|Publisher:||University of Alabama Press|
|Edition description:||1st Edition|
|Product dimensions:||6.30(w) x 9.10(h) x 1.00(d)|
About the Author
Read an Excerpt
Animal, Vegetable, Digital
Experiments in New Media Aesthetics and Environmental Poetics
By Elizabeth Swanstrom
The University of Alabama PressCopyright © 2016 University of Alabama Press
All rights reserved.
Creating and Erasing Worlds of Signs
Philosophy is written in this grand book, the universe, which stands continually open to our gaze. But the book cannot be understood unless one first learns to comprehend the language and read the letters in which it is composed.
— Galileo, The Assayer (1623)
Bits have arrived, they're not going away, and we must learn how to live with them.
— Mark Hurst, Bit Literacy (2007)
What Is Code?
All things that can be read, watched, and listened to online are readable, watchable, and audible because of computer code. Code manages the fuel efficiency of the cars we drive and the energy consumption of our home appliances; it animates and controls the dendrites of network communication as we know it. Yet as crucial as computer code is, we seldom see it. Unlike Neo in the Matrix, code remains hidden to most of us, obscured by its output or masked by its interface. In spite of this invisibility (which is not an inevitable precondition of its existence but is still something more than mere convention), code is a real and material language with the purpose of communicating commands to microprocessors in a computer. A machine-readable program, when compiled or interpreted by a computer processor, executes a series of commands and is called "operational" because it tasks the computer with performing some kind of operation (filing, counting, sorting, calculating, creating, deleting, mapping, etc.). Recently, however, instances of code have emerged from the obscurity of their source files and made their appearance on screens both large and small. When this happens, code becomes "expressive" and functions as an aesthetic trope: visible, readable, sensible. In certain special cases, it serves to emphasize our relationship with the natural world. This expressive type of code has a human audience and allows us to consider how computer code is experienced, that is, the manner in which it functions cognitively, affectively, and phenomenologically.
In this chapter I look at examples of digital art that use computer code in this peculiar way: William Gibson's Agrippa (A Book of the Dead) and Char Davies' immersive art installation Osmose. In each, code instructs a computer, yet it also asserts its presence to the reader. It engages the user by manifesting as a part of the artwork on the screen (as in Osmose) or through its very act of self-erasure (as in Agrippa). The works offer complex ecologies composed of animal, vegetable, and computational agents. In both works, computer code is not merely an invisible set of instructions that control the computation. It is foregrounded and sensible; it both organizes and contributes to the thickly layered topographies of each work and world. Because code in such instances helps foster an ecological model that is rooted in materialism, it will be useful to have a basic understanding of what computer code is, how it operates, and how I am using the term in this context. The following, then, is an attempt to demystify code.
Blockhead and Bole
This should be easier to do than it is. The dictionary definition of "code" isn't all that elusive. A code is "a system of words, letters, figures, or other symbols used to represent others" ("code"). This broad definition works for computer code, a system of symbols that represents other symbols and even whole systems of symbols. But computer code is also something else, something physical. Even its etymology speaks to its materiality: the term "code" emerges from the old French word code, or "system of laws, law-book," which itself emerged from the Latin term codex or caudex, which signified a "tree trunk, wooden tablet or book" as well as "a wooden stump to which petty criminals were tied in ancient Rome" ("codex"). It was also the suggestive word for a "dolt" or a "blockhead." Hence the contemporary term "code" marks an intersection between mineral (the book, the bollard), vegetable (the tree), animal (the blockhead), and digital (the code) realms. The etymologies of the related terms "calculus" and "digital" are similarly suggestive: a calculus in Latin is a pebble used for accounting and, by extension, the name of the counting table at the time of financial reckoning; digitus is the Latin term for the human fingers used for counting; and a certain type of tabella (tablet) is one of the Latin names of the surface upon which very specific, very procedural curses were inscribed.
Several extant critical investigations have grappled with code's puzzling status in ways that simultaneously clarify and frustrate. In his investigation of information aesthetics and creative labor in the Laws of Cool, Alan Liu discusses "transcendental data"; in "There Is No Software," Friedrich Kittler laments the "de-differentiated" nature of code; in How We Became Posthuman, N. Katherine Hayles describes the transmission of data as a process of "flickering signifiers." These frameworks discuss information technology's material nature in great detail; they also demonstrate how high the stakes have been raised for humanist and aesthetic inquiry in the age of information. At the same time, however, the vocabulary that each enlists has paradoxically highlighted code's inaccessibility (transcendence), erasure (dedifferentiation), and instability (flickering) instead of its material properties.
William Gibson, the cyberpunk pioneer who made personal computing, hacking, and coding such popular subjects in the 1980s, once related an anecdote that speaks to this problem. In an author's note to his first trilogy, set in the "sprawl" of the Eastern Seaboard of the United States, he describes his initial conception of how a computer stores information in the following terms: "I'd assumed the data was just sort of, well, held. In a glittering mesh of silicon. Or something" (qtd. in Bukatman 87). Gibson's misconception gives voice to, echoes, and is at least partially responsible for spawning our own initial misconceptions about the material nature of code, a misconception he promotes in his early fiction. Code in Neuromancer is configured as lattices of "intrusion countermeasure electronics," shortened to "I.C.E," abstract, beautiful, and divorced from materiality. But if computer code is not something suspended in a silent, crystalline lattice of silicon — a miniaturized ice cave of Krypton — then what is it? We don't have a common understanding of what it is, how it works, or how to locate it.
According to IIEEE Software, "Whereas hardware is physical, software is intangible. ... It cannot be touched" ("software"). Given that intangible forces are often equated with supernatural ones, it is not too surprising that code in this sense (i.e., immaterial and ethereal) has evolved as an antihumanist — and antienvironmental — force. Dystopian science fiction has made a villain out of code since the dawn of personal computing. In Gibson's Neuromancer (1984), hieroglyphic computer code imprisons the main character on a remote strand of beach, whose silver-hued sands stand in contrast to the way the novel depicts the polluted beaches of the real world's Eastern Seaboard; and "black I.C.E." is the name for code that kills. Code is the means of corporate control and provides a nearly insurmountable blockade against social mobility. Fifteen years later, in the phenomenally popular Matrix film franchise of the late 1990s, green and rain-like code underwrites the structure the matrix, an oppressive computer simulation that hides an environmentally devastated earth from its enslaved human population. Code is what conquers humanity, and every instance of its visibility is a reminder of this subjugation. Code also devalues individuality by objectifying all living subjects.(We might recall the character Cipher pointing to lines of code and saying, "All I see is blonde, brunette, redhead," without referring to any specific person.) In a dystopian science fiction film from 2011, the main character is a hapless airplane pilot who, in spite of his vehement protests, must live the last eight minutes of his life over and over until he can uncover the solution to a terrorist plot on a city-bound train. The name of the film, Source Code, is also the name of the computer program within the film that makes possible this endless loop of dying and suffering. Such works have prepared a negative poetics, have primed us to think of code as abstract, lethal, and controlled by the elite.
But even as it is guilty of some of the dystopian aspects that are expressed in contemporary works such as Neuromancer and the Matrix — particularly in terms of its difficulty and abstraction — code is both local and locatable. But where? We might have a hunch that computer code is "out there" somewhere, remotely, in the cloud, perhaps, or "in there," on or in one's computer, locally, but neither assumption is quite right. Cloud computing is the name of a popular informational product model, but it also provides a suggestive metaphor that fits too easily within notions of transcendence; and although one saves and stores data locally on one's own computer (in a file in a folder in a platter on a hard drive) the notion of it as a self-contained kernel of information is not accurate either. Computer code, at its base, controls electric voltage to store, retrieve, calculate, and manipulate information. This part of its syntax — electric voltage and its absence — gets things done. But in spite of dystopian narratives about the sinister agency of code, the act and execution of coding is thoroughly tied up with human actions. Accordingly, it will be useful to demonstrate how different codes might come to be involved in the execution of a simple command. The best way to do this is not to "see" it — to do so would be difficult, since running code is, at its heart, a series of electrical events — but to see "coding" in practice.
As a way to foreshadow their interplay within Agrippa and Osmose, consider the different valances of code in the following scenarios:
1) A computer programmer is in charge of writing a program to control and monitor CO2 emissions of a new line of hybrid vehicles. The coding language she uses is "Misra C," a powerful, lowlevel language customized for the auto industry. Her final product must resolve the following questions: How will the car's computer receive information from its fueling system? How will it display that information so the driver can make sense of it? How will the system troubleshoot when parts are defective, and how will it communicate its efforts to do so to the driver? Between the program she authors and its execution in the car, several other systems of code are at work. She will use the language of Misra C to write the software; Misra C will compile into machine language; machine language will translate into binary code; binary, in turn, will translate into the presence or absence of voltage, which then instantiates a command (acceleration, use of fuel, etc.). All of this happens sequentially in lines of code but instantly in time; my narration highlights how one system of symbols represents another system of symbols every step of the way. Code refers at once to a highly abstract language (Misra C), to the software that the programmer eventually composes in it, the intermediary machine language, and binary. Coding refers to the work of writing these commands into a set of machine-readable instructions. One is a language, the other a practice.
2) A lepidopterist puzzles over the distinction between the giant-skipper (Megathymini) and other members of the skipper (hesperiidae) family. Although the giant-skipper has enough phenotypical (visibly expressed) traits to suggest that it constitutes an entire skipper subfamily rather than a species, our entomologist isn't so sure. How will he test his theory? Variously and thoroughly: 1. He consults his list of known states and behaviors of butterflies (there are eighty), and checks each off for each butterfly in the form of + (present), - (absent), or ? (unknown); 2. He combines and collates that information into binary code, which condenses the information and allows him to eliminate improbable branches more swiftly than the +-? grid allows; 3. He takes images of each butterfly type, measures the color values of each, and conducts a "pixel-by-pixel image color subtraction" from the images to identify how distant the patterns and colors are from type to type (Grishin). 4. Finally, he consults the GenBank section of the Nucleotide database at the National Center for Biotechnology Information research hub and uses a program called MUSCLE (Μltiple Sequence Comparison by LogExpectation) to collate and contrast the skippers' genotypes. By comparing the DNA sequences of four genes of the giant-skipper to those of the closely related grass-skipper, leonard-skipper, and awl, he determines that the giant-skipper is not, in fact, sufficiently low on the tree of butterfly branching (the nadir of the butterfly genealogy is that point on its family tree where it parts ways with the moth) to count as a subfamily. Instead, it is the striped policeman (Coeliades forestan), who forms the base branch from which all skippers stem. As in our first scenario, code occurs every step of the way in a variety of forms — as letters, symbols, numbers, and colors. As binary, code stands for the absence or presence of a genetic trait. The letters C, A, T, and G denote the occurrence of nucleotides within a genetic sequence. And the color-coding of butterflies allows our scientist to measure variations in hue across his samples. Code here might be difficult and abstract, but it is always in the service of clarifying the properties of natural signs. It provides a way to simplify an overabundance of data.
3) A student of aesthetics (or linguistics, anthropology, sociology, political science, or history — all are applicable) reads Roland Barthes' S/Z, which famously parses Balzac's Sarrasine according to a "volley" of related, interlocking codes (154). Code here does not refer exclusively to an individual language, such as French or English — or Misra C, binary, or DNA, for that matter. Nor does code suggest a single encryption with a single key. Instead, codes for Barthes are patterns that emerge within a text. Although the text itself is composed from a larger language set (French), these patterns nevertheless manage to exceed that text and ricochet through the multiple sign systems that constitute culture at large — and vice versa. Instead of viewing the text as a closed system, Barthes illustrates throughout S/Z one of his most crucial insights about reading in general: "... a text is not a line of words releasing a single 'theological' meaning ... but a multidimensional space ... a tissue of quotations drawn from the innumerable centres of culture" (Image, Music, Text 146). Interpretation is always subjective and interpretative, but it is never conducted within a vacuum of any individual system of signs. Instead, literary codes are porous and refractive; they don't merely "contain multitudes," they generate them.
What insights do these scenarios provide? Each worker in the examples just given would most likely have no interest in the other, at least not on a practical level. Our mythical programmer would probably have little patience for thinking through code's historical implications or its relation to moths; I imagine that her thinking would be pragmatic and oriented toward solving a problem. Our entomologist's approach to code is bound by his interest in the butterfly and its world; it has nothing to do with either hybrid cars or Roland Barthes (something by Vladimir Nabokov might have been a better choice). The student of aesthetics, who unbraids different plaits of codes within a text only to reweave them into an interpretation of his own,has little interest in how to translate his insights into machine-readable commands or how it is that the curvature of a butterfly's antenna might help identify a species.
Nevertheless, these examples from widely divergent domains share points of connection: each illuminates the ways that code, in all of its forms — as process, product, pattern, mark of culture, linguistic signifier, and genetic shorthand — is continuous with the material world. Thinking about code's historical development in this way helps bring the computer back to earth. This should be a fundamental enterprise for something that holds so much power over the way we produce and consume information, not to mention over how it mediates the natural world. As media scholar Roberto Simonowski writes in his essay in Beyond the Screen, "code is an indispensable aspect in every discussion of digital arts. Since it is code [t]hat makes everything happen on the screen or on the scene, everything happening is subject to the grammar and politics of code" (137). Put simply, there is no escaping the fact that all computer technology is dependent upon computer code. It is the most potent sign system of our time.
Excerpted from Animal, Vegetable, Digital by Elizabeth Swanstrom. Copyright © 2016 University of Alabama Press. Excerpted by permission of The University of Alabama Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of Contents
Introduction: Animal, Vegetable, Digital 1
1 Coding: Creating and Erasing Worlds of Signs 19
2 Collapsing: Challenging Boundaries of Bodies and Forms 53
3 Corresponding: Communicating with Natural Agents 84
4 Conserving: Saving Nature through Game Play 112
Coda: Self, System, Ecosystem 142
Works Cited 179