On algorithmic catachresis
Dear friends of Jan Bot,
Today I’m going to talk about Catachresis, a major inspiration for the development of Jan Bot’s movies.
In its original sense, catachresis is a figure of speech used for centuries by poets to convey moods or even describe objects for which there is no word or expression. The term comes from ancient Greek and it means “abuse.” Catachresis deliberately abuses the canon of language, forcing unusual encounters between words or phrases from unrelated domains, giving form to new unique expressions. Take as an example the phrase “to take arms against a sea of troubles,” or, “grabbing a bull by the horns of dilemma,” both crafted by one of the best known catachresis composers in history: William Shakespeare.
The new expressions that come from a catachresis are not intended to be taken in a literal sense. Understanding them requires a knowledge of context, as well as of idiomatic uses of words. Thus the meaning shaped by a catachresis takes place in the field of semantics.
One essential question we explore through Jan Bot keeps a close connection to the notion of catachresis. Given that so-called artificial intelligence nowadays is capable to understand human language and its semantics, could we use AI to let machines create meaningful associations between unrelated words in the way poets do with catachresis?
And given that computers end up processing all their inputs in the form of seamless and plain data, couldn’t we push this question a bit further and explore the ways in which these algorithmic catachreses unfold in the realm of moving images?
Algorithmic Catachreses
Using some of the trendiest AI tools available for the translation of images into words and words into semantic nodes, we have programmed Jan Bot to create meaningful connections between two completely unrelated items: a vast collection of old and unidentified film fragments, and trending topics from today.
Such an impossible task is in itself a formula for catachresis, as it forces our bot to carve out meaning where there is intrinsically no meaning. If a human would be forced to complete this task, he or she would need a considerable amount of imagination to make the connections. But machines do not think on these terms. They convert all the information they perceive into homogeneous sets of data that can be treated equally using the algorithms they have at their disposal. This does not mean that the connections they make will be intuitively comprehensible to humans. In fact, their quirkiness may often give the impression of a random edit, even though that is not the case. Computers’ decisions are strictly guided by logical reasoning.
Take for example the Californian music and arts Festival Coachella that was trending on April 14. Bits & Pieces, the collection of old orphan films that Jan Bot reuses to create its films, certainly doesn’t have images of Coachella. However, by considering “Coachella” as a semantic node instead of just one word that literally means no more than a name, Jan Bot is capable of identifying a series of shots that are somehow related to the location and the festival. These connections range from abstract concepts such as “pop music,” “hot,” “cool,” and “climate change,” to descriptive nouns such as “concert,” “festival,” and “hot springs.” Based on these semantic resemblances, Jan Bot connects shots and intertitles to create a new audiovisual expression.
If we try to match the operation just described to a figure of speech, we can clearly see that here the relation between trending news and old footage does not follow the logic of more traditional figures of speech, such as comparisons or metaphors. Since there is no straight parallel between Coachella and the footage available in Bits & Pieces, Jan Bot forces the meaning of the former to match the imagery of the latter. This collision enabled by the heartless performance of intelligent algorithms gives form to a unique result, a brand new expression. We call it an algorithmic catachresis.
