Extending Genetic Criticism to Audiotexts: A Conversation with Jason Wiens

December 19, 2020
Jason Camlot

Jason Lee Wiens, a CanLit scholar, has been involved in the SpokenWeb Pedagogy Task Force, developing approaches to teaching with sound recordings, over the past year. Wiens’ engagement in the use of sound recordings in the teaching of Canadian poetry has led him to think about the relationship between sound recordings of author’s reading their works (sometimes in advance of their publication) and the published versions of those same poems.  This thought process represents a continuation of some of Charles Bernstein’s reflections upon the differential qualities of texts that become interestingly discernible to us when we bring audiotexts into the mix.  As Bernstein remarks in his introduction to Close Listening: “When the audiotape archive of a poet’s performance is acknowledged as a significant, rather than incidental, part of her or his work, a number of important textual and critical issues emerge. What is the status of discrepancies among performed and published versions of poems, and, moreover, between interpretations based on the text versus interpretations based on the performance?” (Bernstein 7).  Wiens’ interest in such matters has, interestingly, turned his critical attention back to a tradition of textual analysis that emerged in France in the late 1970s, known as “genetic criticism” (La Critique génétique), with its interest in approaching texts as entities whose emergence is traceable through the study of “avant-texts”.  Genetic criticism grew out of “a structuralist and post-structuralist notion of ‘text’ as an infinite play of signs” but also is deeply interested in a more “teleological model of textuality” shaped by the agency of the author.  Its real object of study is less the establishment of an authoritative text from analysis of pre-drafts and adjacent materials, than “the movement of writing” that can be inferred from such analysis (Deppman, et al. 2).  Wiens presented some of his first reflection about the potential (and challenges) of extending genetic critical approaches to audiotexts in his paper “Sounding Difference: Genetic Criticism and Literary Audio Recordings,” delivered at paper delivered at the “Genesis – Genetic Criticism: from Theory to Practice” conference held at the Jagiellonian University (Kracow), and co-sponsored by the Institute des Textes et Manuscrits Modernes (Paris), 12-14 June 2019.  The following discussion explores and extends some of the thinking developed for that paper.

Thanks for participating in the Listening, Sound, Agency Forum, Jason. You have been involved in the SpokenWeb pedagogy task force over the past couple of years. Can you say a bit about how you have begun to work with audiotexts in your teaching, and how this work in pedagogy has led to your exploration of new methods for the study of Canadian literature?

In recent years I have developed three new assignments involving audiotexts.  The first is what I call an “e-locution assignment,” in which students are asked to select a poem from the course readings, record themselves performing a reading of the poem, and then write a paper which reflects on their interpretive decisions in that performance.  They post their recording of the poem to the LMS, and then when we arrive at the poem as a class – they complete these assignments before that day – I can play the recording for their peers, and have them share a bit of what they reflected on in their essay.  This solves the problem of finding a student willing to read a poem on the spot.  It also encourages a new way of close engagement with the text in question.  I have used this assignment in survey courses in which students might select from poems by poets ranging from Coleridge or Dickinson to M. NourbeSe Philip or Jordan Abel.  Given the nature of the work of those latter poets, the performances (and the essays about those performances) need to grapple with certain challenges in oral interpretation, which reflect similar challenges in reading interpretation.  In a senior course specifically on sound and Canadian poetry, I had students complete this assessment selecting from the work of Abel, Philip, Joshua Whitehead, or Rachel Zolf.  For that assignment students had to address questions of voice appropriation, as well as consider the arguments made in Jerome McGann and Lisa Samuel’s essay on “Deformance.”

A second assignment I designed for use in Canadian literature courses asked students, in groups, to attend and record monthly poetry readings at a local bookstore.  The reading series I selected was the monthly Flywheel series at Pages Books in Calgary.  I selected that series because it is venerable (running every month for well over a decade before COVID) and consistent, and because it often features a mix of established and emerging writers.  This project was developed in collaboration with our library, which created a digital collection to store the students’ projects.  The students upload the recording to the digital collection – I had a research assistant help them with this – and then each member of the group selects one text – a poem, story, or whatever – performed at the reading, and then writes a paper discussing that text embedded in a review of the event in general.  These selected recordings are shared with the class as a whole, and they then make up our ‘readings’ or rather listenings for a unit on local writing.

The third assignment I have deployed in Canadian literature courses in recent years is in line with some of the pedagogical approaches of other members of the SpokenWeb, namely close listening assignments working with texts in existing digital archives: the Sir George Williams reading series archived at SpokenWeb; recordings of Canadian poets at PennSound; the Kootenay School of Writing audio archive at kswnet.org; and the recordings of poets such as Robert Kroetsch, Earle Birney, and Alden Nowlan at the University of Calgary.  I ask them to consider the relationship between the written text and the audiotext, usually in terms of what might surprise them about an oral performance, or how an oral performance is also an interpretation of the text in question.  Again, these are completed before we discuss the poets in question, and because I leave it up to them to select the poem they want to work with from among a list of poets, they determine the readings / listenings for this part of the course.

This work in pedagogy has led, first of all, to new ways of thinking about the CanLit canon.  For a number of reasons much of the available online digital audio archive preserves a whitewashed canon.  Were I to try and teach a course consisting entirely of Canadian BIPOC writers, for example, the Sir George Williams or PennSound collections would be of limited use.  This is one of many reasons why SpokenWeb’s work to digitize and network more of the analog audio archive is so important.  The audio archive’s tendency, at the moment, to reinforce a whitewashed canon has implications for scholarly research as well as teaching.

One benefit of both the close listening projects and the recording project would be that it introduces an element of chance into the curriculum, as I cannot determine which writers will read in the reading series nor which poems students may select from online audio archives.  This decentres the instructor to some extent in terms of the composition of the syllabus, displacing those determinations onto the organizers of the reading series or the students themselves.    

These pedagogical approaches have impacted my own thinking about the broader study of Canadian literature by reinforcing a sense that a more comprehensive approach to a writer’s work must take into consideration any audiotextual recordings of that work.

When did you first learn about genetic criticism (La Critique génétique), and what has lead you to consider this approach for the study of audiotexts?

I don’t know if it was termed as such, but I recall an American Literature course in my undergraduate degree in which a guest professor shared with us the fascicle versions of some of Emily Dickinson’s poems, asking us to consider the implications of the many variants between Dickinson’s handwritten manuscripts and the early bowdlerized versions of her posthumous publications.  That sparked an interest in textual histories, variant readings, and indeterminacies of authorship which remained with me in graduate school.  I was working on (mostly) living Canadian writers but was excited to discover that the papers of a number of these writers have been archived and such work could be done with them as well.  However, I don’t believe I started consciously thinking of myself as doing genetic criticism nor reading extensively the scholarship in the area until years after I completed my PhD.

I’ve been led to explore a genetic approach to audiotexts because my research in the area has shown that audio recordings to this point have not been considered as part of the “avant-texte,” which is all of the archival materials that are part of a text’s creative matrix.  In part this is because genetic criticism has tended to focus on 19th century or early 20th century canonical European writers like Joyce or Flaubert, for whom recordings of them reading their work are either limited or non-existent.  But the confluence of a postwar culture of public literary readings with affordable recording technologies means that we have in the audio archive recordings of writers reading works in progress, sometimes from print manuscripts which have otherwise been lost.  I think this opens up all sorts of fascinating questions relating to textual histories and the audio recording as a kind of writing. 

What is genetic criticism?  Can you explain this critical approach, the history of its emergence, its development, and the reasons for its continued use?

Genetic criticism approaches the archival materials of a text’s production, its various drafts, notes, fragments and so on—which it terms the avant-texte—as evidence of the creative process.  As the name implies, genetic criticism is interested in the evolution of literary texts and how we might map that evolution: a text’s genetic sequence, so to speak.  A more comprehensive hermeneutics, from a genetic point of view, reads the “finished” text in relation to its metamorphoses over the course of its history.  Genetic criticism emerged in France in the 1970s—the term critique génétique was coined by the French scholar Louis Hay in 1979—and remains centred in continental Europe through research programs at the Institut des Texts & Manuscrits Modernes (ITEM) in Paris or the Centre for Manuscript Genetics at the University of Antwerp.  In its interest in creative process rather than finished, authoritative textual product, genetic criticism is often distinguished from both traditional philology and textual scholarship. Whereas traditional textual scholarship seeks to stabilize the text by producing an authoritative edition, genetic approaches destabilize any authoritative final product by foregrounding a text’s often messy compositional history.  In recent years there has been some recognition that genetic criticism and textual scholarship share significant common ground, that textual scholarship as advocated by editors and scholars like Jerome McGann embraces or at least accepts textual indeterminacy as much as genetic criticism.  And genetic criticism continues to overlap with genetic editing, as advances in digital technologies (and the expiration of copyright in the case of some authors) increasingly make possible the production of sophisticated genetic critical editions, which incorporate all the known variants of a particular work’s avant-texte.   

What were some of your key considerations and first arguments about the potential of extending ideas and methods from genetic criticism into the study of audiotexts?

Although genetic criticism has always embraced textual indeterminacy, its emphasis on written records at least grounded its investigations in the relative stability of print.  Including the audiotext in the genetic record lends another turn of the screw to the endeavour, introducing further elements of indeterminacy.  What are the implications of attempting to trace textual variants in an audiotext?  Do we read it as a performance of a manuscript, and use it to attempt to recover or reconstruct that manuscript (if the manuscript is not evident in the archive)?  Or do we need to adopt a different approach?  If any oral performance of a text constitutes a variant of the print version, as Charles Bernstein has suggested, does that make audiotextual recordings of performances, even ones that sounds faithful to the authoritative print version, further nodes in a textual history?  

Can you give a specific example or two of how you have applied genetic criticism to a consideration of the Canadian audio poetry archive?

I recently published an essay on Roy Kiyooka’s “the 4th avenue poems” in the collection Pictura: Essays on the Works of Roy Kiyooka, edited by Juliana Pivato.  In that essay I trace the textual history of those poems, both in published versions and as evidenced in the archive, as articulating shifting relationships between poetics and history over the course of Kiyooka’s career as a poet, from the early 1960s to his passing in the early 1990s.  I conducted my research in the Kiyooka fonds at Simon Fraser University, and was generously given access to a taped recording of Kiyooka reading from an early draft of the poems years before their first publication in 1969.  In that essay I included that audiotext amongst the variant texts I investigated, but did not really explore the implications of working with an audio recording.  My paper in Kracow discussed that tape, and other recordings of Kiyooka reading those poems, to raise some questions about how we might incorporate audiotexts into a genetic critical methodology.

Another example of a slightly different genetic approach to the Canadian audio poetry archive would be my close listening of Deanna Ferguson’s recorded lecture “And Weep for My Babe’s Low Station” (also in the SFU archives), in both my doctoral dissertation and more recently in my article “Minutes Over Monuments: Re-Reading the Kootenay School of Writing (as) Archive” published in Canadian Literature a few years ago.  I identified overlapping allusions and social / poetic concerns between the lecture—which was not delivered publicly, and of which no written version has been published—and Ferguson’s poem “Received Standard,” published in her book The Relative Minor, reading the lecture and poem as companion pieces of sorts.  In this case, though, the fact that the lecture—by definition a public utterance—was never delivered publicly lends it a provisionality of interest to genetic approaches.   

In the “Sounding Difference” paper you delivered in Kracow, you anatomize some significant challenges to extending genetic criticism to audiotexts, emerging from such primary complicating factors as “access”, “transcription”, “citation” and “dissemination”.  Can you summarize some of these challenges, and explain how genetic critical approaches may still prove useful despite them?

For the Kiyooka work, I had to travel to the SFU archive and listen to an undescribed recording in order to hear if there was anything relevant to my project on it.  In order to do so, I needed the assistance of SFU library staff, who in turn needed a suitable machine on which to play the recording.  This illustrates a number of problems of access: the audio archive is often site-specific, existing in fragile analog formats requiring obsolete or hard-to-find hardware to play them back.  The audio archive is also comparatively undescribed, though scholars like Deanna Fong have been doing important work cataloguing and describing the Kiyooka recordings at SFU.  But for the moment, it means researchers often need to listen through hours of material in the hopes of hearing something relevant to their project.  This is not a new challenge in archival research, of course, but the additional technological barrier in accessing sound files further complicates the endeavour.  And the pandemic has further demonstrated to us the barriers posed by on-site archival research. These are problems the SpokenWeb project is intended to address, of course: digitizing audio recordings which may currently exist in fragile, obsolete formats; carefully describing the recordings; and making them available in online networks.  I think there is particular urgency with respect to some of the older recordings out there, including the one of Kiyooka that I worked with: we need to determine the provenance of these recordings and the details of the events they record while members of that generation are alive and may still remember these details.

As I note above, when a poet is reading from work in progress, we might assume—or they may make evident in the recording—that they are reading from a manuscript which may be otherwise lost to the archive.  We may hear variants in diction or syntax that distinguish the oral performance from other oral or written iterations of the text, but we cannot necessarily hear line breaks, variable margins, typographical modifications, or other spatial manipulations on the page.  This is of particular relevance when considering the work of a poet, like Kiyooka, who worked primarily in open forms.  This introduces the problem of transcription: we cannot presume to ‘recover’ a lost manuscript through a transcription of an oral performance.  Instead, that element of the avant-texte must be approached in audio form.

The problem of transcription also extends to the other challenges of citation and dissemination.  If we cannot simply transcribe the audiotext into the more user-friendly written form in our scholarly work, nor can we do the same when citing and publishing the scholarship.  This means either that we must be able to cite publicly-accessible digital audio collections, or we must embed into our publications audio clips as we would written citations, limiting publication venues to digital platforms.  Fortunately for genetic scholarship, the main scholarly journal in the field, Variants, is published in digital format.   

What kinds of audio archives best lend themselves to a genetic critical study of audiotexts?

I would say recordings of public readings rather than, say, studio recordings would lend themselves best to genetic critical study.  Writers are more likely to share works-in-progress at a reading, whereas in a studio recording they would likely read from published, perhaps well-known work.  The paraliterary elements of a reading would also be of interest to a genetic critical approach.  A writer’s prefatory remarks, for example, could provide significant contextual information about the genesis of a piece; and audience reaction (or lack thereof) or input in the form of questions or comments may also be of interest.  These would be considered “exogenetic” elements; that is, of relevance to the development of the writing but “exterior” to the writing itself.  

And then there are writers who may have availed themselves of recording technologies to record their thoughts on writing or their ideas for projects.  Or they may have recorded conversations with other writers, sometimes about literary matters, often not: the Kiyooka audio archive is replete with material of this sort.  Such recordings would be the audio equivalents of the journal or of written correspondence, both of interest to genetic criticism. 

What has your foray into a genetic critical study of Roy Kiyooka’s recorded poetry readings revealed to you about his work as a poet in historical context(s)?

It has led to, or rather reinforced, an understanding of a poet who worked a serial, modular practice, continually recording and revising his poetic shaping of historical particulars on multiple scales: personal, local, global.  The audio archive mirrors the written in that his oral performances of the poems were iterations in an ongoing, indeterminate process.  If the writing practice was not entirely distinct from the visual art practice – and I think most would agree it was not – his extension of compositional practice into oral performance involves a similar rejection of firm disciplinary boundaries. 


Jason Wiens is a Senior Instructor and Associate Head, Undergraduate Student Affairs in the Department of English at the University of Calgary.  He has published widely in the field of Canadian literature, including articles on Dionne Brand, George Bowering, Margaret Avison, Sharon Pollock, Daphne Marlatt, Rachel Zolf, Jordan Abel, and the Kootenay School of Writing.  He recently edited a special issue of English Studies in Canada on “Pedagogies of the Archive,” and created an online teaching resource for Oxford University Press’ An Anthology of Canadian Literature in English.

This article is published as part of the Listening, Sound, Agency Forum which presents profiles, interviews, and other materials featuring the research and interests of future participants in the 2021 SpokenWeb symposium. This series of articles provides a space for dialogical and multimedia exchange on topics from the fields of literature and sound studies, and serves as a prelude to the live conference.

Jason Camlot

Jason Camlot’s critical recent critical works include Phonopoetics: The Making of Early Literary Recordings (Stanford 2019), the co-edited collection, CanLit Across Media: Unarchiving the Literary Event (with Katherine McLeod, McGill Queen’s UP, 2019) and  "The First Phonogramic Poem: Conceptions of Genre and Media Format, c1888" in the online journal, BRANCH.    He is the principal investigator and director of SpokenWeb and Professor of English and Research Chair in Literature and Sound Studies at Concordia University in Montreal.