Entertainment about this, because both works are relatively recent. I have an easier time believing that simply forgetting where he had originally heard the material might be the case for William Swanson, whose source was from 1944. In the latter case, the plagiarism charge can be remedied by rewriting with proper attribution, but the copyright charge cannot.

Incidentally, I don't think that copyright has any effect on low-level academic cheaters (like undergraduates writing papers for courses), who really have little concern about whether their work is publishable, and have no economic reason not to cite their sources.
Hi, hope you and Elf are well.

I'm wondering whether your argument accounts sufficiently for the so-called "idea/expression dichotomy." As you probably know, copyright protects expression, but not the underlying ideas expressed. Protection of ideas is within the sole province of patents. So, copyright should not inherently have anything to say about exchanging ideas, and, as noted below, fair use permits most academic quotations. Admittedly, these distinctions become harder to parse in the context of source code, as we've discussed in the past.

While I agree that long(er) copyright terms can incentivize certain types of infringement, I think such incentives apply far more often in the commercial context. It would be awfully hard for the author of an academic article to be held infringing merely by virtue of quoting someone's work for discussion purposes. (Of course, this assumes that we're talking about reproducing a contextually reasonable portion of the third-party work, as opposed to, say, a competing edition of it.) Even with the proliferation of aggressive copyright owners, I'm not aware that anyone seriously disputes this.

I'm not saying copyright owners don't overreach (they surely do), but properly applied, even the existing statutory regime allows for the exchange of ideas, especially in the academic context.

Also, I second Jonathan Bailey's point that clear failure to attribute is wrong completely independent of copyright. I don't agree with the assertion that copyright discourages academic attribution. This result could occur only if there is significant (and widespead) misunderstanding of the scope and nature of copyright and fair use.

Are you saying that potential citations don't happen because of fear that a publisher or author would sue for discussing the now-uncited work, and that the non-citation is a way to lower the risk of being noticed? If so, I don't buy the argument, especially since (as you noted), the hypothetical non-citing author is most likely an academic seeking publication in advancement of a career. Isn't the downside of non-attribution far greater than the risk of such an outlandish copyright claim?

This hypothesis does not apply to academic works. Academia has a citation standard much stronger than copyright. Ideas and expressions are both cited in academia. There is no expiration date for citations.

For commercial and other popular works, the likelihood of source citation decreases both with age of the original work because the original source may be "lost to memory," (becomes part of the general culture?) and decreases with length of copyright term. I'm suggesting that a popular/commercial author might choose to make slight changes or obscure the nonoriginal material, instead of acknowledging that something is derived. Also, commercial/popular works do not have the same citation standards as academia does.

The Internet has increased the availability of fanfic: non-commercial works that derive from commercial works, but have the same potential copyright issues as commercial works.

Software programmers are sometimes expected to work in a vaccuum. Someone who worked on a particular code set for Unix, for example, should not work on the same code set in Linux because there is a fear that the Unix knowledge might "contaminate" his approach to the problem and lead to copyright issues. The same is not (and cannot be) true for other creative works. All authors of teen novels are expected to have previously read other teen novels, for example.
As you note, citation is not typical or necessarily appropriate in works of fiction. The case law is clear that stock plot elements or "scenes a faire" are not protectable. Permission of a copyright owner is required only to the extent that copyrightable elements of a work are used beyond the boundaries of fair use, but the vast majority of the time this is not be the case.

It's true that fan fiction poses more copyright issues, mainly because specific well-defined characters and very detailed plots are used for fan-generated prequels, sequels, etc. Even so, many major copyright owners will tolerate such works to the extent they remain purely noncommercial. This is self-interested on their part, of course, as they see most such works as non-competitive with their market and encouraging further interest in the original work. Even very aggressive rightsholders in juggernauts like Harry Potter and Stars Wars have had to strike some compromises in this area.

On the teen novels point, just absorbing and using a set of conventions and techniques will generally not pose copyright issues.

Finally, with the notable exception of the Dracula example, some of what you're calling "derivative" here may be so only in the colloquial sense, not in the statutory sense of a "derivative work." Extentions of popular copyrighted works into new media actually are derivative works, the creation of which is clearly exclusive to the copyright owner. I don't see that as inherently probematic.

Although I agree with the policy argument for shorter copyright terms, I wouldn't lose much sleep over missing out on extensions into new media. Rest assured, the small percentage of works with great commercial viability will be exploited to the max during their copyright terms in multiple media by those who hold rights in them. Occasionally, good or even great art is created this way.

I wouldn't lose much sleep over missing out on extensions into new media. Rest assured, the small percentage of works with great commercial viability will be exploited to the max during their copyright terms in multiple media by those who hold rights in them

The works that get completely lost are the ones that have either become abandoned by their owners or that their owners cannot be found, are dead and have no obvious heirs, or have multiple potential copyright owners, not all of whom can be found. Just because a work's current commercial relevance is not yet recognized, that does not mean that its cultural relevance or the interest level in it is zero. These kinds of concerns prevent digitally archiving old works, using new technologies to enhance catalogs (see the Google Archive debate, for example), and restore decaying old films.

But, it's all irrelevant to the hypothesis at hand.

Let's take another case related to Dracula as an example. The movie Nosferatu was clearly derivative (in the common speech sense) of Dracula. It trivially renamed all the characters and some of the places so that it wouldn't be legally derivative of it. Nosferatu was still made while the (relatively short) copyright on Dracula was in effect. After the copyright on Dracula expired (1962), there were many derivative works of it (common speech sense) that were also derivative in the copyright sense. At that point, it didn't matter from a legal perspective anymore. From a plagiarism/ethical perspective, isn't it better to acknowledge something's literary history, rather than trying to obscure it?

In the Kaavya Viswanathan case, whoever did the copying effectively "open sourced" the search for the novel's literary roots. They were found because someone recognized identical phrasing. What will probably happen next is that the copied expressions will be trivially rephrased and the book reissued. OK from the copyright perspective, but perhaps not from the plagiarism/ethical perspective.
"The works that get completely lost are the ones that have either become abandoned by their owners or that their owners cannot be found . . . ."

You're absolutely right, and there's an apt name for this problem -- "orphan works." A legislative fix is underway. The Senate Judiciary Committee instructed the Copyright Office to report on this issue, which it has recently done fairly extensively; see: A bill will probably be introduced at some point.

Bram Stoker, it should be noted, was not original in his basic plot, which was (colloquially) derived from centuries-old folktales. It's been 15 years since I read the book and I've only seen clips of the film, but it may be that many of the elements used in Nosferatu were not original or properly protectable by Stoker's estate anyway. I don't know the details, and copyright jurisprudence was fairly sparse at the time.

Of course, public domain works can be copied freely. As you note, there is no statute of limtations for plagiarism. However, I think the concept is often inapposite for fiction.

What does it mean to "acknowledge literary history" when drawing on public-domain works or unprotectable stock plot structures in a work of fiction? I don't think it's unethical to write a novel that reworks a public domain plot or uses unprotectable elements. It would be upright, scholarly, and thorough to include a preface with acknowledgements, but Shakespeare was (by all accounts) a serial plagiarist under this standard. I think it's possible to not cite fictional precedents and still not be a plagiarist. In my view, it's a kind of non-copyright fair use analysis: how much copying was done, how much copied material was original expression, and is the original somehow transformed in the new work?

I don't know many details of the Kaavya Viswanathan case either, but I gather it involved recurring near-verbatim copying of tell-tale expressive phrasing of a work published just a few years ago. That's a trifecta: copyright infringement, plagiarism, and unethical. Oh, and breathtakingly *stupid* to boot.


I think we agree about most of this.

Plagiarism is unethical (and frequently stupid). Copyright infringement is illegal. Not all plagiarism is copyright infringement, and vice versa.

From a creative perspective, few works are truly completely original. And, courts have recognized that by disallowing methods-and-concepts arguments with regard to copyright infringement in literary works (although the story seems to be different for computer software... go figure).

The limited question we seem to disagree about is this one: From a purely ethical standpoint (copyright/public domain aside), which is better for a work of fiction: to copy an expression or to copy an expression then make subtle changes to purposefully obscure its origins.


Aside: today's Boston Globe says that the publisher of "Opal Mehta" will not rerelease the work with minor changes as originally planned. Apparently, there was more verbatim copying from more recent works than had previously been known.
Globe article link.
"[w]hich is better for a work of fiction: to copy an expression or to copy an expression then make subtle changes to purposefully obscure its origins?"

My personal view is that it's never ethical to slavishly copy text and present it as one's own. Doing that and tinkering slightly to obscure the origin is even worse. I view such conduct as fundamentally different than using more abstract forms of literary expression (like an original plot twist or character) as a point of departure.

Post a Comment

<< Home

Links to this post:

Create a Link