A Currency Model of Genre Classification

A Currency Model of Genre Classification

When inquiring over the classification of species, there are solid criteria and solid facts to produce the answer. Is a whale a mammal or a fish? Like fish, the whale lives in water. However, it gives birth like a mammal. The latter fact has led biologists to study the evolution of the whale. They found that it genetically belongs to the mammal family. That's a hard fact. Film genres can't be classified so easily. Or they can be, but there's no reason for anyone to agree with you; there's no fact of the matter. There are obviously no genetic strands linking films back to common ancestors. While there are chains of influences, each film is sui generis to some extent. So how do we go about classifying genres?

Speaking as generally as possible, there are top-down approaches and bottom-up approaches. Top-down approaches are those that begin from general principles, then apply these to concrete cases. These principles will be broad criteria with which the members of the genre can be easily picked out. It would be possible, for instance, to begin with the criterion that all horror films must be designed to produce the emotion of horror within its audience. On this account, Psycho can only be considered a horror film if Hitchcock intended to horrify the audience. If, rather, Hitchcock intended to provide thrills, then it would be a thriller instead. That works for many movies. But surely propaganda films and other genres try to create the emotion of horror as well. One might reply, "But the sole purpose of a propaganda film is not to horrify." This then excludes all films that have diluted purpose, mainly horror comedies like Young Frankenstein and higher-brow horror films like Dreyer's Vampyr, Valerie and Her Week of Wonders, Videodrome, or even Dawn of the Dead, what with its overt social commentary. With Cronenberg it's not even clear that the emotion of horror is his intention at all so much as it is a by-product of the ideas he's presenting. These would be the challenges for a top-down approach: to either refine the principles so that they conform to intuition or to refine intuition to conform to the principles, until one arrives at a point where the genre categories would capture just what one desires, no more and no less: the state Rawls called 'reflective equilibrium.'

The bottom-up approach, on the other hand, rather resembles the scientific approach outlined in the first paragraph. One does not begin from general principles. Instead, based on the actual films, one derives certain features, a set of relevant 'family resemblances', with which to identify the groups to which each film belongs. As with the scientific approach there may also be an historical aspect. This would necessarily involve bringing other media into the investigation, since the horror novel predates cinema. For sake of discussion, we'll only take account of cinema. Returning to horror, the bottom-up approach would look for relevant family resemblances that unite horror films. Let us take as our set of family resemblances the following: tendency to inspire fear and revulsion, characters who experience terror, death, and monsters. A set of family resemblances of this sort can then be taken as either sufficient or necessary for counting a film a horror film. Thus one's criteria or principles are derived from the concrete cases. To say these are necessary is to declare that any film not sharing full family resemblance is not properly of the genre; it merely has characteristics similar to this family. To say these features are sufficient to make a film a horror film but not necessary means there could be films that have some of the features and not others. This would deal with films that are only half within the genre, like Young Frankenstein. Of course, one will begin to run into problems then. If Young Frankenstein is partially horror, then why not the recent Monsters vs. Aliens? It has monsters, obviously. The challenge here is to tweak the set of family resemblances so that there are no obvious howlers. It would be silly to end up calling Titanic a horror film because of the death and the suspense experienced as the ship began to sink; the family resemblances would require tweaking for sure.

The approach I favour is, to keep in the same schema, bottom-down (I guess). I don't believe it's necessary to refer to general principles at all. Let's begin with an example outside of cinema: currency. Money is imaginary: it's pure abstraction. The currency we use are just pieces of paper or coins that indicate an amount of money. These pieces of paper have currency because just about everyone accepts them as valuable. You don't have to worry about a shop saying, "I'm sorry, but this is just a piece of paper. I disagree with you that it's worth anything." There are also cases of a currency of ideas. Take the film Ishtar. More people think Ishtar is a bad film than have ever seen Ishtar. Through the tireless efforts of the critics, Ishtar simply gained currency as a bad movie. Currency is gained through repetition and repetition of agreement to the point that there is no fear of having to defend one's position. You need never worry about defending your view that a five-dollar bill is worth something. It's not an opinion of yours, it's a commonly held view. (Granted in the case of currency there's also the authority of the State.) Similarly, you needn't worry about having to defend the position that Ishtar is a bad film. It has currency as a bad film, so if you express the view that it is a bad film most people who have heard of Ishtar will assent. It's a safe commonly-held opinion on which to rely. Comedians use currency all the time. George W. Bush has currency as stupid, Hitler has currency as evil, Ronnie Corbett has currency as short. This makes these people easy references for jokes on stupidity, evil, or shortness. Comedians require this communal assent: the audience must be on the comedian's side.

I think we have a decent notion of what currency is now. What relevance does it have to genre classification? The approach I favour toward genre classification is that whatever films have gained some currency as one genre or another is sufficient. This approach privileges practice over theory. Where the top-down and bottom-up approaches had to rely on principles of some normative character, my approach is purely descriptive. This is like a dictionary. Good dictionaries contain an "ain't" entry because it's a word people use; it doesn't matter if it's a 'proper' word or not, because it's not the place of dictionaries to tell people how to speak, but to tell people how people generally do speak. Similarly, I think of genre classifications not as telling people how to categorize movies, but as telling people just how people do categorize movies. The importance of currency is filtering out idiosyncrasies. Just as with a dictionary, not just any word anybody says gains an entry. It is only words that fall into general use that gain an entry.

One can only discover what is in currency regarding genre classifications through discourse. There is no controversy discussing The Exorcist as a horror film. You can call it such without having to defend this view. It has currency as a horror film. If horror fans and scholars of the genre can talk more or less uncontroversially about Silence of the Lambs, then it clearly has some currency as a horror film. Similarly, if they rarely or never speak of Blue Velvet in any way other than peripheral, then it seems to lack the currency. Mulholland Dr. is a film that, while sometimes spoken of as a horror film, certainly lacks currency in that it is an opinion one will usually have to defend.

This approach puts considerable faith in common intuition. However, was that not always the case? In the top-down approach, I ended by noting the theorist will have to bring intuitions and principles into equilibrium. Because what good is a theory that classifies a lot of films as horror films when nobody else is willing to accept this view, let alone make use of it? Similarly, the bottom-up approach is based on finding resemblances in films we already intuitively consider as horror. So why not circumvent the notion of principles altogether and rely on intuitions? Everyone will have their different reasons for classifying the genres of films and usually they themselves will not be explicitly relying on intuition. My approach allows all of those reasons to contribute to defining the genre. If someone can successfully defend their position that Mulholland Dr. is partially a horror film, each repetition will yield stronger currency. It's up to individual tastes to decide what has sufficient currency in discourse; to decide, that is, when disagreement with the classification proposed becomes more idiosyncratic than the classification itself.

The guiding notion behind this approach is that genre itself exists for the sake of discourse. It is a pragmatic notion for the sake of discussion in all its forms: criticism, advertising, etc.. A theory of genre that yields idiosyncratic results is without value for discussion. Discussion requires some points of agreement. As Wittgenstein rightly pointed out, one cannot have a personal language. Language is for communication and communication implies, even at an etymological level, more than one interlocutor. Genre ideally is one of those common points of agreement. While some discussions on the boundaries of genre and the location of some films within one tradition or another are fruitful indeed and even necessary, it seems to me best to rely on a currency model of genre classification, which ultimately means relying on the intuition and basic understanding of cinematic traditions the full film-discussing community possesses as a group. It's a democratic view in the tradition of Jean-Jacques Rousseau: most appropriate to a democratic medium like cinema.