Semiotic Review 3: Open Issue - article published November 2013


Shunsuke Nozawa

Abstract:The Japanese society is filled with characters. Starting from an understanding that these characters are creatures of in-between-ness, creatures of semiotic mediation par excellence, my goal in this paper is to tease out patterns of contemporary media ecology that index a modality of semiosis I will call “characterization”: quite simply and literally, transformation of some thing into a character. Rather than, or in addition to, asking the question of what characters mean to people or even what people performatively accomplish through anthropomorphic characters, we might as well consider characterized people and inquire into the condition of this modification, that is, characterization as a semiotic modality. The structure of character-encounter in contemporary media ecology affords a sui generis semiotic ideology of becoming-character irreducible to the anthropocentric semiotic ideology of performance and embodiment. People are not only experiencing characters on street signs, on smartphones, or in anime. They are experiencing themselves as creatures of in-between-ness.

Keywords: Characterization; effacement; semiotic mediation; anthropology; Japan

The World of Characters

The Japanese society is filled with characters. This seems to be a fairly common observation among many tourists, scholars, and the people who live there more generally. A 2002 compilation of characters designed for corporations, groups, goods, and events has the following statement in the introduction to this 206-page hardcover quarto. “One after another,” says the editor, “popular characters have been created and embraced by many enthusiastic fans. Not a day goes by in which you do not see such characters in your daily life, whether […] while riding a train, walking in the city, entering a shop, or buying merchandise” (P. I. E. Books 2002: front matter). Everydayness, or the distributive quality (‘every’) of character-encounter, is also a theme in a more recent photo book by Alt and Yoda that documents “a menagerie of adorable mascots and characters that inhabit the islands of Japan, quietly guiding citizens through all sorts of daily activities and situations” (Alt and Yoda 2007:9).

Here are some samples of what I encountered on one summer day in 2013, rather randomly walking around in the neighborhoods of Shinjuku and Ikebukuro in central Tokyo.

Figure 1. Construction site in Shinjuku. This construction worker character apologizes to passers by for the inconvenience caused by construction work.Figure 1. Construction site in Shinjuku. This construction worker character apologizes to passers by for the inconvenience caused by construction work.

Figure 2. The same construction site. 'No trespassing: staff only'Figure 2. The same construction site. 'No trespassing: staff only'

Figure 3. A tanuki character at the Shinjuku Gyoenmae subway station, calling passengers' attention to various travel-related (“free”) brochures.Figure 3. A tanuki character at the Shinjuku Gyoenmae subway station, calling passengers' attention to various travel-related (“free”) brochures.

Figure 4. Esuzō-kun in a poster displayed at a Shinjuku konbini (convenience store). This elephant-shaped mascot for the Japan Franchise Association announces that this konbini participates in its “Safety Station” campaign and promises that it will act as a sheltering location at times of emergency.Figure 4. Esuzō-kun in a poster displayed at a Shinjuku konbini (convenience store). This elephant-shaped mascot for the Japan Franchise Association announces that this konbini participates in its “Safety Station” campaign and promises that it will act as a sheltering location at times of emergency.

Figure 5. Non-smoking sign in Shinjuku. In Shinjuku and other districts public smoking (for example smoking while walking) is prohibited. Figure 5. Non-smoking sign in Shinjuku. In Shinjuku and other districts public smoking (for example smoking while walking) is prohibited.

Figure 6. “Wi-Fi available.” Otousan, a dog-character for Softbank, one of the major telecommunications companies. This particular notice was found at a Tully’s Coffee in Shinjuku. Figure 6. “Wi-Fi available.” Otousan, a dog-character for Softbank, one of the major telecommunications companies. This particular notice was found at a Tully’s Coffee in Shinjuku.

Figure 7. In a POP (Point Of Purchase) ad, various working characters point customers to weekly special sale events at a local grocery store in Shinjuku.Figure 7. In a POP (Point Of Purchase) ad, various working characters point customers to weekly special sale events at a local grocery store in Shinjuku.

Figure 8. Jikusā a mascot for a neighborhood in Shinjuku, welcoming people. This kind of characterization in particular involves what is known as yuru-kyara, ‘wobbly characters,’ for which see Occhi 2012. Increasingly the yuru-kyara aesthetics is now invoked in ‘official’ institutional settings, apparently so appealing that diverse formal institutions, from Japan Communist Party to Israel, have appropriated it to their own endsFigure 8. Jikusā a mascot for a neighborhood in Shinjuku, welcoming people. This kind of characterization in particular involves what is known as yuru-kyara, ‘wobbly characters,’ for which see Occhi 2012. Increasingly the yuru-kyara aesthetics is now invoked in ‘official’ institutional seMings, apparently so appealing that diverse formal institutions, from Japan Communist Party to Israel, have appropriated it to their own ends

Figure 9. Customer testimonials in a huge wall ad in Shinjuku for an eyeglass manufacturer.	
Figure 9. Customer testimonials in a huge wall ad in Shinjuku for an eyeglass manufacturer.

Figure 10. Poster with a pig-character announcing the summer sale at Shinjuku Mylord, a commercial complex near the Shinjuku station.	
Figure 10. Poster with a pig-character announcing the summer sale at Shinjuku Mylord, a commercial complex near the Shinjuku station.

Figure 11. Echika-chan, a rabbit-like mascot for a commercial complex at the Ikebukuro station.Figure 11. Echika-chan, a rabbit-like mascot for a commercial complex at the Ikebukuro station.

Figure 12. Lightbulb sleeping in a poster calling for saving electricity. This one and the following two posters were all found almost side by side on the same walkway in the Ikebukuro Station on this particular day. Figure 12. Lightbulb sleeping in a poster calling for saving electricity. This one and the following two posters were all found almost side by side on the same walkway in the Ikebukuro Station on this particular day.

Figure 13. Poster for Suica/Pasmo. These are rechargeable smart cards, used as a train/bus fare card as well as at other points of purchase (e.g. konbini, street vending machines) equipped with proper card-reading devices.  Figure 13. Poster for Suica/Pasmo. These are rechargeable smart cards, used as a train/bus fare card as well as at other points of purchase (e.g. konbini, street vending machines) equipped with proper card-reading devices.

Figure 14. Summer sale poster for Parco, a department storeFigure 14. Summer sale poster for Parco, a department store

Some of these characters look cute; some are just weird. Some are smiling while others are concerned, sleeping, alert, and apologetic. Some are named and gendered, others anonymous and gender-ambiguous. While they often involve anthropomorphism, that is, things turned into human form (see Occhi 2010), some are zoomorphic (see Miller 2010). Often we do not even know whether it matters that they belong to a specific species of animate beings. While corporate characters inhabit deeply institutional and collective decision-making processes, other characters, for example the hand-drawing of what appears to be a tanuki, seem to be more spontaneous creations created on the spot, perhaps by a single individual employee. This tanuki is also highly local and not mass-reproduced, in sharp contrast to the Suica mascot, whose distribution is decidedly wider. Some are connected to events (e.g. summer sale), others to more perduring facts and artifacts (e.g. the non-smoking ordinance); yet others are linked to mass media representation (e.g. the Softbank mascot). Very often, interestingly, the characters mediate those moments that directly or potentially involve service labor, which is itself a mediator of value crucial to but consistently undervalued in capitalism: appropriately enough, Alt and Yoda call them “working characters” (2007).

In many cases these characters warn, apologize, explain, testify, and announce; they welcome you, assure you, instruct you, and point you to something. We might call them specialized speech-actants – specialized for producing performative effects whether through conventional verbal signs or through non-verbal signs such as hand gestures. They constitute an interface of objects and spaces that relays signs between other semiotic actants – between pedestrians and a construction site, between smokers and the city, between commuters and the subway station.1

Most basic to an adequate understanding of characters in general, I argue, is that they are creatures of in-between-ness, creatures of semiotic mediation par excellence (see Parmentier 1994). I suggest that we grant them a specific class of semiotic-communicative participants, themselves non-human entities and usually capable of shape-shifting. Like supernatural entities such as ghosts, fairies and Japanese yōkai monsters, they pop up in expected and unexpected moments of everyday life. In short, they are fantastic beings – appearing here, reappearing there. They are fantastic, but they are also real. They are real in the sense that their ubiquity in the everyday life of places like Tokyo is just a plain fact. How can we study this fantastic but real world of characters that is grafted into our kind of world? What is the nature and consequence of this grafting, this layering of worlds? How do people live with characters? It is this relationship between the two worlds, the world of characters and our so-called real world, that is the focus of my paper.

Let me immediately clarify the nature of my argument, however. The question I just posed, “How do people live with characters?”, is actually a foil to what I think is a more interesting question: How do people live as characters? The “working characters” like those noted above exhibit enormous variability in form and function, medium and location, and circulatory scale, and the specific differences are worth further consideration and more careful analysis.2 This paper, however, is not directly about these characters. Rather, my larger goal is to tease out patterns of contemporary media ecology that index a modality of semiosis I will call “characterization”: quite simply and literally, transformation of some thing into a character. (“Becoming-character” might also work.) Rather than, or in addition to, asking the question of what characters mean to people or even what people performatively accomplish through anthropomorphic characters, we might as well consider characterized people and inquire into the condition of this modification, that is, characterization as a semiotic modality. (Here I am particularly inspired by Galbraith’s [2011c] discussion of “maid cafés” as a space that plays out “the contradiction of characterizing self” in concrete form.)

In a nutshell, then, here’s my main argument: the structure of character-encounter in contemporary media ecology affords a sui generis semiotic ideology of becoming-character irreducible to the anthropocentric semiotic ideology of performance and embodiment. People are not only experiencing characters on street signs, on smartphones, or in anime. They are experiencing themselves as creatures of in-between-ness. The latter half of the paper will introduce several concrete fields of practice that exemplify this ambiguous, carefully effaced space of in-between-ness.


In this paper I locate the semiotic ideology of characters in the culture of Japanese subcultural geeks, or otaku, and its historical specificity. Why should we be interested in this subculture? It is because it is a real sociohistorical site where characters are most explicitly and saliently talked about. Otaku people talk about characters all the time: the very reflexive quality of their discourse is itself indexical of their social positionality. They are in a sense ‘native’ theorists, philosophers, and practitioners of characterization. As an ethnographer interested in conditions of virtuality and reality, I find it most appropriate to give prime analytic focus to the otaku conceptualization of characters because their perspective is most crystallized, has already a rich history of reflexive discourse, and, I believe, has its own logic immanent in the culture.

Figure 15. “Dimensionality” (jigen): a subcultural fashion of speaking.Figure 15. “Dimensionality” (jigen): a subcultural fashion of speaking.

As it turns out, the idea of ‘in-between-ness’ suggested above is not only a good conceptual tool to approach otaku culture from outside but in fact it implies a perspective internal to the culture. Here I draw your attention to the idea of “dimension” or jigen. Closely associated with otaku culture, this idiom is something one hears and reads about all the time in otaku cultural texts: everywhere from anime reviews to anime texts themselves, from online forums for gamers to news coverage of dōjin/ fan events. We can rephrase our question about the ‘two worlds’ above in terms of what the otaku call ni-jigen, or the 2-Dimensional, fantastic and virtual world of characters, and san-jigen, the 3-Dimensional, ‘actual’ world of our so-called real life (see Figure 15).

These terms are metaphorical: we are not talking about the ‘literal’ topological status of some objects but a subcultural “fashion of speaking” (Whorf 1956[1939]) that is habitually invoked as part of an interpretative framework. For example, the 3-dimensional materiality of a manga character figurine can still motivate an interpretation of the encounter with it as a ni-jigen phenomenon, while a voice actor performing this character in an anime text with her ‘3D’ voice might bring the distinction to heightened consciousness. This ambiguous nature of the boundary will concern us later – especially in relation to voicing – but at this point I simply ask that you keep this concept in mind as we will be using it throughout this paper.

Sui Generis Realness of Characters

Let us start, then, by following one corollary of this subcultural concept of “dimension”: 2D characters are not a “realist” representation of 3D beings such as humans, though they can be made to function as such a representation. Our usual kind of realist aesthetics posits an asymmetrical relation between sign and reality, as if signs are not real things and real things don’t participate in semiosis. This fools us into thinking that human-form characters are supposed to be a simulation of humans, a supposition that is less incorrect than unhelpful, I believe, for it does not really take us anywhere analytically speaking. In my opinion, such a view places too much faith on the humanness of characters: recall, we are dealing with shape-shifters, and they can fool us into thinking that they represent us.


Figure 16. Lucky★Star (anime series, 2007, based on Yoshimizu Kagami's manga, 2004).Figure 16. Lucky★Star (anime series, 2007, based on Yoshimizu Kagami's manga, 2004).

Consider Figure 16. Observable attributes of these characters, including visual elements, as seen here, like hair colors and types of dress, but also attitudinal patterns and relational patterns (‘sister-of,’ ‘subordinate-to,’ ‘teacher-of,’ etc), are not representative, in the usual realist mode, of human sociocultural identities. These attributes are, first and foremost, characterological tools: they help us, human viewers, as well as characters themselves, to differentiate, classify, and relate different characters.

Consider the designing of character language. Some anime and manga characters (as well as ‘working characters’) are known for their peculiar speech patterns which, or whose aesthetic effects at least, are not reducible to the actual-world distribution of variations. Take the character Lum from Takahashi Rumiko’s vastly popular manga Urusei Yatsura (1978-1987), for example. Her speech so frequently features forms such as the first person singular uchi, the interrogative ke, and the clause-final copulative daccha, that these forms have come to constitute a set of verbal emblems that distinctly point to her and, as far as I can see, no other character. (A character or a person using these forms would most likely be interpreted as enacting a citation – parody, mimicry, etc – of Lum.)

Interestingly, and quite common to this sort of ‘con[structed] language’ for characters, these forms do exist in the actual-world distribution of observable speech patterns of Japanese (see e.g. Miyazaki’s [2004] empirical study of uchi). The form daccha or cha has dialectal distribution in several different regions, including the Niigata region (on the west coast) where Takahashi, the manga artist, is originally from. However, Takahashi is reported to have claimed that Lum’s daccha has nothing to do with her regional dialect. Rather, according to manga artist Yamamoto Atsuji’s report (2011), it is a construction she borrowed from the Sendai dialect (on the east coast) featured in Inoue Hisashi’s novel. The verbal sign daccha is no simple, direct representation of a linguistic reality but a real linguistic effect of layers of figuration, citation, and assemblage. Whatever the form’s origin, one thing is clear: it would be a great disservice to the richness of the character if we reduced her speech somehow to a representation of actual-world speech patterns and sociocultural, human identities they index – as if, that is, she possessed these human identities. Lum is an ogre-shaped extraterrestrial creature who came to our planet with an intention (at least initially) to conquer it. It is perhaps important, for fans’ genuine enjoyment of the work and the character, not to indexically infer, not even for a moment, that she is from Sendai (or from the Earth for that matter).3

Character design – the assemblage of character elements like hair color, type of dress, language, etc – would be nothing if the point of characterization is all about simulating humans. Character design strives to give characters a sui generis reality, one that is irreducible to our kind of reality. To be sure, characters can be made to resemble actual people and represent their social identities in their historicity. Thus in some context, the inference from visual elements, for example, to the actual-world reality is a crucial process in the appreciation of the narrative and a critique of that reality: consider racism (Takamori 2012). But, then, characters can animate the nonhuman perfectly as well. My point is that humanity is not their essence: dolls in Rozen Maiden, penguins in Penguindrum, bacteria in Moyashimon, deities both from ‘real’ myths (Saint Seiya) and from ‘mythic’ myths like H. P. Lovecraft’s myths (Haiyore! Nyaruko-san), monsters in Pokemon and Yugioh, familiars and mediators in numerous “magical girl” works.

That humanity is not essential is even the central premise of Humanity Has Declined (Jinrui wa suitai shimashita) (anime series, 2012; based on Tanaka Romeo’s light novel). Here is Anime News Network’s plot summary: “It has been several centuries since human population has declined. Food has become harder to find and what little sources the humans have are considered highly valuable. The most prosperous species on the Earth are ‘Fairies,’ 10 cm tall creatures with high intelligence and a great love for sweets. A nameless girl, the main character, became a UN arbitrator between the humans and the fairies and had returned to her hometown to help her grandfather” (Anime News Network, n.d.).

But perhaps the most important premise is that the humans – us, the ‘old’ or ‘former’ human species (kyū-jinrui), as we are described in the story – have largely given up. Humanity has already declined. There is no war over those scarce resources; we are too tired for such a grand and exciting activity and very much ready to disappear and be taken over by the fairies, the ‘incumbent’ or ‘present’ human species (gen-jinrui). The honorable “UN arbitrator” has over time become a perfunctory, superficial bureaucratic position. The grandfather, also working for the UN, spends much of his time on his hobbies. Fascinated by the radical alterity of the fairies, and rather indifferent to her fellow old-humans – her experience of school bullying may be the factor – the protagonist becomes a de facto fieldwork researcher on the culture of fairies. In fact, while in college she studied anthropology (a study of jinrui), which, however, in this case has come to mean a study of the fairies, the present-humans. Miyadai (1995) has noted two forms of “the end” in contemporary Japanese subculture: “endless everyday” (owarinaki nichijō) and “post-nuclear war community” (kakusensō-go no kyōdōtai) (see Cassegård 2007:194-5).4 Humanity Has Declined is, interestingly, not a story of post-apocalyptic utopia/dystopia, neither an endless everyday nor post-rupture solidarity, obviously because it does not have a dramatizable rupture, a point, but just a slow decline, a line, but also because such anxieties and hopes are decidedly human projections. The old-humans have pretty much given up on themselves. The story decentralizes humanity, its science, its progress, and its civilization, in a way that troubles the figuration of rupture and eschatology. And the fairies’ strange ontology haunts us and invites us to rethink the relationship of anthropology to the fantastic.

The “fantastic,” Victoria Nelson suggests, “incorporat[es] elements by turns supernatural, antirepresentational, and grotesque” (2001:75). Vsevolod Emilevich Meyerhold speaks of a ‘realist’ puppet theatre director who “embraces naturalism and wants his puppets to imitate real people, until finally he realizes, with some regret, that the simplest solution is to replace his realistic puppets with real human actors” (Manning 2009:320; see also Proschan 1983; Bogatyrev 1983[1973]). Like this director’s search for realist representation in the realm of the fantastic-qua-antirepresentational, asking questions of ‘resemblance’ in our usual humanistic terms is kind of missing the point. Or let me put it this way: the ‘realist resemblance’ claim does not ask the question we are interested in here in any analytically new and interesting way: What are characters? What do they desire? How do they inhabit and haunt media? What are the conditions and effects of becoming-character? Against the background of anthropocentrism, we might propose character-centric realism, a view which does not reject the former but offers itself up as an alternative in order to disorient our general human-centered thinking and symmetrize the field of investigation for both humans and non-humans.

If anthropology (in our real world as much as in the world of Humanity Has Declined) can illuminate or at least recognize special and important roles played by fairies, ghosts, gods, angels, the dead, and other fantastic and liminal actants (including liminal objects like dolls and feces as well as liminal humans such as novices in a rite of passage and spirit mediums in a séance), then it might as well do the same with characters. My sense is that anthropology has, at least sometimes if not always, been a secret fan of the general notion of “symmetry” of explanation that Latour (e.g. 1993) and others speak of – perhaps secretly to itself.

Media Mix, Movement-in-Society, and Character (Non-In)Consistency

The character-centric perspective helps us better understand the logics of contemporary media ecology on which, I argue, the life of characters depends. In this section, drawing largely on the existing scholarship, we turn to examine the form of media “convergence” (Jenkins 2006) known in Japan as “media mix.”

The term “media mix” has been used since the 1980s to refer to “the cross-media serialization and circulation of entertainment franchises” (Steinberg 2012:viii; see ibid for an extensive analysis of the term’s historicity). Ito also explains it in the following way: the media mix creates “a synergistic relationship between multiple media formats, particularly animation, comics, video-games, and trading card games” (2010:86). Pokemon and Yugioh are perhaps the most visible examples of globally successful media mix enterprises. But examples of media mix just abound. It is nowadays quite rare to see a popular cultural production that is not, in some sense, media mix.

Ito makes the following, crucial observation:

Unlike with US origin media, which tends to be dominated by home based media such as the home entertainment center and the PC Internet, Japanese media mixes tend to have a stronger presence in portable media formats such as Game Boys, mobile phones, trading cards and character merchandise that make the imagination manifest in diverse contexts and locations outside of the home (ibid).

Portability is crucial to the organization of Japanese media mix, and, in fact, it is crucial to the life of characters as well. The focus on portability requires us to go beyond the question simply of the “what” of representation and inquire into the “when” and “where” of popular cultural encounter: the overall scheduling, or “chronotopic” organization (Bakhtin 1981), of character-encounter as a species of semiotic eventfulness.

Media mix must presuppose or must create a situation where characters can move beyond one media format and beyond one narrative context. Animated characters do not simply ‘move’ on a screen; they must be able to move out of the frame, and enter into other media forms and other narrative contexts. So movement-in-society is what is at stake (see LaMarre 2009). Characters’ life is maintained through processes of “decontextualization” and “recontextualization” (Bauman and Briggs 1990), which create a diversity of contexts of character-encounter. As Steinberg (2009) aptly suggests, the portable aspect of Japanese media mix makes it possible for people to encounter characters “anytime, anywhere,” saturating them with a feeling of “24/7” accessibility – a lot like convenience stores or vending machines, one might add, sociotechnical actants that are also rather ubiquitously distributed in Japan.


Figure 17. Media Mix, Movement-in-Society, and Character (Non-In)ConsistencyFigure 17. Media Mix, Movement-in-Society, and Character (Non-In)Consistency

But for this movement-in-society of characters to happen, people must be able to be intuitively convinced that the character they encounter in a manga, for example, is the same as, or at least not completely different from, the character that appears in its anime version. So interdiscursive consistency is another important aspect. Or more precisely: it is character (non-in)consistency that contributes to the convincing force of characters as characters (and not, or not necessarily, as a realist simulation) (Figure 17). For we are not talking about the absolute condition of sameness and difference but the sufficient condition of character malleability across different contexts. People must be sufficiently convinced in the semiotic event of character-encounter: “Ok, the way this character behaves and looks in this context, I’m ok with that. Based on my experience of other contexts in which I have encountered it, I’m sufficiently convinced that this behavior and this look is – here’s the magic phrase – not out of character.” Characters need to be sufficiently “in character". How simple is that!5

As Azuma (2009) points out, it is precisely these (non-in)consistent and malleable attributes or “elements” (yōso) of characters that people are consuming and producing in the otaku cultural realm. These elements are carefully typologized and classified, organized into “databases.” These databases can be implicit. People can implicitly cue a classificatory scheme by invoking more or less salient classificatory distinctions, like the attitudinal contrast between tsun-distancing and dere-intimate. But databases can also be explicit, as some fans have managed to create online databases that explicitly list and gloss various elements. This in a sense recalls the hybrid between the “encyclopedic” and the “ludic” modes that Foster (2009) identifies in the discourse of Japanese yōkai monsters.

The take-away point of this database theory is that people are affectively attracted not necessarily or not always by narrative “world” as such but by character elements that are organized at the database level, what Azuma calls “grand non-narrative.” That is, characters live on beyond specific narratives as an ensemble of elements and a site of affective engagement. This affective dimension is often described as moe, characterological empathy (see Galbraith 2009 for more extended discussion). On the other hand, the database view does not make narratives disappear totally. Quite the contrary: it generates a special relation of dependence to narrative. The potential pattern of combining different elements to create new characters requires new narrative imagination and narrative eventfulness (see Ōtsuka 2010). Decontextualized databases always afford narrative recontextualization, and decontextualization and recontextualization always form a feedback mechanism. Nothing is forever decontextualized, and no single context exists that is self-sufficient without others. It is character non-inconsistency that mediates between these processes (Figure 17).


As characters ‘move’ in society, this motivates people to fantasize their own ‘movement.’ Here, let us recall the otaku idiom of jigen, “dimensionality” (Figure 15). It is interesting to note that while the two realms, 2D and 3D, are kept separate, people fantasize what I will call cross-dimensional travel, where a double desire is expressed: keeping the boundary intact while at the same time ‘traveling’ to the other dimension. This ambiguous moment is often playfully encapsulated by the expression 2.5 jigen, or “2.5D”: not completely virtual 2D but not entirely actual 3D, either.

An often-noted example of this double desire is anime “sacred place pilgrimage” (seichi junrei). The phrase is part of a religious register. But it is re-appropriated in otaku culture to mean the fan practice of visiting actual-world places depicted in otaku cultural texts like anime that, by virtue of this depiction, “consecrate” these places for fans.6 “The ‘sacred place pilgrimage,’” Imai writes, “exemplifies this activity of seeking for the nexus of the otaku’s aspired ‘center’ (i.e. the anime/ two-dimensional world) and real life” (2011:18). Furthermore, this practice often involves online sharing of images of the place among fellow pilgrims, and thus it anchors real life (3D) in two sites of 2D, anime and online world. Imai continues: “In order to proceed further along the path through this nexus, [pilgrims] […] objectify themselves by uploading their images on websites and become part of the circulation of visual consumption in the same way as the anime characters, thereby making it possible to plunge into the two-dimensional world” (ibid:19; my emphasis).

While these pilgrims explicitly address the fantasy of 2.5D, the idea resonates with other practices in a more figurative way, but this figurativeness underscores all the more vividly the far-reaching applicability of this fantasy. In what follows, I attend more closely to three further examples.7

Dating Sims

“Dating simulation games” or dating sims refers to a type of simulation/ adventure game with a romantic theme. The objective is for the player-protagonist to achieve a romantic or sexual relationship with one of the characters in the game. While the view that sees this as a male-centered genre is certainly dominant, there are many dating sims for women, too, where the protagonist is a woman and the target romantic subjects are men. In fact, many dating sims today are not pornographic; rather, they focus more on intimacy and relationship. Moreover, sometimes romantic targets are not even human, as in Hatoful Kareshi (freeware version 2011), where the protagonist is a high school girl and all other characters, her potential targets, are male pigeons. In 2011 Japan Racing Association, an organization for horse racing, produced a dating sims where three female characters (the trainer, the breeder, and the jockey) interact with the protagonist, a colt; a sequel was produced in 2012. One thing is clear: if one claims the point of dating sims is just about male heterosexual masturbatory satisfaction, one would perhaps need to do more research. It is more productive for our discussion to align these games to Tamagocchi devices and Pokemon games as they both take part in the condition of “techno-intimacy” (Allison 2006; Galbraith 2011a). If anything, these games invite us to reexamine the normative assumption of humanistic sexualized desire (in this vein, see Nishimura 2013; Galbraith 2011b).

From the outlook, Loveplus, a popular dating sims first produced by Konami in 2009 for Nintendo DS, seems to follow the genre convention rather faithfully. However, unlike conventional dating sims, Loveplus does not end with a successful romantic accomplishment, but that is just a beginning. The game is not about ‘finding a date’ but about ‘staying in a relationship’: a subtle but crucial difference (see Galbraith 2011a for more extensive discussion). The game play follows the everyday life of the characters and the player. The mobile platform of Nintendo DS allows players to be ‘with’ their partner anytime, anywhere, a 24/7 kind of experience. Conventional dating sims are already suggestive of a cross-dimensional travel ‘to’ the world of characters, but in Loveplus, with the mobile device you can travel ‘with’ your character ‘to’ actual places. In fact, some local communities have positively responded to such a possibility. Konami once developed a tourism campaign for a spa resort in Atami specifically aimed at Loveplus players and their character-partner (Figure 18).


Figure 18. Promotional ad at (no longer available; last accessed July 2010).Figure 18. Promotional ad at (no longer available; last accessed July 2010).

I should note that the seeming weirdness of all this does not get lost on the part of those who participate in games like this. In fact players are quite ironical about it, though probably not really cynical. They can take jokes and humors; they joke about themselves. That is, players are not deluded. This is actually quite important to note. The general “moral panic” claim (see Kinsella 1998 for a discussion) that games like this make their users pathological – “They can’t distinguish between reality and fantasy!” – is, if anything, indicative of the insufficient, outdated, and deluded understanding of popular cultural aesthetics and contemporary media ecology on the part of those who make such claims. Otaku people are highly aware of the way the mainstream culture has stigmatized them, and they playfully twist the terms in which normalcy is imagined in the society, what conventionally goes in the name of normal.

Let me take a little detour here to elaborate on one particular joke, a well-known one, before returning to the foregoing discussion of dating sims. As we will see, the joke, an invocation of stigma, bespeaks a particular allure of 2.5D that haunts these games.

Niconico, a massively popular Japanese video-hosting site, has in recent years come to occupy a crucial node of popular cultural circulation and participation. The site’s main contents are user-generated videos that create fabulously complex structures of subcultural allusion and which are, therefore, largely hard to understand as cultural text without sufficient knowledge of otaku culture. Figure 19 shows a screenshot of one Niconico video, with its “crawling” comments superimposed on the video, a default feature on all Niconico videos (see Johnson forthcoming for analysis of this user interface). This particular video claims itself to be a slideshow of “ghost photographs” (shinrei shashin) but it is actually a parody made by a Niconico user of the genre of TV “horror story” shows where such photographs are often featured. The video starts with a photograph of some deserted house with a woman standing by the window, looking down and her long hair hiding her face. “According to the expert,” the video’s voiceover narration says, “this is a spirit of Sadako,” the ghostly character from the vastly popular J-horror work Ringu. Calling her “Sadako-chan” with the diminutive suffix, however, the narration readily cues a play frame. “If she ever follows you home,” it continues, “place a handful of salt at the four corners of your room so that she would not be able to escape. This way, you could transform your room into an exciting love nest.” Here the joke depends on the stereotypically pathologized “room” of a lonely, single, male otaku (see Galbraith and Christodoulou 2012 for the historicity of this chronotope).

Figure 19. Playing with the “Gross Face”Figure 19. Playing with the “Gross Face”

The video ends with what appears to be a blank screen, as shown in Figure 19. The narrator says, continuing with the eerie, dark tone reminiscent of the horror show genre: “Do you see this?8 The face of some ugly person is mysteriously captured here….” The viewers immediately respond with laughter. Because the glossy computer screen with which they view the video functions like a mirror when it blacks out, the joke is that otaku viewers are horrified by their own “gross” (kimoi) face. “Who’s this ugly one?”, one comment readily takes up the play frame.

While Sadako’s cross-dimensional movement (here as well as in the Ringu narrative series) points to the historicity of haunted media (Fisch 2010; Sconce 2000), the “gross face” in online world might contribute tragicomic laughter to the hauntology of media in a unique way. The gross face haunts normalcy. It ridicules the arrogance of “normal” people who act as if they are worthy of spectacular self-display, as if they have something important and beautiful to show, while the otaku saturate a space of commiseration, this video for example, with laughter as resounding as it is self-stigmatizing.

Note, then – back to dating sims – that these games usually do not show the protagonist/ player’s face and body in the game frame. The shadowed, effaced figure of the protagonist is iconic of the blackened screen that haunts and is haunted by the gross face, and therefore it bespeaks the general ludic sensibility of otaku culture. The “weirdness” with which dating sims players are well aware that they are often associated is part of the larger structure of what it means to play (and talk about) these games, just as the “grossness” of otaku face is part of the larger structure of playful talk in virtual communicative sites like Niconico.

In this vein, let me introduce a video clip, titled “Going on a date with Nene-san around the Kasai Rinkai Kōen station,” as an example of such playful reflexivity. One can perhaps read it also as a critical commentary with a tragicomic (and rather melodramatic) spin. Basically replicating the narrative structure of the promotional video for Zoot Woman’s “Memory” which it uses as a material for mashup, the clip is made by an anonymous amateur video producer, and it is posted initially on Niconico on June 20, 2010.

[,; the original upload: (requires membership sign-in)

The game player goes out with his device on a date with the Loveplus character Anegasaki Nene to the area around the Kasai Rinkai Kōen station in west Tokyo, a popular destination for couples and families. Initially the player still retains the sense of distinction between the inside and the outside of the game frame, as demonstrated by the way the video creator manipulates game frame and video frame. The creator embeds the ‘2D’ animated Nene on the actual-world ‘3D’ landscape – hand-clapping for a street performer, enjoying the aquarium, sitting at the table across the player/ protagonist, etc – but only within the frame created by the player’s Nintendo DS. As soon as the game frame is taken out of the video frame, Nene disappears; as soon as the game frame is reintroduced, she reappears – demonstrating the ambiguous presence and absence of characters. However, perhaps after so many dates, the player gets disoriented: is she really inside the frame? Am I really having a relationship with a virtual character? Then she pops out of the frame, holding his hand. But this brief moment of contact is immediately betrayed. He realizes that he has himself become a character in the game: he sees himself inside the frame while Nene walks away in the actual world, free from the game frame, in multiple bodies. He gets trapped in a virtual-actual limbo, and loses his control. Does he now regret his move? Or is he in a state of bliss, trading a death in the actual with a life in the virtual? But where is Nene?

Voice Acting, Vocaloids, and the Chronotope of “Person-Inside”

The playful disorientation of ‘inside’ and ‘outside’ across 2.5D is also a crucial theme in Japanese voice acting. Similar to puppetry, voice acting indexes a curious space of “split” between characters and actors (Proschan 1981:528). This is most directly expressed by the idiom of naka no hito, ‘person inside,’ an expression that otaku cultural participants often use to designate professional voice actors, seiyū. A metaphorical extension of costumed performers, those ‘inside’ the costume, this expression is, like “dimension,” another example of the widely circulated subcultural ‘fashion of speaking.’ I will argue below that the expression indexes the paramount importance of “effacement-work” – a concept I borrow from and develop with Gretchen Pfeil to suggest a counterpuntal move from Goffman’s “face-work” (1955). That is, rather than acts of disclosure and relations of identity, it emphasizes acts of opacity and relations of layering. Let me elaborate.

In considering Japanese voice acting, it is instructive to contrast it to our usual notion of theatrical acting. Singin’ in the Rain (dir. Stanley Donen, 1952) concludes with a spectacular salvage of dismembered entities into an organic whole: “That’s the girl whose voice you heard!” The revelation satisfies both the desire of the audience (in the story) to know the truth of identity and that of the cinematic audience to see that desire satisfied. This conventional theatrical desire depends on a body capable of preserving an organic whole provided by actors. Here characterization as an aesthetic technique is a property of actors and operates on the modality of embodiment and identity. In contemporary North American context, such a perspective often motivates the question of ‘performing the self,’ and accordingly a hierarchical relation is mapped on the terms of ‘acting’ between inauthentic acting(-out) and authentic disclosure of identity (see Lemon 2009 for a fascinating counterexample in Russia; see also Carr 2010).

Voice actors, on the other hand, live disembodiment everyday. As a matter of their professional labor, they disengage their voice from their body. Their artistic virtuosity is measured by the skills with which they hide their body ‘inside’ their characters. They often describe the space of their labor as “in front of the microphone” (maiku mae). The enclosed space of maiku mae – that is, the sound studio – is iconic and indexical of naka no hito, and it points directly to the necessary effacement of the body. The actors’ body is necessarily effaced in the narrative text in which their characters appear.9


Figure 20. Effacement-work. Mizuhashi Kaori (center) at a promotional event for Puella Magi Madoka Magica, animating her character, Tomoe Mami.Figure 20. Effacement-work. Mizuhashi Kaori (center) at a promotional event for Puella Magi Madoka Magica, animating her character, Tomoe Mami.

  But voice actors are also effaced or obscured from the public gaze at large, too. When they step up on a public stage, for example promotional events, and are asked to “do the voice” of their character for the audience, it is not unusual to see them deploy techniques of semiotic interruption: hiding one’s face with one hand, asking the audience to close their eyes, showing their back to the audience, etc. More commonly, voice actors tend to rely on the existing lines of the character as opposed to ‘speaking’ freely in the character voice. It should be clear from these examples that the ‘true’ identity of naka no hito is ordinarily not a secret at all in the usual sense of classified, confidential information or in the psychoanalytic sense of repression or interdiction. Rather, it is made elaborately irrelevant through effacement-work.10

It is worth noting here that the expression naka no hito seems always to carry a playful quality. Part of this quality derives from the fact that, while a standalone expression, the phrase is often used in a well-known collocation: naka no hito nado inai! “No way there’s person inside!” Fans have traced this collocation to Yoshida Sensha’s nonsense-comedy manga Utsurundesu (1989-1994). In the manga, the character Kawauso kun, an otter, appears in disguise with another character simply called Shita no Hito (‘person underneath’), who is also disguised with a mask. Kawauso kun uses Shita no Hito as a way to give himself human physical attributes. But of course, everyone can literally see that there is Shita no Hito underneath holding up Kawauso kun to make him look like a human, and yet he keeps denying that by saying: Shita no hito nado inai, “There’s no one underneath.”

Just as in this manga, the statement “No way there’s person inside!” (absence, irrealis) is usually apprehended in comedic contrast to the counterclaim, “Of course there is person inside, what are you talking about” (presence, realis). These incongruous statements form a comedic couplet underscoring ambiguous determination between presence and absence and, reminiscent of the comedic genre of manzai, suggest that the absurd is always ready to haunt (and bring laughter to) the real.

 Figure 21. VocaloidFigure 21. Vocaloid

Vocaloid, a singing voice synthesizer for computer music, provides a case that helps us better understand both the modality of disembodiment and the ludic quality of effacement-work. As often noted, it is the power of anthropomorphic characterization that has made this technology globally popular, rendering a technical fact into a bundle of moe-inducing character elements, a “media platform that is part software technology […] part cultural idea […]” (Condry 2011a). At the center of this bundling is, of course, the voice. A Vocaloid consists of three layers: Score Editor, Synthesis Engine, and Singer Library (utagoe raiburari). Score Editor functions like a traditional sequencer (composition data processing), and Synthesis Engine like a traditional synthesizer (sound data processing). But the distinctiveness of a Vocaloid’s voice is generated really at the level of the database, Singer Library (developed by a specific developer). For each Vocaloid, its distinctive database is constructed from voice samples taken from a specific source.11

Figure 22. Notable VocaloidsFigure 22. Notable Vocaloids

These sources are often casually described as naka no hito, ‘person inside.’ Indeed, some of the known sources are professional voice actors, like Shimoda Asami (for Rin and Len), Nakajima Megumi (for Megpoid), and Fujita Saki (for Hatsune Miku). But we must note certain funniness in talking about these seiyū as the naka no hito of their respective Vocaloid character. Is it not strange to say that they ‘perform the character’ in the same sense in which we usually attribute the idea of performance to acting in general? As it turns out, the naka no hito metaphor refers not to the relation of “performance,” an actor embodying a character, but to the relation of “animation” in Silvio’s (2010) sense, an actor giving a life-force, a spirit, to a character. Note that the metaphor sometimes refers even to the relation between the operator and the operated, as in the ‘pilot’ of a robot who operates it from inside its body (a crucial chronotope in the post-Mazinger Z Japanese mecha anime and manga, recently recontextualized in Pacific Rim, a 2013 film directed by Guillermo del Toro). This is why it makes sense to otaku cultural participants to use the metaphor even in this case of a high degree of disembodiment.

When these seiyū “do the voice” of their Vocaloid characters in public events, therefore, such an action is almost always framed in a self-consciously playful way. During one segment of a radio show in 2010,12 Fujita Saki, Miku’s naka no hito, was asked by the audience to do the voice of Miku, and she accordingly provided her voice for the audience. When this segment of the show was uploaded on Niconico, comments responded to Fujita’s impromptu ‘impersonation’ with applauds and laughter. That is, the listeners immediately understood the funniness of talking about this act as ‘impersonation,’ which it was not. It was not an act of impersonation (monomane) because they all knew Fujita is Miku’s naka no hito and she does do her voice. But by the same token, it was nonetheless funny to treat it as though it were an act of impersonation, because Miku’s voice is indeed already technologically removed so many degrees from the actor and her human presence. (Fujita even mimicked the robotic disfluency characteristic of Miku’s speech; Vocaloids are not designed to produce human spoken sounds.)13

“Is it Miku? Is it Fujita? But wait, aren’t they the same in fact? But then why am I asking this question?” This sense of ambiguity is well articulated especially for its ludic quality by one of the comments to the video. This commenter came up with a curious expression soto no hito ‘person outside’ to refer to Fujita, the actor (Figure 23). That is, this commenter and numerous others who cite the phrase frame Miku as though she has her own sui generis power with which to animate – ensoul, operate – Fujita from within. The commenters appreciated the funny uncanniness of similitude in Fujita’s voicing by playfully inverting the spatial relation of inside and outside.


Figure 23. Naka no hito/ soto no hito. Screenshot of Niconico's video interface, showing the video and the comments a moment after Fujita's '(non-)impersonation.'Figure 23. Naka no hito/ soto no hito. Screenshot of Niconico's video interface, showing

the video and the comments a moment after Fujita's '(non-)impersonation.'


Figure 24. Ludic disorientationFigure 24. Ludic disorientation

Effacement and Characterization in Online Videos

My final example comes from Niconico, the Japanese video-hosting site I have already mentioned a few times. With the site becoming increasingly popular since its inception in 2006, some of its users who habitually post their videos there, especially music performance videos, have attained a certain degree of virtual fame. For example, a singer who goes by the name Isaji has several of his singing videos with over one million views. He has so far remained – or, I would like to say, has managed to remain – amateur and a pure online presence, while some of the other Niconico singers and musicians have become professional, letting their virtual fame leak into the realm of the actual.

What is interesting about these amateur Niconico video producers, in sharp contrast to, say, YouTube videos, is that their self-presentation is often cloaked in elaborate camouflage: masks, costumes, digital blurring, avoidance of closeup, use of text-to-voice applications, compositional arrangement that places the head off frame, and so forth. While there are multi-million viewers for Isaji’s videos, no one really knows what he looks like. In fact, his videos do not visually represent his ‘3D’ body at all. So here is a seemingly curious situation: millions of viewers, but nobody has seen the face.

Figure 25. Effacement-workFigure 25. Effacement-work

But in fact this is not curious at all. As I have discussed elsewhere (Nozawa 2012), on Niconico this kind of camouflage is in fact normative and expected. For example: cooking videos. I know no Niconico cooking videos (with a very few exceptions, which prove the rule) where people appear within the frame and directly speak to the camera, unlike similar videos on YouTube and other cooking-related sites (which are, largely, a simulation of TV cooking shows). The hands and the arms are the only body parts that are frequently visible. In one ordinary cooking video, the uploader14 is beating eggs in a stainless bowl. At one point during this routine, we see a little drawing of a butterfly conspicuously overlaid on the video just where the bowl is depicted (Figure 25). The reasoning here is that the stainless bowl is showing a reflection of the uploader’s face. Even though we can easily imagine that the reflection would be blurry, upside down, and for all intents and purposes visually unrecognizable, the uploader felt it just appropriate to hide the reflection by superimposing the butterfly image upon it, and, as indicated by the comments, the viewers also shared this feeling. The crucial thing to note here is that her action is not simply or even primarily motivated by and read as an effort to protect her privacy (though that is obviously a factor) but it is motivated by and read as a habitual pattern of action, a sense of propriety developed over time for and by the participants in this virtual communicative world. If you accidentally spill some of the beaten eggs on the kitchen table, even a tiny drop, you wipe it; if you accidentally get your face exposed in your video, even a blurry unrecognizable bit, you mask it. Acts of camouflage, opacity, and anonymity are just plain normative in Japanese virtual communication. Anonymity is not a hindrance to communication, but it is the normative condition of communication in this situation.

How does Isaji, the legendary Niconico singer, efface his presence and succeed in staying virtual? – By consistently using a manga character, Abe Takakazu from Yamakawa Junichi’s homoerotic work, as his avatar. I will not go into the nature of this character in the manga’s context or why this character has been chosen. It is enough for us here to note that every time Isaji makes a video appearance on Niconico, or every time someone mentions him, this character is invoked. Every time this character is invoked on Niconico and in related interdiscursive contexts, Isaji is now a crucial part of the meaning of the image. Isaji is this character: he lives on as this character in the virtual world. Just as in the butterfly image in the cooking video, the use of the avatar is not really a way of hiding the real identity of Isaji as a “3D” person, nor is it an attempt by Isaji to identify with the avatar, in this case, Abe. Though some of his real-world attributes have in fact been known (for example he lives in Miyazaki prefecture; he has videos featuring his singing in the regional dialect), this has not thereby resulted in shattering a fantasy. Knowing who Isaji the person is in the actual world is separate from enjoying communicating with his online presence. Everyone knows him as a character on Niconico; and that is sufficient for this character’s (non-in)consistency.

Recall the state of disorientation brought forth by voice acting: “Is it the character? Is it the actor? But are they not the same? Then why am I asking this question at all?” A similar sense of ambiguity is entertained here as well. In the second episode of science fiction comedy anime Kiddy GiRL-AND (2009) – note, this is a commercial production for TV broadcast – main characters Q-feuille and Ascoeur encounter a human-shaped “security system” that looks just like Abe Takakazu (Yamakawa’s manga character). Upon recognizing the face Ascoeur says, immediately and in an utterly matter-of-fact way, “Oh, that’s Isaji.” The voice that comes from “the security system” is performed by a professional voice actor but the voice is obviously a phonic homage to Isaji, for the voice actor mimics the singer’s characteristic low tone voice. In matter of weeks after the broadcast, Isaji (who had not been informed that his online name would be appropriated in a TV broadcast this way) uploaded a short video on Niconico that recycles the scene in question, adding, however, his own voice-acting for “the security system,” mimicking the professional voice actor’s voice, which is a mimicry of his own voice. There is no question about it: Isaji has successfully characterized himself. He is a character, as real as his ‘3D’ being but irreducible to it.


The meta-meta-....discursive spirals that envelop Isaji are much more complicated than I could explore here, but I chose his case because he is perhaps the most emblematic character on Niconico whose semiotic effectiveness, as a character, depends on layers of effacement. This effacement-work is hardly peculiar to him and is more generally observed among Niconico participants and in other sites of Japanese virtual communication.

Examples like these tell us that people are not simply consuming and producing “2D” characters. As “3D” beings, they are realizing themselves as characters, transforming themselves into fantastic but real beings through cross-dimensional travel. The interstice of 2.5D affords characterization as a distinct mode of semiosis irreducible to notions of embodiment and identity. Just as it is moot to define the essence of anime characters by their ‘realistic’ resemblance to the human order, participants in virtual communication normatively efface their 3D social identities in order precisely to participate in these communicative events. Effacement-work is a semiotic technique of characterization. Indeed I would go so far as to say that characterization, understood as effacement-work, defines semiosis more adequately. For semiosis is a process of creating layers of signs, not a process of exposing something behind them.


I would like to conclude with a few comments for further research beyond the present argument.

  1. Of particular interest to those interested in Japan: In this paper I have not been able to explore the more general cultural ideology of characters (kyara) in contemporary Japan. The term has recently been used to mean “persona,” an index that typifies personal demeanor. Today one hears expressions involving kyara as frequently in everyday conversation as in mass media messages: kyara o tsukuru (lit. creating a character, viz. forcing an unexpected/ unwarranted persona on oneself), kyara ga kowareru (a character one has so far maintained is breaking down), kyara o tamotsu (staying ‘in character,’ keeping oneself characterized in a specific way), etc.

    Here let me just mention one issue related to the ideology of kyara: stigmatization. Events of character “animation” (Silvio 2010) rely, in principle, on non-embodiment: kyara are most productive when they are prosthetically recontextualizable like cosplay outfits. Thus, it is sometimes deeply problematic when embodiment happens in such a way that a negative kyara is fixed in a relation of identity. Bullying in school and workplace is often seen as resulting from rigidly fixing a character to a person, especially ijirare kyara (the character of the ‘teased’). On the other hand, an intriguing counterstrategy to cancel such characterological fixity is suggested in the subcultural idiom of debyū, ‘debut,’ a hyperbolic reference to biographical transitions. The term is a relatively new addition to popular cultural slang and is directly linked to school context (e.g. ‘high school debut’) though it is applicable to other contexts, such as going to an urban center for the first time (e.g. ‘Tokyo debut’). Ideally, ‘debut’ at these transitional moments allows one to ‘reset-to-default’ one’s past kyara and to obtain new ones, whose newness, however, is to remain effaced. For, as Goffman (1963) said, when an effort to erase one’s stigma becomes known to others, this knowledge works to produce a new or related stigma. I hope that a more adequate analysis of these processes of stigmatization might follow and complicate some of the claims made in this paper.15

  2. Of particular interest to linguistic anthropologists: Japanese sociolinguists have recently focused on “character-language” and the intersection between sociolinguistic variation (heteroglossia) and the subcultural mode of characterization; see for example Kinsui’s (2003) work on yakuwari-go (‘role language’), Tanaka (2011) on hōgen kosupure (‘dialect-cosplay’), Sadanobu (e.g. 2011) on hatsuwa kyarakuta (‘discourse character’). While these works are still only available in Japanese, an interesting linkage may be made to the Anglophone literature on “enregisterment” (Agha 2007). In this vein, Goffman’s Frame Analysis (1974) already features the concept of “figure,” which, especially in the form of non-“natural” figure, comes closest to the notion of characterization developed here. Recall that Goffman in fact offered the term “character” as an alternative (1974:523; see Hastings and Manning 2004). The Japanese material presented in this paper may be useful for a re-reading of Goffman and contemporary linguistic anthropology more generally from a character-centric perspective.

  3. Lastly, the present paper proposes a move away from anthropocentrism to explore ideas and practices afforded by character-centric perspectives, and I argued that this move requires a theoretical and ethnographic focus on effacement-work. There is an implicit political stake in exploring characterization at this historical moment. When our contemporary predicament prompts us to constantly engage a politics of anonymity and privacy – when, for example, we find that in Canada it is now potentially a crime punishable by imprisonment to participate in a public protest with one’s face covered with a mask16– a semiotics of characterization-as-effacement could make significant contribution from a distinctly ethnographic angle. If political domination operates by securing its own secrecy while demanding that we disclose our naked humanness, that we disrobe unconditionally, that we dismantle our opaque façade – a hoodie in Florida, a veil in Paris – then characterization has tactical value in political theory and action. This should also cue a renewed appreciation of anthropology’s longstanding interest in such themes as masks and tricksters (see Coleman 2010, 2012), and help reorganize the wealth of anthropological thought to consider anew a politics of characterization in its relation to power, to capitalism, and to technology.


The paper is based on a series of lectures. I thank Karl Friday, David Slater, and James Dorsey for their kind invitation. Thanks to William Feeney, Michael Foster, Patrick Galbraith, Keiko Nishimura, Debra Occhi, Ayako Takamori, Satsuki Uno, and anonymous reviewers for comments. Special thanks to Paul Manning, the editor, for convincing me to write this paper.


Agha, Asif. 2007. Language and Social Relations. New York: Cambridge University Press.

Allison, Anne. 2006. Millennial Monsters: Japanese Toys and the Global Imagination. Chicago: University of Chicago Press.

———. 2009. “Pocket Capitalism and Virtual Intimacy: Pokemon as Symptom of Postindustrial Youth Culture. In Figuring the Future: Globalization and the Temporalities of Children and Youth. Jennifer Cole and Deborah Durham, eds. Pp. 179-95. Santa Fe: School of American Research Press.

Alt, Matt, and Hiroko Yoda. 2007. Hello Please!: Very Helpful Super Kawaii Characters from Japan. San Francisco: Chronicle Books.

Anime News Network. n.d.

Azuma, Hiroki. 2009. Otaku: Japan’s Database Animals. Jonathan E. Abel and Shion Kono, trans. Minneapolis: University of Minnesota Press.

Bakhtin, M. M. 1981. The Dialogic Imagination: Four Essays. Michael Holquist, ed. Caryl Emerson and Michael Holquist, trans. Austin: University of Texas Press.

———. 1986. Speech Genres and Other Late Essays. Caryl Emerson and Michael Holquist, eds. Vern W. McGee, trans. Austin: University of Texas Press.

Bauman, Richard, and Charles L. Briggs. 1990. Poetics and Performance as Critical Perspectives on Language and Social Life. Annual Review of Anthropology 19: 59-88.

Bogatyrev, Petr. 1983[1973]. The Interconnection of Two Similar Semiotic Systems: the Puppet Theater and the Theater of Living Actors. Semiotica 47 (1-4): 47–68.

Carr, E. Summerson. 2010. Scripting Addiction: The Politics of Therapeutic Talk and American Sobriety. Princeton: Princeton University Press.

Cassegård, Carl. 2007. Shock and Naturalization in Contemporary Japanese Literature. Leiden: Brill.

Coleman, Biella. 2010. Hacker and Troller as Trickster. Social Text

———. 2012. Our Weirdness is Free. Triple Canopy 15.

Condry, Ian. 2009. Anime Creativity: Characters and Premises in the Quest for Cool Japan. Theory, Culture & Society 26 (2-3): 139-63.

———. 2011a. Miku: Japan’s Virtual Idol and Media Platform. MIT Center for Civic Media.

———. 2011b. Post-3/11 Japan and the Radical Recontextualization of Value: Music, Social Media, and End-around Strategies for Cultural Action. International Journal of Japanese Sociology 20 (1): 4-17.

Fisch, Michael. 2010. Ringu/ The Ring: Tracing the Analog Spirit in a Digital Era. Refractory: A Journal of Entertainment Media.

Foster, Michael Dylan. 2009. Pandemonium and Parade: Japanese Monsters and the Culture of Yōkai. Berkeley: University of California Press.

Galbraith, Patrick W. 2009. Moe and the Potential of Fantasy in Post-Millennial Japan. Electronic Journal of Contemporary Japanese Studies.

———. 2011a. Bishōjo Games: ‘Techno-Intimacy’ and the Virtually Human in Japan. Game Studies 11 (2).

———. 2011b. Fujoshi: Fantasy Play and Transgressive Intimacy among ‘Rotten Girls’ in Contemporary Japan. Signs 37 (1): 211-32.

———. 2011c. Maid in Japan: an Ethnographic Account of Alternative Intimacy. Intersections: Gender, History and Culture in the Asian Context 25.

Galbraith, Patrick W, and Androniki Christodoulou. 2012. Otaku Spaces. Seattle: Chin Music Press.

Goffman, Erving. 1955. On Face-Work: an Analysis of Ritual Elements of Social Interaction. Psychiatry: Journal for the Study of Interpersonal Processes 18 (2): 213-31.

———. 1963. Stigma: Notes on the Management of Spoiled Identity. Englewood Cliffs, N.J: Prentice-Hall.

———. 1974. Frame Analysis: An Essay on the Organization of Experience. New York: Harper & Row.

Hastings, Adi, and Paul Manning. 2004. Introduction: Acts of Alterity. Language & Communication 24 (4): 291-311.

Imai, Nobuharu. 2011. Anime ‘Sacred Place Pilgrimages’: the Potential for Bridging Traditional Pilgrimage and Tourism Activities through the Behavior of Visitors to Anime ‘Sacred Places’ – an Analysis of ‘Votive Offering Tablets’ (ema) at Washinomiya Shrine, Saitama Prefecture. Isaac Gagné, trans. University of Hokkaido Center for Advanced Tourism Studies 1: 87–111.

Ito, Mizuko. 2010. Mobilizing the Imagination in Everyday Play: the Case of Japanese Media Mixes. In Mashup Cultures. Stefan Sonvilla-Weiss, ed, Pp. 79-97. Wien: Springer–Verlag.

Jenkins, Henry. 2006. Convergence Culture: Where Old and New Media Collide. New York: NYU Press.

Johnson, Daniel. In press. Polyphonic/Pseudo-Synchronic: Animated Writing in the Comment Feed of Nicovideo. Japanese Studies.

Katsuno, Hirofumi. 2011. The Robot’s Heart: Tinkering with Humanity and Intimacy in Robot-Building. Japanese Studies 31 (1): 93-109.

Kinsella, Sharon. 1998. Japanese Subculture in the 1990s: Otaku and the Amateur Manga Movement. Journal of Japanese Studies 24 (2): 289-316.

Kinsui, Satoshi. 2003. Virtual Nihongo: Yakuwari-go No Nazo [Virtual Japanese: Exploring Role-Language]. Tokyo: Iwanami Shoten.

LaMarre, Thomas. 2009. The Anime Machine: A Media Theory of Animation. Minneapolis: University of Minnesota Press.

Latour, Bruno. 1993. We Have Never Been Modern. Cambridge: Harvard University Press.

Leavitt, Alex, and Andrea Horbinski. 2011. Even a Monkey Can Understand Fan Activism: Political Speech, Artistic Expression, and a Public for the Japanese Dōjin Community. Transformative Works and Cultures 10.

Lemon, Alaina. 2009. The Emotional Lives of Moscow Things. Russian History 36 (2): 201-18.

Manning, Paul. 2009. Can the Avatar Speak? Journal of Linguistic Anthropology 19 (2): 310-25.

Miller, Laura. 2010. Japan’s Zoomorphic Urge. ASIANetwork Exchange 17 (2): 69-82.

Miyadai, Shinji. 1995. Owarinaki Nichijō o Ikiro [Live the Never-ending Everyday]. Tokyo: Chikuma Shobō.

Miyazaki, Ayumi. 2004. Japanese Junior High School Girls’ and Boys’ First-Person Pronoun Use and Their Social World. In Japanese Langauge, Gender, and Ideology: Cultural Models and Real People. Shigeko Okamoto and Janet S. Shibamoto Smith, eds. Pp. 256-74. New York: Oxford University Press.

Nelson, Victoria. 2003. The Secret Life of Puppets. Cambridge; London: Harvard University Press.

Nishimura, Keiko. 2012. Where Program and Fantasy Meet: Female Fans Conversing with Character Bots in Japan. Transformative Works and Cultures 12.

Nozawa, Shunsuke. 2010. Real Nowheres, Real Nobodies: a Techno-Aesthetic of the Everyday. Presented at the Association for Japanese Literary Studies conference, Yale University.

———. 2012. The Gross Face and Virtual Fame: Semiotic Mediation in Japanese Virtual Communication. First Monday 17 (3)

Occhi, Debra J. 2010. Consuming Kyara ‘Characters’: Anthropomorphization and Marketing in Contemporary Japan. Comparative Culture 15: 77-86.

———. 2012. Wobbly Aesthetics, Performance, and Message: Comparing Japanese Kyara with Their Anthropomorphic Forebears. Asian Ethnology 71 (1): 109-32.

Ōtsuka, Eiji. 2010. World and Variation: the Reproduction and Consumption of Narrative. Marc Steinberg, trans. Mechademia 5 (1): 99-116.

P. I. E. Books. 2002. Character World. Tokyo: P. I. E.

Parmentier, Richard. 1994. Signs in Society: Studies in Semiotic Anthropology. Bloomington: Indiana University Press.

Peters, John Durham. 1999. Speaking into the Air: a History of the Idea of Communication. Chicago: University of Chicago Press.

Proschan, Frank. 1981. Puppet Voices and Interlocutors: Language in Folk Puppetry. Journal of American Folklore 94: 527–55.

———. 1983. The Semiotic Study of Puppets, Masks, and Performing Objects. Semiotica 47 (1-4): 3–44.

Sadanobu, Toshiyuki. 2011. Nihongo Shakai Nozoki Kyarakuri [Exploring the Culture of Japanese through Characters]. Tokyo: Sanseidō.

Sconce, Jeffrey. 2000. Haunted Media: Electronic Presence from Telegraphy to Television. Duke University Press.

Silvio, Teri. 2010. Animation: the New Performance? Journal of Linguistic Anthropology 20 (2): 422-38.

Steinberg, Marc. 2009. Anytime, Anywhere Tetsuwan Atomu Stickers and the Emergence of Character Merchandizing. Theory, Culture & Society 26: 113-38.

———. 2012. Anime’s Media Mix: Franchising Toys and Characters in Japan. Minneapolis: University of Minnesota Press.

Takamori, Ayako. 2012. ‘The Cute Savage’: Visual Semiotics and Racialized Iconography. Paper presented at the 3rd Mechademia Conference, Dongguk University, Seoul. December 2.

Tanaka, Yukari. 2011. Hōgen Kosupure no Jidai: Nise Kansai-ben kara Ryōma-go made [Dialect Cosplay: From Pseudo-Kansai Dialect to the Language of Ryōma]. Tokyo: Iwanami Shoten.

Whorf, Benjamin L. 1956[1939]. The Relation of Habitual Thought and Behavior to Language. In Language, Thought, and Reality: Selected Writings of Benjamin Lee Whorf. John B. Caroll, ed. Pp. 134-59. Cambridge: MIT Press.

Yamamoto, Atsuji. 2011. Cha. Atsuji-ya Nikki.

Yano, Christine Reiko. 2013. Pink Globalization: Hello Kitty’s Trek Across the Pacific. Durham: Duke University Press.


1 In this paper I do not wish to elaborate further on the expression “speech-actant,” for this seems to me to deserve more careful discussion, except to note that I use it here simply to fancy a conceptual space that speech act theory and actor network theory might use as their rendezvous point.

2 Excellent scholarship on specific characters, especially high-profile ones, and their sociohistorical, economic, aesthetic significance is now available from a variety of theoretical perspectives; see e.g. Allison 2006, 2009, Condry 2009, 2011b, Fisch 2010, Leavitt and Horbinski 2011, Miller 2010, Occhi 2012, Steinberg 2012, Yano 2013. See Occhi 2010 for “less famous, yet ubiquious characters” similar to the “working characters” mentioned here.

3 Doing so would in fact bring you into the realm of the counterfactual, because her being alien is more factual than her being human. Accordingly, a fan may indeed be able to generate a (meta-)fantasy spinoff: “What if she is actually from Sendai… from my own neighborhood!” I would not be surprised if such a derivative (dōjin) work has already been produced in the past.

4 I thank Patrick Galbraith for bring my attention to Miyadai’s formulation here.

5 But of course the limit of character (non-in)consistency is no simple matter, like all contextually determined (indexical) properties. That people can be convinced of a character’s consistency means that they may also be convinced otherwise or that they may invoke different criteria.

6 While some of these places are themselves religious sites, for example, shrines, they also encompass what I call “real nowheres,” locations that exist in our actual world but which, unlike religious or historic sites, are so utterly ordinary and unremarkable that they often escape our everyday recognition; Nozawa 2010; cf. Bakhtin 1986:47.

7 Consider also techniques like rotoscope and prescoring, as well as augmented reality environment and related geo-media interfaces. These are often cited as being productive of technosocial hybrids that play with and build upon the fantastic interstice of 2.5D.

8 This phrase is itself a playful citation of the emblematic phrase used in the horror show voiceover narration: owakari itadakeru darōka, asking the viewers if they could ‘recognize’ weird and strange things in what they are seeing.

9 Such a habitual and necessary engagement with effacement suggests that what Silvio (2010) calls “ensoulment,” following Nelson (2003), is in fact more important to voice acting than “embodiment.” Voice actors like Hayashibara Megumi and Sawashiro Miyuki, as well as fans, often invoke spiritualist terms like soul, summoning, channeling, etc as part of their practical theory of characterization. The question of ensoulment in relation to voice and character is a complex one, and I’d like to give it a separate treatment. For spiritualism and technology see Peters 1999; Sconce 2000; see also Katsuno 2011. See also contributions to the 1983 issue of Semiotica to which Proschan 1983 serves as an introduction.

10 The culture of voice acting has undergone a lot of changes over the past few decades. The aesthetics of effacement as noted here coexists with a more recent tendency to require voice actors to expose themselves (‘face-exposing’ or kaodashi) more often and actively to the media for promotional and other purposes. But I believe the chronotope of naka no hito still underscores the general semiotic ideology of this culture, in so far as the definition of voice acting constitutively entails effacement, in a similar way that that of ventriloquism entails dislocation. For kaodashi see Nozawa 2012.

11 An enormous (and, in principle, infinite) range of characterological possibility is suggested by the fan practice known as or associated with jinriki vocaloid, ‘manual Vocaloids.’ Fans disassemble some existing sound data (e.g. anime character voice, but, technically, anything) into some unit of metricalization – a process equivalent to the creation of Vocaloid’s “Singer Library” – and reassemble them into sung texts. While Vocaloid™ is proprietary software, a free/shareware program called UTAU has been developed that supports this sort of playful sono-phonic collage. This turns any sonic information (which does not have to be ‘human’ at all) into a phonic (‘voicing’) character technically equivalent to Hatsune Miku (and other Vocaloids). One online list of jinriki vocaloids includes more than 70 characters based on sound sources as diverse as anime, TV commercials, gay porn, and miscellaneous sound bites (

12 Episode 473, Mikako @ Payo Payo. Interview with Takahashi Mikako. Animate TV. August 8, 2010.

13 This whole interactional pattern of “Oh such a good monomane! – No, wait, it’s not monomane because it’s the voice actor him/herself!” is in fact a common joking routine often observed in the context of ‘doing the character voice.’

14 The expression ‘uploader’ (upunushi) technically refers to the one responsible for uploading some content online and therefore its referent could be different from the content producer or its participant. But it often elides these differences and is used to contrast with shichōsha, ‘[video] viewers.’

15 I thank William Feeney for introducing to me this problem of characterological fixity in cases of bullying. As expected, popular cultural texts are highly reflexive about these processes, tragicomically transforming stigma into characterological and narrative resource (e.g. consider Kimi ni Todoke and Watashi ga Motenai no wa Dou Kangaetemo Omaera ga Warui!)

16 See