Levels of Abstraction

In light of this week’s experiment with iconography and reading Neil Cohn’s list of “Conventional Representation in VL” I was thinking about levels of abstraction in comics. Conventional iconographic elements (such as hearts, arrows, skulls, etc.) are really just an extreme level of representational abstraction.

Comics are no stranger to different levels of abstraction across artists–even in early strips Krazy Kat is certainly more abstract than Little Nemo (I’m reading collections of both right now, so they are on the mind)–and McCloud makes a nice chart in Understanding Comics (52-3) detailing the myriad of styles from realist to stick figure. But in almost all cases these styles are consistent in a work.

If one accepts the idea that words are the highest level of abstracted image, then most comics exist on two levels of abstraction: the words and the pictures. These two levels’ distance from each other (on an abstraction level) will vary greatly, but not within the same work. Most artists will not vary their styles with the same work, though there are exceptions.

Most common would be, of course, the iconographic symbols used in comics. These exist on a different level of abstraction than both words and most comics images, closer to the former than the latter. We don’t see these things as particularly jarring, perhaps because we read them much like letters and words. They have an accepted, often learned, meaning (like words) and thus are read in a different way than pictures.

Two devices often used in manga also show a mixing of abstraction levels: 1) the “superdeformed” look and 2) what McCloud calls “masking” (I have no idea who coined the term, but I think that is where I first heard it). Superdeformed refers to a style wherein (mostly) characters are drawn in squat, big headed, cartoony style. In manga this style is often inserted into a work that is otherwise in a more realistic style, usually for the purposes of comedy. It is visually jarring, but also an effective means of switching the narrative’s emotional register. “Masking” is the term for the use of abstracted cartoony characters existing in realistic backgrounds. The former is a juxtaposition of styles, while the latter is an overlapping (at least in the commonly used way). These are two quite different uses. In one there is a stylistic inconsistency, abstraction levels change even while representing the same character. In the other consistency is maintained.

Personally I think the former is more interesting. Visual consistency in comics is a barrier that is not often broken (oh, I’m generalizing, I’d love to hear examples that prove me wrong), perhaps dating back to serialization of strips, where artists needed to maintain consistency so the audience could easily identify the strip and follow along with it. Conventional narrative art (I include written art here) is tied to a sense of consistency and smoothness. It is something that comics, more easily than many arts, can break down through the use of visual style.

What would happen if a comic changed levels of abstraction at will, not randomly, but rather as an element of narrative and expression? It’s something I’ll have to look for in my reading. Similar to a experimental prose writers move away from the confines of realistic (or even modernist) convention, I think comic artists can also explore the further reaches of experimentation. So much of the now touted graphic narratives are still very much narratively living in a conventional past… that, though, is probably a rant for another day.

(Comments are, of course and as usual, most welcome. Recent conversations with others have certainly aided my thoughts and opinions on the “pictureless comic” concept.)