How to Read Ulysses By the Numbers
A novel and its numbers: a novel as its numbers. And why not, if, as Pound puts it, “number is the source of all things.” It was the Pythagoreans who first started suggesting that an elaborate system of numerical resemblances exist between nature and the cosmos. The Pythagoreans believed that numbers not only possessed attributes (reason, opinion, harmony, and justice) but also, Aristotle writes, “since it seemed clear that all other things have their whole nature modeled upon numbers, and that numbers are the ultimate things in the whole physical universe, they assumed the elements of numbers to be the elements of everything, and the whole universe to be a proportion or number.” A lot happens to get numbers from the cosmos of the fifth century BCE to the computer of the mid-20th CE, but the basic idea that a number is something to be seen, heard, and read remains to this day. And that’s just one part of the equation.
We also need to remember that at one point in human history, numbers were only spoken words; the rhythm of a nursery rhyme an example of the unity between the two (Eenie meenie minie mo; One two, buckle my shoe; and so on). For Friedrich Kittler, the complete separation of number and numeral occurred with writing, and it remained that way until Alan Turing imagined a universal computing machine that could process an infinite set of coded instructions made of ones and zeroes.
The novel is not a computer built to generate calculations or solve complex equations. It is a precoded machine, part of a much more expansive “media history of numbers,” as Kittler once called it, that can help us understand how literature can be computable and cosmic, practical and mystical, lettered and numbered all at once. Though filled with words, novels also carry numbers (sometimes in the form of letters), and no matter how elaborate or incomplete the fictional worlds inside them, these numbers are there behind the arrangement.
Stuart Gilbert was the first reader to apply this idea to Ulysses. In his 1930 study, he treated the novel as a sequence of thematic variables that could be put into an equation to help decode “the structure of the book as a whole.” “Ulysses is like that of all epic narratives, episodic,” he explains on the first page. “There are three main divisions, subdivided into chapters or, rather, episodes.” The meaning contained within these subdivided episodes, he continues, is “implicit in the techniques . . . in nuances of language, in the thousand and one correspondences and allusions with which the book is studded.” As much as Gilbert’s numbering here and elsewhere included passing statements about the structure, the network of allusions, and events and timeframes of the plot (“Mr. Bloom’s day begins, like Stephen’s, at 8 a.m., when he is preparing his wife’s morning tea at their house, No. 7, Eccles Street”), all of it was backed up by a schema that Joyce passed on to him, one identifying the Homeric title and character but also the hour, organ of the body, art, color, symbol, and narrative technic.
For John Kidd fifty years later, this schema was evidence enough that a more elaborate numerical pattern lay in waiting. In his preface to an abandoned Dublin edition from the 1990s, he wrote: “It should surprise no one that Joyce, who assigned a reigning color and organ of the body to each chapter of Ulysses, and constructed a huge scaffolding of Homeric correspondences, would count letters, words, sentences, paragraphs, and pages.” Kidd never ended up publishing the results of his theory, and his plan to recuperate the original layout was abandoned. But if Gilbert’s publication of that schema made it impossible for readers to ignore the extent of Homer’s presence after the 1930s, the reverse was happening to the numbers since each resetting of the novel, beginning with the fourth impression in 1926, effectively erased any patterns encoded in the lines, pages, and paragraphs of the original. Facsimile editions are now widely available in print and digital formats, but that impulse to go a-counting has not gained traction in the meantime.
The practice of reading Ulysses by numbers that was recommended by Gilbert, promised by Kidd, and mastered by Hugh Kenner is not new of course. Numbers have long played a role in the production and reception of literary works and have been part of a numerological tradition prizing their allegorical and symbolic value for millennia. Take Dante’s intense admiration for nine as one powerful example. Looking back on the meaning of events in his life, as he does in Vita Nuova, he discovers the number everywhere. He is nine years old when first setting eyes upon the nine-year-old Beatrice, his intercessor, and though Florence is a small town, he only sees her again nine years later, long enough to change the direction of his life, before she dies on the “ninth day of the month.” According to Dante, the nine heavens counted by Ptolemy were in perfect alignment on the day of her earthly birth making her a miracle. And here is his proof: “the number three is the root of nine, for without any other number, multiplied by itself it gives nine, as we plainly see that three times three is nine. Therefore, if three is the sole factor of nine and the sole factor of miracles is three, that is, Father, Son, and Holy Ghost, who are three in One, then this lady is accompanied by the number nine so that it may be understood that she was a nine, or a miracle, whose root, namely of the miracle, is the miraculous Trinity itself.”
Dismiss it, call it nonsense, refuse to buy into the logic, and still, for Dante, this number is a fact, not just a figure: Beatrice, the embodied number, nine, is a miracle. Literary history is filled with all kinds of numerological puzzles, some more elaborate than others, and they can be found in works by Homer, Catullus, Vergil, Dante, Petrarch, Boccaccio, Spenser, Shakespeare, Milton, and Pope. R. G. Peterson, in fact, has identified four distinct phases in the history of literary numerology—Alexandrians (ancient world), Aristotelians and Thomists (Middle Ages), Platonists and Pythagoreans (Renaissance), and Palladians (eighteenth century)—but he acknowledges, as do several others, that there was a rapid decline at the end of the eighteenth century, inspired by a rationalism and empiricism suspicious of the mystical charge numbers can carry. But if listeners and readers have been trained at different moments over the millennia to recognize the presence of numbers in different literary works, it was not a skill that carried over into the discipline of literary criticism until the middle decades of the 20th century when numerological criticism was in its heyday and led to the uncovering of a wide range of symmetrical constructions with the critics Alastair Fowler, A. Kent Hieatt, and Marie-Sofie Røstvig leading the charge. Not only did they take the numbers seriously, but their intensely close readings of some major, mostly English poets, revealed that it was irresponsible to ignore them any longer. The motivation to write by numbers may change over time, along with the audience’s ability to decode them, but R. G. Peterson is right to suggest that the literary use of measurement and the desire for symmetry is bound up with “deep human feelings about number, centers, circles, and limits—from the writer, like the early Greek philosopher, responding to the immediate circumstances of nature.”
In the 1960s, at a moment when numerologically minded critics were searching for symmetry, others were looking ahead to a future of machine-assisted reading. This newly emerging field of so-called humanities computing was planning to use computers to uncover patterns, symmetries, structures, and correspondences of several works at once. A few attempts were made to apply computers to the study of style, but the most successful results were confined to the compilation of concordances and word lists. Computers, in other words, were saving time through their capacity to compute, but they were not going to revolutionize the interpretation of literature or modify anyone’s perspective on literary history.
Fast forward to the early 2000s and Franco Moretti’s ambitious call for distant reading. By that time, humanities computing had already begun to evolve into a mode of computational criticism interested in conducting quantitative experiments on corpora that contained thousands of works with the goal of going even bigger. Operationalizing was fast becoming a new keyword for quantitative literary analysis, and it described the process by which concepts could be transformed into a series of operations that allow us to measure objects. To know the concept required having the measurement to test it, and it could involve the sentence or paragraph but also work across genres and archives.
Living in a quant-obsessed computer age, it’s tempting to think that numerology belongs somewhere in the cabinet of literary-critical curiosities: fun to show off now and again but nothing to take too seriously. And it’s no surprise to discover that the Digital Humanities community pretends it never existed. That is a mistake because it would require ignoring a legitimate practice in a shared literary critical history and also one that we need nowadays to remind us that numbers have a qualitative dimension. To confront a number and to decipher what it means is not, strictly speaking, a quantitative exercise. There are values, some of them empirical, but history has taught us that they can have a mystical or aesthetic connotation.
Yet instead of seeing these two critical traditions as somehow opposed to each other, it can be more productive to consider what they share. For starters, numbers. Numerology is, in effect, a mode of computational reading adapted from a much earlier tradition of biblical exegesis that not only recodes words, lines, pages, and paragraphs as discrete and sequential units but also involves counting them to discover sequences, ordinal and cardinal positions, and sums. In addition, the numerological approach believes that the meaning of a text cannot be found only in what the words or the structures that contain them say. The numbers, both alone and together, can speak. The possibility of a message hidden within them requires that someone is there who can identify, decode, and interpret.
That same impulse drives computational literary analysis (CLA). In the most basic terms, CLA counts literary texts. The major difference involves the fact that these texts are digitized, which means they can be coded by an extensible markup language (XML) and fed into different processing programs. Like numerology, CLA believes literary works are not made of words alone or, to put it another way, that the words in literature mean more than they say. And whether analyzing syntax, genre, style, or narrators, a computational approach treats literature as data for all kinds of formal and statistical analysis. I am not arguing here about the validity of numerology, and I have no interest in trying to convince anyone that CLA is really numerology in a digital age. Rather, I want to reorient how we understand the relationship between the two precisely because Ulysses is a novel that holds them both in suspension—the symbolic and the empirical, the qualitative and the quantitative, the abstract and the concrete, the mystical and the rational. The numbers of Ulysses are never just numerological or computational: they are both. And that is precisely the reason that the numerological and the computational can inform each other whenever reading by numbers is at stake.
Take “number 7 Eccles Street” by way of an example. It’s the address of Leopold and Molly Bloom, but does that exact number really matter? What if it were 6, 14, or, why not, 9? For the numerologist, that number 7 might correspond with the days of creation and rest, but so what? How far does this connection take us in our interpretation of a fictional day in Dublin? A computational reading, on the other hand, would instead see that seven as a quantitative fact. Molly and Leopold live in the seventh unit located on Eccles Street, a real street in Dublin. And that’s of potential interest for someone interested in the historical or topographical dimension of the novel, who might want to know that there was actually a 7 on that street in June 1904. But why argue over the validity of one or the other when they’re both correct? Seven is a number in a novel, and it can have a qualitative and quantitative dimension. So even if it is charged with symbolic meaning, this digit also serves a quantitative function that contributes to the realism.
Ulysses may seem like a strange choice for anyone going in search of numbers. For one thing, it is a modernist novel, which means it belongs to an ironic moment in literary history that largely lacked any patience for sacred correspondences even if they came from an author who had renounced Catholicism but whose numerical sensibility kept his appreciation for its rituals and symbols. But Ulysses is also a novel, belonging to a genre without a particularly strong numerological tradition. A lack of rigorous rules for formal composition may be one reason, but Alastair Fowler has argued that prose is also to blame since it teaches readers “to move relatively quickly along the semantic line.” If readers of verse are conditioned to meditate on words and lines in search of mathematical patterns, novel readers can be found speeding along into nothingness. Perhaps that’s a fitting end, given György Lukács’s definition of the novel as the “epic of a world that has been abandoned by God.” But if God is absent from Ulysses, critics such as Ernst Robert Curtius have claimed that number has a “form-bestowing factor in the divine work of creation” making Dante’s Commedia an obvious precursor, with Kidd arguing that the “numeric patterns” make Ulysses as “consummately ordered” as Spenser and Dante.
All of this numerical ordering makes Ulysses seem like a literary relic from a bygone age. Some of his contemporaries may have been steeped in the mysticism of Eastern religions, but Joyce builds primarily on Greek, Christian, and Hebrew models with a specific literary lineage. Dante’s influence looms large, but that’s also part of an intertextual strategy for advertising his own oversized ambition. To write by numbers by 1922 was one way to insert yourself into an established tradition. Numbers, then, were not simply ornaments or offhand allusions. They could be adapted to set the foundation for a fictional work and insert the novelist into a preexisting literary-historical chronology.