An Essay on Character Protection and Copyright Law
The idea for this essay came from a number of places. I am first and foremost a comic book geek. In particular, I am a huge fan of “X-Men” and its related characters. One such “X-Men” story that comes to mind is the seminal “Days of Future Past.” In that story a future-version of Kitty Pride, an X-Man, goes back in time to prevent her dystopian future from occurring by preventing the assassination of an anti-mutant politician. I bring this up because at its core, the story is about how our decisions today have monumental effects on humanity’s future. This essay, in many ways, is my postcard to the future. But rather than preventing a dystopian future, I am trying to promote a copyright regime where creativity is encouraged whilst not stifled by over-protection.
This paper seeks to explain the protection afforded to characters and concludes that copyright law alone is insufficient to adequately protect characters. First, this paper asks if we should even protect characters, ultimately deciding that character protection is an important incentive. Secondly, I will outline and explain how characters are currently protected in the current copyright regime and explain what the current infringement standard is. Thirdly, I will analyze Feldman’s essay Finding a Home for Fictional Characters: A Proposal for Change in Copyright Protection, and look at what he believes are the problems with current character protection. This paper argues that Feldman’s assertions are correct in many ways, however, I believe that he underestimates the ability of trademark and unfair competition law to correct copyright’s inability to adequately protect characters. I will then talk about the policy implications of granting characters their own independent copyright protection, ultimately concluding that independent protection would cause more harm than good. Finally, I will offer an analysis of other proposed solutions to the character protection problem, namely what Schienke calls a “copymark.”
Why Protect Characters?
Before getting into the nuances of the current protection of characters and subsequently what type of protection should be afforded to them, I think a basic question should be answered: should we even protect characters? I would answer the question in the affirmative for a number of reasons.
Firstly, and most importantly, I believe characters should be protected because I subscribe to the incentive-based justification for copyright. The theory can be summed up as follows: because mental creations take considerable time and effort, authors will only create a work if they are assured that they can benefit from its creation. This is where copyright comes in, by giving an author an exclusive right to exploit their work, copyright gives authors the incentive to create. The theory does pose some problems, as Chisum, Ochoa, Ghosh, and LaFrance state, the key question is, “will anyone create new literary and artistic works if other can copy those works without permission and without paying compensation?” One is not hard-pressed to find examples of people creating works without such an incentive. Some authors write books for critical acclaim or for social change. However, my argument is not that creative works would not happen without copyright, my argument is that copyright creates more of an incentive for authors to create.
The next question is how this incentive applies to the protection of fictional characters. I think the incentive justification is equally applicable in the realm of character protection. In 1954 the Harvard Law Review already realized the importance of character protection stating:
Since in many fields the chief value of a character to its author lies in the character's adaptability to new situations, rather than its use in one work, denial of this protection might cause such authors to lose much of the potential value of their creation. At present the courts appear to give broad protection to literary works generally, and, assuming that this is to be continued, there would seem to be no reason why copyright should not prevent the use of a character in new works, provided that the character is sufficiently delineated to assume a concrete and independent existence.” The Protection Afforded Literary and Cartoon Characters Through Trademark, Unfair Competition, and Copyright., 68 Harv. L. Rev. 349, 357 (1954).
This is more than true in the modern world. Characters, in many ways, have surpassed their source material and have become institutions in their own right. To go back to my “X-Men” example, a character like Wolverine is far larger than its humble beginnings as a generic super-agent in the 180th issue of Incredible Hulk. The character has now been featured in seven feature films, children’s cartoons, video games, and all types of merchandise. Clearly characters can and have become more famous than the source works they derive from. This provides an enormous incentive for authors to create, and thus characters should be protected.
Current State of Character Protection in Copyright Law
In the current copyright regime, characters are offered some protection. Like any other work, characters must meet the minimum standards for copyrightability. Section 102(a) of the Copyright Act lays down the prerequisites for copyright protection:
Copyright protection subsists… in original works of authorship fixed in any tangible medium of expression, now known or later developed, from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine, or device.
Thus, to attain a copyright in a work, it must both be “original” to the author and “fixed in a tangible medium of expression.”
Case law has clarified what these terms mean, and it might take a book to explain all of the intricacies and issues involved with these terms. For the purposes of this essay, it is sufficient to know that the “originality” standard is quite a low threshold. The Court in Feist Publications, Inc. v. Rural Telephone Service Co. held that:
Original, as the term is used in copyright, means only that the work was independently created by the author (as opposed to copied from other works), and that it possesses at least some minimal degree of creativity.
Furthermore, a work is “fixed in a tangible medium of expression” when it is “sufficiently permanent or stable to permit it to be perceived, reproduced, or otherwise communicated for a period of more than transitory duration.” (Section 101)
However, characters, must meet another standard in order to be protected. Michael D. Scott in his book, Scott on Multimedia Law, summarizes the two inquiries a court engages in when deciding whether a character is copyrightable:
In order to determine whether a character might be protected under copyright law, a court usually conducts one of the following inquiries. First, the court looks to whether the character is ‘sufficiently delineated’ to merit independent protection. The fundamental question is what constitutes sufficient detail for the character to be more than just a stock character. Under the second inquiry, the court looks to whether the character is just a part of the ‘story being told’ or is itself the story. In the James Bond novels the character of James Bond constitutes the story being told and is not merely a ‘chessman in the game of telling a story.’ The ‘story being told’ inquiry has been criticized as being too difficult to achieve, and the sufficient delineation test has been criticized as being too unclear. Still, courts rely on these analyses to determine whether protection should be given to the character and generally do not question the basic principle that a character may be copyrighted.
2014 WL 1832414
However, these two inquiries don’t address the wide breadth of types of characters that could be protected. Some characters are very much a product of the medium they are created in, a cartoon character or a comic book character is immediately recognizable whereas literary characters don’t immediately have that visual cue, even if described in a literary work with sufficient detail. This begs the question: should an illustrated character be judged by the same standard as a literary character? I don’t believe they should. Because of the wide array of works that copyright protects, I believe there is value in a flexible standard. The court in DeCarlo v. Archie Comics Publication seemed to agree that illustrated characters should be judged with a different standard stating:
[T]he difficulties of distinguishing distinct attributes of a literary character from its embodiment of more general ideas and themes (the test for copyrightability offered by the Second Circuit in Nichols v. Universal Pictures Corp.) do not arise, at least to the same degree, with visual images. “While many literary characters may embody little more than a protected idea, a comic book character, which has physical as well as conceptual qualities, is more likely to contain some unique elements of expression.
DeCarlo v. Archie Comic Publications, Inc., 127 F. Supp. 2d 497, 505 (S.D.N.Y. 2001) aff'd, 11 F. App'x 26 (2d Cir. 2001)
I believe that this statement stands for the assumption that visually-depicted characters have a lower bar to meet in either the “sufficiently delineated” standard or the “story being told” inquiry.
In sum, copyright unequivocally offers some form of protection for characters. First, a character must be both original and fixed in a tangible medium of expression. Next the character must satisfy one of two inquiries. The character must either be “sufficiently delineated”, that is to say it must be more than just a stock character. The second inquiry asks whether a character is more than just a pawn in the story or whether the character is the “story itself.” If it is more than just a chess piece in a story, then it warrants protection. However, how the character is represented also plays into whether or not it is protected. If a character is visually depicted, the court is more likely to protect it because its expression is more readily perceived. This is the current state of character protection in copyright, the next number of sections offer a critique on whether or not this protection is adequate.
Criticism: Why Copyright Alone Isn’t Sufficient to Protect Characters
The next section is really a critique of David B. Feldman’s essay, “Finding a Home for Fictional Characters: A Proposal for Change in Copyright Protection.” In that essay Feldman argues that fictional characters are second-class citizens in the world of intellectual property in that existing statutory schemes provide inadequate protection for them.
In arguing for an independent subject matter category in the Copyright Act, Feldman asserts three problems with current copyright protection. First, he argues that the lack of express copyright protection complicates infringement analysis by forcing the court to argue that the allegedly infringed character is copyrightable. Secondly, Feldman asserts that making a character’s copyrightability dependent on the underlying work it originated from makes it difficult to determine ownership. This is especially true when the creator has separated the character from the original work in which the character first appeared. Finally, he argues that alternative protections such as trademark and unfair competition are inadequate to protect fictional characters.
Although I agree with most of Feldman’s arguments, I do not believe that an independent subject matter category is necessary to protect characters, in fact, I believe trademark and unfair competition make up for the inadequacies with copyright law. The following sections will outline Feldman’s arguments while providing my own insight on the matter.
Problem One: Inconsistent Adjudication due to the Standard Infringement Test
Feldman calls his theory on the problems with the standard infringement test the “delineation paradox.” The test he is critical of is the “delineation” standard (discussed above), which originated in Nichols v. Universal Pictures, a case wherein a motion picture was held not to have infringed another film that had a similar plot and some similar “stock” characters. Feldman summarizes the inquiry as follows: “First, is the original character's expression sufficiently delineated to be copyrightable? Second, is the infringing character's expression substantially similar to that of the original character?” David B. Feldman, Finding A Home for Fictional Characters: A Proposal for Change in Copyright Protection, 78 Cal. L. Rev. 687, 691 (1990). Feldman’s problem with this standard is that it looks nothing like standard infringement analysis. Rather than asking if the second character had copied the original, and then asking whether that copying infringed protected expression, the courts have fixated on the threshold copyrightability question and never even get to the standard infringement question.
Feldman’s problem with this “mixing up” of the standard infringement question is that the standard has both the potential for over- and under-protection. He says this because in some cases the criteria has determined that the characters and the underlying work are unprotectible (See Giangrasso v. Columbia Broadcasting System). In other cases the courts have found the character sufficiently delineated even though the appropriation did not affect the underlying work (See Anderson v. Stallone). This inconsistency, Feldman argues, muddles the infringement adjudication process. In short Feldman asserts that, “While the pertinent factors in any substantial similarity analysis will necessarily vary, that variation should not be considered in determining whether the original expression of a character should be afforded copyright protection.” David B. Feldman, Finding A Home for Fictional Characters: A Proposal for Change in Copyright Protection, 78 Cal. L. Rev. 687, 696 (1990).
What Feldman calls inconsistency, I call flexibility, a feature that I believe is needed to protect such a wide-array of media in which characters appear. One of Feldman’s arguments insists that the over-protection of a visually-depicted character like a cartoon and the under-protection of say literary characters is a failing of the copyright system. But I do not believe this is the case, as a general matter literary characters are more likely to be seen as a stock character.
Let’s take the example of a detective novel, featuring a hard-boiled, cigarette smoking, whisky drinking detective. Most people wouldn’t believe that the “hard-boiled” archetype warrants protection. But what if we were take those traits and draw a comic book character that had a trench coat, fedora, and covered his face with a white mask emblazoned with a black Rorschach blot test? I think there would be a very strong argument for protection of that character. Again, the court put it best in DeCarlo v. Archie Comics, “While many literary characters may embody little more than a protected idea, a comic book character, which has physical as well as conceptual qualities, is more likely to contain some unique elements of expression.” DeCarlo v. Archie Comic Publications, Inc., 127 F. Supp. 2d 497, 505 (S.D.N.Y. 2001) aff'd, 11 F. App'x 26 (2d Cir. 2001). This flexibility is what copyright law needs, and I think the case law has developed this flexibility at least in the context of graphically depicted characters.
Problem Two: Characters Treated as Components of the Works They Appear
Feldman’s initial assertion is that Courts have “viewed characters as components of the works in which they appear.” I agree with Feldman’s hypothesis, I would point to the “story being told” inquiry as an example of the courts looking at characters primarily as a component of the work in which they appear. That inquiry looked not at how sufficiently delineated a character was, but whether the character was in itself the story, or whether it was merely a chess piece in the narrative. The “story being told” inquiry is just one of many examples of this limited view of characters. Feldman looks at three problems with viewing a fictional character as a component of the works in which they appear. These problems occur when trying to separate the character from the work, when the fictional character migrates through different media, and when a character is “pure”, that is to say that it has no “work” to originate from.
Separating the Character from the Work
Feldman’s argument is that treating a character as part of a work creates a “dangerous hurdle” in litigation in that it creates an unnecessary inquiry as to whether characters are not only substantially similar, but whether they are qualitatively substantial in order to find infringement. Feldman uses the Southern District of New York’s decision in United Artists v. Ford Motor Co as an example of this hurdle. In that case United Artists sued Ford because they included in their commercials an animated cat. Plaintiffs argued that Ford had copied “Pink Panther,” their character used in the title sequences in the film of the same name and the subsequent sequel. Although the court held that Ford’s animated cat and the “Pink Panther” were not substantially similar, Feldman argues that if they had, it would have still been unfair because Plaintiff would then need to prove that the “Pink Panther” was a qualitatively substantial element of its movies.
What his argument boils down to is that the length of the character’s appearance is dispositive on the question of infringement, and that is a problem. If this was true, then I would certainly agree with him, however, it isn’t. The test he refers to is both a qualitative and quantitative inquiry. As the court in Narell v. Freeman stated, “Quantitatively insignificant infringement may be substantial only if the material is qualitatively important to [a]…work.” Narell v. Freeman, 872 F.2d 907, 913 (9th Cir. 1989). Thus, a character with very limited screen time may be protected if they have a significant impact on the film. Let’s set aside any comic book iterations of “Superman” and focus on the 1978 film starring Christopher Reeve. Marlon Brando’s portrayal of “Jor-El” takes up an insignificant amount of the film. However, his character sets the entire stage for the film, sending his son “Kal-El” to earth before his planet is destroyed. Although “Jor-El” only took up as much time as the “Pink Panther” on screen, he had a great impact on the film.
Furthermore, is this requirement that a character be a quantitative and qualitative part of a work to be protected really a problem? I think in granting someone a copyright for any work, we are giving them a monopoly on something that society believes warrants protection. If by just being in a work a character becomes protected we hurt the commons. The extra qualification that the character need to be either qualitatively or quantitatively important encourages people to create characters that make an impact, perhaps even more creativity.
Character Migration Throughout Different Media
Feldman’s next argument is that there seems to be a lot of ambiguity as to the protection of characters when the original work they appeared in enters into the public domain but subsequent works in other types of media have not. He uses the court’s decisions in Burroughs v. Metro-Goldwyn-Mayer, Inc and Silverman v. Columbia Broadcasting System as examples of the types of problems that occur when you treat fictional characters as part of the works that they derive from.
The first of these two cases, Burroughs v. Metro-Goldwyn-Mayer, Inc and Silverman v. Columbia Broadcasting System concerned the character of Tarzan. Burroughs, the creator of Tarzan, introduced the character in the book “Tarzan of the Apes” and transferred all his rights in the book to ERB who in turn licensed the Tarzan character for use in an original screenplay (and the right to make remakes of that film). When termination was available to the heirs of Burroughs, they sent ERB notice. MGM at the time was making a second Tarzan remake. The court found that the ERB/MGM agreement conveyed a nonexclusive license to the Tarzan character.
Feldman’s analysis does not seem to have a problem with the outcome of the case but rather what would have happened if Tarzan was deemed uncopyrightable. The court turned to the question of whether Tarzan was independently copyrightable according to the “delineation” standard. They ultimately concluded that Tarzan was independently copyrightable. Feldman’s main contention is that if Tarzan was not deemed independently copyrightable, then MGM would have retained its remake rights. Thus, in limited circumstances, copyright law poses a risk of authors losing their characters due to licensing to other media.
I have a couple of problems with this argument, it seems like his argument isn’t that copyright doesn’t offer authors protection, it is that the technical nature of the protection could result an author losing their characters due to licensing. As an initial matter, the “Tarzan” example resulted in the author retaining the rights to the character, and serves as a poor example of an author’s licensing causing his or her control of a character due to licensing in other media.
Furthermore, the Burroughs case stands for the proposition that if a character is not “sufficiently delineated” than authors stand to lose their characters when licensing the characters to other media, a proposition that I do not necessarily disagree with. Feldman’s problem with the Tarzan example is that if the character was not “sufficiently delineated” than Burroughs would have lost their character due to licensing. Again, I don’t see the problem with this. When asking if a character is sufficiently delineated, one is essentially asking if a character is more than just a stock character. This seems like a pretty low bar to reach to gain protection. In fact, the court decided that “Tarzan” was sufficiently delineated based upon the following description: “Tarzan is the ape-man. He is an individual closely in tune with his jungle environment, able to communicate with animals yet able to experience human emotions. He is athletic, innocent, youthful, gentle and strong. He is Tarzan.” Burroughs v. Metro-Goldwyn-Mayer, Inc., 519 F. Supp. 388, 391 (S.D.N.Y. 1981) aff'd, 683 F.2d 610 (2d Cir. 1982). If a character is unable to meet such a low bar, should we even protect that character? I believe not. We should protect characters that deserve protection, and I think that the “sufficiently delineated” standard is not too much to ask of an author.
The second of these cases, Feldman argues, compounds the issue even more. The court in Silverman v. Columbia Broadcasting System was posed with the question as to the extent of rights conferred to characters that were now in the public domain. Plaintiff wanted to create a Broadway show based on characters of an earlier radio show. Although the characters were in public domain, the court ruled that plaintiff could use the characters in their play. However, any character traits added to the characters in works that had not entered the public domain could not be put into their musical. Feldman seems to have a problem with the intricacies involved with having serialized characters, because elements of characters are added over time there may be some problems when earlier stories enter the public domain whereas others don’t. If the first Spiderman story enters the public domain, do all the character elements of subsequent stories featuring the character enter public domain too? I can see how this posed a problem to Feldman in the 1990’s, luckily for us, we are very close to an answer to this question.
The recent litigation concerning the “Sherlock Holmes” character sheds some light on the problem Feldman poses. The seventh circuit in Klinger v. Conan Doyle Estate, Ltd. were asked whether an editor could produce new unauthorized stories featuring Sherlock Holmes when the first initial stories entered the public domain. The problem here is that there were still 10 remaining Sherlock Holmes stories written by Sir Arthur Conan Doyle that had not entered the public domain. The estate of Sir Arthur Conan Doyle, the creator of Sherlock Holmes, argued for post-expiration protection of Holmes. In discussing the policy implications of allowing such protection, the estate argued that, “It may take a long time for an author to perfect a character or other expressive element that first appeared in his early work. If he loses copyright on the original character, his incentive to improve the character in future work may be diminished because he'll be competing with copiers.” Klinger v. Conan Doyle Estate, Ltd., 755 F.3d 496, 501 (7th Cir. 2014).
The court disagreed with the Conan Doyle Estate, providing some insight as to the extent of copyright protection for characters whose original stories or works had entered the public domain. The court held that Sherlock Holmes had entered the public domain and plaintiffs were free to use the Holmes character. There is still a question as to whether the additional features of a character that are not in the public domain are still protected. The court answered this by adding that “these additional features, being (we may assume) “original” in the generous sense that the word bears in copyright law, are protected by the unexpired copyrights on the late stories.” Klinger v. Conan Doyle Estate, Ltd., 755 F.3d 496, 502 (7th Cir. 2014). Thus, case law has provided new guidance to the problem of what aspects of a character enter the public domain when the original work the character originates enters public domain.
Feldman’s last argument is that copyright protection affords no protection to “pure” characters. His argument is that the 1976 protects works and not characters, thus it offers no protection to artists who do not include their character in a work. He uses the case of CBS v. DeCosta to illustrate the problems of this. That case concerned DeCosta’s “Paladin”, a character that DeCosta dressed up as for rodeos, parades, horse shows, and other public appearances. DeCosta put his “Paladin” image on business cards with the accompanying phrase “Have Gun Will Travel.” DeCosta sued CBS after they made a television series featuring the “Paladin” Character” called “Have Gun Will Travel.” The court ruled that the character was not protected because DeCosta failed to register his photographs. His final argument is that all of DeCosta’s appearances after he handed out his photos were “arguably copies” that should have warranted protection.
Feldman states, “The injustice in demanding that actor Paul Rubens incorporate his Pee Wee Herman character into a film or television show before he can establish any rights in the character is readily apparent” David B. Feldman, Finding A Home for Fictional Characters: A Proposal for Change in Copyright Protection, 78 Cal. L. Rev. 687, 701 (1990). But is this really an injustice? I don’t believe that this is too much to ask of an author, otherwise, where would we draw the line? To use Pee Wee Herman as an example, that character has been featured in movies, plays, a television series, television appearances and certainly warrants copyright protection. But without incorporating the character into a work, what standard would we use to protect the character? I believe that asking protection for a pure character without incorporating it into a copyrightable work would be unworkable under the current copyright regime. Furthermore, with modern technology it would be easy to protect a pure character. If the “Paladin” character had been around nowadays, DeCosta would have only needed to video tape one of his performances to be incorporated into a “work.” Thus, I believe that granting “pure characters” copyright protection creates more problems than it solves.
The Problems with Alternative Protections
My initial thought when reading Feldman’s essay was that, in my view, it seemed like trademark and unfair competition seemed to make up for most of the deficiencies that Feldman asserts. However, he does make some arguments about how these alternative protections are deficient. I argue that these alternatives certainly are sufficient to make up for some of the holes in character protection in copyright law.
Trademark Law and Unfair Competition
I think Feldman’s argument against trademark law really boils down to one thing: trademark is used as a source indicator, and this provides inadequate protection for characters. He states, “trademark provides a legal home only for those well-known fictional characters whose names or visual images readily identify a single source of authorship and who have had significant exposure to the general public.” Thus, trademark protection for characters really depends on the character’s popularity and the public awareness of the character’s creator. Feldman seems to argue that this means that there is a very narrow range of characters in which trademark could protect.
My problem with this argument isn’t that trademark has a very narrow range of characters, my problem is that Feldman treats trademark like it should offer characters a complete protection. However, I think trademark law is only appropriate to fill the gaps of copyright law. It is true that trademark serves as a source indicator, and thus protects a very narrow range of characters. Furthermore, the protection it provides is quite extensive. Copyright is based on the premise that all works will eventually enter the public domain, however, trademark rights can last in perpetuity. Thus, a character like Spiderman or Wolverine would certainly be protected in trademark. They are both popular characters, and most of the public knows that it is Marvel Comics that produces their comics and films. It would seem that these characters could be protected in perpetuity. I don’t believe that this is an ineffective means of protecting characters.
Is Independent Protection for Characters Warranted?
Feldman’s essay is really arguing for the independent subject matter category of characters within Section 102 of the copyright act, and although I do think that characters warrant protection, I do not believe that this is necessarily the answer. In this next section I will look at the implications of creating a separate category for character protection within the copyright act. Firstly, will Feldman’s suggestion fix the problems that he has with character protection right now? I believe the answer is a firm maybe, and that alone is not sufficient to warrant creating a new subject matter category in Section 102. Furthermore, will creating this new subject matter category encourage more creativity and production of works? Again, the answer is likely that it won’t.
As an initial matter it is worth noting that the subject matter categories are not meant to be an exhaustive list of copyrightable “works”, they are merely a guideline. Section 102 states that “[w]orks of authorship include the following categories.” 17 U.S.C.A. § 102 (West). Nowhere in the statute does it say that protection in works is limited to those categories. Feldman seems to argue for a subject matter category that is settled in case law as already being protected to a large degree.
The first question is whether creating another subject matter category will have much affect at all. Feldman argues that:
The threshold copyrightability requirement leads to muddled inquiries regarding delineation and infringement. Moreover, attaching copyrights only to the work and not to the character limits protection for characters that move into new works and new mediums and leaves pure characters unprotected. The inadequacy of alternate protections, such as trademark, unfair competition, and the right of publicity, compound the failure of copyright to provide adequate protection. Creating a separate category of copyright protection for fictional characters could solve these problems.
David B. Feldman, Finding A Home for Fictional Characters: A Proposal for Change in Copyright Protection, 78 Cal. L. Rev. 687, 710 (1990)
The operative word here is “could.” I think Feldman is correct in stating that in creating an independent subject matter category for characters, the “sufficiently delineated test” and the “story being told” test might be replaced by standard infringement analysis. Certainly recognizing characters as an independent subject matter would not harm what I am calling the “character protection regime”, but what is to stop courts from applying the same tests once the category is recognized. If recognized as an independent category, there would still be the problem of defining what type of “character” would warrant protection. In creating the definition for a “character” Congress would likely look at the tests already laid out and discussed in this paper. Thus, recognition of an independent subject matter category for characters does not automatically fix the problems Feldman asserts.
However, I do agree that creating an independent subject matter category for characters does not necessarily harm the copyright regime. I say this because after looking at the copyright law in Japan, which does not have “subject matter categories”, the case law does lead to a rather confused application of copyright as it pertains to characters. As Kenneth L. Port states in his essay “Copyright Protection of Fictional Characters in Japan”:
“The scope of material protected by the Copyright Act is limited to ‘works’ (chosakubutsu) defined as ‘productions in which thoughts or sentiments are expressed in a creative way and which fall within the literary, scientific, artistic or musical domain.’…Therefore, to be copyrightable, a work must express thought or sentiment in a creative way and come within the literary, scientific, artistic, or musical domains.”
Thus, Japan does not have the nine subject matter categories that the United States does.
The next question is: what effect does this have on their protection of characters? Kenneth L. Port summarizes the Sazae-san Case wherein a famous Japanese cartoon artist sued the Tachikawa Bus Company because they had placed the faces of three of the leading characters of his cartoon strip on their tour buses. Their test for infringement was: “if a person were to look at the defendant’s buses and in his/her mind immediately connect those figures on the bus with the plaintiff’s copyrighted figures, that would sufficiently establish a copyright infringement.” (Port at 215). Furthermore, an Osaka court held that, “Popeye was an abstract idea when not contained within one of the originally copyrighted comic strips. Therefore, copyright law was not available to protect Popeye independent of those comic strips.” Id. at 225. What we see here is a very limited protection of characters in Japan. The “Popeye” case stands for the proposition that (in Japan), characters are not protectable when used outside of the context in which they were created.
So is this lack of protection attributable to a lack of subject matter category for characters? The answer has to be maybe. I believe there are more systematic differences between the two copyright regimes. One of which just happens to be a lack of subject matter categories. There is certainly a correlation (in this one case) between a lack of express protection for characters and thin protection for those characters. However, correlation does not necessarily mean causation, and without a more extensive comparison between Japanese and American copyright law I am not willing to give a conclusive answer.
However, even Kenneth L. Port does not believe that creating specific protection for characters in Japanese Copyright law would alleviate the problems of character protection. He states, “Even if an amendment was added to provide copyright protection to ‘(x) fictional characters independent of their work in tangible form,’ it would not clarify the issue. Judicial standards are still needed to determine under what factual setting fictional characters will receive copyright protection when someone other than the original author tries to use the character and when a character is an ‘idea’ rather than a ‘tangible expression’.” Id. at 222. Thus, even if specific protection were granted to characters in the Japanese context, scholars believe that it would not solve most of the problems of protecting characters.
My next inquiry is whether or not creating this independent subject matter category would further the incentive based justification for copyright law. Admittedly, I believe I am being a little unfair on Feldman’s assertions. Feldman is arguing for consistent and adequate protection for characters in the United States copyright regime. In that regard, Feldman does have compelling arguments in support of working towards a more uniform standard with regards to protection of copyrightable characters. I am being unfair because I believe we have completely different goals. I, on the other hand, am looking as to whether adding an independent subject matter category for characters would increase production of “works.” I think the inquiry can be summed up as follows: will creating a new subject matter category in section 102 of the copyright act create more incentive to create? I think any answer to that question would be purely conjectural, however, I would answer the question in the negative.
The adoption of the 1909 Copyright Act is illustrative because it shows the effect of introducing subject matter categories on “creativity.” The first Congress enacted a statute granting copyright protection only to “maps, charts, and books” in 1790. (Cite 1790 Copyright Act). Furthermore, the duration of this copyright was 14 years, with a renewal period of 14 years if the author was alive at the end of the first term. Although the 1790 Act was amended 1831, 1870, and 1891, the most drastic change occurred in 1909 when Congress enacted a comprehensive new copyright law. Among many changes, was the listing of 11 categories of works that warranted protection including books, periodicals, dramatic compositions, musical compositions, maps, works of art, photographs, and pictorial illustrations. (cite 1909 Act). Thus the 1909 saw a jump from 3 categories of works offered copyright protection to 11 categories.
Boldrin and Levine in their book “Against Intellectual Monopoly” believe that the substantive changes in the 1909 Act did not lead to more creation of works. They state:
We might well begin by asking how well the 1909 revision of copyright worked. Did it increase the rate at which books and other copyrightable new products were produced in the U.S.? Apparently not. Even abstracting from the general increase in literacy over the century and the enlargement in the number of items that are copyrightable (do not forget that, for example, software products are now copyrighted) the increase in the registrations/population ratio is miniscule in the forty years following the 1909 Copyright Act.
(Cite Boldrin and Levin, Chapter 5, p.2-3)
Boldrin and Levine’s research point out that in 1900, the percentage of works registered per population was 0.13% whereas in the years following the adoption of the 1909 Act only rose to 0.14% in 1925 and stayed at that same percentage in 1950. I am uncertain as to whether this information is conclusive, after all, the research is based on registered works and does not account for works created and not registered (which would be a near-impossible undertaking). Furthermore, it is also based on a percentage of population. As the population of the U.S. increased it does mean that the actual amount of copyrightable works did in fact increase.
However, Boldrin and Levine’s study does provide compelling evidence that, at least in the past, adding subject matter categories did not increase “creativity.” In that example, Congress added 8 more categories of copyrightable works, and Boldrin and Levine would suggest their effect on creativity was negligible. What would be the effect of adding just one more in the modern context? I would agree that the effect would be negligible. Again, this is purely conjectural, there are certainly many more differences between the 1909 Act and the adoption of the 1976 Copyright Act. Again, we are looking at the law’s effect on creativity more than 100 years ago. With modern technology we have been not only exposed to new ways to create works, we also have more avenues to display and circulate those works. I have no idea how this would affect the analysis. However, there is something to be said by the fact that an expansion of subject matter categories in the copyright context did not necessarily result in more creativity.
I do not believe that we need a new category for characters in section 102 of the Copyright Act. I do not believe that the supposed answer lies in an overhaul of our copyright regime, the next section explores an interesting alternative proposed by Gregory S. Schienke.
I have argued against the creation of an independent subject matter categories of characters in Section 102 of the copyright act, however, I think that a more viable solution to character protection is the creation of what Gregory S. Schienke in his essay “The Spawn of Learned Hand-A Reexamination of Copyright Protection and Fictional Characters: How Distinctly Delineated Must the Story Be Told?” calls a “copymark.”
Schienke shares the same concerns with creating an independent subject matter category for characters. One of the more glaring problems is the application of the “fixation” requirement. Again, Feldman wants to find protection for “pure” characters, or characters tht may not be fixed in a work or a tangible medium of expression. The “fixation” requirement is one of the most important prerequisites to attaining a copyright in a work. As Shienke states, “If an equitable application of the law is the reason for such creation, how could a pure character be included? In addition, unless there was a statutory definition for the “threshold of delineation” for fictional characters, what would prevent a person from asserting their assumed rights in stock characters?”Gregory S. Schienke, The Spawn of Learned Hand-A Reexamination of Copyright Protection and Fictional Characters: How Distinctly Delineated Must the Story Be Told?, 9 Marq. Intell. Prop. L. Rev. 63, 83 (2005). Schienke’s solution is independent protection of characters outside of the realm of copyright, a sort of protection that is in-between a copyright and a trademark.
Shienke suggests that because characters really do exist in two legal worlds, both copyright and trademark, that it makes sense to create statutory protection with elements of both bodies of law. His requirements are as follows:
The owner of a fictional character wishing to apply for copymark protection would be required to show that the character: (1) originated in a work available for copyright registration; (2) had been in use in commerce; (3) had been in use for a minimum of five years; and (4) is famous.
Gregory S. Schienke, The Spawn of Learned Hand-A Reexamination of Copyright Protection and Fictional Characters: How Distinctly Delineated Must the Story Be Told?, 9 Marq. Intell. Prop. L. Rev. 63, 86-87 (2005)
I do think that a “copymark” is a more viable alternative solution to the character protection problem for a number of reasons.
Firstly, these requirements are firmly rooted in existing copyright and trademark law, making case law available to clarify any issues with the statute.
Furthermore, the first requirement assures that “the character has authorial roots of the kind implicated by the Constitution. The requirement is necessary to avoid the potential abuse by certain trademark owners of “double dipping”: applying for copymark protection for a trademark.” Id. Feldman would argue that this requirement does not address the problem of treating a character as part of the work. However, the standard fixes most of the problems Feldman asserts, for one there is no determination of whether the character is “sufficiently delineated” or whether the character is “story being told.” These tests would not come into play because all that is required is that the character originate from a copyrightable work.
The final three requirements are taken directly from settled trademark law, and imply that the character must be used in commerce, used for a minimum of five years, and the character must be famous. These requirements are important because they are a sort of qualitative requirement for this rather extensive protection of characters. I say this because I believe that by basing protection on whether the character is both famous and used in commerce for 5 years we are only protecting characters that the public have deemed worth protecting. This is good, because it stops protection of “stock characters” and does not hurt the commons. I say this because I believe that “stock characters” rarely become famous and are usually not used in commerce the way that full-fledged characters do. Thus, a character like Spiderman would certainly warrant protection under a “copymark” but a character that Spiderman saved from a dastardly villain would not get the same protection.
My only problem with the “copymark” is that like a trademark, a “copymark” would last in perpetuity. I believe that the commons would be hurt if the characters appearing in works never entered public domain. The trade off with most of IP (except trademark law) is that in exchange for protection over the IP, that IP will enter the public domain for the free dissemination of the public. I do not think that the public is served by keeping these characters protected in perpetuity. The free flow of ideas is necessary for innovation, and that is why I propose an amendment to Schienke’s “copymark” requirements. I would add the caveat that the “copymark” would last only as long as the work in which the character originated in has not entered public domain. This makes sure that the character will eventually enter the public domain.
Characters are worth protecting. Not only is there an economic incentive in creating a famous character, I believe that there is an emotional one too. People might relate to the overall story of a character, but at its core, people relate to characters. Thus, the law needs to find a way to protect characters that the public deems worth protecting. Do I believe that copyright law is insufficient to protect these characters? Yes and No. I do believe that Feldman’s arguments are compelling in that the current character protection regime does result in some inconsistent applications of the law. However, the current copyright regime’s requirements that a character either satisfy the “sufficiently dileneated” test or the “story being told” test is not a high bar to meet. The law needs to strike a balance between creating an incentive to create works while making sure that these works eventually enter the public domain. I do not believe that creating an independent subject matter category for characters is the answer to that problem. This may be a cop-out but I would suggest the “if it ain’t broke, don’t fix it” approach. My research has not come across any glaring examples of insufficient protections of characters, and if we were to look for more adequate protections, I believe we should look outside the realm of copyright to a modified version of what Schienke calls a “copymark.”
My love of comic book characters made me look at the state of copyright law in terms of character protection. This begs the question: where is the law going? The case law seems to suggest that the “sufficiently dileneated” test and the “story being told” test are here to stay for the time being. Although there seemed to be some press regarding the recent litigation over the Sherlock Holmes case, the fact that the Supreme Court has denied certiorari seems to indicate that, at least for now, the law governing character protection will remain stagnant. Is this dormancy a good thing? Hard to say, but I believe that current copyright law will be sufficient until any further innovations occur.