Looking back from this digital age toward the early twentieth century, I continue to see traces of code and programming appearing in a range of literary works. In my previous post, I followed John Cayley to define code rhetorically. Writing of any kind qualifies as code when it is addressed to other writing. An instance code operating in digital media functions as code because the instructions or information it contains is addressed to other information, operating like a machine. The code works like a machine because it is understood and acted on without the intervention of a human interlocutor. Humans can read and understand the code, but only by adopting the rhetorical stance of, eavesdropping on, the non-human agents.
As I’ll be examining in my next couple of posts, modernist authors were obsessed with representing this type of nonhuman expression, and as such they were participating in changes in how all of culture began seeing language and information. This change amounts to a materialist view of how language shifts from a transparent expression of human interiority, into a material and resistant medium. Far from destroying its potential to express ideas, this way of seeing language opened up entirely new fields of aesthetic and conceptual potential in which information had an intrinsic capacity to animate itself, automatically connect with other information, and to operate independently of humans. The literary spaces in which these experiments with the materiality of language take place are part of a larger twentieth century revolution in information and technology resulting in the code, programming, algorithms, and other types of automated information processing at work in digital media.
In every genre and format, modernist authors continually insist on this new understanding of language. The primary symptom of these changes is an author working to eliminate the human voice from their work. This achieves the effect whereby language operates as code–automatically–by addressing itself to other language, and interpreting these texts requires overhearing these non-human exchanges. 
From the perspective of scholars (as opposed to the authors), an essential part of comparing literary language to code requires also understanding the affordances of print media. This includes fields such as book history, printing history, comparative media studies, textual scholarship, circulation studies, and genetic analysis. These fields are important not only because the authors themselves figured these material concerns into their works, but also because these fields illuminate how a print text is, no less than a digital computer, a processual medium that is assembled in stages and operates according to encoded instructions.
The analogy remains a loose one, but assuming a text consists only of the version we read in a new paperback edition is similar to assuming the information we receive on the computer screen both begins and ends with that interface. Stacked infrastructure, information economies, technology manufacturers, and the physical processes at work in the machine itself, combine to produce the media experience of using a digital object. Understanding these factors requires both mental and physical labor to reverse engineer these objects as we use them. A similar critical engagement is necessary with texts appearing in print. Confining our study of literary works to the trade paperback edition we read and teach for coursework, research, or pleasure not only limits our interpretive purview, but misunderstands what a print text is and how it fits into media history.
To use an extreme example, Joyce’s Finnegans Wake demonstrates the ways in which the study of texts requires the examination of the entire media object–manuscripts, drafts, versioning–just to get a sense of what is taking place on the pages of the final version. Without the wealth of scholarship performed by genetic critics, our understanding of the Wake would be significantly different. Simply knowing that Joyce “wrote” the Wake by accumulating discrete bits of information through associational connections with established “nodes”, (as David Hayman calls them in The ‘Wake’ in Transit) alters our interpretation of the text as a whole. The connections Joyce made between the nodes and information he collected is a perfect example of the active the role language came to play in modernist literature. Joyce integrated this information according to relationships it bore to other information, a section of text mentioning rivers, for instance, would prompt Joyce to collect all the names of rivers from references such as the Encyclopedia Brittanica and weave them into the text. The process itself was human, Joyce and his network of information gatherers were reading texts and recording information, but the instructions for what to collect derived from relationships between elements of information. As if Google Search were just a bunch of people responding to queries and collecting information.
The same could be said of the puns the Wake is famous for: the portmanteau words are assembled according to connections between characteristics the words share. Semantic, orthgraphic, phonic features of words, any type of informational relationship, provided Joyce with instructions for agglutinating them together like LEGOs. According to Jean Michel Rabate, Joyce’s “associative and collective method implies the enlisting of many other people and induces a mechanical linguistic process” . This basic network of information gathering expanded into an “automatic word machine systematized by Joyce in the winter of 1930” which was “designed to radically alter a previous text. The number of layers piled up on the first-draft version is staggering” . Something took place between that first draft version and the final version we read and that decision by Joyce–the development of that word machine–is essential to arriving at a full picture of what the Wake was and what it became. The changes that took place tell us a great deal about Joyce’s process as a writer, but also about the way texts come together and depend on these organizational and compositional substrates to become what they are.
As I keep fleshing out these ideas, I will be analyzing some of the texts I feel best exemplify code or programming as well as an examination of the widespread trend in experimental modernist texts of removing the human voice from the text.
 I’m not suggesting that the authors themselves are eliminated from the creative process, an obviously untrue position. Instead, I’m arguing that the authors are expressing the language’s potential to interact with other language. Joyce and Pound, with their separate but similar penchants for quotation, are obvious examples of authors who did not “create” their works wholesale, but were involved in connecting elements of information according to patterns inhering between them. But Stein, H.D., Williams, and others were also involved in similar aesthetic experiments in which language takes an active role in determining the text’s composition and function as art. I will be expanding on each of these authors.
 Rabate, Jean Michel. “The Fourfold Root of Yawn’s Unreason: Chapter III.3.” How Joyce Wrote Finnegans Wake: A Chapter by Chapter Genetic Guide. Eds Luca Crispi and Same Slote. University of Wisconsin Press, 2007. 394.
 Ibid, 398.