After including the Generative Literature Project in my Experimental Writing course during the Fall 2014 semester, three senior undergraduates remained mesmerized by the perceived novelty of a generative, digital novel. For the following semester all four of us shared our frustrations, questions, and perplexities, which later drove our inquiry into the nature of the novel in its digital future. Many of those started as definitional questions around the confines of a novel while some others were reactionary, addressing why many are fearful of literature’s migration to digital platforms. As it turns out, much of the criticism we read to spur thought on our questions addressed materiality of the codex in conjunction with literature’s responses to digital technologies. We each read from a corpus of essays I chose and students augmented. With each week of reading we all wrote responses to the ideas we encountered, compiling and rearranging them as our collaborative essay developed. Each section of the essay had a “parent” author that worked to consolidate, develop, and edit more heavily than others in the group. This final segment of our four-part series considers how the generative nature of the Generative Literature Project affects the genre of the novel, namely the murder mystery.

Realism in the Mystery Novel

While authorship in The Generative Literature Project may be shared, the concept originated with Michelle Kassorla and Frederick Cope, and with that concept came only one rule: all of the action takes place in a specific story universe, which centers around the fictional Theopolis College campus. Using a familiar setting as a backdrop for our characters may make the piece a successful murder mystery. Matthew Levay claims “the detective genre is most captivating when it grounds its mysteries in the circumstances of everyday life; avoiding hyperbole for a grittier and more naturalistic depiction of the world as it currently exists” (9).  In keeping with this principle, the project creators gave us a blueprint of the President’s room at the college and left themselves open to any questions about the setting of the college environment.  It is important also to note, though, how outside of the box some of our personal stories got (with the moving from one country to the next, the pet monkey, and the strip clubs).  Perhaps what will come of the work will be a farce or parody of the murder mystery, but regardless it demonstrates the bias towards realism in fiction even should some of our work fall into the hyperreal.

Paul Cobley sets the stage for the challenges of crime fiction and so makes for a good place to start. He identifies two points of stagnation in the genre: the focus on the murder as the crime and the conservative view of realism. The Generative Literature Project still invests in the centrality of the murder and solving the case as the narrative satisfaction, but we wonder if that will be the experience in the gamified version or if it will be an immersion into individual characters. Such characters often come out as stock in detective and crime fiction, but will ours? While the Generative Literature Project participates in Cobley’s first conceit, it could be productive to consider how our generative novel departs from the second. With the Generative Literature Project, none of the contributors were constrained by a demand to make the narrative appear “realistic.” Rather, we were given very few constraints and full agency over what we chose to contribute to the project, so long as characters inhabited and event unfolded in the fictional world given to us by the project’s creators. Where the environment and setting of our murder mystery is realistic, the overall narrative is given opportunity to stray from the expectations placed on the genre in the form of open-ended contribution.  Cobley claims that crime fiction is too proud of its realism and conflates that realism with objectivity (289). W.H. Auden argued that crime fiction actually desires escapism, and the clean resolution is an “antidote to depression or opiates for the reading public” (Colbey 289). Transformation of the version of realism that exists in crime fiction, and even the centrality or necessity of that realism, may be one important contribution to the genre from the Generative Literature Project.

After discussing the significance of realism to the genre as a whole, Cobley then turns his attention to popular Nordic crime fiction and credits its success to the reader’s perception of realism and the intense attention to detail. These are two of Roman Jakobson’s five frameworks for realism. The others are an author’s perception of verisimilitude, characteristic features of 19th-century realism, and the motivation to realize poetic devices in the text. The somber tone and its departure from melodramatic realism of past or pulp crime fiction set Nordic crime fiction apart and supposedly helped to earn its popularity. This turn arguably takes the genre away from realism as a style to verisimilitude as a function. While realism has its established literary style, verisimilitude is the action of following the shifting doxa of a culture.

Moving to verisimilitude over realism requires that authors and readers adopt Jakobson’s first corollary of deforming/distorting the realistic style rather than his second corollary that upholds realistic style. For Roman Jakobson, realism occurs when the author intends to display the work as true to life and is perceived as such by the person judging the work. Achieving realism then is accomplished in two ways according to Jakobson’s corollaries. The first has the author deform the norms of realism as a more accurate view of reality. The second upholds the limits of traditional realism’s style to conserve how reality ought be constructed (Cobley 287). Often the former is achieved by exposing the perceived realism of the narrative as a construct often through metatext, directly addressing the reader to break the fourth wall, or bending the rules of realism. By exposing realism as a construct, such narratives claim to be more authentic to reality than narratives that uphold realism as a style accurate to life.

With the Generative Literature Project, we supplied the material of the story, but cannot dictate whether or not the end product will skew to realism as a style rather than verisimilitude as a process or goal. The created product will pay close attention to detail, given all of the ocular and tangential artifacts we produced, but at the same time, the programmers could cut all this out. Cobley’s essay leads to the question of whether multimedia or multimodal elements are a step towards verisimilitude or if it marks a development in the style of realism. Postmodern tradition dictates that such devices pierce the guise of realism, and expose the constructed nature of text to follow Jakobson’s first corollary, but as time marches on, wouldn’t that piercing of realism just seem like a more profound realism?

Curiously, this question seems impossible to answer at this stage of the project. We have made the content, but the organization and structure of that content may have more influence over how the piece will approach realism — as style or as function? The raw material is ripe to craft in either direction. The conflict of writing a murder mystery as we did was that while new media and generative or collaborative work often resists or refuses closure, the murder mystery demands it.

Closure in the Mystery Novel

The implied promise of closure is what coauthors risk as they switch from one author to another. This fear of fouling the plot was ever present in constructing the Generative Literature Project even though our class was confined to only one character’s development. Students had to frequently juggle the larger timeline of the story and our individual desires for what would happen in the narrative pieces we each contributed. Elise Takehana, who read all of the students’ pieces, felt responsible for announcing overlaps, divergences, or convergences between character work from different students. The drive for narrative consistency no doubt came in part from the expectations of the literary genre the Generative Literature Project is based on: the murder mystery and detective mystery.

In his essay “The Long Wait,” Theodore Martin defines the detective novel as one with a promise of resolution extended over the passage of time. While endings are not always satisfactory, closure is the central pleasure of detective fiction. However, Martin also notes that postmodern fiction removes the idea of closure from the genre. This postmodern detective genre is about disappointment in the closure or a lack of it, so it’s really not about the end, but “the uncertain distance between expectation and fulfillment, the persistent gap that makes writing-and reading-take place” (168). Openness comes from leaving gaps while closure comes from full disclosure by the end; however, for Eyal Segal, openness and closure aren’t opposites, but two ends of a sliding scale (162).

For the detective story, the convention is to have a mystery that the detective (for the Generative Literature Project, the detective is the reader/player of the gamified novel) then systematically investigates. The detective is curious about the past, and the reader is in suspense regarding the solution of the mystery (Segal 163–4). Also standard is the fair play principle that claims or clues should be clearly described and the reader shouldn’t be willfully deceived; but then how can the mystery remain mysterious (173)? This can come from making the detective eccentric or cryptic or by manipulating point of view (177–8). The detective dramatizes the gaps to encourage the reader to recreate the detection as well. For the Generative Literature Project, this equates to the imperative on enhancing the reader, user, or player or the novel to carry the central character’s role — the detective.

In The Poisoned Chocolates Case, author Anthony Berkley made the decision to avoid substantial closure in favor of a plot twist occurring in the final chapter and ending the novel with the question, “what the devil do we do?” (176). After a potential murderess is convincingly shown to be one of the Crime Circle’s own, finding enough proof to press charges seems difficult due to the brilliance of the set up, and the suspect walks out confident they will be unable to prove her guilt. The disappointment of being unable to have closure within this fictional world is offset by the events creating that circumstance: a convincing plot twist combined with the under appreciated member of the Circle being the one to pinpoint the murderess.

If in crime novels there exists a move away from exposing the answers and an embrace of ambiguity, would fostering a crime novel in a generative context allow for the crime story to even more fully embrace this seeming lack of resolution?  This would be the case because even the authors (the people both writing the characters and potential plot lines/encounters/stories and the people putting them together) have no idea of its ending.  Does this push further the concept of mystery and deepen the ways a reader can get lost in this mystery and obscurity?

A further complication of the Generative Literature Project would be the potential for multiple simultaneous murderers. As with Berkeley’s The Poisoned Chocolates Case, having multiple solutions undermines the veracity of any one solution. Not only that, but there will be any number of investigations, seemingly one per player/reader. With each character in the Crime Circle proceeding with their own investigation of the same unsolved mystery, the reader is given a substantial amount of often fruitless information per chapter. The anticipation as each respective character began their speech and presented their facts supporting their hypothesis, rose neatly with each chapter. Even with the disappointment by the end of each chapter and understanding that, though this information once seemed correct and valid is now unimportant, the characters still kept the passage of time and therefore kept anticipation regardless of circumstance.

The tension in detective fiction comes from the output of information in regards to the context of the story. The anticipation of closure, directly affected by the information released to the reader as they navigate through the detective process, tends to steadily build with each surprise and enlightenment. As Martin points out,  “While the detective narrative promises to tell us what we want to know, its true function is to remind us again and again, that, as readers embedded in time, we do not know everything yet” (169). But does turning this into a game change any of this? On that same note, is the Generative Literature Project fated to produce an extreme or exaggerated murder mystery, since it may all be about anticipation without resolution because it is generative? Will this bring what Martin calls the “motivation to realize poetic devices” or will this fizzle interest in solving the mystery? Perhaps that fear of the fizzled interest is what motivated the gamification of the Generative Literature Project. Maybe instead it’s adopting Martin’s claim that “the detective genre is not to be found at the end of these stories, but in their middles” (176). The story then becomes about the passage of time. The motive to find “what’s next, what’s next” is about both its fulfillment and its deferral. Are we going to get all “middle” or endless “middle”? Such an aesthetic and narrative effect doesn’t correspond to the tastes of typical readers but perhaps it will become so with the typical digital reader. By chance, would the Generative Literature Project be the opportunity for media to raise the literary value of a narrative without the print codex platform? While the singular and obsessive mentality of the detective may keep mysteries a genre fiction separate from high literature, one must wonder if multiplying the detectives through the media output of digital, gamified novel, help make the mystery become more literary? Do two “low culture” elements equal a high literary piece of narrative experimentation?


Victorian England saw the rise of popular fiction, stories meant for the masses. This started the rise of crime fiction and detective stories as well as the “Penny Dreadful.” Characters like Sherlock Holmes and Sweeney Todd started their lives in pulp fiction, looked at as low brow stories for commoners. With the passage of time, they have been elevated. Arthur Conan Doyle’s detective has become a pillar of fiction itself. Composer Stephen Sondheim adapted the story of Sweeney Todd into a musical, which is considered by many to be his masterpiece. These pieces were categorized as genre fiction but now, with the passage of time and the evolution of the culture in which they exist, they are revered and analyzed as artistic and literary achievements.

Matthew Levay, in his analysis of Gertrude Stein’s relationship and work in crime novels, writes about how “…the boundaries of both modernism and genre fiction hinged upon the question of how literary culture might be enriched by a more nuanced approach to the connections between highbrow and lowbrow forms” (4).  With this in mind, does the choice to make the Generative Literature Project a crime novel of sorts reflect the melding of experimental literature of the popular more genre-based aesthetic of crime-fiction?  Also, would a generative novel have worked as well/better with any other genre of writing as its counterpart?  Will this piece when it is completed represent a new step in the potential of crime-fiction created by the embrace of digital literature and a generative production basis?  This seems to coincide with Stein’s own reading of Edgar Wallace, as it is “much better to make an old thing alive than to invent a new one” (Levay 8).  In this sense, are we creating something new, or showing something older in a new light and granting it new life through its new foundational place of existence?

Will the Generative Literature Project result in a game, a novel, a pedagogy experiment, even a work of literary significance? This cohort of students worked through experimenting with writing because we were in part like Jonathan Safran Foer in his defense of Tree of Codes appropriation and excision of Bruno Schultz’s collection of stories, The Street of Crocodiles. Foer die-cuts Schultz and makes a new story not to improve on Schultz but to love and venerate literature and language. We want an intimacy with literature and language. We want to be part of how it grows and changes. We want literature to be alive in what we do with language. Considering the Generative Literature Project as a novel in digital form raised a number of challenges and questions for all the students involved, but the process of questioning the territory literature will command in the digital realm was one that absorbed us all and one we could only begin to explore here. Joe Bray, Alison Gibbons, and Brian McHale in their Introduction to The Routledge Companion to Experimental Literature announced the call and our work attempted to answer:

What is literature, and what could it be?  What are its functions, its limitations, its possibilities?  These are the sorts of questions that “mainstream” literature, at all periods — commercial bestseller literature, but also the “classics” once they have been canonized, domesticated and rendered fit for unreflective consumption — is dedicated to repressing.  Experimental literature unrepresses these fundamental questions, and in doing so it lays everything open to challenge, reconceptualization, and reconfiguration.  Experimentation makes alternatives visible and conceivable, and some of these alternatives become the foundations for future developments, whole new ways of writing, some of which eventually filter into the mainstream itself.  Experiment is one of the engines of literary change and renewal; it is literature’s way of reinventing itself. (1)

For fifteen weeks we tried to channel living literature, to be a part of its fringe and project what its possibilities might be. If literature is to stay alive, it must, like other living things, change and evolve. These changes and evolutions are not to be considered progressive or as improvement, but evidence only of its sentience and its deep connection to the world, to the cultures, and to the people that it touches.

While this fourth and final installment considered the effect of generative structures on the murder mystery genres, previous installments laid out foundational ideas we explored in the nature of the novel in its original codex form (Part I), defined traits of digital, multimedia literary expression (Part II), and considered the impact of moving the novel from the codex to digital platform via paratextual implications (Part III).

This article is the eighth in a series of reports on the Generative Literature Project, sponsored by Hybrid Pedagogy Publishing.