In this paper, Whalen proposes a course of study into the textuality of computer-generated novels, specifically the corpus of work generated for NaNoGenMo. Given the scope of this corpus, Whalens' intention is to use text analysis techniques such as topic modeling, frequency analysis, stylometrics, and other varieties of machine reading to explore these questions about the textual characteristics of computer-generated fiction.
Natural language generation (NLG) – the process wherein computers translate data into readable human languages – has become increasingly present in our modern digital climate. In the last decade, numerous companies specialising in the mass-production of computer-generated news articles have emerged; National Novel Generation Month (NaNoGenMo) has become a popular annual event; #botALLY is used to identify those in support of automated agents producing tweets. Yet NLG has not been subject to any systematic study within the humanities.
This paper offers a glimpse into the social and literary implications of computer-generated texts and NLG. More particularly, and in line with the ELO 2018 Conference’s 'Mind the Gap!' theme, this paper examines how NLG output challenges traditional understandings of authorship and what it means to be a reader. Any act of reading engages interpretive faculties; modern readers tend to assume that a text is an effort to communicate a particular pre-determined message. With this assumption, readers assign authorial intention, and hence develop a perceived contract between the author and the reader. This paper refers to this author-reader contract as ‘the hermeneutic contract’.
NLG output in its current state brings the hermeneutic contract into question. The hermeneutic contract’s communication principle rests on two assumptions: that readers believe that authors want them to be interested in their texts, and that authors want readers to understand their texts. Yet the author of a computer-generated text is often an obscured figure, an uncertain entanglement of human and computer. How does this obscuration of authorship change how text is received?
This paper will begin with an introduction to, and brief history of, NLG geared towards those with no previous knowledge of the subject. The remainder of the paper will review the results of a series of studies conducted by the researcher to discern readers’ emotional responses to NLG and their approaches to attributing authorship to computer-generated texts. Studies have indicated that a sense of agency is assigned to an NLG system, and that a continuum from authorship to generation is perhaps the most suitable schema for considering computer-generated texts. Who is responsible for the text? Are computer-generated texts worthy of serious literary analysis? What do computer-generated texts reveal about human creativity and lived experience?
The paper will conclude with an argument for why consideration of the social and literary implications of NLG and computer-generated texts is vital as we venture deeper into the digital age. Computer-generated texts may not just challenge traditional understandings of authorship: they may engender new understandings of authorship altogether as readers explore the conceptual gap between human and computer language production.
Computer-generated texts may not just challenge traditional understandings of authorship: they may engender new understandings of authorship altogether.
2×6 consists of short “stanzories”—stanzas that are also stories, each one relating an encounter between two people. Appearing in English, French, Spanish, Russian, Japanese, and Polish, the stanzories are generated by a similar underlying process, even as they do not correspond to one another the way a translation typically does to a source text. These sixfold verses are generated by six short computer programs, the code of which is also presented in full. These simple programs can endlessly churn out combinatorial lines that challenge to reader to determine to whom “she” and “he,” and “him” and “her,” refer, as well as which is the more powerful one, which the underdog. Generating 2×6 is a simple process, and readers are invited to study the programs and even modify them to make new sorts of text generators. Reading the output can be much more difficult, as the text that is produced crosses syntax with power relations and gender stereotypes, multiplying those complexities across six languages.
‘’ was derived from Gertrude Stein, . It was first published in Picador New Writing 1995.
"Neuromancing Miss Stein" was freely adapted from a from a loose draft resulting from computer-aided analyses [using Brekdown] of letter-group frequencies in two samples of text, one from Gertrude Stein's The Autobiography of Alice B.Toklas, and William Gibson's Neuromancer. The index and frequency tables from these analyses were then blended, and the draft text regenerated from the resulting combination. First published in the 1995 print book Picador New Writing 1995, the story was later posted to the web by the author.
"Carousel" was freely adapted from a from a loose draft resulting from computer-aided analyses [using Brekdown] of letter-group frequencies in two samples of text, one from Henry Miller's Tropic of Cancer and the other from Yasunari Kawabata's The Master of Go. The index and frequency tables from these analyses were then blended, and the draft text regenerated from the resulting combination. First published in the 1998 print book _Different Hands_, the story was later posted to the web by the author.
Since the rise of the mainframe computer, literary authors and critics alike have expressed anxiety about the computer’s ability to write narrative prose and poetry as well humans, or better. This lecture situates the contemporary digital literary practices of reading, writing, rewriting, and performing computer-generated texts within a broader social and historical context, dating to long before the advent of the computer. Christopher Strachey's 1952 Love Letter generator and Theo Lutz's The Castle generator are discussed in depth.
In human terms, a generation refers both to a group of individuals of approximately the same age having similar ideas and attitudes, and to the period of time between one such a group and the next, which is roughly thirty years. Two human generations have passed since the first generation of computers. Many more generations of machines have passed since then, each supporting yet more generations of operating systems and softwares capable of being programmed to generate a wide variety of computer-generated texts.
Not quite a short story, not quite a stage play, ‘Once upon a Tide’ is just one of those moments in literature when time … stands … still. When plot advances by simply refusing to budge. One of those waiting times, slack tides, great hollows within which heat intensifies, cold deepens, night thickens, fevers rage, or the sun continues its relentless blaze. Tension builds, and still nothing happens; neither the sight of a sail on the horizon nor the slightest breath of wind. It is within these long stillnesses that sailors’ yarns unravel. In Conrad’s Heart of Darkness (1899), the entirety of Marlow’s tale is recounted in one evening whilst sitting utterly still on the deck of a ship moored on the Thames. In the pitch dark and the heavy night air of the river, the narrator strains to discern meaning: ‘I listened, I listened on the watch for the sentence, for the word, that would give me the clue to the faint uneasiness inspired by this narrative that seemed to shape itself without human lips…’.
In fiction, these long feverish pauses eventually break. In a variable text, however, we may hover forever within the tense and nuanced relation between reading, listening, watching, and waiting for the sentence, the word, the clue…
(Source: J. R. Carpenter, The Junket)
There is no logical reason to cause Conrad-esque characters to speak Shakespearian dialogue. The compulsion to do so is born of reading and re-reading sea stories across genres and across centuries. The reader of ‘Once Upon a Tide’ is encouraged to do the same – read and re-read, aloud if possible.
‘Once upon a Tide’ is a variable, restless, shifting narrative. Turns of phrase, stage directions, and lines of dialogue from Shakespeare’s The Tempest (1610-11) are randomly, repeatedly, and somewhat enigmatically recombined within a close, tense, ship-bound setting reminiscent of Joseph Conrad’s The Secret Sharer (1910), or The Shadow-Line (1916). On the deck of a ship off the shore of an island, two interlocutors are closely observed by a narrator who remains hidden from view.
Not quite a short story, not quite a stage play, ‘Once upon a Tide’ is just one of those moments in literature when time … stands … still.
(Source: J. R. Carpenter, The Junket)
Once upon a perigee tide we sailed past a lagoon, our ship charmed. On the forecastle deck two stout men sat mending nets. From their looks I wondered that they had come from calmer shores, certainly none so desolate as these.
Once upon an apogee tide we piloted past a cliff, our ship a brave vessel. On the poop deck two mean boatswains squated twisting tales. From their looks I assumed that the pair had set out from foreign shores, certainly none so harsh as these.
Once upon a spring tide we piloted past a delta, our ship dashed all to pieces. On the tween deck two old strangers loitered mending nets. From their apparel I reasoned that both had sailed from braver shores, surely none so harsh as these.
The Montaigne Machine is a work of electronic literature that invites users to participate in the creation of multimedia personal essays. The essays generated by The Montaigne Machine each center on a specific topic taken up by the inventor of the genre, Michel de Montaigne. The essays combine text from Montaigne’s famous Essais, first published in 1580 and here translated into English, with original text from each visitor who uses the machine. These texts are placed within an image that has been uploaded by a photographer on Flickr, designated as available for remixing, and most recently tagged with a term appropriate to the essay’s topic. The resulting essay is a collaboration, perhaps even a conversation, across time and media by three artists.
(Source: http://conference.eliterature.org/media/eric-lemay-montaigne-machine)
The term ‘writing coastlines’ implies a double meaning. The word ‘writing’ refers both to the act of writing and to that which is written. The act of writing translates aural, physical, mental and digital processes into marks, actions, utterances, and speech-acts. The intelligibility of that which is written is intertwined with both the context of its production and of its consumption. The term ‘writing coastlines’ may refer to writing about coastlines, but the coastlines themselves are also writing insofar as they are translating physical processes into marks and actions. Coastlines are the shifting terrains where land and water meet, always neither land nor water and always both. The physical processes enacted by waves and winds may result in marks and actions associated with both erosion and accretion. Writing coastlines are edges, ledges, legible lines caught in the double bind of simultaneously writing and erasing. These in-between places are liminal spaces, both points of departure and sites of exchange. One coastline implies another, implores a far shore. The dialogue implied by this entreaty intrigues me. The coastlines of the United Kingdom and those of Atlantic Canada are separated by three and a half thousand kilometres of ocean. Yet for centuries, fishers, sailors, explorers, migrants, emigrants, merchants, messengers, messages, packets, ships, submarine cables, aeroplanes, satellite signals and wireless radio waves have attempted to bridge this distance. These comings and goings have left traces. Generations of transatlantic migrations have engendered networks of communications. As narratives of place and displacement travel across, beyond, and through these networks, they become informed by the networks’ structures and inflected with the syntax and grammar of the networks’ code languages. Writing coastlines interrogates this in-between space with a series of questions: When does leaving end and arriving begin? When does the emigrant become the immigrant? What happens between call and response? What narratives resonate in the spaces between places separated by time, distance, and ocean yet inextricably linked by generations of immigration? This thesis takes an overtly interdisciplinary approach to answering these questions. This practice-led research refers to and infers from the corpora and associated histories, institutions, theoretical frameworks, modes of production, venues, and audiences of the visual, media, performance, and literary arts, as well as from the traditionally more scientific realms of cartography, navigation, network archaeology, and creative computing. "Writing Coastlines" navigates the emerging and occasionally diverging theoretical terrains of electronic literature, locative narrative, media archaeology, and networked art through the methodology of performance writing pioneered at Dartington College of Art (Bergvall 1996, Hall 2008). Central to this methodology is an iterative approach to writing, which interrogates the performance of writing in and across contexts toward an extended compositional process. "Writing Coastlines" will contribute to a theoretical framework and methodology for the creation and dissemination of networked narrative structures for stories of place and displacement that resonate between sites, confusing and confounding boundaries between physical and digital, code and narrative, past and future, home and away. "Writing Coastlines" will contribute to the creation of a new narrative context from which to examine a multi-site-specific place-based identity by extending the performance writing methodology to incorporate digital literature and locative narrative practices, by producing and publicly presenting a significant body of creative and critical work, and by developing a mode of critical writing which intertwines practice with theory. (Source: Author's Abstract)