Bestsellers and a Modernist’s Library: Hits and Misses

2 August 2021

Emily Bell, University of Antwerp

Despite the broadening of approaches to push boundaries in modernist studies – linguistic, cultural, national, critical, formal – single-author studies are alive and well. Author-centricity has been a methodology in its own right: the life of an individual justifies critical study whether it is biographically-oriented or not. This longstanding tradition of author-centred modernism has been subject to critique since structuralist interventions (and ensuing subfields) widened the scope of literary studies to explore the cultural matrices and circumstances of textual production. But how do we continue to justify scholarship that spotlights individuals already centralised in our narratives of modernist culture?

Happily, I work on James Joyce: I am part of a project at the University of Antwerp which is digitally reconstructing Joyce’s library over his lifetime by uniting extant documents with reading sources found through genetic research. Equally happily, this project is as much about Joyce as the people around him – readers, writers, patrons, family, and friends. Authors’ familial and social networks are often integral to their cultural reach and exposure. Several aspects of these social relations are contained in the bibliographic network – in the provenance, authorship and readership of the library texts.

This project does not subscribe to an image of the isolated author in their ivory tower bureau surrounded by shelves of leather-bound volumes. On the contrary, using Joyce’s library as a case study, we might be able to demonstrate that the canon (as represented by an author like Joyce) is not independent of nor immune to a more broadly-defined modernist culture. Combining author-centricity with bibliographic focus paradoxically encourages a branching out in often unanticipated directions, illuminating those subfields of modernism we seek to equalize in our study.

In practical terms, Joyce’s archival presence is the envy of many other writers – and those who work on them. Seven libraries worldwide house books from his surviving personal library; his papers are dispersed more widely still. Documentation, in most cases, is a result of contemporary cultural value – hence, in general, the woefully late critical attention afforded to women writers, queer writers and writers of colour. Joyce certainly enjoyed a high level of recognition by the time of his death in 1941. In many respects, it therefore isn’t possible to recover or examine underrepresented voices because, historically, institutional efforts did not cater to writers that had not already been pedestalled by the powers that be. 

If we want to think more broadly about what Joyce was reading – especially in the 1920s and 1930s, when the Finnegans Wake notebooks were being filled with anything from pulp to Plutarch – it would serve us to know who else might have been reading such materials. A method for doing this could be to compare the contents of Joyce’s library with bestsellers from the time. There are many caveats here: for one, the commercially-driven model of bestseller lists cannot profess correlation in any real sense with reading records. Thus, this is really a hypothetical exercise, because neither dataset is large enough to draw out conclusive nor convincing argumentation. Nevertheless, let us hypothesise.

This data derives from matches between Joyce’s library and the annual bestseller lists of the top ten fiction books as recorded by the American magazine Publishers Weekly for the years 1900 up to 1941; there are also nonfiction lists of between six and ten titles for every year from 1917 onwards, plus 1912 and 1913. It is worth highlighting that Joyce isn’t on any of these bestseller lists – unsurprising since Ulysses was banned in America until 1933 – though Virginia Woolf appears on the list for 1937 with The Years. This is some indication that so-called ‘High Modernism’ is not widely represented in the historical record of a commercial, reading public, but is not totally absent either. The matches I sought operate on two levels of the metadata: the authors and the titles (see figure  below).

Of the authors represented in the bestseller lists, we have 361 different names: twenty of these authors are found in Joyce’s library, corresponding to 5.54% of the authors of historical bestsellers. These name matches are, in nearly every case, the usual suspects from this period: Ernest Hemingway, Rudyard Kipling, Virginia Woolf, Joseph Conrad, H. G. Wells, and so on. There are a few known figures who don’t appear in Joyce’s library which could provide some pushback on our expectations of his extensive reading: Aldous Huxley, Somerset Maugham, Thornton Wilder and Eugene O’Neill. At least, this could indicate some launchpads for source hunting in Joyce studies, or it might humbly remind us of our overreaching assumptions about the reaches of Joyce’s reading.

There are only five works found in both datasets (the historical bestsellers and Joyce’s reconstructed library), four of which are nonfiction works. Of the 580 distinct titles in the bestseller lists, these five equate to less than 1%. They are: Creative Evolution by Henri Bergson (nonfiction list, 1912); Queen Victoria by Lytton Strachey (nonfiction list, 1922); Gentlemen Prefer Blondes by Anita Loos (fiction list, 1926); the first volume of Our Times by Mark Sullivan (nonfiction list, 1926); Nijinsky, the diary of Vaslav Nijinsky, edited by his wife and published in 1934 (nonfiction list, 1934). These title matches broadly represent works that have significance for the writing process of Finnegans Wake rather than Ulysses (owing to dating and usage in the Joyce library). This could tell us any number of things: that Joyce read more widely and more attentively to popular writing for the Wake; that, generally, he had greater access to books later in his career; that he was more interested in nonfiction than fiction; that, in fact, he was not particularly interested in the books occupying the minds and bookshelves of the average reader. But this data alone proves none of these things. 

This has shown some connections between Joyce’s library and the broader culture, providing insight into the contemporary reading public and snapshots of the popularity of texts. We are, however, comparing a single author against an entire reading public, so we shouldn’t be too hasty in our conclusions: how many books had the average American read of these lists anyway? As mentioned earlier, this argument is hypothetical, showing how we could (and should) compare Joyce’s library with other bibliographic datasets. If we were able to compare the content of this Joyce library with Virginia Woolf’s library (now at Washington State University Library) or with W. B. Yeats’s library in the NLI, this analysis could be triangulated against the data on contemporaneous bestsellers in Britain from The Bookman magazine. So, what percentage of Woolf’s library contains those popular titles – are they the same as Joyce’s? Is Joyce representative of owning or working with a “Modernist’s” library in this specific sense, in his reading of popular literature? The painting of such a picture requires simultaneous work on the background and foreground: the fine details of instances of Joyce’s reading stand side-by-side with emphasis on the meta-bibliography of his library in order to illuminate the large-scale reading network in which he and others participated.

Comments are closed.

Blog at

Up ↑

Create your website with
Get started