Fabula and Sjužet in “Wandering Rocks”
Joyce wrote the Wandering Rocks with a map of Dublin before him on which were traced in red ink the paths of the Earl of Dudley and Father Conmee. He calculated to a minute the time necessary for his characters to cover a given distance of the city. (Budgen)
This is a web project that presents episode 10 of Ulysses (“Wandering Rocks”) as a data visualization that highlights its multiple and clashing spatiotemporalities.
As the project matures, you can see it in action on github pages.
Fabula and Sjužet?
“Fabula” and “sjužet” are two words that have entered into narrative theory via Russian Formalism. Both “фабула” and “сюжет” can mean “plot,” but early 20th century Russian critics narrowed their definitions. Fabula, in its most basic sense, is the historical order of the events in a story. Sjužet (from the French “sujet”) is the narrative order of the story.
For example, in Ulysses, episode 3 (“Proteus”) is after episode 4 (“Calypso”) in the novel’s fabula, as the events in the former take place starting at 11:00 on Bloomsday, while the events in the latter cover 8:00–10:00. However, “Proteus” is before “Calypso” in the sjužet, as it is the third episode, and “Calypso” is the fourth. Reading Ulysses’s fabula would force reading episode 4 before episode 3. Reading its sjužet reverses the process.
These contrasting and perhaps contradictory temporalities (or, better, spatiotemporalities) separate the novel from the world that the novel creates. When discussing story/discourse, histoire/discours, and fabula/sjužet, Alex Woloch adds that
All these dichotomous terms refer, finally, to the essentially divided nature of the literary text, as it is torn between form and content; between the signified and the signifier; between the text’s linguistic world—words, sentences, chapters—and the imagined world that we grasp at through the text. Discourse [sjužet] points us to the narrative’s actual language and structure; story [fabula] to the fictional events that we reconstruct through the narrative (38).
The world generated by the novel has a spacetime that interacts with the aesthetic means by which the world is uncovered in the plot. The elements of the fabula are abstracted out and exist in a spatiotemporality we could call “geohistory,” one where time moves only in one direction. Hence the fabula of the Dublin created in “Wandering Rocks” could actually begin with the 8th Earl of Kildare, whose late 15th century rule of Ireland comes up in conversation between Lambert and O’Molloy, as those acts are a cause whose effect (the conversation), bubbles up on Bloomsday. For the purposes of this visualization, however, the timeline of the fabula marches in its solitary direction from 14:55 to 15:58 on 16 June 1904.
The elements of the fabula, however, provide the foundation for the sjužet. Here, the building blocks of the novel world’s geohistory are recombined with aesthetics in mind. Within the sjužet, the fabula is reordered, and the timeline is no longer a persistent march forward.
Though Ulysses moves linearly for the most part (save the clock’s resetting at the beginning of “Calypso” as noted above), in “Wandering Rocks,” it’s clear how different fabula and sjužet can be. Because of Joyce’s meticulous design for the chapter, we can see it as a network of crisscrossing timelines. The first moment in the episode is the same in both fabula and sjužet; Conmee checks his watch. Similarly, the last is the same in both as well; Artifoni finally catches a Dalkey tram just as the viceregal cavalcade passes him by. But these events bookend something that jumps back and forth, underscoring the different modes of spatiotemporality offered by fabula and sjužet. Yet because of the necessary temporal component of spatiotemporality, it doesn’t suffice to think of “Wandering Rocks” with a fixed map showing everyone’s journeys. The information is incomplete.
Furthermore, “Wandering Rocks”’s focus on minor characters also highlights the tension between the world of the novel (where the fabula is) and its art (where the sjužet is). The episode takes the reader out of the frame of The Odyssey, of the story of Stephen and Bloom. Instead, it showcases Dublin and its inhabitants, a side story populated by masses of minor characters not typically given the space to breathe in a novel.
“Wandering Rocks” is the most obviously complexly, mechanically intertwined episode in James Joyce’s 1992 novel Ulysses. It also stands outside of the rest of the novel in its use of spatiotemporality, focus on secondary characters, and lack of an analogue in The Odyssey.
Ulysses, after all, has structural similarities to The Odyssey, but in the latter, Odysseus is only told about the danger of sailing through the Wandering Rocks. He takes his chances, instead, with Scylla and Charybdis. As a result, the episode provides Joyce a bit of space in which to play outside the frame of the rest of the novel, which is, of course, already a very forgiving frame.
Richard Ellmann notes that the episode brings “the city of Dublin even more fully into the book by focusing upon it rather than upon Bloom or Stephen,” the two main characters of the novel (452). In its difference from the rest of the novel on these terms, it also reflects back on the rest of the novel. Stuart Gilbert adds, “In its structure and its technic (‘labyrinth’) this episode may be regarded as a small-scale model of Ulysses as a whole” (227).
As a novel-within-a-novel, then, the episode is broken up into 19 distinct sections. Each centers around, more or less, a different character. Some move around, with the greatest distance covered in the first and last sections. But other characters remain in one place during the course of their section. Furthermore, each section except the last features at least one “intrusion,” which acts like a jump cut taking us to another location but at the same time. Because the intrusions typically refer to characters who show up elsewhere in the episode, it becomes possible to plot out a timeline of events, which is what this visualization does. We know person x was at place n to meet person y, but over at m later to meet person z. But through the intrusions, we also know where y and z were at the same time, though separate from one another. These collisions are nowhere clearer than in the final section, which features a viceregal cavalcade riding from Phoenix Park to the Mirus Bazaar in Ballsbridge, thereby riding straight through downtown Dublin, where nearly every character in the episode has an opportunity to remark on the cavalcade as it rides past.
A brief description of each of the 19 episodes:
Father Conmee walks from his church to Artane, taking the tram for a part of his trip.
Corny Kelleher tends to a coffin and speaks to a constable.
A onelegged sailor walks to Eccles Street, where he is given money including a coin tossed by Molly Bloom from her window.
Katey and Boody Dedalus arrive home. Maggy offers them soup.
Blazes Boylan orders a package at Thornton’s to be sent to Molly and leers at the shopgirl.
Stephen Dedalus and Almidano Artifoni talk by Trinity College. Artifoni misses his tram.
Miss Dunne does some work and gets a phone call from Boylan.
Ned Lambert gives a tour of the council chamber of St. Mary’s Abbey to Hugh Love. J. J. O’Malloy accompanies.
Lenehan and M‘Coy walk from Crampton Court to Temple Bar to the Liffey. Lenehan tells a story about groping Molly.
Leopold Bloom decides which books to buy for Molly.
Dilly Dedalus runs into Simon Dedalus and asks him for money.
Tom Kernan walks from completing a business deal to the Liffey where he just misses the cavalcade.
Stephen looks over some books and runs into Dilly, who has bought a French primer.
Simon, Father Cowley, and Ben Dollard walk along the Liffey toward the Ormond Hotel.
Martin Cunningham, Jack Power, John Wyse Nolan, and Jimmy Henry walk from Dublin Castle to Kavanagh’s winerooms.
Buck Mulligan and Haines order mélanges at D. B. C. while John Howard Parnell plays chess.
Artifoni still chases his tram, while Cashel Boyle O’Connor Fitzmaurice Tisdall Farrell walks along Merrion Square.
Master Patrick Aloysius Dignam starts to bring porksteaks back home.
The viceregal cavalcade leaves Phoenix Park and drives through downtown Dublin on the way to the Mirus Bazaar.
Spacetime, Nothing but Spacetime…
Did it start with Bergson, or before? Space was treated as the dead, the fixed, the undialectical, the immobile. Time, on the contrary, was richness, fecundity, life, dialectic (Foucault 70).
The relationship between space, spacetime, and the novel has been at the center of my work for almost a decade now. And still I often start with this maybe throwaway quote by Michel Foucault from an interview that seems often cited instead for its final kicker, “Geography must indeed necessarily lie at the heart of my concerns” (77). But the solution to geography’s second-class status compared to time isn’t solved by doubling down on geography, unless that is done just to bring geography up to the same speed as time.
In literary study, we have a nearly century-old model for thinking literature spatiotemporally (as opposed to just temporally or, in a more postmodern, Jamesonian vein, just spatially). Mikhail Bakhtin defines the chronotope as that which demonstrates the “intrinsic connectedness of temporal and spatial relationships that are artistically expressed in literature.” Based on Albert Einstein’s Theory of Relativity, the chronotope “expresses the inseparability of space and time” (84). Many critical geographers (Doreen Massey, Edward Soja, David Harvey) have insisted on merging the two for decades now, but doing so continues to present challenges to literary study, at least when thinking about the worlds a novel creates.
Authors of aesthetic texts have followed Aristotle’s command that a work have a beginning, middle, and end for over two millennia. There is, then, an intrinsic temporality provided merely by how we consume the text. Even edge cases like Choose your Own Adventures or Rayuela still provide a linearity on the fly rendered temporally. A sjužet remains; it’s just more reader-driven and divorced from the increment variable we typically call “page number.”
Roland Barthes writes that “for those of us who are trying to establish a plural… the reading must also be plural, that is, without order of entrance.” Reaching this height of plurality or writerliness is a summit that I’m still trying to work through, but I think that a good way in is precisely through the multiple spatiotemporalities present in any (?) novel, at least. Barthes’s encouragement to read “the text as if it had already been read,” which helps to manhandle and interrupt the text, trips up the dominance on a temporal-only kind of thinking, which both allows for space to catch up and join its dimensional sibling but also allows the reader to see that there is not just one time—or one spacetime—in a novel (15).
Focussing on the competing, colliding, contradicting, and circumventing spatiotemporalities in “Wandering Rocks” actually helps shattered the text, with what feels like literally hundreds of obvious points of entry providing another rereading, another time through, another production.
No matter what I’ve called the spatiotemporalities in novels before (“sites,” “разноречие” and “разномирность,” and so on), as I try to build out my idea of “everyday criticism,” any kind of working over a text that allows these variable points of entry and variable productions, these writings by readers, provide yet newer means of pushing the plurality.
There are three distinct datasets in this project, and all of them rely on the previous work done by Don Gifford and Ian Gunn and Clive Hart.
The data in the first dataset are what I call “instances.” These are instances
of a place’s being mentioned in the text. I wrote a web application,
NYWalker, that makes it easier to add
these instances by hand. They include a page number, a sequence, and a specific
location. To this data, I also added times, using Hart’s measurements. Hart did
not time each instance of a place’s being mentioned, so I guessed a time for a
lot of them. The dataset is further split into two explicit
spatiotemporalities, in that there is an “exterior” spacetime (where actors
are) and an “interior” spacetime (on what actors or the narrator remark). These
distinctions are local to this visualization and are not part of the
data, which is available for
The second dataset is built on the first. It includes “collisions.” These are incidents that clearly take place at a specific place and time and feature multiple actors but may not be explicitly noted in the text with an “instance” (though a nearby instance often provides a clue for location and time!). An instance is Conmee’s boarding a tram on Newcomen Bridge. A collision is Conmee’s (actor 1) looking at the awkward man (actor 2) while on the tram.
The third dataset is mostly invisible at this time. It is made up of the paths taken by around 30 different actors in the episode. Currently, only two are visible: Conmee’s and the cavalcade’s. These two are available because, apparently, Joyce relied on them in constructing the mechanics of the episode. I drew all of the paths by hand and most rely on a lot of speculation, such as how someone walks from O’Connell Bridge to “King Billy’s horse” on College Green. The narrator does not tell us, after all, how the Breens took the trip.
There are several famous geographical “blunders” in the episode, where the narrator (or a character) misidentifies a place. Three deserve special comment. important here. First, the Poddle river acts as an actor and not a place. This permits its “misplacing” by the Wood Quay with no worry. Next, Farrell confuses the Merrion Hall with the Metropolitan Hall. Finally, Stratton’s image welcomes the cavalcade on a bridge spanning the Grand Canal that is referred to as the “Royal Canal Bridge.” In both later instances, I treated these as misidentifications, not as references to places far removed from the spatiotemporal logic of either actor’s location and movement.
In one last detail, Gifford writes that Thomas Court, the location of the unfindable “mansion of the Kildares” that Love mentions, “was the main street of the walled city of medieval Dublin. It is at present a series of streets including Thomas Street” (268). I read the reference as one to, instead, the Liberty of Thomas Court and Donore, a manor right outside the walled city and the location of the current Dublin street Thomas Court.
This project is available on GitHub.
The text in these various tabs was written in Markdown and is converted to HTML on the fly by Showdown. The overall aesthetic is farmed out to Bootstrap, which is extended with the Bootstrap Toggle extension to make the handsome switches. The colorscheme on the map, of course, is from Color Brewer 2.0.
If the project looks a lot like Chris Whong’s NYC Taxis: A Day in the Life visualization, that’s not a coincidence. The project proved to me that it was possible to do a time animation usefully using a combination of D3 and Leaflet (in fact, I found it by just googling “d3 Leaflet animation” and clicking through the link on Zev Ross’s top Google hit). Mike Bostock’s OG “D3 + Leaflet” demo describes how the various dots and paths are converted from GeoJSON to SVG objects.
I also consulted two books rather extensively. Scott Murray’s Interactive
Data Visualization for the
Web is a
great introduction to how weird D3 is. Once I read and understood everything in
that volume, Elijah Meeks’s D3.js in
Action served as handy
secondary resource. Both books could use updates, I feel, including sections
about, among other things, something like Leaflet. Meeks’s chapter on
geospatial information visualization, for example, rather startlingly answered
very few of my questions, largely because of the difficulty I encountered using
d3.geoPath with Leaflet instead of creating a sui generis map where I provide
even the basemap/shapes (see below).
Of course, I also used StackOverflow extensively, but that goes without saying.
Let’s get technical…
For those seeking to emulate this project, by far and away the most complicated
aspect for me is/was figuring out how to control what a
Point GeoJSON object
would look like using the
d3.geoPath generator. It doesn’t draw a simple
<circle> SVG object, but, rather, creates a circle in the SVG
(visible under attribute
d). Hence, the points don’t respond to the usual
changes in the
r attribute like one might expect (or like every D3 tutorial
This mostly became an issue because there are two maps in this visualization,
meaning two different
d3.geoProjections. I rewrote the
whole project using
<circle>s with on-the-fly conversions for each, but it slowed the
site down. Bostock’s example, in contrast, streams points, which is, if I
understand correctly, asynchronous and much faster. So I had to stick to
d3.geoPaths and figure out how to manipulate
Actually, this video from
DashingD3js, despite its
hilariously dull repetition, was what finally made me understand what was going
on with the path and why changing the
r attribute wasn’t doing anything!
I had hoped to have animated paths like in the NYC Taxi visualization above. Whong has it kind of easy in this case, because the path is pre-defined (or at least built w/ an api call), there’s a start point, an end point, and a duration. The animation is made up of only one piece, then—that specific trip from A to B. Here, rather, I have paths covered by people where they have to be in many different, specific places at many different, specific times. That’s a big project of splitting up the paths in the data I am not quite yet ready to do. Building up the other datasets was enough work for now. But in the future, however, I’d love a little horse emoji representing the Cavalcade as it rides to the Bazaar, say.
This Is Great and All, but So What?
My task is to move, to shift systems (Barthes 10).
This is the hardest tab to write. Building this visualization involves (involved) giving in to tendencies that I am trying to avoid in my scholarship. Given that one of the goals was to learn the visualization/animation technology, however, it could not have been otherwise.
Do these animations, despite being highly neat, actually tell us much about “Wandering Rocks”? I think the answer is yes, but perhaps not for the what might seem obvious reasons. I give below a few conversations between my inner hater and myself, conversations that I think provide insight into the massive and potentially unexpected problems this kind of digital work creates—and, again, they’re much more complicated than the initial responses from digital skeptics. Those “OK, So?”s tend to be much easier to parry.
Great, just what the world needs, another deep dive on James Freakin’ Joyce. There are lots of novels with fractured fabuly and sjužety. Similarly, there are lots of novels that provide extensive geographical detail, making mapping movements through a city possible. Why go back to such a canonical novel like Ulysses? Well, its very canonicity means that a lot of the work has already been done. Most of the geocoding for this section was just double-checking the work already done by Don Gifford and Ian Gunn and Clive Hart. Additionally, the timings would have been impossible for me to include without Hart’s own walking around in Dublin pretending to be an old woman or onelegged sailor, stopwatch in hand.
Generally, I aim to make spatiotemporal investigations of novels that are not simply obvious hits like The Novel of High Modernism, but sometimes it’s worthwhile to build on a strong, already existing foundation of scholarship and interest. For example, the geographical “blunders” in the episode have already been dissected by generations of Joyce scholars, meaning when I stumbled upon them independently, I could rely on those scholars’ expertise and either accept that confidently or give reasons for rejecting it. Finally, the novel’s canonicity and popularity helps boost this project’s profile. I think the ideas and troubles hinted at in how I talk about this project are important to any scholars interested in digital scholarship that frames spatiotemoporality for any novel (or perhaps any aesthetic text). Because a lot of people care about Joyce, perhaps this specific visualization will give those ideas more visibility.
Friend, Ulysses takes place in Dublin in 1904. You have a 2017 basemap. This is a valid concern, but perhaps not as valid as it is clever. And it is certainly one that merely putting a c. 1904 georectified basemap underneath the dots and lines would not solve. Johanna Drucker encourages data-handling humanists to reconsider the value of georectification, especially with historical maps, because the process merely “reconciles spatial data and maps… with a given standard, such as Google maps” (76–77). Georectification is the imposition of a specific, GIS kind of thinking on data (or “capta,” to use Drucker’s term) that were not generated with GIS in mind. As Drucker continues, “the greater intellectual challenge is to create spatial representations without referencing a pre-existing ground” (77).
The information within the novel Ulysses takes a troubling trip to your computer screen: I capture a chunk and identify it as a datum (or captum). I then spatiotemporally locate it through consulting some combination of Gifford, Gunn and Hart, Wikipedia, and Google. Following a conversion, perhaps with Google’s help, to Cartesian coordinates based on a measurement of the Earth’s shape from 1984, I line those coordinates up with a contemporary map of Dublin, making the risky guess that, as far as downtown is concerned, streets have not changed their shape all that much. Then Leaflet and D3, two software packages, combine to recalculate those coordinates into coordinates on an SVG plane that you see as little exploding dots. Every step adds new assumptions about how space and time work and move the capta farther from their source. Furthermore, as soon as the jump is made to the digital, a faux-precision dominates, where, for example, something like the immense idea of “America” is reduced to a teeny exploding dot with its center in Kansas. These issues remain unsolved, and this project fails Drucker’s challenge. But adding a georectified historical map would be an even bigger blunder.
You took preëxisting information and just added color and made it bounce. Drucker distinguishes between information visualizations that produce the knowledge they draw and those that merely display information (3). It seems at first inarguable that this visualization does only the latter. Any new insights into “Wandering Rocks” seem destined to evoke the small surprise of the merely interesting: “surprising—but not that surprising” (Ngai 145). We have the delayed judgment of “isn’t it interesting that Boylan’s section is the first to intrude on Conmee’s command of the fabula from the of the episode?” It strikes me that moving that glimmer of interestingness somewhere else requires a deeper look. Or perhaps more rereading.
As far as deeper looks are concerned, I have an idea for a different form of visualizing the tension between fabula and sjužet (hint: a two-dimentional (time|plot)line); that visualization, however, would be exclusively temporal, thereby violating the canonical rule of spatiotemporal thinking, in a Bill Murray singing voice: “Spacetime, nothing but spacetime.” The progress explicit in Fabula and sjužet lure us into thinking of them as strictly temporal terms—did it start with Bergson?—and it’s all too easy to treat them exclusively as such. But we have to avoid that.
Still, on the other hand, the aesthetic of the merely interesting does have “the capacity to produce new knowledge” (Ngai 171). Maybe there’s a way through below.
Rereading draws the text out of its internal chronology (“this happens before or after that”) and recaptures a mythic time (without before or after); rereading is no longer consumption, but play (that play which is the return of the different). If then, a deliberate contradiction in terms, we immediately reread the text, it is in order to obtain, as though under the effect of a drug (that of recommencement, of different), not the real text, but a plural text: the same and new (16).
That sense of “mythic time” is especially appealing to me here. One way to read it would be to think of it as erasing time from the picture into the “Messianic time, a simultaneity of past and future in an instantaneous present” (Anderson 24). For Benedict Anderson, however, this time is replaced by the Benjaminian “homogenous, empty time” that is “measured by clock and calendar” (Anderson 24; Benjamin 264) and existing outside the frame of reference. This new, empty time (and space), of course, is explicitly tied to the rise of the novel (Watt 21–27).
Instead, to me, that “without before or after” signals that there is no one spatiotemporality in the text. The emptiness of that framing, homogenized time, is challenged by “Wandering Rocks” on its own, and that challenge becomes more acute through the play of this visualization, which relies on yet undoes the centrality of a ticking clock to the episode. Clocks and their mechanicity are, of course, vital to “Wandering Rocks.” Frank Budgen calls Joyce an “engineer” in plotting the episode (123). Stuart Gilbert focuses on the ticks and tocks of the clocks and other “references to mechanical movement” scattered throughout the episode (Gilbert 234). The episode begins, after all, with Conmee’s checking his watch and seeing he has enough time to get to Artane. But then it closes with Artifoni finally catching a tram and boarding it… the last moment of the episode providing the breath catching pause of just-in-timeness (in contrast to Hitchcock’s being out of time in missing his bus at the start of North by Northwest) that shows that the framing clock is not enough.
Conmee’s leisurely having all the time in the world is a promise made by the regulating ticks of a clock outside the frame. But it is an empty promise. Gilbert describes how the sections of “Wandering Rocks” “interlock like a system of cog-wheels or the linked segments of an endless chain [that] may be described as ‘mechanical’,” but that system of cog-wheels breaks down throughout for the characters inhabiting the world in Ulysses. Kernan, the Ulsterman probably most excited to see the cavalcade, just misses it. Master Dignam misses the boxing match he’d just found out about by several weeks. And for all the just-in-time success of the tram’s closing its door right behind Artifoni, he has still been chasing it or its follower for about five kilometers, having missed it at College Green and not boarded at Merrion Square. Considering this, the fabula is no longer a meticulous machine moving, and Budgen and Gilbert’s insisting on putting Joyce central to the episode recasts the author not as a machine builder or systems designer but, rather, as instead a mischievous puppeteer, the deus persistently torturing these minor characters by letting the machine tick past them. Joyce cackles with enjoyment over the delays, the misses.
This visualization is not about Joyce’s omnipotence rendered wicked, however. It is about the spatiotemporalities within “Wandering Rocks.” The sjužet’s recombinating of the fabula already confounds the time “measured by clock and calendar,” because now time is measured alongside lines of prose alongside distance. Flipping back to the fabula causes a new rereading. Stepping forward and backward another. Clicking on dots another. The text is always the same, but now it is plural. New. Interesting? Maybe even more.
Online resources and inspiration
Bostock, Mike. “D3 + Leaflet.” The canonical D3 demo with Leaflet.
DashingD3js. “D3 Geo Path.” A tedious and repetitive video that finally made me understand
Forster, Chris. “On Joyce’s ‘Wandering Rocks’ & Simultaneity.” Forster includes two visualizations, but sadly the commenter’s links to an animated visualization have rotted.
Gunn, Ian, and Mark Wright. “O, Rocks! she said. Tell us in plain words.” Includes a 3d vizualization of “Wandering Rocks,” where the z coordinate stands in for time.
Idleworm. “Dublin c. 840 – c. 1540.”
Mappery. “1608 Dublin Historical Map.”
Reeve, Jonathan. “corpus-joyce-portrait-TEI.”
Trustees of Boston College. “Walking Ulysses.”
Wikipedia. “The Civic Survey of Dublin and Environs (1922–23).” Includes tramlines.
Wikipedia. “Historical Maps of Dublin.”
Print resources consulted
Adams, Robert Martin. Surface and Symbol: The Consistency of James Joyce’s Ulysses. New York: Oxford UP, 1962.
Anderson, Benedict. Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso, 2006.
Bakhtin, Mikhail Mikhailovich. The Dialogic Imagination: Four Essays. ed. Michael Holquist. trans. Caryl Emerson and Michael Holquist. Austin: U Texas P, 1981.
Barthes, Roland. S/Z. trans. Richard Miller. New York: Hill and Wang, 1974.
Benjamin, Walter. Illuminations. ed. Hannah Arendt. trans. Harry Zohn. New York: Schocken, 1968.
Blamires, Harry. The Bloomsday Book: A Guide through Joyce’s Ulysses. London: Methuen, 1966.
Budgen, Frank. James Joyce and the Making of ‘Ulysses’ and Other Writings. London: Oxford UP, 1972.
Drucker, Johanna. Graphesis: Visual Forms of Knowledge Production. Cambridge, MA: Harvard UP, 2014.
Ellmann, Richard. James Joyce. Oxford: Oxford UP, 1983.
Foucault, Michel. “Questions on Geography.” In Power/Knowledge:Selected Interviews and Other Writings, 1972–1977. ed. and trans. by Colin Cordon. New York: Pantheon, 1980.
Gifford, Don. Ulysses Annotated. Berkeley, CA: U of California P, 1989.
Gilbert, Stuart. James Joyce’s Ulysses. New York: Vintage, 1955.
Gunn, Ian and Clive Hart. James Joyce’s Dublin. New York: Thames and Hudson, 2004.
Hart, Clive and David Hayman. James Joyce’s Ulysses: Critical Essays. Berkeley, CA: U of California P., 1974.
Harvey, David. Justice, Nature, and the Geography of Difference. Cambridge, MA: Blackwell, 1996.
Huang Shan-Yun. “‘Wandering Temporalities’: Rethinking Imagined Communities through ‘Wandering Rocks.’” JJQ (49.3–4), 2012. 589–610.
Jamson, Fredric. Postmodernism, or, the Cultural Logic of Late Capitalism. Durham: Duke UP, 1999.
Joyce, James. Ulysses. New York: Vintage, 1961.
Joyce, James. Ulysses: The Corrected Text. New York: Vintage, 1986.
Massey, Doreen. “Politics and Space/Time.” New Left Review (196.1), 1992. 65–84.
Meeks, Elijah. D3.js in Action. Shelter Island, NY: Manning, 2015.
Murray, Scott. Interactive Data Visualization for the Web. Sebastopol, CA: O’Reilly Media, 2013.
Ngai, Sianne. Our Aesthetic Categories: Zany, Cute, Interesting. Cambridge, MA: Harvard UP, 2012.
Soja, Edward. Thirdspace: Journeys to Los Angeles and Other Real-and-Imagined Places. Cambridge, MA: Blackwell, 1996.
Watt, Ian. The Rise of the Novel. Berkeley: U of California P, 2001.
Woloch, Alex. The One vs. the Many: Minor Characters and the Space of the Protagonist. Princeton: Princeton UP, 2003.
© 2017 Moacir P. de Sá Pereira. Part of #nyudh and Columbia’s Group for Experimental Methods in the Humanities.