Joyce’s Paper Database

Joyce’s Paper Database: Presentation for BH and DH Conference at The Center for the History of Print and Digital Culture at the University of Wisconsin–Madison

In his recent book, Plain Text, Dennis Tenen proposes a new way of reading the different layers through which digital media deliver text.  This approach, which he calls computational poetics, is “a strategy of interpretation capable of reaching past surface content to reveal platforms and infrastructures that stage the construction of meaning.  … computational poetics breaks textuality down into its minute constituent components.  It is a strategy of microanalysis rather than macroanalysis” (loc198).  Tenen calls for looking at these latent, tiny spaces in order to more fully describe the processes that determine how texts appear.

Somewhat perversely, in this paper, I am advocating a nearly identical approach, but for the study of print texts, in particular James Joyce’s Ulysses.  Texts like Ulysses (I’m thinking of Finnegans Wake, Flaubert’s Bouvard and Pecuchet, Pound’s Cantos, and more recent works like Anne Carson’s Float and Susan Howe’s The Telepathy of Archives: Spontaneous Particulars and Debths) lend themselves to this approach because they feature unusually robust organizational and I would argue, even programmatic, networks existing below the surface of the page.  Like digital forensic or media archaeological approaches to digital texts, understanding how and why these print texts appear as they do requires investigation of the material and encoded processes that authors use to produce them.  I will begin by outlining the material structures Joyce developed to organize the information he integrated into Ulysses, and then I will make a case that understanding the relationship between these structures and the text as art object involves reverse engineering the “surface content” to illuminate the invisible encoding that guided Joyce’s composition of the novel.  


Material Structure:

In Joyce’s Creative Process and the Construction of Character, Luca Crispi performed the most comprehensive census of the papers Joyce used to compose Ulysses as they are scattered in libraries throughout the Western hemisphere.  He identifies five main document forms the avant-texte takes: “For Ulysses, there are five basic kinds of manuscripts: Notes, Drafts, Faircopies, Typescripts, and Proofs” (Crispi Joyce’s Creative Process 281).  Each document type serves certain functions and each of the stages is tied to Joyce’s idiosyncratic compositional tactics, or to the demands of publishers, printers, and other individuals involved in the final versions of the text.  

The first stage is Joyce’s information gathering methods.  Joyce collected a great deal of information from conversation, letters, and by sending acquaintances, friends, and family on information gathering errands.  He recorded many of these details onto small slips of paper he carried with him.  In addition, Joyce’s letters are saturated with requests for materials from a myriad of sources, and, together, these efforts comprise a complex network of information accretion.  

After gathering the information, Joyce developed methods for organizing it and these evolved from the 1917 Subject Notebook into the famously chaotic Notesheets [show slides].  During the seven year composition of Ulysses, Joyce’s notes took multiple forms and were organized according to several principles, but by the time Joyce was relying on the Notesheets, he was entering elements of information in the form of lists of phrases organized under episode headings that were based on correspondences with The Odyssey.  (With some exceptions, each episode of Ulysses corresponding with a chapter of Homer’s text.)  By late 1920, Joyce had developed the Linati and Gilbert schemas which consist of a set of categories derived from the Homeric associations for each episode [show slide].  The final destination for the information entered into this system was the versions of Ulysses themselves.  Studying these systems is complicated because they evolved over time, but also essential because changes to the latent systems correspond with changes taking place on the pages of Ulysses.  Thus, like the examination of digital media, confining our attention to the text on the surface ignores crucial details about how and why that surface text appears as it does.

The most obvious change took place in 1919 as Joyce wrote the “Cyclops” episode when Ulysses changed from a basically narrative text about the lives of a small set of characters, to one that dramatically increased in size and complexity and foregrounded extreme stylistic experimentation.  These changes on the surface of Ulysses follow changes in the note systems Joyce used to manage the information.  The Subject Notebook is the first surviving version of these notes.  Joycean and genetic critic Wim Van Mierlo describes:

Predating all other surviving manuscripts and notebooks for Ulysses, the Subject Notebook gives us a glimpse of what Ulysses may have looked like in late 1917, at a moment when Joyce began redrafting what he had written since 1912 or 1914 to begin the process of reshaping the book in the form it was finally to assume in 1922.  

The Subject Notebook merits particular attention because it reveals several things about Joyce’s early organization methods.  The notebook consists of a series of categorized lists, some that are recognizable in final text of Ulysses, others that are not [show slide again]. 

The Notesheets are difficult to make sense of on their own because, apart from the episode headings, they have little internal organization.  First, though Joyce entered this information as he accumulated it, the Homeric structure remains the primary organizational principle, each episode with a title from The Odyssey having its own set of notes.  Next, the largest units of information appearing in the Notesheets are complete sentences.  The majority of the material consists of what amounts to a list of smaller words and phrases Joyce identified as having relevance to the episode under which they are filed.  These characteristics reveal a process in which the final versions of the text are comprised of a multitude of discrete elements Joyce gathered from various sources, recorded onto the Notesheets, and arranged onto the pages of Ulysses according to the Homeric framework.  

The role the schema played is somewhat more ambiguous.  It is difficult to determine where (or if at all) in the composition process Joyce used the schema, but the categories of association they outline do reveal that Joyce’s initial Homeric framework extended beyond simple character and event correspondences [show slide again].  They reflect detailed conceptual and formal categories according to which Joyce accreted information and then conceived of the episodes.  

The transition from the Subject Notebook to the Notesheets illustrates an important point about how Joyce used these systems to compose Ulysses.  As Ulysses expanded in size and informational density, the lists in the Subject Notebook no longer served to organize the information he collected.  In order to cope with this expansion, he developed the Notesheets which allowed him to collect items that corresponded to each individual episode.  Paired with the secondary categories of association as formalized in the schema, Joyce was able to dramatically amplify the specificity and volume of information he collected and organized.  This shift from the Subject Notebook to the Notesheets is something akin to the difference between a spreadsheet and a database.  The Subject Notebook is a single two dimensional table with rows and columns.  The Notesheets and schema operate like a database with a table for each episode with space for Joyce to enter elements into the secondary categories of information.  The increased capabilities of the latter system demonstrates that Joyce was developing a kind of information machine which helped him compose Ulysses.  As he encountered the limitations of early versions of this machine, Joyce was forced to increase its power to house and organize information.  However, as with digital media object, understanding of this machine is incomplete without examining the rules by which it operates.  


The associations in Ulysses begin with the Homeric correspondences.  According to the parallelism between Ulysses and Homer’s Odyssey Bloom is Odysseus, Stephen is Telemachus, and Molly is Penelope.  The figures Odysseus meets in The Odyssey appear as twentieth-century Dubliners: the Citizen in the “Cyclops” episode is Polyphemus, Gerty McDowell is Nausicaa, Garrett Deasy is Nestor, etc.  The events that take place on June 16th, 1904 also participate in the parallel: the conversation in the National Library is Scylla and Charybdis, the lying-in hospital in the “Oxen of the Sun” episode is Helios’ island, Bella’s brothel is Circe’s lair, the visit to the Cabman’s Shelter is Odysseus speaking with Eumaeus in his hut, etc.  

Joyce used elements from The Odyssey fluidly, creating character and setting correspondences contributing to the narrative of Ulysses.  The Odyssey also determines the text in more complex and granular ways.  Joyce identified individual themes, motifs, images, and ideas from throughout Homer’s text and used them as the basis for integrating other information through associations.  Joyce’s insight was that his initial impulse to construct Ulysses on narrative correspondences with The Odyssey opened into a hugely powerful logic of associations that possessed the seemingly limitless potential for integrating more and more information through continually expanding relationships.  When Joyce began to focus on these associations rather than the narrative–as he was composing the “Cyclops” episode–the high-modernist narrative dominated by Stephen and Bloom’s stream of consciousness began to be crowded out by the escalating amount of non-narrative information he integrated into the text according to this programming.

By way of example, I’ll focus on the correspondences surfacing throughout the “Aeolus” episode.  The Homeric correspondences center on Aeolus, the warden of the winds, who helps Odysseus by confining unfavorable winds in a bag.  Odysseus’ men let the winds out of the bag and their ship gets blown back to Aeolia where Aeolus refuses to help them any further.  The Homeric correspondences in the “Aeolus” episode are stylized and overt: the characters in the Freeman’s Journal newspaper office are ceaselessly blowing, breathing, speaking, sighing, and shouting.  The office is constantly being blown through by newspaper business, the newsboys, and both Bloom and Stephen.  Under the direction of Miles Crawford, the counterpart of Aeolus, the office is the source of voices moving in every direction, within the building in which the events take place as well as the network of news traveling throughout Dublin, the rapidly circulating activity of the news business resembling Aeolian winds.  The connection is complex, but intuitive, and it is emblematic of the type of informational relationship through which Joyce generates characteristics of Ulysses from the Homeric elements.  Joyce has taken an element, in this case the wind motif from The Odyssey, and used it as a catalyst for generating each of these aspects of the text.  However, the correspondences expand further and determine aspects of the episode that go beyond diegesis.  This passage exemplifies each of these characteristics [show slide]:


Mr Bloom, breathless, caught in a whirl of wild newsboys near the offices of the Irish Catholic and Dublin Penny Journal, called:

–Mr Crawford! A moment!

Telegraph! Racing special!

–What is it? Myles Crawford said, falling back a pace.

A newsboy cried in Mr Bloom’s face:

–Terrible tragedy at Rathmines!  A child bit by a bellows! (U 7.963-965).  

The most conspicuous characteristic of the text are the headlines, here “RETURN OF BLOOM”, that interrupt the narrative sixty-three times.  They are defined by the over-clever distillation of the events in the section of text that follows them, like a newspaper headline.  The headlines are joined by a panoply of rhetorical devices Joyce added to the episode.  In James Joyce’s ‘Ulysses’, Stuart Gilbert says, “The thirty-two pages of this episode comprise a veritable thesaurus of rhetorical devices and might, indeed, be adopted as a text-book for students of the art of rhetoric” (189).  Gilbert provides a list of nearly 100 rhetorical devices alongside examples from the episode in which they appear.  Though less conspicuous than the headlines, the rhetorical devices are woven into the episode’s most realist details: [refer to quotation].  Gilbert lists this as an example of the “Anticlimax” rhetorical device, presumably anticlimactic because it’s not a terrible tragedy to be bit by a bellows.  These single line of “Aeolus” are determined by multiple pre-established factors.  The wind references from Homer expand into the newsboy’s affiliation with the newspaper, he is yelling in Bloom’s face, mentioning a bellows, and manifesting a rhetorical device all in this brief passage.  

“Aeolus” demonstrates how Joyce used Homeric elements to gather and present information in Ulysses.  It was during large-scale changes in Ulysses after he wrote the “Cyclops” episode that Joyce added the non-narrative elements such as headlines and rhetorical devices.  Through associations, the wind motif expands into the secondary categories of elements that appear in the episode.  Though the interposition of the headlines and rhetorical devices alter how the text operates as fiction, we can trace the appearance of the headlines and rhetorical examples back to the original Homeric correspondence.  Their origins derive from the progression from Aeolus, to wind, to breath, and then into the associational relationship between circulation, speech, and the art of rhetoric.  Portions of the text, in this sense, are the product of non-narrative processes.  They are on the page not because they participate in the story, but because they associate with motifs or images from The Odyssey.  In other words, accounting for their presence on the page requires looking beneath the surface of the text to study the associational logic Joyce used to compose it.  

As in Tenen’s description of “computational poetics”, in order to fully understand the text appearing on the pages of Ulysses, it is necessary to dismantle, layer by layer, the series of processes working to produce it.  It’s important to note that one can certainly read Ulysses, like any text on a digital platform such as Kindle or a laptop, without understanding these processes.  It’s also true, however, that the presence of many of the aspects of the text, in “Aeolus” the headlines, the rhetorical devices, the emphasis on breath and lungs, is determined by the series of categories Joyce derived from the Homeric correspondences.  Like a text read in digital formats, the integration of these elements is the result of an informational programming not visible to the reader at the interfacial level.  At a conference whose aim is to consider the overlap of book history and digital humanities, the structures Joyce built and the programming by which they operate have much in common with the functions of digital texts.  

It’s necessary to point out that the latent systems I have outlined have been exhaustively written on and studied in fields such as textual genetics and what has come to be called comparative textual media studies.  Scholars in textual genetics for instance, have been studying these manuscripts for decades in order to trace each element in the Notesheets onto the pages of Ulysses much in the way that Tenen describes computational poetics.  Many scholars have also compared Joyce’s work with digital media, particularly hypertext, dedicating an entire online journal, called Hypermedia Joyce Studies, to these types of comparisons.  This is all to say that there is widespread belief that Joyce’s methods appearing in early twentieth century print have things in common with the technical media that have come to define the twenty first.  The empirical and material focus of textual genetics and the conceptual abstraction of media comparison done on Joyce’s work need each other to reach conclusions, just as studying the “code” of a digital object also requires understanding the “body” of the device that actuates it.  My contribution to these discourses is in some sense simply a literal application of digital methods to print texts.  Ulysses, the text and its latent structures are one text machine complete with an interfacial surface, material substrate, and a concealed programming.

Our current conference’s CFP asks “is ‘dh’ a continuation of some of the most ‘traditional’ scholarly work in the humanities: bibliography, textual criticism, and book history?”  Based on the work done on Joyce’s composition process, I would have to say that DH is a continuation of textual studies in print.  I would add, however, that the digital pane through which we view print in the twenty first century allows us to recognize functions not normally associated with non-digital textual media.  Here I’m thinking of the complex, material, and evolving data management structures Joyce developed to augment his memory, or the Homeric programming he employed to compose Ulysses.  If DH is a continuation of BH, then it also provides us with an expanded vocabulary with which to describe normalized features of print media.  



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

UWMadison iSchool Daily Digest

What you need to know to be in the know


UW-Madison | Digital Humanities | Research Network


News & Information for Students Admitted to the UW-Madison Information School MA Programs


researching, writing, triathloning


Manuscripts, materiality, method: a research network

Text and Medium: Intro to Digital Humanities

ENGL 2393, The University of Tulsa

Posts on the Penman

An Iowa Joycean's Musings on the Hilarious, Irreverent Irish Genius

an online business card, entry point, and space for musings

Ragman's Circles

Talking about what the matter was

Modernist Fragmentation and After

International Postgraduate Conference, Princeton University, September 29-30 2016

This and That Continued

Jenny Diski - Writing and stuff.

Parrots Ate Them All

A blog about Stevie Smith and the art of the aphorism.

New Readia

WordPress site of Nicholas M Kelly, PhD, Lecturer in Rhetoric at the University of Iowa, literature, media, and digital culture scholar

Stewart Varner

scholarship, libraries, technology

Media Theory of/for the present

graduate seminar, UC Santa Barbara (Winter 2016)

%d bloggers like this: