Tuesday, January 21, 2014

How the visual language of comics could have its roots in the ice age

Psychologist and comics obsessive Neil Cohn believes cartoons have a sophisticated language all their own and a heritage that goes back to cave art. Here he breaks it down

Neil Cohn's love of comic strips began in his family's attic. In one of his earliest memories, he recalls his dad climbing the stairs and pulling down a box of 1960s Batman and Superman books that he had stashed away from his own childhood. To Cohn's four-year-old self, it was as if they'd been imported from a strange and foreign place. "They had this kind of mystery to them," he says. Instantly he was hooked.

It was not long before he became a compulsive comic artist himself; in his teens he even started his own mail-order comic company. As he set about his creations, he would often wonder how the brain makes the huge cognitive leap to piece together a story from the fragmentary, stylised pictures on his drawing board.

Now a psychologist at the University of California, San Diego, Neil Cohn is finally getting the chance to answer that question, as he carefully dismantles comic strips such as Peanuts. His theory, presented in The Visual Language of Comics (Bloomsbury) next month, is provocative. At a neural level, he says, the pictures of comic strips are processed as another form of language, with their own vocabulary, grammar and syntax.

"Human beings only have three ways to convey our thoughts," he explains. "We create sounds using our mouths; we can move our bodies with hands and faces; and we can draw things… My idea is that whenever these meaning-making channels get structured in a coherent sequence, then you end up with a type of language." If he is right, the hidden logic of cartoon panels could provide new vistas on art, language and creative development.

Cohn's theory builds on a growing acceptance that the brain's language toolkit is a kind of Swiss army knife for many different kinds of expression, such as music or dance. In some ways the ties with art should be stronger, however – since, unlike music, pictures encode a definite meaning. "Drawing has always been about communication – to express an idea in your head to other people," says Cohn.

The drive to tell stories with pictures certainly has deep roots. Stone age paintings in places such as the Chauvet cave in France seem to show scenes of galloping horses and pouncing lions, using techniques that would be familiar to graphic artists today. More advanced picture narratives appeared in works such as the Bayeux tapestry and Paupers' Bibles. In some indigenous Australian cultures, sand drawings are used as a regular part of discourse; in fact, drawing is so entwined with speech in the language of these cultures that you can't be considered fluent if you don't know the appropriate pictures.

Before Cohn began his research, however, few serious analyses of comic strips existed. One writer who dissected the way strips are constructed was Scott McCloud in in his landmark book Understanding Comics: The Invisible Art, published in 1993. "That book really got me thinking about it, as a teenager," says Cohn. His interest would remain a hobby, however, and after school he embarked on an undergraduate course in Asian studies at the University of California, Berkeley. As part of his degree, he decided to take a linguistics course. Suddenly his musings took a whole new shape. "I noticed that all these things that happen in language were the same as the things I understood in comics," he says.

Comic strips do, after all, have the basic structure of language, with a hierarchy of elements that can be combined with infinite variety. The building blocks of this hierarchy are a "visual vocabulary" of signs and symbols. That might include speech bubbles, motion lines, or stars to represent the throw of a punch. Even the characters' anatomy is highly stylised: cartoon hands, eyes and noses can look almost identical from strip to strip, even when these are by different artists. "If you look at the bits and pieces, they're all systematic," Cohn says.

At the next level comes the cartoon pane, which combines elements of the vocabulary to build new meanings. It's difficult to find an exact analogy to the English language here; according to Cohn, the cartoon strip pane helps to divide our attention into units, like a word, but it is perhaps closer to the words in agglutinative languages such as Turkish or Inuit, in which stems and suffixes are strung together in a single term to give a more complex meaning. (In one Inuit language, for instance, a single word, angyaghllangyugtuqlu, encapsulates the sentence "also, he wants a bigger boat"). Similarly, a single panel of Peanuts can represent "Charlie Brown gets ready to swing his baseball bat" through its signs.

Governing the hierarchy is a set of rules that Cohn dubs "narrative grammar". Just as spoken or written grammars contain different types of word – nouns, verbs, adjectives and so on – the narrative grammar includes different types of panel. Among these are: establisher panels, which set up a scene; initial panels, which create a tension; peaks, which show the climax; and release panels, which undo the tension. Each has its own characteristics, and, like the words in a sentence, they have to follow a certain order. Crucially, Cohn argues that these panes can also be grouped into separate clauses that are embedded in larger structures – so a whole string of panels might represent an "establisher" clause in another sequence. That leads to recursion – the property that allows us to say: "He thought that he said that she said…" – which is thought to be one of the defining characteristics of language.

Despite having no background in linguistics or psychology apart from his undergraduate course, Cohn realised that he needed to add flesh to the bones of his theory with experimental evidence. "The only way to make the comparison between visual language and spoken language is to see what the brain is doing," he says.

That would be easier said than done. Cohn had to wait five years for a place on a postgraduate course at Tufts University, Massachusetts, under the linguist Ray Jackendoff, who is known for groundbreaking work comparing music to language. With additional supervision from the psychologists Phillip Holcomb and Gina Kuperberg, he set about testing his ideas.

To do so, he turned to some classic psycholinguistic experiments from the 1970s and 80s that had attempted to pick apart the way the brain processes grammar. Measuring how quickly subjects are to pick out certain words in a sentence, researchers found that grammatical sentences are quicker to process, even if they are nonsensical, such as "green ideas sleep furiously", than something ungrammatical such as "Picnic strike ideas quiet launched".

If comic books are built from a visual language with its own grammar, Cohn suspected that the same must happen when we look at the pictures of Peanuts – so he set about picking apart Charlie Brown's antics. In some cases he would take pictures from different strips but arrange them so the establishers, initials and peaks all followed the right order. In others, the pictures of the panes were scrambled completely. Recording his volunteers' reaction times as they answered questions on the strips, he found that the grammatical sentences did seem to be processed more quickly – just as you would expect if the brain were using underlying grammatical rules to try to make sense of the confusion.

Further experiments played more subtly with Snoopy's grammar. For instance, in one study Cohn placed blank panels in the middle of a "clause", interrupting the flowing sequence. "I compare it to moving a comma to the wrong place in a sentence," he says. Measuring his subjects' brain activity using electrodes placed on the scalp, he found the same pulse of activity after 600 microseconds – known as a P600 signal – that seems to come as the brain tries to grapple with violations of written or spoken grammar.

Clearly, there is still a long way to go before comics enter neuroscience textbooks, but Cohn's work is already attracting serious interest from comic artists and cognitive scientists alike. He regularly gives talks at Comic-Con events, and this year he won a $10,000 award from the Cognitive Science Society in the US, based on the strength and originality of his doctoral thesis.

Cohn is passionate about the way his theory could influence art education. He points out that children naturally absorb language through imitation and mimicry. But that's not how we are taught art, where individuality is championed. "Our culture is suppressing the biological desires for imitation." The result is that we never learn a fluent visual vocabulary, except a few simple symbols, such as stick men.

A better approach, he says, would be to tap into children's innate language instinct by actively encouraging them to mimic others' drawing. He speaks from experience: from the age of eight, he obsessively copied figures from Disney until he was fluent in every aspect of Mickey Mouse's world. "I was obsessed," he says. "By third grade I was teaching my class to draw them."

For further evidence you need only look to Japan, where comics are more central to mainstream culture. According to studies by Brent Wilson at Penn State University, nearly all six-year-olds surveyed were already able to draw complex narratives to a high level, whereas even by the time they had reached 12, less than half the children he surveyed in western countries were able to do the same. Clearly, that's a different skill to producing the Mona Lisa – but it's still an enticing idea that comics might kickstart artistic development.

For the future, Cohn hopes to investigate the different visual languages that have developed across the world, by building large dictionaries of the elements and their grammars. He has already found that manga comics tend to focus less on a wider scene and more on the individual characters than American comics. Cohn is also interested in studying indigenous Australian sand drawings in more detail. His early analyses suggest that this language, which evolved over perhaps thousands of years, does seem to follow some of the same rules and grammars as western comics. Ultimately, such studies may help to find universals that govern all visual languages. Cohn also wants to apply a historical linguistics approach to comic books – examining how symbols such as motion lines first arose, for instance.

Whatever happens, it's clear that no matter how long he studies them, the fascination he first felt as a child when poring over his dad's old comics will never be quenched. "Each study raises so many new questions about the brain," he says.

Robson, David. 2014. “How the visual language of comics could have its roots in the ice age”. The Guardian. Posted: November 24, 2013. Available online: http://www.theguardian.com/science/2013/nov/24/comics-language-neil-cohn-cartoons-grammar

No comments: