The lending library cards reveal the activities and addresses of 651 lending library members: what they borrowed and purchased, when they renewed and canceled their memberships, where they lived. This information promises to answer important questions about interwar Paris and twentieth-century literature.
But the logbooks reveal that the Shakespeare and Company lending library had thousands of members. The cards represent only a fraction. What, then, do the cards actually tell us? Do the books on the cards reveal the reading practices of the Lost Generation? Do the renewals—or lack thereof—illuminate the transience of expatriate life? Do the addresses capture where the lending library community lived and how it moved around the city? The information on the cards is deep: over 22,000 activities for only 651 members. The information in the logbooks is broad: over 11,000 activities for 5,700 members.1 How do they fit together?
To understand the relationship between the cards and the logbooks—or more precisely, between the members with borrowing activity and the members represented in the logbooks—we created a series of charts. The first chart—based on information from the cards—shows members with borrowing activity by month over the life of Shakespeare and Company, 1919 to 1941:
The Shakespeare and Company Project has information about the borrowing activity of, on average, twenty-three members per month between 1919 and 1941. If we focus on each decade separately, we notice a stark difference. For the 1920s, we have information about the borrowing activity of seventeen members per month. For the 1930s, the number rises to twenty-nine.
The second chart—based primarily on information from the logbooks—shows the lending library members by month:
From 1919 to 1926, the number of members steadily increases, reaching its peak in January 1926 with 240 members. In 1930, at the beginning of the Great Depression, the number declines precipitously. The chart also reveals the “seasonality” of Shakespeare and Company membership: winter was the busiest season, summer the slowest. Why are there gaps in the chart? Some logbooks have been lost—if they ever existed. For example, the Shakespeare and Company Project doesn’t have logbooks for most of 1930 and all of 1931, 1932, and 1937.
The third chart makes the gaps in logbook coverage explicit:
The chart is important insofar as it helps prevent misinterpretations. We can see, for example, that the dip in members in 1928 is not a dip at all: we are missing the logbooks from January and February 1928. Likewise, the downward slope at the beginning of 1930 and 1937 reflects the trailing off of members from the preceding month.
The fourth chart overlays the first two charts to show the relationship between members with borrowing activity and the members represented in the logbooks:
The Shakespeare and Company Project has information about the borrowing activity of 11% of lending library members. For the 1920s, the percentage is lower: only 6%. But in the 1930s, the percentage is higher: 23%. Why do some months show more members with borrowing activity than total members? Information on the cards compensates for gaps in logbook coverage.
Why are there relatively so few lending library cards in the Sylvia Beach Papers? Were some cards lost? Did Beach destroy them? (We know she used some cards as scrap paper.) Did she save some cards and not others? We suspect that she was more likely to save the cards of famous members and friends. But we know that F. Scott Fitzgerald was a member and we do not have his cards. We are also missing the early cards of Ernest Hemingway: his first extant card is from 1925, although he joined the lending library in 1921. The first chart shows that cards from after 1932 are more likely to be preserved. Beach’s relationship with James Joyce offers one explanation: in 1932, she gave up her right to publish Ulysses and refocused her attention on the lending library. Perhaps she wanted to highlight the influence of Shakespeare and Company beyond Joyce’s masterpiece. Perhaps she was simply less overwhelmed. Membership patterns reveal that members with cards joined the lending library for, on average, a month longer than members without cards. This fact is not surprising: Beach likely would have discarded or reused cards of short-term members, especially members only visiting Paris.
The Project has a trove of information about lending library members. That information is more complete for members with cards. For example, we have linked 40% of members with cards to VIAF records, but less than 2% of members without cards. This imbalance is due to various factors: first names and addresses on the cards make it easier to identify members, the preponderance of cards of famous members. (Famous members are more likely to be listed in VIAF, which only includes people associated with library records.) But the imbalance also reflects our research priorities. The earliest phases of the Project focused the cards.
At this point in the Shakespeare and Company Project, we don’t yet know how the relationship between the cards and the logbooks impacts our ability to make generalizations about life in interwar Paris or literary history. But we are eager to test the Project’s potential and will continue to add essays to the analysis section of the site.
This article uses data from Shakespeare and Company Project Dataset: Lending Library Members, Books, Events, version 1.0.
1. The lending library likely had fewer than 5,700 members. Members with the same name—Adams, Fournier, Smith—appear multiple times in the logbooks. We only identify these members as the same person when we have sufficient evidence. See our FAQ for more information.
Cite this document
“The Shakespeare and Company Lending Library Cards in Context.” Shakespeare and Company Project, version 1.5.6. Center for Digital Humanities, Princeton University. March 9, 2020. http://shakespeareandco.princeton.edu/analysis/2020/03/shakespeare-and-company-lending-library-cards-context/. Accessed March 25, 2023.