The Unoriginal Sin: Differences Between Plagiarism and Copyright Infringement

Plagiarism, which many people commonly think has to do with copyright, is not in fact a legal doctrine. True plagiarism is an ethical, not a legal, offense and is enforceable by academic authorities, not courts. Plagiarism occurs when someone – a hurried student, a neglectful professor, an unscrupulous writer – falsely claims someone else’s words, whether copyrighted or not, as his own. Of course, if the plagiarized work is protected by copyright, the unauthorized reproduction is also a copyright infringement.

By the way, I cribbed every word of that first paragraph from Black's Law Dictionary, which, in turn, was quoting (with attribution) from copyright guru Paul Goldsteins' book Copyright's Highway.   If I hadn't bothered to mention Professor Goldstein, I would have been guilty of the sin of plagiarism, but not the actionable offense of copyright infringement.  (As a matter of copyright law, my quotation from Goldstein is, I trust, safely within the bounds of "fair use.")

In the words of Merriam-Webster Online, "plagiarism" is the act of steal[ing] and pass[ing] off (the ideas or words of another) as one's own; use [of] (another's production) without crediting the source."   By contrast, "copyright infringement," occurs "when a copyrighted work is reproduced, distributed, performed, publicly displayed, or made into a derivative work without the permission of the copyright owner."  (See the definitions section of the U.S. Copyright Office website.)  Similar, but not exactly the same.

Many acts of plagiarism are not copyright infringements.  And many acts of copyright infringement do not arise from plagiarism.  This post seeks to clarify the differences between the two transgressions -- differences that writers (and sometimes judges) often tend to blur.  Indeed, even Professor Goldstein engaged in a bit of blurring; it would have been more precise to say that "if the plagiarized work is protected by copyright and the copying is substantial, the unauthorized reproduction may sometimes also be a copyright infringement."

Copyright infringement is a legal offense against property rights, whereas plagiarism is an ethical failure to honor one's intellectual forefathers and foremothers.  Plagiarism does not amount to copyright infringement unless (a) the plagiarist has republished copyrightable expression of another, and (b) the amount of copied expression exceeds the boundaries of fair use.  For example, facts and ideas are not protected by copyright; only original "expression" is.  So, an academic who harvests facts from another scholar without giving due credit may be a  plagiarist, but, if she expresses those facts and ideas in her own words, she is not an infringer.  Or to take another example, works first published in the U.S. before 1923 are no longer in copyright.  Consequently, a novelist who lifts sentences and scenes from the short story sequence Winesburg, Ohio (first published in 1919 and now in the public domain in the U.S.) without crediting Sherwood Anderson would also be a plagiarist, but not an infringer.  When I pass off as my own original work ideas or public domain expression or sufficiently brief snippets of expression taken from others (e.g., my quote taken from Goldstein, devoid of quotation marks), I may have  -- depending upon the context -- committed plagiarism because, in the words of Judge Richard Posner, "readers of the new work are invited to think that those features are the inventions or discoveries of the plagiarist."  Yet I could not successfully sued for copyright infringement.

Conversely, if I quote too lengthily from the copyright-protected works of others, I may be a copyright infringer, even though I have carefully and fully acknowledged the author and work that I am quoting.  Thus, for example, in one well-known copyright case, the poet Ian Hamilton was found to have committed infringement by quoting too extensively from J.D. Salinger's unpublished letters in a biography of Salinger, even though Hamilton meticulously footnoted each quotation.

Examples of plagiarism are legion.  The Wikipedia offers a long, varied, and fascinating list of plagiarism controversies.  (Did you know that Helen Keller was caught up in a plagiarism scandal?)  Caslon Analystics of Australia has published another inventory of alleged incidents of literary plagiarism.  Wikipedia has a discussion of the theory and history of plagiarism here.  And Caslon Analytics's useful overview of plagiarism (again, with an Australian slant) may be found here,

I have just finished reading Judge Posner's book entitled The Little Book of Plagiarism, which provoked this post.  Judge Posner touches upon many of the famous unoriginality controversies of recent years, including those involving Doris Kearns Goodwin, Joe Biden, Stephen Ambrose, Laurence Tribe, and Kaavya Viswanathan.  (As Judge Posner points out, an uncommon number of such scandals seem to implicate Harvard personalities, not because plagiarism is especially common at Harvard, but because exposing the lapses of the academic elite gives the press and readers an extra shot of schadenfruede; see the Harvard Plagiarism Archive here for a master list of such controversies since 2002.)

Judge Posner offers an especially good explanation as to why some unacknowledged copying is plagiarism, and some is not:

A judgment of plagiarism requires that the copying, besides being deceitful in the sense of misleading the intended readers, induce reliance by them. 

* * * *

The reader has to care about being deceived about authorial identity in order for the deceit to cross the line to fraud and thus constitute plagiarism.

I am also indebted to Judge Posner for teaching me the word "cryptomnesia," which refers to "unconscious plagiarism, a sin of neglect rather than intention and, therefore, less blameworthy."  When caught in the act, almost all plagiarists, including Helen Keller, plead cryptomnesia.  But Judge Posner cites research suggesting that cryptomenesia is almost always a fallacious excuse when the copying consists of "verbal passages of more than trivial length."

The legal remedies for copyright infringement are powerful and plainly set forth in federal law.  (See Chapter 5 of the U.S. Copyright Act.)  But, as per Professor Goldstein's observation quoted above, the legal remedies, if any, for plagiarism are by no means clear-cut.

For many years, authors (and other creative people) sought to use the federal Lanham Act as a basis for litigation against those who used their work without proper authorization or credit. That legal strategy reached a dead end when the Supreme Court held in the 2003 case of Dastar v. Fox that the Lanham Act cannot be used to "creat[e] a cause of action for, in effect, plagiarism -- the use of otherwise unprotected works and inventions without attribution."  Law geeks may wish to take a look at this law review article -- and this one, too -- deploring how Dastar and its progeny effectively eliminated the Lanham Act as a remedy for failure to give credit where credit is due.  Without resort to the Lanham Act, victims of plagiarism -- who do not also have a viable copyright claim -- are left with few means of legal recourse.  In commercial settings, a buyer of a book riddled with plagiarism might theoretically seek to use consumer deception statutes to achieved redress, but there is no economic incentive to sue to recover the price of a book, and a plagiarized author may not even have standing to assert such a claim. See this earlier post discussing some less-than-satisfactory post-Dastar options, including ethical complaints to professional organizations and academic institutions.

After Dastar, shaming (either in private communications with the plagiarists' supervisors or in published accusations) remains the most effective tool for attacking plagiarism and plagiarists.  Timothy Noah of Slate offered this original idea for formalizing the shaming process in the book publishing industry; alas, it is not likely ever to be adopted.  And beware: all too often, accused plagiarists respond to shaming efforts with libel lawsuits against their accusers, as discussed in this article from The Chronicle of Higher Education.

In Europe, the legal doctrine of droit moral or moral rights gives authors the ability to claim a right of attribution in their works.  However, even in Europe, droit moral lawsuits are not routinely used to provide a remedy for garden variety plagiarism.  The United States recognizes droit moral or moral rights only with respect to certain limited categories of visual arts.  (See this brief summary of protections accorded by the Visual Artists Rights Act  in the United States.)  Writers do not have equivalent protections for the "right of attribution."

Interestingly, in June 2011, the Third Circuit Court of Appeals seems to have imported into the Digital Millennium Copyright Act ("DMCA") what is, in essence, a kind of cause of action for plagiarism in certain narrow circumstances.  The kinky case involved two New Jersey shock jocks, who hired a photographer to take a picture of them, apparently nude, with their manliness concealed by a sign bearing their station's logo, WKXW.  The photograph originally appeared in New Jersey Monthly.  Later, however, the station scanned the photograph, stripped out the photographer's credit line (without permission), posted it on the station's website, and invited fans to manipulate the image.  The photographer sued for copyright infringement and, based upon some on-air comments that the shock jocks made about him, defamation. The district court dismissed the photographer's claims, but the Third Circuit reinstated them, concluding, among other things, that the photographer's credit line was "copyright management information," and the digital manipulation and removal of the credit potentially violated the DMCA. Here is the always-entertaining Eriq Gardner's summary of the case.  And you  can see a portion of the image in question in this analysis of the case by New York lawyer Peter Fakler. You can read the entire decision here.  Of course, the DMCA does not provide a tool for addressing plagiarism outside of the context of digital manipulation, but it is an interesting new tactic in the age-old wars over plagiarism.

This New York Times article suggests that plagiarism is on the rise among college students and others, who find it tempting and easy to cut and paste into their papers the expression of others available the web.  Plagiarism.org's website cites a national survey published in Education Week that found that 54 percent of students admitted to plagiarizing from the internet.  Schools have responded by requiring the use of plagiarism-detection software, such as Turn-It-In, created by IParadigms.  

My firm recently handled case where plagiarism detection intersected with alleged copyright infringement.  Several students from McLean High School in northern Virginia tried to turn the tables on IParadigms, and sued the company for copyright infringement.  In checking for plagiarism, Turn-It-In software compares student papers not only to a huge textual database drawn from the internet, but also to an archive consisting of virtually all past papers submitted by students for scanning and review by the Turn-It-In software.  This archive is important to the review process because it enables the software to determine when students have copied from each other rather than copying from the internet.  The database of student papers is not published in any way, but merely resides on computers in the form of digital files used to carry out the comparisons.  The students contended that, in maintaining copies of their papers in these digital files, IParadigms infringed their copyrights in those papers.  In submitting their papers, the students had agreed not to assert any claims against IParadigms, but they argued that the agreement was against their will and otherwise unenforceable.  The courts disagreed and found that the archiving of the students' papers was effectively consented to and, in any event, a "fair use." Here is the opinion of the Fourth Circuit Court of Appeals affirming the lower court's decision of non-infringement.  (My firm represented IParadigms.)

One of the things I do in my legal practice is review manuscripts before they are published for potential legal problems.  The legal vetting task is fairly clear-cut when it comes to identifying issues such as libel or invasion of privacy.  However, it is often nearly impossible for a lawyer (or an editor) to spot potential copyright infringement when an author has plagiarized his sources, failing to use quotation marks or endnotes.  As a result, some publishers are now sometimes using anti-plagiarism software to detect unauthorized copying in manuscripts, particularly in high profile books by celebrity authors.  According to this article in The Chronicle of Higher Education, academic journals are similarly "turn[ing] the anti-plagiarism software that professors have been using against their students on the professors themselves."

As Judge Posner is at pains to point out, what constitutes plagiarism and what does not is not always clear.  Many universities publish useful online resources for identifying and preventing plagiarism, including this one maintained by the UCLA libraries.  And here is the American Historical Association's Statement of Standards of Professional Conduct, which includes a fine discussion of plagiarism, and the sage advice: "No matter what the context, the best professional practice for avoiding a charge of plagiarism is always to be explicit, thorough, and generous in acknowledging one's intellectual debts."

Which reminds me, I almost neglected to mention that the headline to this post is a knock-off of the title of this great article by Roy Peter Clark on plagiarism in the newsroom, which originally appeared in The Washington Journalism Review. Clark's article is filled with examples of journalistic plagiarism and offers good counsel on distinguishing between permissible re-purposing and unethical, intellectual fraud.

Actually the phrase "unoriginal sin," has been used by many other writers in connection with articles about plagiarism.  (See this Google search report.)  Likely, this is the result of independent creation -- numerous writers separately coming up with the same pun, though it wasn't at all original on my part.  Suffice it to say that not every lifted phrase or unattributed joke is a sin.  Remember, Judge Posner's point that there must be some element of deceit and the originator and/or the reader/listener must have some reason to care about the undisclosed repurposing.  This well-known anecdote illustrates the irrepressible tradition of passing off the jokes of others as one's own:

After hearing his friend James McNeill Whistler make an especially witty remark, Oscar Wilde said, "My God, James, I wish I had said that."  To which Whistler replied, "You will, Oscar, you will."

Copyright in Fictional Characters: Can I Have Don Draper Make a Cameo Appearance in My Novel?

I'm throwing a dinner party in my novel. My guest list includes Don Draper, James Bond, Jack Ryan, Scarlett O'Hara, Dolores Haze a/k/a Lolita, and Elizabeth Bennett.  I don't expect my guests to say or do anything at my fictional party.  The question is:  Can they simply show up at the dinner table without my infringing the copyrights of Matthew Weiner, Ian Fleming, Tom Clancy, Margaret Mitchell, Vladimir Nabokov, and Jane Austen or their heirs?

Before trying to answer that (somewhat surprisingly complicated) question, let me introduce you to one of my guests.  Jack Ryan is known to millions of readers as a tough former Second Lieutenant in the Marines, a onetime CIA agent, and ultimately President of the United States. He appeared as a character in Tom Clancy's first novel, The Hunt for Red October, which was published in hardcover in October 1984 by the U.S. Naval Institute Press, a small publisher that was then primarily issuing works on naval history.  Ryan later reappeared in many other Clancy novels.  But did you know that Jack Ryan was the subject of allegedly infringing use by none other than . . . Tom Clancy?

Clancy was a true unknown when he wrote Hunt for Red October -- an insurance agent who daydreamed of becoming a novelist.  The New York Times tells the back story here.  Prior to Red October, Clancy had published no fiction, but he had published a non-fiction article on the MX missile in the Naval Institute's Proceedings magazine.  So when he completed Red October he offered the novel to the Naval Institute; its headquarters were, after all, just a few miles away from his home near Annapolis.

The Naval Institute had published a few previous novels on naval themes and offered Clancy its standard contract, which included an assignment of the copyright to the publisher -- not uncommon for academic books at the time, but rare for novels.  Clancy happily signed the agreement and received a $5,000 advance.  Red October surprised both author and publisher when it turned out to be a colossal bestseller, selling many millions of copies (after Ronald Reagan praised it as "the perfect yarn").

For his second book, Clancy jumped ship and signed with a big New York publisher.  But the earlier copyright assignment to the Naval Institute almost torpedoed the new book.  Why?  Because the copyright assignment arguably carried with it the rights to the characters in Red October.  The Naval Institute commenced an arbitration claiming that, as the Times explained, the Naval Institute's copyright ownership in Red October gave it "a continuing interest in the Jack Ryan character, and it should therefore receive a percentage of the profits from Patriot Games and The Cardinal of the Kremlin and from any films or miniseries made of them."

The Red October arbitration was settled on undisclosed terms.  But the point is that fully drawn literary characters are generally subject to copyright protection, and the copyright owner will often be able to prevent others from using the character in other works without permission.  But the legal principles are by no means simple.

There are two separate but related questions here.  First:  Is a particular character protected by copyright?  Second:  Is the particular use made by someone other than the copyright owner infringing?

Here is what famed judge Learned Hand said 80 years ago in Nichols v. Universal Pictures Corp. about distinguishing between literary characters that are and are not protected by copyright:

If Twelfth Night were copyrighted, it is quite possible that a second comer might so closely imitate Sir Toby Belch or Malvolio as to infringe, but it would not be enough that for one of his characters he cast a riotous knight who kept wassail to the discomfort of the household, or a vain and foppish steward who became amorous of his mistress. These would be no more than Shakespeare’s ‘ideas’ in the play, as little capable of monopoly as Einstein’s Doctrine of Relativity, or Darwin’s theory of the Origin of Species. It follows that the less developed the characters, the less they can be copyrighted; that is the penalty an author must bear for marking them too indistinctly.

Judge Hand's analysis still holds true today.  As copyright guru Paul Goldstein puts it:

Fully realized characters in literature are little different from fully defined personalities in daily life, and it is no surprise that the test of protectibility that courts apply to literary characters is closely akin to the criterion that individuals apply in daily life to determine whether they in truth know someone. A literary character can be said to have a distinctive personality, and thus to be protectible, when it has been delineated to the point at which its behavior is relatively predictable so that, when placed in a new plot situation, it will react in ways that are at once distinctive and unsurprising.

Every reader of Catcher in the Rye comes away with the feeling that she knows Holden Caulfield, and so it was unsurprising to many copyright lawyers when, a little over a year ago, a court enjoined publication an unauthorized sequel by Fredrick Colting that told the story of a 76-year-old Holden Caulfield.  Here is the appeals court decision in that procedurally complicated case, which ended in a settlement prohibiting further publication in the United States and Canada.  Holden Caulfield is, in Goldstein's phrase, "fully delineated," and like other fully delineated characters (e.g., James Bond, as the court ruled in this lawsuit) he is surely protected by copyright, while, by contrast, sketchily defined, stock characters may not be (as the court found in this case).  (See also this account of a 1998 lawsuit to block publication of Lo's Diary "a distaff version of Nabokov's Lolita.")

But the fact that a character is copyright protected does not necessarily mean that every use of his/her fictional persona is an infringement.  A second work that invokes a copyright-protected character must copy some significant amount of expression in order to be an infringement.  The mere mention of the name of a copyrighted character ought not to be a copyright infringement because names, standing alone, are not copyrightable. (But see the discussion of trademark and unfair competition law below.)  A copyright infringement occurs only when a quantum of protectible expression has been copied, and the copying is not excused by the doctrine of fair use.  (Note that this post deals primarily with openly acknowledged use of another writer's character, not the situation where an author merely borrows certain traits or characteristics from a character.)

With this principle in mind, it would seem to follow that a fleeting appearance of another writer's fictional character as a dinner guest in my novel should not qualify as a copyright infringement.  However, as soon as I start to make Jack Ryan or Don Draper talk like, act like, or look like Jack Ryan or Don Draper in my novel I may be treading on thinner ice.  And if I make Jack Ryan or Don Draper an important character in my book, I'm begging for trouble.

There are several other important considerations here.  First, there are some copyright owners who, regardless of viability of their claims, will not hesitate to sue me at the drop of a hat if I use their characters in any way without obtaining permission (which they will never give me).  There is, in other words, a practical risk in inviting other writers' characters to my fictional dinner party, even if those characters keep their mouths shut and do nothing.  Frankly, it may not be worth it to me to take any risk of provoking a lawsuit arising from my imaginary soiree, even if I have the better part of the argument on copyright.  It's always safer to invite some nameless characters of my own creation.

(And Elizabeth Bennett can still attend the dinner, since she has been in the public domain for a long, long time.  Seth Grahame Smith, author of Pride and Prejudice and Zombies, is surely thankful for that.)

Second, the owner of the rights in a famous fictional character will also likely seek to invoke other legal theories -- particularly trademark and unfair competition laws -- when trying to protect her characters from my unauthorized use.  If the name of a character has acquired secondary meaning -- i.e., if the name is sufficiently well known that readers associate it exclusively with a particular author's work -- then the owner may be able to claim that my unauthorized use of the name creates confusion among readers, who may mistakenly conclude that my use of the name is authorized, approved, or licensed by the owner.  Confusion as to source, authorization, or  endorsement is the essence of an unfair competition claim. There might be ways to get around the trademark infringement/unfair competition argument, particularly with the use of prominent disclaimers that make very clear that Matthew Weiner has not authorized me to mention Don Draper.  But, still, the unfair competition argument is a complicating factor when referencing the characters of others.

Concern about characters as trademarks may -- or may not -- have figured into some of the choices made in The Wind Done Gone, by Alice Randall.  Randall's novel was, of course, a reinterpretation of the characters and events in Gone with the Wind.  Scarlet O'Hara appears in the novel, but she is referred to only as "Other."  And Rhett Butler is called "R."  Place names, such as Tara, were also altered.  That said, Scarlett and Rhett are "fully delineated" characters whose fictional lives are continually referenced (albeit not by their full names) in Randall's novel.  Small wonder then that the owner of the copyright in GWTW sued Randall's publisher, Houghton Mifflin.  As you can see from the complaint, the plaintiff alleged both copyright infringement and unfair competition, claiming, among other things, that the characters were both copyright-protected and had acquired secondary meaning.  It was a hard-fought case, with an inconclusive ending.  The trial court found that The Wind Done Gone was likely a violation of copyright and issued a preliminary injunction.  But the appeals court, in this decision, found on First Amendment grounds that it was improper to preliminarily enjoin publication of TWDG in part because Randall's work could qualify as a "parody" of GWTW.  But the appeals court did not rule out the possibility that the publisher of TWDG could ultimately be liable to the copyright owner for money damages.  The case settled before there was any final decision on a host of interesting legal issues.

These days, the unauthorized use of characters occurs all of the time in the context of fan fiction, which typically involves inventing new stories using familiar characters from literature, comic books, movies, or television programs.  An entire subculture has built up around fan fiction, and recently FanFiction.Net was the 747th most trafficked website in the United States.  Although each work must be judged on its own, suffice it to say that, if the issue ever reached a court, many works of fan fiction would be deemed derivative works that infringe upon the characters and stories on which they are based.  Chilling Effects has a great Q&A on the legal issues relating to fan fiction.  And the Wikipeida entry on the legality of fan fiction also makes for interesting reading.  Georgetown Law School professor Rebecca Tushnet wrote an excellent law review article on the subject.  And here is a list of other law review pieces discussing the legal issues.

Fan fiction writers have two things going for them.  First, as discussed in some of the references above, many (but by no means all) copyright owners have turned a blind eye to infringing fan fiction, on the theory that they do not want to alienate some of their most enthusiastic followers. Second, fan fiction writers rarely seek to profit from their work, which is helpful in arguing the defense of fair use, but by no means provides immunity from an infringement claim.  In short, writers of fan fiction, who publish their work on the web or elsewhere, should be aware that they are taking some degree of legal risk, unless, like Pride and Prejudice and Vampires, their works are based upon public domain sources.

So can I have Don Draper make a cameo appearance in my novel?  As you can see, the answer is a thoroughly equivocal "it depends."  A writer contemplating using the characters of another writer would well-advised to seek out the counsel of an experienced intellectual property lawyer.  (See this prior post on free and low-cost resources.)  If you're publishing with an established house, you should discuss your particular use of other writers' characters with your editor and the in-house attorney.

There are several helpful discussions on the web concerning copyright in fictional characters, including this one by Denver lawyer Lloyd Rich, this one by California lawyer Ivan Hoffman, and this by Phoenix lawyer Jasmina Zecevic Richter.

Titles and the Law: Can I Call My Novel "The Great Gatsby"?

I just finished reading Nemesis by Philip Roth.  When I was originally looking to buy a copy online, I was struck by the fact that there are roughly a dozen earlier novels by the same title that are still in print -- including works by such well-known authors as Agatha Christie and Isaac Asimov.  A legal problem for Roth?  Nah.

Let's take a look at three legal theories that authors and publishers have sometimes asserted (with varying degrees of success) in lawsuits to protect book titles.

1.  Copyright Law.  U.S. courts have unanimously held that titles are not copyrightable.  Here is one such case.  Indeed, the Copyright Office will not register a claim of copyright in a title, as it explains in Circular 34:

To be protected by copyright, a work must contain a certain minimum amount of authorship in the form of original literary, musical, pictorial, or graphic expression. Names, titles, and other short phrases do not meet these requirements.

Given the rationale for non-protectability, my wonderful copyright professor in law school, the late John Kernochan, once posed the question whether the inventive 26-word title of Peter Weiss's play The Persecution and Assassination of Jean-Paul Marat as Performed by the Inmates of the Asylum of Charenton Under the Direction of the Marquis de Sade aka Marat/Sade would be copyright-protected. After all, the title is as long as some poems.  While it would certainly be worth a try to argue that a lengthy and highly expressive title is copyrightable, suffice it to say, the courts and Copyright Office have never, to my knowledge, recognized an exception of the sort that Professor Kernochan hypothesized.  (But see this case finding that the famous short phrase of movie dialogue "E.T. phone home" was likely protected by copyright.)

2. Trademark Law. Here is where it starts to get complicated. The Patent and Trademark Office ("PTO") will not register the title of a stand-alone literary work (including a fiction or non-fiction book, song, movie, or video game). But the PTO will register as a trademark a designation for a series of works. Think of [Insert here almost any subject you can think of] For Dummies. "For Dummies" is a federally registered trademark for a series of "non-fiction books, guides, manuals, and catalogs on a wide variety of topics." Similarly, "Harry Potter" is a federally registered trademark for a series of novels for young people, among many other items.  But, if there had been only one Harry Potter novel and it had not become the basis for a series of books and movies and countless merchandising opportunities, trademark registration would not have been an option.

If you do manage to obtain a federally registered trademark for your series of books, you might have a plausible basis for a trademark infringement claim arising out of the publication of a book with a confusingly similar title, particularly if it's on a confusingly similar subject. Trademark infringement cases involving a registered mark for a series of books are not especially common, but they do come up from time to time. The Fishing Hot Spots case is one example.  A federal trademark registration gives you a significant advantage in any litigation over a title because it carries with it a legal presumption of exclusive nationwide rights to use the name in connection with the goods for which it's registered.  The defendant might still be able to prove that his use creates no likelihood of confusion, but it will be an uphill battle.

3.  Unfair Competition or Passing Off.  But it gets even more complicated.  Although the PTO won't register a trademark for a stand-alone literary work, the federal courts will nonetheless sometimes invoke the Lanham Act to protect unregistered titles from confusingly similar uses by others.  However, the courts will only extend protection to titles that have achieved "secondary meaning."  Here's how the Second Circuit Court of Appeals explained "secondary meaning" in Rogers v. Grimaldi: 

The purchaser of a book, like the purchaser of a can of peas, has a right not to be misled as to the source of the product. Thus, it is well established that where the title of a movie or a book has acquired secondary meaning — that is, where the title is sufficiently well known that consumers associate it with a particular author's work — the holder of the rights to that title may prevent the use of the same or confusingly similar titles by other authors . . . .  Indeed, it would be ironic if, in the name of the First Amendment, courts did not recognize the right of authors to protect titles of their creative work against infringement by other authors.

Similarly, most states have laws forbidding "passing off," which is a form of unfair competition.  While the laws vary from state to state, here, again, in order to prevail, a plaintiff-author will likely have to prove that her title has acquired "secondary meaning" and/or that the defendant adopted his confusingly similar title with the intention of deceiving book buyers into believing that his book was somehow connected with plaintiff's book.

So, to simplify quite a lot, you may be able to use the federal or state unfair competition laws to protect against the publication of a subsequent book by another author with a title that is confusingly similar to your title.  But first you must be able to prove to a court's satisfaction that a substantial portion of the book-buying public would assume that any book by that title must have been written by you or must have been published by your publisher.  That's actually a hard row to hoe.  Only occasionally will the title of a single work in a single medium be deemed to have achieved secondary meaning.  On the other hand, if the author has published a series of books using the same title, it is more likely to be found to have achieved secondary meaning, even if it has not been registered as a trademark, as in this case concerning the protectability of William Bennett's title The Book of Virtues.

Let's consider how these principles might play out in various situations.

Given that Nemesis is a fairly obvious title for a work evoking a mood of menace or suspense, given that by the time Roth published his novel there were already several other books called Nemesis in print, and given that readers are accustomed to distinguishing among novels by their authors as well as their titles, it is virtually unthinkable that the Asimov estate or the Christie estate could prove that the public associated the title Nemesis exclusively with Isaac Asimov, Agatha Christie, or anyone else.  And, for a host of reasons, a reader looking to buy Christie's Nemesis is unlikely to end up buying Roth's book by mistake.  So, Roth doesn't have to lose any sleep about the unoriginality of his title.  Similarly, if I were to call my novel Nemesis, Roth probably could not successfully pursue a claim against me.

But what if I were to entitle my novel Swamplandia?  Karen Russell recently released a popular and well-regarded novel by that distinctive name. There does not appear to be a single other book listed on Amazon, in-print or out-of-print, called Swamplandia.  Under the circumstances, a court might be tempted to conclude that the only reason I adopted the title Swamplandia was to confuse book buyers into mistakenly buying my book when they really meant to buy Russell's. I fear that Karen Russell might have a plausible claim against me for unfair competition -- at least if I were to publish a book called Swamplandia right now. But if, like most books, Swamplandia eventually fades from the consciousness of the book-buying public, it might lose its "secondary meaning," and another author could then safely adopt the same title -- or one that is very similar. That's effectively what the court decided in the venerable case of International Film Service v. Associated Producers, 273 Fed. 585, 587 (2d Cir. 1921) (involving the title "Broken Doll").

So, given the passage of time since F. Scott Fitzgerald published his little novel, could I call my novel The Great Gatsby?  As a matter of copyright law, no problem.  And "The Great Gatsby" does not appear to be a federally registered trademark for books or movies (although, curiously, there is a "Great Gatsby's" registration for auction services).  As a result, there probably would be no claim for infringement of a registered trademark.  Nevertheless, given the distinctiveness of the title, the enduring fame of Fitzgerald's novel, the fact that the work has been adapted for the movies, and a host of other factors, I would be cruising for a lawsuit if I chose The Great Gatsby as the title for my novel unless (and maybe even if) the cover was filled with wall-to-wall disclaimers.

That said, there are several other books currently in print by authors other than Fitzgerald that do have "Great Gatsby" in the title -- namely, non-fiction books of literary criticism, such as this one.  If I were using "The Great Gatsby" in my title to refer to Fitzgerald's book, it might qualify as nominative fair use and therefore not be an infringement. In other words, if I make it clear enough that my book is about The Great Gatsby, but that it is not Fitzgerald's book or some kind of sequel, then my "nominative fair use" defense will likely succeed.  (It would probably help to have a subtitle or prominent disclaimer on the cover that further clarifies this is not The Great Gatsby, but a book about The Great Gatsby.)

The same general legal principles that apply to recycling of book titles hold true as well for recycling song titles and movie titles.  However, as described in this piece in Slate, the Motion Picture Academy of America has a system designed to minimize duplication of movie titles, including a procedure for arbitrating title disputes.  Still, "video doppelgangers" are common, as discussed in this Entertainment Weekly article.  As to the music industry, Kate Kiefer of Paste magazine posted this blog about duplicate song titles.  And this site maintains a list of duplicate song titles, with some 3,092 entries.  The "most recycled song title"?:  "Home."

In short, using a title that has previously been used on another book is, more often than not, perfectly lawful.  In a relatively rare instance where your title is likely to create confusion with a previously published book, your publisher will likely urge you to adopt a different, safer title -- like Nemesis.