I.    INTRODUCTION

¶ 1         Trademark dilution is a right awarded to famous trademark holders to prevent the use of the same or similar marks on non-competing goods or services.¹ Trademark infringement, in contrast, requires competition; if there is no competition between the parties, there can be no infringement.² The concept of dilution was created to plug this "gap."³ That is, Kodak camera company could bring a suit for trademark dilution against a company making "Kodak" brand cameras,⁴ but such a suit would not doctrinally satisfy the requisite elements for trademark infringement because there would be no competition between a bicycle manufacturer and a camera manufacturer.

¶ 2         Japanese trademark dilution jurisprudence is in a state of confusion. In an otherwise highly rational, highly developed system, this scattered jurisprudence is unexpected.⁵ Perhaps the cause of action is still too new,⁶ or perhaps Japanese courts, much like American courts,⁷ are reticent to recognize the full scope of the dilution right. It is more likely, however, that the confused status of Japanese trademark dilution law indicates a judiciary at odds with the legislature. This jurisprudence is not scattered because of a lack of technical skill by the judiciary, but to rein in a right the judiciary sees as inconsistent with the purposes of trademark protection in Japan.

¶ 3         The Japanese Legislature first recognized something resembling a dilution cause of action in 1993. In that year, the Japanese legislature ("Diet") amended the Unfair Competition Prevention Act (hereinafter "UCPA").⁸ The two provisions of the UCPA that courts have used to deal with trademark dilution are as follows:

¶ 4         Article 2-1: As used in this Law, the term "unfair competition" shall mean:

(1) the act of using the Goods or Other Appellations⁹ (as used hereinafter, "Goods or Other Appellation" shall mean a name connected with a person's business, trade name, trademark, mark, the container or package of goods, or any other appellation of goods or businesses) which is identical with, or similar to, another party's Goods or Other Appellation that is well-known among the consumers,¹⁰ . . . and causes confusion with the goods or business of that other party's.

(2) the act of using Goods or Other Appellations of another that are identical with, or similar to, another person's famous Goods or Other Appellations . . . .¹¹

¶ 5         The most striking omission from these two statutory provisions is the word "dilution" (kishakuka) itself. Japan has no statute containing the term "dilution" that provides a cause of action for an owner of a famous trademark to prevent the subsequent use of that trademark on dissimilar goods in a non-competing manner. Therefore, although Japan is a civil law country, and courts must only apply the law as stated, rather than generate new options through their opinions,¹² by definition any dilution case in Japan is judge-made law.

¶ 6         Although Article 2-1-2 of the UCPA is considered to be the "dilution" provision, not Article 2-1-1,¹³ in many important cases that reference dilution directly, courts rely on Article 2-1-1, and not Article 2-1-2.¹⁴ Article 2-1-1 of the UCPA has been broadly utilized to prevent the use of any well-known registered or unregistered appellations of source.¹⁵ Litigants have relied on it to prevent the use of everything from shelving shaped like a beehive¹⁶ to the importation of pharmaceuticals bearing the word VIAGRA.¹⁷ As such, it is also relied upon to prevent dilution.¹⁸

¶ 7         Japanese courts often do not make the critical distinction between a provision that requires a showing of confusion (Article 2-1-1) and one that does not (Article 2-1-2).¹⁹ Furthermore, Japanese courts do not distinguish between "causes confusion" and "an apprehension" of confusion.²⁰

¶ 8         In the twelve years since inception of the UCPA,²¹ there have been relatively few reported decisions that squarely address dilution.²² In the same time period, there have been over 1,300 opinions issued regarding federal trademark dilution in the United States ²³ and countless other state court claims. That is, United States courts have been inundated with trademark dilution claims. This is not the case in Japan.

¶ 9         Why are there so few opinions in Japan? In a society with one half the population of the United States²⁴ and the second largest economy in the world,²⁵ would not one expect more trademark dilution cases?

¶ 10         The point of this article is to directly address this paucity of trademark dilution cases in Japan. This article concludes that first, there are more dilution cases brought in Japan than are reported (although that number is probably negligible). Second, because the UCPA is vague and does not even use the word "dilution," Japanese courts do not or cannot distinguish between fundamental elements of the cause of action. Third, the defensive trademark system, underutilized as it is, provides an alternative to dilution protection. Finally, the theoretical understanding of trademark law renders dilution claims superfluous.²⁶

¶ 11         The result is that the dilution cause of action is very unsure in Japan. This lack of certainty may prevent some entities from recognizing and pursuing their rights under the statute. The confused state of Japanese dilution protection indicates that Japanese courts do not understand the cause of action or are reticent to apply it as it is written. Given the high degree of technical skill of Japanese judges,²⁷ it is not likely that they are simply making serious errors when applying dilution law to any given set of facts. Rather, it appears that Japanese judges have an inherent distrust of the notion of dilution. Japanese judges seem reluctant to apply the language as written and, instead, seek other options to attempt to confine the expansion of the trademark right.

II.    WHAT DILUTION STATUTE?

¶ 12         The language of Article 2-1-1 of the UCPA is vague and broad, and clearly contemplates preventing the use of someone else's well-known appellation of source. As such, for both clauses 1 and 2 of Article 2-1 of the UCPA, the initial inquiry involves determining what an "appellation" is. Infringing use can take the form of any appellation of source; however, Japanese courts have yet to broaden the understanding of this provision to sensory marks.²⁸ Although the scent of thread was recognized as a valid and registrable trademark in the United States, it would not be possible to protect such indicators of source in Japan.

¶ 13         In the SEIROGAN TOI A case,²⁹ the Osaka District Court concluded that each of the component parts of the name were common descriptive terms, but the combination of these terms, the long period of use, and the remarkable success of the product made it appropriate to call what the plaintiff had an "appellation" (hyoji).³⁰

¶ 14         In another case, a middle school started using the term Aoyama Gakuin Chugakko to refer to its school. Aoyama Gakuin University is a Methodist college that started in Tokyo in 1874,³¹ a mere six years after Japan was opened to the outside world.³² The middle school was enjoined from using that name after the college established that its name amounted to an "appellation" of source (hyoji).³³ It is not surprising that this school was able to establish that their name was an appellation after 127 years of exclusive use.

¶ 15         The core of Article 2-1-1 seeks to prevent the use of a "well-known" (hiroku ninshikisareteiru) "good or other appellation" ("shohintohyoji") - either registered or unregistered. However, what constitutes a "good or other appellation" is not entirely apparent. It is clear that the statute contemplates a good itself as indicating source because it uses the term "shohintohyoji," which would literally be translated as "good or other appellation." On the other hand, the statute defines shohintohyoji in a parenthetical as "a name connected with a person's business, trade name, trademark, mark, the container or package of goods, or any other appellation of goods or businesses." Despite the implication of the word shohintohyoji, its given definition seems to preclude the protection of "goods" themselves unless the good in question is a "container or packaging of [other] goods."³⁴

¶ 16         Article 2-1-1 also requires the plaintiff to show actual confusion - "causes confusion" "kondowo shojisaseru."³⁵ In order for confusion to be established, the parties must be in competition with each other. As such, it does not appear to be the appropriate provision for preventing dilution; however, as stated above, several courts have relied upon this provision. Article 2-1-2, a provision that allows the holder of a famous appellation of source to enforce that right against third parties regardless of competition, is a more appropriate provision to apply to a likely dilution setting.³⁶

¶ 17         In 1992, the Japanese government commissioned a study of the need to have a "dilution-like" statute. It concluded as follows:

In the current information society, product or business appellations are being more broadly used via a variety of different media. These brand images are becoming well known. These independent brand images have specific customer appeal. As such, they have attained an independent value of their own. Through use of these famous appellations, for example even when there is no confusion, such a user can gain the customer appeal of this famous appellation even though he did nothing to achieve it. This is known as "free riding." As a result, consumers might become confused that this newcomer is the same as the company that endeavored to obtain high trust and repute. In this case, the good image of the firstcomer would be damaged. This is known as "dilution." [In such a case in the past,] courts have simply presumed confusion even though the facts of the case made confusion an impossibility. Judges have deemed such a conclusion to be appropriate; however, they have come to question the notion of presuming confusion. Therefore, frankly speaking, it is appropriate to create a new cause of action where confusion is not necessary in order to protect famous appellations.³⁷

¶ 18         Several things become apparent from this summary. Most striking, of course, is that the contemplated dilution cause of action is somewhat different than the typical American notion of dilution, where actual harm must be shown to establish dilution when marks are not identical.³⁸ Although brand identity is a growing phenomenon in Japan, it is certainly not the case that it did not exist prior to 1992, as the report of the intellectual property policy committee suggests. Nevertheless, in the United States, while brands might have an independent identity, they do not have an independent monetary value.³⁹ Trademarks themselves are not property in the American system, as such, they cannot have an "independent value." Any value a trademark has is directly tied to the goodwill behind the mark and thus can never be separated from that goodwill.⁴⁰

¶ 19         Dilution is a cause of action somewhere between blurring and tarnishment.⁴¹ To find a party liable, this report would require that the "good image" of the first-comer actually be damaged somehow by the party engaged in dilution. Actual damage is required. This is also a distinction that seems lost on Japanese courts.

¶ 20         Article 2-1-2 is incredibly broad in its scope. Any use of anyone's famous good or other appellation, regardless of confusion and regardless of competition between the parties, creates a cause of action. To establish "dilution" under Article 2-1-2, the plaintiff must establish the following:

1. use of the plaintiff's goods or other appellation by the defendant;

2. the plaintiff's appellation is famous;

3. the defendant's appellation is the same or similar to the plaintiff's.⁴²

¶ 21         There appears to be little case law on the first element.⁴³ Essentially, this element contemplates the obvious: that the defendant has to be using the plaintiff's appellation. This includes the possibility that the defendant may be using the appellation for something other than source denoting purposes. It is only the source-denoting purpose that is protected.⁴⁴

¶ 22         Fame, on the other hand, deserves serious thought and attention, and Japanese courts have responded to this challenge. Fame becomes relevant due to another key distinction between Article 2-1-1 and Article 2-1-2. Article 2-1-1 applies to well-recognized (hiroku ninshikisareteiru) goods or other appellations, while Article 2-1-2 applies only to famous (chomei) goods or other appellations. For an appellation to be well-known, it need not be known nationwide⁴⁵ - simply being known by most people in one region of Japan will suffice.⁴⁶ However, for a mark to be famous, and thus protected under Article 2-1-2, it generally needs to be known nationwide, such that even someone in a completely different market would recognize it.⁴⁷

¶ 23         To determine how "famous" a mark must be to trigger Article 2-1-2, courts examine several elements.⁴⁸ The first question involves the degree of fame. The second asks "for whom must the mark be famous?" This second question can be further broken down into questions of regional fame and field-specific fame. Shoen Ono describes the requirement that the appellation be famous in the following terms: "[t]o be famous, an appellation must achieve an exacting standard... it must appear to be a striking or remarkable appellation."⁴⁹ Even though the degree of fame is supposed to be an exacting standard,⁵⁰ if the appellation is famous enough, Japanese courts will presume confusion.⁵¹ In the Levi Strauss & Co. v. K.K. Edwin Mfg.,⁵² the court found that the fact that a mere 18.3% of respondents nationwide could correctly identify the stitching on the rear pockets of Levi's jeans was adequate to establish that the mark was famous.⁵³ This decision begs the question that if 18.3% satisfies the test for a famous appellation,⁵⁴ how exacting could the standard be? A clear declaration of the standard by the court would certainly be desirable. Given the status of dilution cases in Japan, however, this seems very far off indeed.

¶ 24         Although the basic, black letter rule about fame for the purposes of Article 2-1-2 is that appellation must be nationally famous in Japan,⁵⁵ some important commentators feel differently. For example, if a product is used only in northern Japan, one could still consider its appellation as famous, depending upon the characteristics of the product and of the purchasers.⁵⁶ There are others who feel that even though the appellation is not famous nationwide, as long as the defendant is using the same or similar appellation where it is famous, this use should suffice for the fame element.⁵⁷ It is clear that the standard for fame for the purposes of Article 2-1-2 is quite high. Although there may be some narrow exceptions, nationwide fame is generally required.

¶ 25         Japanese commentators also discuss a "field of use" of fame.⁵⁸ Under this theory, there are fields of use where an otherwise famous mark would not be famous. While this is directly inconsistent with the purpose of dilution prevention in the first place, there is case law to support this theory. For example, in K.K. Toraya v. K.K. Kurokawa,⁵⁹ the plaintiff had used TORAYA on, or in connection with, many kinds of candies and bread. The defendant intended to use TORAYAKUROKAWA on or in connection with the manufacture and sale of Japanese sweet bean jelly, jewelry, textiles, and beauty salon massages. The court held that the plaintiff's appellation was famous in the field of sweets but not in other fields.⁶⁰

¶ 26         Holdings that presume confusion even when there is no competition seem to weaken the fundamental purpose behind dilution protection in the first place: to prevent a non-competitor from using a famous appellation on a non-competing good or service. Taking these holdings on their face, it's not hard to see why dilution rationale has been unpopular in Japan.

¶ 27         There is some debate over what is required in addition to the level of fame an appellation must have before it is protected.⁶¹ Some argue that an exceedingly high brand identity is required.⁶² Tatsuki Shibuya lists the following as essential elements required for protection: (1) specific distinctiveness; (2) fame; (3) very high quality goods or services; (4) unique positioning or status; and (5) abstract distinctiveness.⁶³

¶ 28         It is difficult for Americans to conceive of the difference between "specific distinctiveness" and "abstract distinctiveness." Under American law, any "abstract" distinctiveness of an appellation of source is unprotectable; however, if you apply it to the traditional dilution rationale, it becomes a bit clearer. Specific distinctiveness is when you hear the word KODAK and know that the manufacturer of the goods or services has something to do with cameras. However, if the word KODAK is used on bicycles, there is no longer specific distinctiveness because the Kodak company does not sell bicycles. Expectation of some sponsorship of the bicycle company becomes the "abstract distinctiveness" which Shibuya argues is an essential element.⁶⁴

¶ 29         Others require that the mark be "particularly striking" (tokubetsu kenchosei).⁶⁵ This would be a standard even in excess of "exceedingly high brand identity." If this standard were used uniformly, a truly "famous" appellation would be rare. Additionally, in order to establish that a mark has been specifically diluted, some commentators require much more than the claim that the appellation was famous.⁶⁶ They also require a positive determination regarding the appellation's originality, uniqueness, and impression.⁶⁷ In other words,in addition to showing that the mark is famous, the plaintiff must establish that the junior user's goods or services are inferior.⁶⁸

¶ 30         Overall, Japanese commentators conclude that in order to be protected by Article 2-1-2 of the UCPA, a mark has to be truly remarkable. It would seem that very few marks could carry this burden. This may, of course, be why the number of cases is so small. However, the cases that have been brought and succeeded in establishing fame involve appellations other than Japan's most famous marks such as Sony or Matsushita. Much lesser appellations have succeeded in establishing fame. Therefore, in reality, this "exacting standard" is not that exacting. That is, the mark must be remarkable but it does not have to be among the strongest marks in the world.

¶ 31         In order for the provisions of Article 2-1-1 or 2-1-2 to apply, the second-comer's appellation must be the same or similar to the first-comer's.⁶⁹ However, because confusion is not required for an Article 2-1-2 violation, the element of similarity is given extra weight.⁷⁰ Similarity is determined by comparing the sound, meaning, and appearance of the two appellations to determine if there is an apprehension that the two could be confused by a customer.⁷¹

¶ 32         In determining similarity, the analysis should be broken down into three steps. The first step is to determine the essential elements of the plaintiff's appellation. The second step is to determine the essential elements of the defendant's appellation. The third step is to compare the essential elements of the two appellations.⁷² The other non-essential elements should be disregarded.⁷³

¶ 33         Determining the essential elements of an appellation is done by omitting the portions of the appellation that are common, descriptive terms. The remainder should then be compared on an overall basis to determine if they are the same or similar in sound, meaning, or appearance.⁷⁴

¶ 34         Although the statute only allows the holders of "famous" appellations to prevent the dilution if the second-comer's appellation is the same or similar to the famous appellation, courts seem willing to apply this in a manner that is not as strict as the black letter law seems to imply.

III.    REPRESENTATIVE CASES

¶ 35         There are relatively few trademark dilution cases in Japan.⁷⁵ Below is a description of some representative cases that would amount to precedent if Japan was a common law system.⁷⁶

¶ 36         Negurosu is an electronics manufacturer. Matsushita is a competitor but operates in a geographically different market. Matsushita adopted the shape of an electrical switch which had been manufactured and sold by Negurosu for quite some time. The question was whether this electrical switch could constitute a well-known⁷⁸ good or appellation for purposes of Article 2-1-1 of the UCPA and as such form the basis of an injunction preventing the future sale and distribution of Matsushita's electrical switch.

¶ 37         The Tokyo High Court held that the switch was a well-known good or other appellation, and ordered Matsushita to cease sales of the switch and pay damages in the amount of approximately $140,000. Regarding the dilution claim, however, the Tokyo High Court stated as follows:

Matsushita argues that after it developed its product it used the same or similar shape on or in connection with the sale of a plurality goods sold to a plurality of consumers and that this situation continued for some time. In such a situation, if there was dilution, the product's shape would have ceased to be an appellation of source and it would no longer be a famous good or other appellation. To be sure, the type of dilution that Matsushita argues for is possible; however, in that case, it would not be sufficient to simply use the same product configuration on the same goods. In Matsushita's argument, actual sales for a period of time would be required before dilution could occur. In the case at bar, [Negurosu's products] have a limited share of the marketplace. They do not have an overwhelming presence in the marketplace and they have not been used for an extended period. Therefore, we cannot say that through dilution Negurosu's fame of its appellation has been diminished.⁷⁹

¶ 38         Of course, this is not a traditional dilution case even though the court labels it as such. The court discusses "dilution" but analyzes this case under Article 2-1-1 of the UCPA which is not a dilution statute because it requires the plaintiff to show confusion.

¶ 39         Additionally, Article 2-1-1 says "causes confusion" (kondowo shojisaseru), and not "an apprehension of confusion" (kondono osore), which is used as the test for determining trademark infringement.⁸⁰ The court here determines that the test for dilution should be an apprehension of dilution without recognizing that this case falls under Article 2-1-1, as the parties are in competition with each other.

¶ 40         In the United States, the Supreme Court requires a showing of actual dilution if the marks are not identical.⁸¹ Although Congress is doing its best to overturn this outcome,⁸² at least for the time being, the United States applies a more restrictive standard to dilution than Japanese courts apply.⁸³

¶ 41         In this rather odd case, an entertainer took the stage name of Takachi Noboru, represented by kanji characters that also read "kochi tokyu" or Tokyu Coach, and started performing in January of 1993. The plaintiff, Tokyo Kyuko Dentetsu, had used the appellation "Tokyu" since 1922. Its use resulted in that appellation becoming well-known as part of the "Tokyu Group" set of marks. Based on Article 2-1-1 of the UCPA, the plaintiff sought an injunction to prevent the use of "Tokyu."

¶ 42         This case is ripe with Japanese puns and a little complicated to explain in English. The plaintiff, among other activities, operated a bus line called Tokyu Coach beginning in 1976. Another reading of the kanji characters representing the defendant's stage name is "kochi tokyu" or "Tokyu Coach," precisely the appellation used by the plaintiff for thirty years in different characters but resulting in the same pronunciation.

¶ 43         In an opinion that is said to feel the effects of the American dilution rationale,⁸⁵ the Tokyo District Court found the appellations confusing and issued the injunction. As a general rule, for the appellations to be confusing, the parties must be in competition with one another. In a one-sentence conclusion, the court observed that it was not possible to believe that the parties were not in competition whatsoever (using this double negative).⁸⁶ That is, because the court elected to use the double negative, it appeared like it was backing into this conclusion. The court clearly was bending over backwards to find competition when none really existed. As such, enough competition was shown for establishing the Article 2-1-1 cause of action, and the court does not have to address any dilution claim.⁸⁷

¶ 44         The defendant claimed that he was engaged in parody and that he had come to rely on the acquiescence of the plaintiff. In response to this argument, the court stated:

However, Article 2-1-1 of the UCPA includes the right to prevent acts which amount to the free-riding of a well-known appellation whose value has been built up over time through unusual effort and the right to prevent the dilution of such well-known appellation. The defendant claims that he adopted the stage name so he would have a name that was distinctive and was easy to remember. However, the stage name is not a natural person's name, was not a name the entertainer was assigned at birth, and is a name he chose himself. As such, the conduct of promoting the stage name as a famous name using someone else's well-known appellation or a portion thereof is conduct . . . which the UCPA prohibits. There is no evidence that the plaintiff specifically allowed this conduct and no evidence that the plaintiff permitted this alleged custom. Therefore, we cannot employ the defendant's theory.⁸⁸

¶ 45         Although this analysis of dilution is repeated in the case a number of times, there is no discussion of what "dilution" might mean or what the parameters of the claim might be. The reader is left taking judicial notice that there is something called dilution (kishakuka) and, whatever it is, it happened in this case.

¶ 46         Gibson Guitar had sold its Les Paul design for decades in Japan. Fernandes and others adopted a similarly shaped electric guitar. Although Gibson prevailed on its claim of dilution in the United States,⁹⁰ it failed in its attempts to enjoin Fernandes in Japan. Specifically addressing a dilution claim, the Tokyo High Court stated as follows:

Gibson's Les Paul guitar became well known among rock musicians and fans in Japan at least by 1973. We also have found above that this product configuration has come to represent Gibson's goods in the music industry. We have also found, however, that since at least 1973, Gibson's product configuration no longer represented the source or origin of any good. That is, for over twenty years more than thirty different brands were being used to sell guitars that resembled appellant's product configuration. During this period, Gibson admits that it took no steps to curb such use. Therefore, it is impossible to say that consumers recognize the source of the Les Paul guitar simply by looking at the good. Thus, consumers would not remember the particular source of the guitars merely by looking at their shape.

On this point, Gibson argues that the dead copies being manufactured and sold in Japan were knowingly placed in the stream of commerce as dead copies. The manufacturers were passing these products off not as their own goods but as those of Gibson's. As such, these copies in the stream of commerce diluted Gibson's ability to use the product configuration to identify it as the source of the goods. Therefore, Gibson argues that the Les Paul shape lost its ability to identify Gibson as the source of its guitars.

The reality, however, is that the consumers recognize Gibson's guitar as one of the many copies and, as such, Gibson's guitar configuration does not function to identify a particular source of the goods; rather it now identifies many sources of the particular goods.

Gibson argues that consumers, upon being confronted with the Les Paul guitar, confuse the configuration of the goods with the source of the goods. This, however, simply is mistaken. Even if consumers, when confronted with the Les Paul guitar, know that it originated with Gibson, and even if they know a copy from the original, they do not believe that all guitars shaped like the Les Paul model originated with Gibson.⁹¹

¶ 47         Normally, a dilution cause of action arises when there is no competition between the parties. Here, although the parties were in direct, head-to-head competition in the marketplace, the court was detered. Instead, the court concluded that guitars shaped like the Les Paul guitar had become so popular in the marketplace and that the shape of the guitar had become generic.⁹² It appears that one lesson from this case is that dilution in Japan does not apply only to non-competitors. Because the UCPA in Article 2-1-2 is silent on the role of competition, this court presumed that it meant that the statute applied to competitors and non-competitors alike.

¶ 48         Yoshinori Kobayashi is a famous and extremely controversial author of manga in Japan. Some manga are far more than mere conventional comic books. Some manga are serialized, sophisticated, and tell quite elaborate, mature stories.⁹⁴ In keeping with Kobayashi's general persona, his manga are extremely controversial as well, such as including alternative histories of major events such as World War II. In the manga at issue in this lawsuit, titled "On Taiwan," Kobayashi depicts Taiwanese women begging to become Japanese soldiers' sex slaves in order to advance their social standing.⁹⁵

¶ 49         Kobayashi is also a lead editor of a history textbook published by the Japanese Society for History Textbook Reform.⁹⁶ When this textbook received Ministry of Education approval, riots broke out in China and Sino-Japanese relations reached their lowest level in post-war history.⁹⁷ Kobayashi has an extremist, right-wing agenda, but he is also making record sales of his manga in Japan.⁹⁸

¶ 50         In Kobayashi v. Uesugi, Uesugi wrote a book that criticized Kobayashi's manga "Gohmanism Sengen."⁹⁹ Uesugi's book was titled "Datsu Gohmanism Sengen."¹⁰⁰ "Sengen" means "manifesto" or "declaration." "Gohmanism" is more controversial, as Kobayashi coined the term. Literally, gohmanism means "arrogant-ism." Therefore, literally translated, Kobayashi's title means "Declaration of Arrogant-ism." Uesugi's book title means "Ex-Declaration of Arrogant-ism."

¶ 51         What this really means is that Kobayashi believes only that he is right (he seems, at least, to believe in his manga). The manga stands for the notion that one should be proud of being arrogant because it is only through this arrogance that people might actually see the "true" story of Japan's history. Gohmanism would have us believe that World War II was fought for the noble cause of freeing Asia from white colonists, that sex slaves in Korea and China thank their persecutors for improving their status in life, and that war crimes in the rape of Nanking did not happen.¹⁰¹ Essentially, this amounts to Japanese ultra-right wing extremist vacuousness.

¶ 52         Uesugi's book consisted of a panel by panel discussion of the accuracy of Kobayashi's claims. Uesugi pointed out the oddities as well as the outright myths perpetuated by the characters in Kobayashi's work. Kobayashi did not take kindly to being criticized; and consequently sued Uesugi for copyright infringement. Kobayashi claimed that "gohmanism" acted as an appellation of source, and the mark had become famous.¹⁰²

¶ 53         The district court agreed with Kobayashi that "gohmanism" was a good or other appellation. They disagreed, however, that Uesugi had "used" that good or other appellation as required by Article 2-1-2 of the UCPA. On this point, the Tokyo District Court concluded:

"Datsu Gohmanism Sengen" is the title of the defendant's book. Since the defendant's book title includes the title of the plaintiff's work, we conclude that the defendant's use is a good or other appellation.

However, if one includes someone else's good or other appellation in explaining the goods, contents or characteristics of that person's work, we will not call such conduct the "use" of that good or other appellation or a similar appellation thereto [for purposes of the Unfair Competition Prevention Act]. . . . The defendant's use of "Gohmanism Sengen" was to explain the content of his book. As such, we do not recognize the argument that the defendant used the plaintiff's appellation or an appellation similar thereto.¹⁰³

¶ 54         If this case arose in the United States, the fair use doctrine would apply.¹⁰⁴ Uesugi is making a nominative fair use of Kobayashi's appellation.¹⁰⁵ He is not using it in order to identify the source; he is using it to describe. Coupled with the "datsu" part of Uesugi's mark, this case is very similar to the "www.walmartsucks.com" cases¹⁰⁶ under the Uniform Domain Name Dispute Resolution Policy¹⁰⁷ in the United States.

¶ 55         Instead of a "fair use" analysis, this case focuses on whether including someone else's appellation in the title of a book can be considered "use" at all under Article 2-1-2 of the UCPA. Concluding that such conduct does not constitute "use" of a good or other appellation as required by Article 2-1-2, the court dismissed Kobayashi's claim.¹⁰⁸

¶ 56         In this, Japan's first domain name case, the defendant had begun using the appellation JACCS in various places on its website and in its domain name, including registering the domain name "www.jaccs.co.jp." The plaintiff was a large, extremely well-known credit card company commonly called JACCS. The defendant claimed that it had legitimate commercial interest in the domain name because it stood for "Japan Association of Cozy Cradle Society." One product the defendant sold was toilet seats.

¶ 57         The case is an example of the use of Article 2-1-1 to enjoin the use of the domain name at issue under a dilution theory. The parties were obviously not in competition. As such, any injunction that might have issued in this case would be justified only under a dilution rationale. However, in enjoining the action, neither the District Court nor the High Court used the term "dilution" in their opinions.

¶ 58         The defendant claimed that the statute required the "use" (shiyo) of another's good or other appellation. Because "jaccs" merely appeared in its domain name, the defendant argued that it was not "using" the plaintiff's trademark in commerce. The court dismissed this argument and held that registering and having an active website that included the plaintiff's appellation was "use" of that appellation for purposes of Article 2-1-1.

¶ 59         In this case, the defendant's stitching in its "501" jeans around the rear pockets closely resembled the stitching of Levi's "505" jeans. The court, quite accurately, found that this was both an infringement and a violation of Article 2-1-1, but not a violation of Article 2-1-2. As the parties were clearly in competition, Article 2-1-1 was the more appropriate provision; however the court appropriately refused to label the defendant's conduct "dilution" (using the English term, not the Japanese term) even though the plaintiff had demanded as such.¹¹¹ Therefore, it used the correct statutory provision to partially enjoin the defendant, but it used the wrong justification in its explanation.

¶ 60         In this case, a company by the name of J-Phone started providing cellular telephone services in February 1997.¹¹³ Later that year, the appellant, K.K. Daigyotsusho, registered the domain name "www.j-phone.co.jp" with JPNIC (Japan Network Information Center-the equivalent of ICANN in the United States¹¹⁴). Even though J-Phone's services preceded Daigyotsusho's advertisements by only a few months, the court found that J-Phone had advertised extensively in newspapers, television, radio, and magazines.¹¹⁵ The court found that its advertising campaign had reached all corners of Japan, sufficiently establishing that the plaintiff's product as well-known.¹¹⁶

¶ 61         Interestingly enough, the appellant found usages of "j-phone" in New Zealand and Australia for telecommunication companies dealing with Japan. As such, the appellant argued that "j-phone" was an ordinary term for a Japanese telephone company. However, the lower court decided (there is no mention of this argument on appeal) that such use was negligible and done in a foreign country. Therefore, no weight was given to this argument at all.¹¹⁷

¶ 62         Finally in this case, the defendant argued that, as a result of the lower court's ruling that it could not use the appellation "j-phone" in a domain name or on its web site, the court had restricted the appellant's freedom of expression (hyogennojiyu)¹¹⁸ and the holding was unconstitutional.¹¹⁹ In response to this claim, the Tokyo High Court stated that for a judge "[to] prevent unfair competition or the apprehension of unfair competition is only natural. Through the regulation of unfair competition, even if speech is restricted to some extent, that is not unconstitutional."¹²⁰ Therefore, the freedom of speech argument was dismissed by the High Court.¹²¹

¶ 63         Finally, in a case that proposed to draw a distinction between the "well-known" provision from Article 2-1-1 of the UCPA and the "fame" provision from Article 2-1-2 , the Tokyo District Court found VOGUE Magazine well-known but not famous and, therefore, not diluted under Article 2-1-2.¹²³

¶ 64         The plaintiff was the publisher of a magazine whose name, although a common term in French that means "in style," was considered by the court to have acquired a distinctiveness as used on or in connection with the magazine.¹²⁴ The defendant owned a condominium complex called "La Vogue Minami Aoyama."¹²⁵

¶ 65         The Tokyo District Court held that even though the appellation "vogue" is a term in common usage in the United States and France, it was a rather distinctive appellation in Japan. As such, the court determined that the appellation was well-known for purposes of Article 2-1-1 of the UCPA. However, since the magazine's market was limited to those who read fashion magazines, the court held that it was not famous for purposes of Article 2-1-2 and, therefore, not diluted.

¶ 66         This was a case that is based on confusion in the broad sense. There was absolutely no competition between the plaintiff and the defendant. Any potential overlap between the two marks was only that rather wealthy, stylish people lived in the defendant's condominium. Such association was speculative at best and unstated in the case. Competition was presumed.

¶ 67         If anything, this was an Article 2-1-2 case. VOGUE was either both famous and diluted or neither. The creation and reliance on this middle standard by Japanese courts, when nothing to this effect is stated in the legislation whatsoever, muddled things interminably.

IV.    WHY SO FEW CASES?

¶ 68         There have been very few cases dealing with dilution in Japan. This is likely due to multiple factors. First, there are probably more cases than are reported, but even so that number is likely negligible. Second, the statute is vague and does not even use the word "dilution," thus Japan's civil law judges are not provided with much guidance. Third, the defensive trademark system, underutilized as it is, provides an alternative to dilution protection. Fourth, the theoretical understanding of trademark law renders the dilution claim superfluous.

¶ 69         One of the biggest problems researching Japanese trademark dilution is the fact that judges do not use a standardized term when they mean "dilution." It is convenient that the United States has adopted such a standardized term. In the United States, should one want to search all dilution cases, all one needs to do is search the word "dilution" in the Westlaw or Lexis databases. To be sure, that number will be large; however it will be inclusive and conclusive.

¶ 70         On the other hand, the Japanese courts do us no such favor. There is no standardized term in use by all Japanese judges for the concept of "dilution." Courts might refer to "dilution" in romanized letters, "dilution" in katakana (Japanese script indicating the word is of foreign origin), or the Japanese term for dilution (kishakuka); they might use the term "free rider" (furii raida) in katakana, the Japanese term that means "free rider" (tada nori), or many other possibilities. Most significantly, Japanese courts have also found dilution without naming it as such, usually under a theory of presumed confusion because they found confusion "in the broad sense."

¶ 71         Therefore, because Japanese courts do not use one standardized term for the cause of action, searching is extremely difficult.¹²⁶ On the other hand, if one searches all of the above usages, one will most likely reveal most of the cases. That is, there are not hundreds of cases using yet some other indication other than the words given above. Therefore, it is possible that there are other cases, but that number is not large.¹²⁷

¶ 72         Another problem researching Japanese case law in general is the fact that there is no official reporter where all or substantially all opinions are collected. Rather, there are multiple "magazine" type publications that publish one or sometimes two trademark cases every two weeks. Therefore, it is possible that some trademark dilution cases exist but simply go unreported in any of the seven or eight journals where one would expect to find them. Although that is the case, there is a grand tradition in Japan to discuss important cases through study groups and the like on the faculty level. These professors closely watch, hear about, and discuss the various pending cases. It is, therefore, possible that some unreported and important judicial decisions exist, but rather unlikely that a vast number of these decisions exist.

¶ 73         As indicated above, Japan has no statute that contains the word "dilution" or any of its derivatives in English or Japanese. Therefore, the fact that this law is to prevent something known as "dilution" is perceived intuitively from the fact that there is no other, more specific provision of the law that allows for this cause of action. Additionally, Article 2-1-1 is also used to prevent a host of other acts considered to be "unfair" under the statute. As a result, Japanese courts cannot keep these causes of action straight. Japanese courts routinely interchange, with no discernable pattern, the causes of action that (1) prevent a well-known appellation from confusing use, and (2) prevent a famous appellation from any use by a third party.

¶ 74         Furthermore, Japanese courts do not seem to recognize the substantial difference between "causes confusion" and an "apprehension of confusion." Japanese courts sometimes require confusion in Article 2-1-2 cases - even though the statute only says "use" (shiyo) - and sometimes even presume confusion for purposes of Article 2-1-1 if the appellation is famous enough.

¶ 75         As such, it seems that it would be exceedingly difficult to predict when a client is likely to prevail in Article 2-1-1 or 2-1-2 cases, especially if one considers its famous appellation to be diluted by non-competing use. The scattered reasoning demonstrated by Japanese courts on the issue of dilution does not seem to provide the predictability one would expect from an otherwise highly developed legal system such as Japan's.¹²⁸

¶ 76         Although other developed nations' dilution statutes do not necessarily use the term "dilution" either,¹²⁹ they are far more specific. The courts in these other countries, outside of the United States, seem to have well-defined and consistent concept of dilution.¹³⁰ Therefore, it seems that Japanese judges find something inherently flawed about the dilution rationale.

¶ 77         The American instinct is to dismiss Japanese judges as uninformed or ignorant. This conclusion is fostered by the ambiguity of the UCPA itself. However, I categorically deny the notion that Japanese judges are just missing the boat. The Japanese judiciary is a well-educated, well-trained group of career jurists. They are not political appointees like United States federal judges. They enter the judiciary upon graduating from the Legal Research and Training Institute.¹³¹ District and Appellate Court judges have had no other job in their life other than as a judge. They are supported by a very well-paid and motivated research and support staff.¹³² The idea that Japanese judges are just uninformed or ignorant of trademark dilution is simply not believable.

¶ 78         Under the defensive mark registration system,¹³³ "widely recognized" trademarks can be registered¹³⁴ to protect identified goods or services¹³⁵ other than those listed in the original registration. This system does not exist in the United States. Additional goods or services need not be similar to the original goods or services, and the registrant need not have any use for these additional goods or services nor any intent to use the mark on these additional goods or services.¹³⁶ The requirements are only that the mark be "widely recognized by consumers" (hiroku ninshiki sareteiru) and that there be an apprehension of confusion (kondo no osore) if the mark is used on these additional goods or services by a third party.¹³⁷ It is not difficult to see why this has been referred to as a "super registry."¹³⁸

¶ 79         Whether a trademark is "widely recognized" (or "famous")¹³⁹ for purposes of a defensive mark application is determined by referring to Guidelines for Examination of Trademarks published by the Japanese Patent Office. According to this manual, the determination of a registered trademark's degree of fame is made by referring to the following items:

(1) Duration and area of use;

(2) The range of goods on which the mark is used;

(3) Extent of advertisement used or other methods of publicizing the mark;

(4) The nature of the user's business and scale of use of the mark; and

(5) The amount of recognition the mark already enjoys within the Patent Office.¹⁴⁰

¶ 80         Not surprisingly, most likely because the strict nature of the requirements to prove a mark is widely recognized,¹⁴¹ very few marks have been registered as defensive marks. Furthermore, the JPO and courts seem overly restrictive in granting defensive mark registrations.

¶ 81         The JPO and Japanese courts require that the registration of the mark a defensive mark must be "identical" to the primary registered trademark.¹⁴² Therefore, the Tokyo High Court affirmed an examiner's rejection of "Mercedes-Benz" as a defensive mark application for string, rope, or netting because it was not identical to its principal mark, Mercedes-Benz as registered for use on automobiles. That is, the marks do not share the same font and the primary registered trademark is in all upper case letters.¹⁴³ The court held that "if the plaintiff [Mercedes-Benz] wanted to prevent the use of its mark in this fashion, they ought to elect to register the mark as an associated trademark. Attempting to interpret this as a defensive mark registration is absurd (sujichigai)."¹⁴⁴

¶ 82         Given the specific language of the Trademark Law,¹⁴⁵ it is difficult to determine who is being absurd. It appears on the surface that this is a perfect example of a famous mark that deserves defensive mark registration status. However, the court also read strictly the requirement that there be some apprehension of confusion as to source in the class of goods under which defensive mark registration is sought. Because there was none here, the court found the associated mark doctrine to be the best source of protection for Mercedes-Benz.

¶ 83         The JPO also denied 3M Corporation's defensive mark registration application for its SCOTCH trademark, which is registered in Japan for use on tapes and other adhesives, to prevent its use on kitchen utensils and other kitchen products. The Examiner claimed that there was no apprehension of confusion because 3M used SCOTCH in upper case letters on some goods, while on others the mark appeared in lower case letters, and therefore the marks were not identical.¹⁴⁶

¶ 84         However, the Tokyo High Court held that SCOTCH was, indeed, well-known and reversed the JPO's decision. The court determined that the appropriate standard should include an analysis of the trademark as it is actually used. If consumers would presume that the principal mark and the mark that is subject of a defensive mark application identify the same source, then the mark should be considered identical and appropriate for defensive mark registration.¹⁴⁷

¶ 85         The defensive mark system is already in place for a "famous" registry, making Article 2-1-2 somewhat superfluous.¹⁴⁸ This super registry afforded by the defensive mark system is the precise mechanism third party users of appellations might rely upon to determine when someone else's appellation is famous. Of course, this is only analogous to a super registry for marks subject to dilution. However, because Japan has such a registry in place for famous marks for purposes of the defensive mark system, the leap to a registry for appellations subject to dilution protection is not that great.

¶ 86         The defensive mark system is somewhat redundant to dilution protection because it already provides an avenue by which registrants of famous marks might find additional protection. Of course, it applies only when there is an apprehension of confusion and therefore is not precisely responsive to a traditional dilutive setting; however. Because Japanese courts are inconsistent in the requirement of competition, or not, one would expect a Japanese attorney zealously representing the interests of his or her clients to be very interested in the defensive mark system when the client is the owner of a famous mark.

¶ 87         However, all that being the case, the defensive mark system is a perfect example of how countries could require registration on a super registry before dilution protection is granted.¹⁴⁹ Because the Japanese have the defensive mark system already in place, it is not as much of an extension to require such a registration of anyone claiming protection under Article 2-1-2 of the UCPA.

¶ 88         This system is drastically underutilized.¹⁵⁰ It is underutilized because it provides protection that few actually desire. The system requires the would-be defendant's mark to be identical to the would-be plaintiff's mark. Defendants do not often adopt identical copies of famous marks. Of course, if they do and they do it with the intent to deceive, the defensive mark system is there to help prevent such conduct. However, this probably does not happen very frequently.¹⁵¹ This defensive mark system would help Fuji Film prevent the use and registration of Fuji for bicycles, if they registered for a defensive mark. However, as there were only sixty defensive mark registrations in all of Japan in 2004, it appears that this kind of conduct is rare.

¶ 89         The system also provides enough time for the would-be plaintiff to file for the trademark application for a defensive mark, rather than just initiating a lawsuit. Nearly all of the Article 2-1-2 cases reviewed for this article were ripe for a defensive mark registration. Rather than registering, the plaintiffs elected to file for an injunction and damages under Article 2-1-2 of the UCPA.

¶ 90         Therefore, there are redundancies built into the Japanese system. These redundancies themselves may be providing a chilling effect on the frequency with which lawsuits are filed under Article 2-1 of the UCPA. The existence of the defensive mark system provides an alternative route for protecting famous marks.

¶ 91         Dilution protection, as defined herein, is not a surprising expansion of the law, as it was in the United States.¹⁵² Courts in Japan frequently engage in protecting trademarks even without a showing of competition between the parties. They do this by calling it "confusion in the broad sense of the term."¹⁵³

¶ 92         Japanese courts generally approach confusion as "narrow" (kyogi) confusion and "wide" (kogi) confusion.¹⁵⁴ That is, Japanese courts consider narrow confusion to exist when two parties are in direct competition. Confusion in the wider sense of the term (kogi) is used to prevent conduct that may also be prevented under a dilution rationale.¹⁵⁵

¶ 93         Therefore, when "Mitsubishi Building Company" uses the appellation "Mitsubishi," it is not unreasonable to believe that all companies named "Mitsubishi" are related, even though at that time, the larger and famous Mitsubishi did not have a "building" company in its quiver.¹⁵⁶ Also, when the defendant used the appellation "Womanpower" for temporary secretarial services, they were deemed to be in competition with "Manpower," even though there was no direct evidence that this was, in fact, the case.¹⁵⁷ The court stated that there was an apprehension that people would believe Womanpower was a subsidiary of Manpower and, even though there was no direct evidence of their competition in the marketplace, the court deemed them in competition. In both of these representative cases, the court simply deemed the parties to be in "competition" with each other even though there was no direct evidence to that alleged and important fact.¹⁵⁸

¶ 94         The holding in Japanese Womanpower K.K. v. Manpower Japan K.K., seems to allow for a plaintiff to "bridge the gap."¹⁵⁹ That is, although the defendant may not be in competition with the plaintiff at the time of the case, there is reasonable expectation that they may actually be in competition in the future. Instead of using this potentially more appropriate analysis, however, the courts in these cases merely deem competition to be present. Worse yet, if the holder of a famous mark can be presumed to be in competition with anyone who might come along, then there is really no need for a dilution statute at all and it is not surprising in the least that Article 2-1-2 is underutilized.

V.    CONCLUSION

¶ 95         Given the overall circumstances of trademark protection in Japan, it is not surprising that there are very few traditional "dilution" cases. The statute is too vague to be of any real guide - it does not even mention the word "dilution." The statute allows an injunction and damages in the event that a famous mark is used and does not require the plaintiff to establish confusion. While this cause of action sounds like dilution, it would help Japanese courts immensely if that word was used, so that the Japanese legislature's intent could be more easily determined.

¶ 96         Judges in Japan do not stick to an establish vocabulary when addressing dilution-like cases. They use terms like "free-riding" or other terms to describe what would be a dilution case if it were raised in the United States. This lack of a standardized vocabulary leads to further confusion regarding the cause of action and makes it appear that even fewer cases are raised.

¶ 97         The nature of trademark infringement also makes dilution redundant. The defensive mark system in Japan provides an alternative to trademark owners who desire to protect their famous marks from use by non-competitors on non-competing goods. However, even this system, in place in Japan since 1959,¹⁶⁰ is underutilized.

¶ 98         The other redundancy regards the trademark right itself. In Japan, the trademark right can be conceived of "broadly." Under the so-called broad conception, confusion and competition are not necessary. When the mark is famous, Japanese courts sometimes will presume confusion.

¶ 99         The state of confusion over this cause of action and the resulting diminished number of cases may be interconnected. Through this confusion, Japanese judges either intentionally or intuitively apply a brake to the application of dilution law in Japan.

¶ 100         On the other hand, perhaps the apparent lack of dilution in Japan is not an aberration. Perhaps it is not that there is no dilution in Japan; perhaps the issue is whether there is really much actual dilution anywhere. Perhaps it is the United States that is the aberration. Maybe, it is just that, in the United States, we see dilution everywhere and over utilize the Federal Trademark Dilution Act.

¶ 101         One way or the other, Japanese trademark dilution law is in a state of confusion and disarray. Such confusion does not help in developing a coherent and predictable body of law. The rest of Japanese trademark law is highly predictable and certainly coherent. That trademark dilution law is not stands out as an oddity.

ENDNOTES