So, last week I went on a bit of a philosophical/theoretical tangent with a mind to explaining where much of what I have to say not only in today’s post but throughout the run of this blog comes from conceptually. I would strongly suggest, dear reader, that you go back and read it before advancing any further. Don’t worry, I’ll still be here when you get back.
Neil Cohn recently published a monograph of his visual language (VL) research, The Visual Language of Comics: Introduction to the Structure and Cognition of Sequential Images, so I thought now might be a good time to consider a slightly older essay in the volume Manga: An Anthology of Global and Cultural Perspectives titled “Japanese Visual Language: The Structure of Manga.” I should say from the get go that I am not terribly sympathetic to strictly linguistic approaches to comics (as such methodologies seem unnecessarily limiting and grossly misleading as a result), and I am even less sympathetic (bordering on hostile) to sweeping judgments of culture (particularly when such judgments exhibit a crypto-exoticism) that are based in the seemingly unassailable “objectivity” of methods most commonly found in the sciences. I have on good authority that Hannah Miodrag’s recent book Comics as Language: Reimagining Critical Discourses on the Form does much to undermine the “comics as language” arguments that float about here and there, but seeing as I have yet to get my hands on it, I have no idea whether I am simply repeating here what she has already said. If this happens to be the case, I apologize in advance.
Before I get into my primary gripes with Cohn’s piece, I should note a general problem I have with claims of “Japanese this” and “Japanese that.” To call something Japanese in a provisional sense in order to simply narrow the field of discussion (e.g. Japanese lyric poetry as opposed to all lyric poetry) is one thing, but to then use that discussion to make essentialist claims about the culture at large is to ignore just how ill-defined the adjective “Japanese” truly is. Sure, it may be easy enough to go about our lives with the facile assumption of Japanese racial and cultural homogeneity, but when we say “Japanese,” do we mean it in some ethnic sense so as to include, say, Japanese people living abroad and generations descended from emigrants (to Brazil or the US, for instance)? Do we recognize how this potentially excludes the Ainu or Okinawans or zainichi Koreans? Do we mean it in a national sense (i.e. citizens of the nation of Japan) and thereby include naturalized citizens but exclude ethnically Japanese citizens of foreign lands, whose own sense of Japanese identity is inextricably linked to that of their country of residence? Do we dare broach the fact of an entire cottage industry of popular nonfiction in Japan that tries to identify an essential (and generally racial) Japanese-ness that, in practice, often has the adverse effect of functioning as a kind of prescriptive cultural identity?
Which brings me back to Cohn’s essay. There is a serious methodological problem with his argument in regard to manga, so today I would like to look more closely at two aspects of that argument: how he constrains the range of what might be considered manga in the first place (down to more or less a stereotype) and how his own linguistic metaphor/analogy not only undermines his argument for a kind of generic “manga-ness” but also runs the risk of simply inventing the very thing he wishes to find out of a selective reading of certain empirical observations.
Why the What Question Still Matters
We have in Cohn’s essay a perfect encapsulation of why how one defines manga has lasting importance and why it is important to leave the what question open, as something to which we must at least tacitly respond each time we use this term, so as not to find a definitive answer but to think through useful and meaningful possibilities. Interestingly, the sense of manga is something Cohn addresses at the very beginning of his essay.
“[T]he word manga has come to have two meanings outside Japan. Some use it to designate Japanese “comics,” the sociocultural objects, and often the industry and community surrounding them. However, others use “manga” to name this visual language itself — loosely conceived of as an “aesthetic style”… Since the conflation of these ideas can be confusing and inappropriate in this chapter the term manga will be used in the first sense–to designate a sociocultural artifact–while referring to the system of graphic expression as Japanese Visual Language.” (p. 188)
This seems rather cut and dry, except Cohn keeps finding himself coming back to the what question in order to make necessary qualifications.
“To the extent that representations from genres can be grouped into recognizable ‘styles,’ each constitutes a type of ‘dialect’ or accent of [Japanese Visual Language], since their patterns reflect varying degrees of similarity among a group of authors. It should be no surprise that these divisions fall into separate genres.” (p. 190) Bear in mind that what Cohn refers to here as “genres” really ought to be demographics (i.e. shōnen, shōjo, josei, etc.), since that is the sense in which he is using it, and here we encounter our first problem, for what Cohn is talking about is not the full range of styles that fit within those demographic categories but stereotypes. Are Anno Moyoko’s seinen manga fundamentally different, stylistically speaking, from her josei manga? Does a shōnen gag manga have anything meaningfully in common with a shōjo gag manga? Which conventional system is more salient in determining the individual title’s structure, the gags or the demographics? I am not trying to say that the stereotypes of, say, shōjo manga are complete figments–this month’s Bessatsu Margeret titles all look frighteningly similar–but rather that identifying this fact is not interesting precisely because it is obvious. The problem is that the stereotypes-as-conventions overwhelm any easily observable breaks with the expected and push them to the periphery.
It’s worth noting that most of Cohn’s examples of the phenomena he wishes to identify are drawn himself. I cannot say for certain why this was done (reprint rights?), but it has the rather convenient effect, intended or not, of allowing him to literally invent his visual evidence ad hoc. I’ll leave aside for the moment the fact that many of these supposedly Japanese specific emblems actually have their historical origins in American animation to admit to the one example taken from an actual published comic, Koike and Kojima’s Lone Wolf and Cub (Kozure ōkami).
Following from his breakdown of two pages of the 20th volume of Lone Wolf and Cub, on p. 197 Cohn references an earlier study that “examined the panels in various American comics and Japanese manga [sic] to see what types of panels they were using.” You can probably already imagine what the results were: there is a statistically significant difference between American and Japanese comics according to an arbitrary coding scheme. I, being a stickler for details concerning methods, tried to track down this study, which incidentally has gone poof due to a massive redesign of Cohn’s website. I did manage to find this paper, which seems to have the same methods and results, so I’ll assume for the time being that it and the 2005 study Cohn cites are sufficiently similar.
A few peculiarities in this study stand out. 1) Despite the claim in the stimuli section that all the American and Japanese comics were “mainstream titles from a variety of genres,” all the titles listed in the appendix are either shōnen or seinen (in fact, mostly seinen) with an emphasis on action/martial arts. An extremely limited range not just of manga but even of manga for men/boys is being used as representative of all Japanese comics. Even more worrying than that is 2) the statement in the procedure section that “[b]ecause of peculiarities of some authors, certain panels were excluded from the coding because of their redundancy. For example, in Frank Miller’s The Dark Knight Returns, a rhetorical device of a newscast is used throughout the narrative, often taking up several panels per page. Since these panels are thematically restricted in their spatial properties to a lateral viewpoint of a newscaster’s bust, their redundancy would have biased the analysis.” I can’t even begin to explain how Looney Tunes I find this to be: actual observable phenomena on a comic book page were removed simply because they failed to conform to the methodology. Instead of concluding that there might be a flaw in the methods themselves, the researchers felt it more appropriate to simply alter the texts rather than account for the significance of phenomena that challenge their conceptual frame. As far as I am concerned, all of the talk of p values and standard deviations that follows is rendered moot, because the unchallenged assumptions that frame the experiments are themselves already incredibly biasing. No mountain of statistics can conceal that fact.
When Heidegger introduces his distinction between present-at-hand (i.e. world as objective reality distinct from self) and ready-to-hand (i.e. world as totality of interrelations) in section 15 of Being and Time, he notes how these are not simply unrelated worldviews but that presence-at-hand has its basis in the breakdown of readiness-to-hand, meaning (as is made clear[er] in section 16) things become conspicuous as “objects” of theoretical speculation when their interrelatedness to all other entities in the world breaks down. In order to render manga more amenable to scientific observation, these interrelations have been broken and obscured prior to Cohn’s analysis. The task, then, is to see if we might restore them.
Language, Center and Periphery
Of all the formulations of language that have popped up over the years, I have grown increasingly fond of the adage Max Weinreich supposedly heard from an audience member at one of his lectures: a language is a dialect with an army and a navy. In one simple quip, Weinreich reminds his reader that any attempt to think of distinctions between languages purely in structural (i.e. formal) terms will always come up against the fact that language[s] are subject to and participate in social conditions that language itself cannot entirely account for. As above, there’s nothing wrong with provisionally identifying something as Spanish or English without accounting for how they don’t quite map onto any particular culture or nation, but the moment you try to make claims for some essential difference between them, you need to bring those social conditions back into your conceptual frame and, for example, account for things like the large numbers of people in the US who grow up speaking English and Spanish and whose cultural background is largely divorced from the European countries that give those languages their designations.
I mention this, because Cohn uses a rather careless metaphor to explain what he sees as the relationship between some core/standard Japanese Visual Language and its “dialects” (i.e. the stylistic conventions of what he refers to as “genres”).
“Speakers of the Tokyo and Kyoto vernaculars remain intelligible enough that the broader patterns are labeled as ‘Japanese,’ while the differences are thought of only as unique aspects of ‘dialects.’ Like the way Tokyo-ben is considered the ‘standard’ dialect of spoken Japanese, this stereotypical ‘manga style’ can be considered the ‘standard’ dialect of Japanese Visual Language since it uses a common model for drawing people shared by a broad range of ‘visual speakers.’” (p. 189)
Something I noted last week is how over the course of working on this blog I’ve come to realize that formalist theories of comics/manga are profoundly sexist and that these sexist assumptions are so pervasive as to infect the work of even those who have tried to take, say, shōjo manga seriously. In formalist Japanese manga studies discourse (e.g. Natsume, Takeuchi, or Yomota), the basic features of manga in toto are first identified in comics for men/boys and only thereafter are the stylistic conventions of many shōjo/josei manga seen as variants thereof. An honest question: why isn’t it the other way around? Why aren’t shōnen/seinen the variants?
Cohn’s claim that the patois of the Kanto plain is “standard” speaks directly to this problem in the way it asserts such a status without even touching upon how that vernacular came to be “standard” in the first place, when historically it was the West of Japan that was its cultural center. As the center of Japanese government (as well as the Imperial household) over time shifted to the East/Edo, so too did all of its major, centralized cultural institutions–in the modern day education especially. It is through the relative uniformity of educational curricula that Kanto-ben is enforced as the “standard” dialect, because the people who write the curricula hail from Tokyo and its environs. Tokyo-ben-as-standard is a function of that region’s cultural and political hegemony in Japan and not of its supposed status as a base from which, one supposes, even incommensurable “dialects” such as Okinawan simply deviate.
This is why above I pointed to the lunacy of using a limited range of comics for men/boys as representative of all manga. It reinforces the invisibilized sexism/parochialism in seemingly “objective” formalist approaches to comic structure, perhaps unintentionally, but it does so precisely because in those “objective” approaches the assumptions about what manga is or might be embedded in the methodology never get exposed and interrogated.
In two weeks: Is the range of manga texts available in English indicative of Orientalism?
contact me: email@example.com