This school semester (sans-finals) is finally over and this time round, I took my first honors-level class on Japanese popular culture. After much contemplation, for my research paper, I decided to write on the inherent importance of moe-elements when it comes to character merchandise. It was a long and arduous process to see this paper to completion, but it was well worth it and exceedingly fun and educational.
Character Merchandise and the Autonomous Moe Character
It is an undeniable fact that character merchandise plays an integral role in the anime industry. It takes nothing more than a simple stroll through Akihabara, often referred to as the otaku mecca of Japan, to get inundated by countless stores like Animate and Gamers that dedicate most of their floor space to nothing but character merchandise. Citing the Digital Contents White Paper published by the Digital Content Association of Japan and the Ministry of Economy, Trade and Industry (DCAJ and METI 2005), Ian Condry notes that “the market for licensed merchandise based on fictional characters is ten times that of anime itself” (Condry 2013, 72), showing the importance and prevalence of character merchandise. The scope of this paper will be limited to fictional human female characters and merchandise that are specifically targeted towards the male anime otaku market. Unlike other forms of character merchandise, like those of robots, there is a peculiar change that occurs to the female characters’ image upon becoming merchandise. Throughout the course of this paper, I will show that moe-elements in the characters’ design are integral to character merchandise, regardless of the strength of the anime’s narrative, backing up my arguments with several contemporary primary sources that consist of anime, characters and character goods.
1. Autonomous Characters, Moe and Database Consumption
In his paper, Marc Steinberg uses the following definition of character merchandise or character goods provided by the World Intellectual Property Organization (Steinberg 2009, 115):
Character merchandising can be defined as the adaptation or secondary exploitation, by the creator of a fictional character or by a real person or by one or several authorized third parties, of the essential personality features (such as the name, image or appearance) of a character in relation to various goods and/or services with a view to creating in prospective customers a desire to acquire those goods and/or to use those services because of the customers’ affinity with that character. (World Intellectual Property Organization 1994, 4)
Essentially, character goods are physical commodities that depict the characters’ likeness on them and include figurines, phone straps, notebooks and key chains. For the sake of this paper, the above definition will be used. In addition, Marc Steinberg notes that the character itself is “a named, visual figure that possesses recognizable visual attributes” (Steinberg 2009, 128). As they are nothing more than a collection of visual traits, the character can easily be transposed across different media. For instance, a character that originates from an anime could also appear in video games, on lunch boxes and figurines. This is what is known as the autonomous nature of characters.
One of the most notable examples of character merchandise is the Testuwan Atomu stickers which were sold in Japan in 1961. They were a resounding success (Steinberg 2009, 119). From the case study of the Tetsuwan Atomu stickers, Marc Steinberg proposes several reasons as to why they were so successful. Firstly, it was due in part to the popularity of the original anime series. Secondly, and most importantly, there was a distinct lack of a disjuncture between the image of Atomu on television and on the sticker. According to Marc Steinberg, it is necessary for a characters’ appearance to remain consistent between media in order for them to be successful. As the stickers were literally “traced from the cels of the anime series” (Steinberg 2009, 122), the disjuncture was significantly minimized and this contributed to their success. He cites other examples of media that failed to do well due to having a large gap between the representations of the character, such as the live-action Atomu show. Finally, character merchandise serves to remind fans of “their favorite character and his narrative world” (Steinberg 2009, 124) by allowing them to bring the character into reality in the form of everyday items that they can be with at all times.
While that might be the case for Tetsuwan Atomu and other similar franchises targeted towards a younger demographic, like Pokémon, the same cannot be said for character merchandise that feature human female characters, targeted towards a more niche audience, namely the male anime otaku. There is an added factor at work that has not yet been considered. Contrary to Marc Steinberg’s claims, there are instances where the image of the female human character is altered upon becoming character merchandise. In order to understand the reasons behind this phenomenon, we have to look at a new form of consumption that the post-1990 otaku engage in, database consumption – a term coined by Azuma Hiroki, and their increasing fixation on moe-elements (Azuma 2009, 54).
Figure 1: Hirasawa Yui from the anime K-On! as an example of a moe character design.
The term moe or moe-elements, as defined by Azuma Hiroki, are desirable elements that are mostly visual such as cat ears, bells, maid uniforms, tails and loose socks (Azuma 2009, 42). The definition of moe and moe-elements is still up for debate and various scholars have their own interpretations of moe and they range from being sexual in nature to promoting sexist ideologies. However, in this paper, moe will be defined along the lines of Azuma Hiroki’s usage of the term. Moe-elements are elements that are commonly found desirable by the otaku such as specific outfits and accessories and a deformed style of character design with non-human, exaggerated proportions that seek to portray a cute, child-like image that does not contain a sexual element (see figure 1).
The rise of database consumption and the increased importance of moe-elements were preceded by a decline in narrative consumption as Japan moved into the postmodern era. In the 1970s, just before the beginning of the postmodern age, the Japanese people were still interested and searching for a grand narrative to latch on to (Azuma 2009, 54). The grand narrative, which governed the modern society, is defined as “a system that is consolidated for the purpose of organizing members of a society into a unified whole” (Azuma 2009, 28). In a postmodern society however, the “grand narratives break down and the cohesion of the societal entirety rapidly weakens (Azuma 2009, 28). As Japan entered the postmodern era in the 1970s, companies then began to take advantage of the lack of a grand narrative, and the fact that people were still searching and interested in one, to sell products. However, it was impossible for them to sell the grand narrative itself to consumers, one of the reasons being that it was not economically feasible. So instead, companies sold consumers pieces of the grand narrative. Ōtsuka gives us an example of anime like Mobile Suit Gundam (Ōtsuka 2010, 107). By piecing together small narratives available in each episode, consumers were able to get a hold of the grand narrative (Ōtsuka 2010, 107) and it is the grand narrative that was consumed. This form of consumption is called narrative consumption.
The issue here is that, as time passes, the forms of consumption change. Those who grew up in the postmodern era see the world as nothing more than a database. The database is literally a database of information that has replaced the grand narrative. Characters are now “derived from the database of moe-elements” and are just a remix of desirable elements found within otaku culture (Azuma 2009, 42). Compared to the 1980s otaku, those of the 1990s cared more for the characters and less for the narratives in anime (Azuma 2009, 47). In other words, works are now no longer seen as or desired for a window to a grand narrative, but for the moe-elements, or desirable elements. This is considered to be database consumption. According to Azuma Hiroki, the narrative has now been reduced to nothing more than a “surplus”, equivalent to the settings and illustrations (Azuma 2009, 41).
2. The Importance of Moe-elements in Character Merchandising
Returning to character merchandise, this section will deal with my main argument on how moe-elements in character designs are integral to character merchandise. Firstly, to lend credence to Azuma Hiroki’s theories of database consumption is the character of Hatsune Miku. Hatsune Miku was originally conceived by Crypton as a character to represent their Vocaloid series of voice synthesizer programs and has become exceedingly popular both inside Japan and out. Using pre-recorded voice samples, these voice synthesizer programs allow users to create their own songs from scratch just by inputting their own music and lyrics. Hatsune Miku has no narrative to speak of. Being nothing more than a character, her popularity is indicative of the importance and strength of moe-elements and the reduced significance of narratives. While one might attribute Hatsune Miku’s popularity to the voice synthesizer program itself, Ian Condry thinks otherwise. He believes that the rabid fan activity and merchandise in her image cannot be explained by the voice synthesizer program. Instead, “the phenomenon around Miku shows that the character, more than the music software, is the platform on which people are building” (Condry 2013, 63). This is in reference to fan art and fan created merchandise based on the character as opposed to the synthesized music.
As mentioned previously, according to Marc Steinberg, it is in the nature of characters to be autonomous – being able to travel easily from one media to another. Adding on to his point, it is the lack of a narrative that encourages and facilitates this nature. Compared to characters with narratives, characters without narratives would not be bound specifically to one world or context. Otaku consumers would not need to know the original narrative to be attracted to the characters and can consume said merchandise for the moe-elements alone.
However, even though Azuma Hiroki’s theories do account for the popularity of many contemporary anime franchises, there are exceptions to this rule. Upon closer inspection, one can realize that the theory of database consumption cannot be applied to all genres of anime and that there are anime that are still being consumed mainly for their narratives. Slice-of-life anime like K-On! and Lucky Star that portray the daily lives of high school-aged female characters have little to no narrative and can attribute their success to database consumption. In contrast, equally popular anime like Puella Magi Madoka Magica, a series about girls dealing with the consequences of taking on the life-threatening role of a magical girl in exchange for a wish, and Shingeki no Kyojin, a series that depicts the remainder of humanity struggling to survive the onslaught of gargantuan, flesh-eating humanoid beings called Titans, are extremely plot-heavy.
On the surface, Puella Magi Madoka Magica and Shingeki no Kyojin seem similar in that both series seek to hook the viewer through their narratives that are peppered with complex plot twists, mature themes and cliffhangers. Yet, there is a clear, noticeable difference between the two on the merchandising front – when their female heroines are concerned, Puella Magi Madoka Magica has far more character merchandise than Shingeki no Kyojin. A search for the respective series’ character merchandise on AmiAmi, one of the biggest online retailers of anime character merchandise, yields a result of 345 items for Kaname Madoka (female protagonist of Puella Magi Madoka Magica) and only 54 items for Ackerman Mikasa (female protagonist of Shingeki no Kyojin). This list might not be all-encompassing or exhaustive, but it is indicative of the fact that Kaname Madoka has far more merchandise than Ackerman Mikasa. This difference is the result of the presence of moe-elements in the character designs, or the lack thereof in Shingeki no Kyojin’s case.
Figure 2: Kaname Madoka (left) and Ackerman Mikasa (right).
As can be seen from figure 2, Madoka embodies many of the archetypal moe-elements that are favored by otaku. These elements include the more child-like proportions with a larger head and eyes, small nose, frilled cute clothes and exposed thighs. In contrast, Mikasa is more realistically proportioned, with a smaller head and eyes, more pronounced nose and wears plainer clothing – none of which are elements that the otaku would consider to be moe.
It is indeed possible to associate the amount of merchandise with the air dates for the two series. One can easily argue that Puella Magi Madoka Magica (2011) aired much earlier than the anime adaptation of Shingeki no Kyojin (2013, though 2009, if taking the manga into consideration), allowing more time for the Madoka merchandise to get churned out. However, the differences do not end there.
Figure 3: A Mikasa key chain, an example of Mikasa character merchandise.
Unlike the Madoka merchandise, an overwhelming majority of the Mikasa merchandise on sale do not portray her as she appears on screen, but instead renders her in a more moe, super deformed style with exaggerated proportions – reinstating the larger head and eyes and removing her nose altogether (see figure 3).
Figure 4: Kusanagi Suito, from The Sky Crawlers, as she appears in the anime.
Figure 5: Kusanagi Suito as she appears in figure form.
This practice is not limited to Shingeki no Kyojin as the relatively realistic female protagonist designs of anime like The Sky Crawlers and Summer Wars receive a moe-makeover upon entering the character merchandise market (see figures 4 and 5). Kusanagi Suito’s proportions are once again made more disproportionate, her facial features are made rounder and cuter and her eyes are made much bigger. From the examples and analysis given above, we see that the characters’ on-screen and on-merchandise appearance vary greatly. This proves that Marc Steinberg’s point on the importance of minimizing the disjuncture between the images of the character across different media (Steinberg 2009, 122) is not entirely accurate.
However, there is still a minute number of Mikasa merchandise that does go about replicating her image as she appears on screen. In cases such as these, Marc Steinberg’s other theories on character merchandise can account for their existence. For example, this merchandise serve to allow fans to bring Mikasa out of the fictional world in which she inhabits and into reality, allowing them to be surrounded and reminded about their favorite character and narrative wherever they go (Steinberg 2009, 124).
Even if an anime series has a strong narrative and is consumed for its narrative, the presence of moe-elements in the character designs is what enables the characters to distance themselves from the narratives and become autonomous – facilitating the creation of character merchandise and proving that moe-elements are integral to character merchandise.
This is not to say that the narratives do not account for anything. What the strength of the narrative does, is to boost the sales of the actual series itself, be it in manga or disc media forms. As a testament to its popularity, the first volume of the Puella Magi Madoka Magica Blu-ray discs sold 53,000 copies in the first week of its release and holds the record for being the highest selling television anime release in first-week Blu-ray sales (Anime News Network 2011). Similarly, all 11 volumes of the Shingeki no Kyojin manga have constantly appeared on the Oricon charts, selling over 20 million copies as of September 2013 (Anime News Network 2013).
To further my case on the importance of moe with regard to character merchandise, we turn our attention to Studio Ghibli. Despite their fame and popularity, the plain and unassuming female protagonists of the Studio Ghibli films receive painfully little merchandise, especially when compared to the merchandise of Ghibli’s animal characters like Totoro from Tonari no Totoro. Miyazaki himself has been known to hold a strong distaste of otaku consumers and moe character designs. In a talk with Murakami Ryu regarding heroines, Miyazaki said that:
They immediately become the subjects of rorikon gokko (play toy for Lolita Complex guys). In a sense, if we want to depict someone who is affirmative to us, we have no choice but to make them as lovely as possible. But now, there are too many people who shamelessly depict (such heroines) as if they just want (such girls) as pets, and things are escalating more and more. (The Hayao MIYAZAKI Web)
Here, Miyazaki is making the assumption that the otaku who are attracted to moe characters seek to objectify them by treating them as pets or objects to buy and he is repulsed by it. The interview occurred shortly after the release of Miyazaki’s earlier works like Tenkū no Shiro Rapyuta and Rupan Sansei: Kariosutoro no Shiro that were “not so different from what is typically considered otaku fare”. Now, Studio Ghibli aims to “distinguish its manga films from television anime and to avoid association with “subculture” audiences (otaku) who become obsessed with them” (Lamarre 2009, 98) by deliberately making the designs of their female characters not moe. While it is true that many of the female protagonists from the films of Studio Ghibli are indeed children, their proportions are far closer to those of actual humans to be considered moe, most notably in terms of their facial features and eyes. The clothes that they wear are more often than not plain and non-descript – nothing that would be deemed moe-elements. Therefore, removing moe-elements from the character designs severely limits the amount of character merchandise that can be produced, highlighting the importance of moe-elements once again.
Figure 6: Several figures of Ayanami Rei from Evangelion.
Another integral role that moe-elements play in character merchandising is the inherent ability of moe-elements to facilitate the increased creation of merchandise. For instance, a myriad of figures can be easily created by simply remixing or changing the moe-elements (see figure 6). As fans consume character merchandise for their moe-elements, or for a specific combination of moe-elements, a simple remixing of moe-elements would be sufficient to bait fans into purchasing them, even if the figures are non-canon. This illustrates the decreased importance of the narrative in relation to character merchandise and the importance of moe-elements.
Finally, moe-elements in character merchandise can serve as an ingenious marketing tool to attract even more otaku viewers to watch an anime series. Even if an anime series does not have the moe character designs that are consistent with what otaku want, the moe-elements added into character merchandise would be more than enough to pique their interest and curiosity. Seeing the moe merchandise would influence them to watch the anime from which the merchandise originated from, aiding the marketing and promotion of the anime. This is yet another aspect in which moe-elements play an important role in character merchandise.
In conclusion, through the course of this paper, I have shown that moe-elements are the most important factor when dealing with character merchandise of human female characters targeted towards a male anime otaku market, regardless of the strength of the original narrative. While there are anime series that are consumed for their narratives, the same cannot be said for their character merchandise as illustrated using the examples of Puella Magi Madoka Magica and Shingeki no Kyojin. By creating characters that are not moe, producers can limit the amount of merchandise that can be created. They can also use moe-elements as effective merchandising and marketing tools, creating multiple variations of the same character and promoting the original anime series at the same time.
Through this research into character merchandise targeted towards the otaku, one can get a better understanding of the consumption trends and habits that drive the otaku and the societal factors that influence their tastes and preferences. In addition, it allows us to further our knowledge on the production side of the industry and the relationship that they share with the consumers – how they are able to capitalize on the otaku demographic and at the same time become shaped by them.
Anime News Network 2011. Madoka Magica 1 Sells 53,000 as #1 TV Anime BD in 1st Week. http://www.animenewsnetwork.com/news/2011-05-03/madoka-magica-1-sells-53000-as-no.1-tv-anime-bd-in-1st-week (accessed October 2013)
Anime News Network 2013. Attack on Titan Manga Sells 20 Million Copies. http://www.animenewsnetwork.com/news/2013-09-14/attack-on-titan-manga-sells-20-million-copies (accessed October 2013)
Azuma, Hiroki. 2009. Otaku: Japan’s database animals. Trans. Jonathan E. Abel, and Shinon Kono. University of Minnesota Press.
Condry, Ian. 2013. The Soul of Anime Collaborative Creativity and Japan’s Media Success Story. Duke University Press
Digital Content Association of Japan (DCAJ) and Ministry of Economy, Trade, and Industry (METI). 2005. Dejitaru kontentsu hakushō 2005. Tokyo: METI
Lamarre, Thomas. 2009. The Anime Machine A Media Theory of Animation. University of Minnesota Press.
Ōtsuka, Eiji. 2010. World and Variation: The Reproduction and Consumption of Narrative. Trans. Marc Steinberg. In Mechademia 5: Fanthropologies, ed. Frenchy Lunning, 99-116. University of Minnesota Press.
Steinberg, Marc. 2009. Anytime, Anywhere: Tetsuwan Atomu Stickers and the Emergence of Character Merchandising. Theory Culture Society 26: 113-138.
The Hayao MIYAZAKI Web. Why heroines in Miyazaki works A collection of short excerpts. http://www.nausicaa.net/miyazaki/interviews/heroines.html (accessed October 2013)
World Intellectual Property Organization. 1994. Character Merchandising. Geneva: World Intellectual Property Organization.
(Yes, I know the formatting for the references is off, but I can’t figure out how to do it on WordPress.)