Numbers

Among the data points we gathered, the year number has been the most useful orienting information, since it allows us to match the timeline of the cards with that of the collections (i.e. catalogues and inventory). The fact that many cards feature time stamps – for receipt of the deposit and/or monthly subscriptions, or for the return of borrowed books, for example – facilitated the recognition, yet intrigues remain. Given that the monthly subscription section only have space for 12 months, it is reasonable to assume that readers get a new card during each year of active membership. Indeed, most cards only date one year’s activity, occasionally extending into the next year, as some books borrowed at the end of a year are to be returned in the next calendar year, contributing to entries like “1923-1924” in the spreadsheet.

In other cases, however, inconsistencies emerge: some cards feature no year number at all, only month and day. While these dates are usually handwritten and thus understandably more cursory than the stamped ones with the neat day-month-year formula, it still leaves one to wonder how these cards were to be organized and distinguished from one another if the reader ends up being a long-term member. Other cards, such as GARF no. 564, feature dates ranging from 13 August 1926 to 16 April 1930. Ten different dates are stamped or scribbled somewhat arbitrarily across the recto side of that card. Upon close examination, one can deduce that the reader paid the deposit in 1926, paid monthly subscription once in 1928 and 1929 each, and then twice in 1930. Perhaps because the sporadic months in which he paid subscription throughout the three years never overlapped, the librarian opted for convenience rather than formality, saving material and labor resources instead of issuing a new card each year for this reader. We therefore observe a system designed to be meticulously ordered along a strict chronology (i.e. January to December of each calendar year, to be renewed every year), dissolving at the hands of librarians during actual practice.

This dichotomy invokes questions on the relationship between the library and the librarian, and the latter’s agency in shaping institutional practices. Here again we may reference Derrida, who is concerned with inscription technology’s relationship to archives. “The technical structure of the archiving archive also determines the structure of the archivable content even in its very coming into existence and in its relationship to the future. The archivization produces as much as it records the event,” wrote Derrida. Thus archiving technology - the choice of orthography, the (non-)issuance of new cards, etc. - determines the very institution of the archivable event, informing as well the conception of the future (e.g. an expectation of whether the member will be active this year).

In the absence of an official guideline amongst the existent archival materials, it could be difficult to distinguish the decision-making of the individual versus the institution. This would not be a problem for Shakespeare & Co., since Sylvia Beach simultaneously embodies both functions, yet for Turgenev Library, we remain ignorant of the identities of the various librarians. We can glean differences in their record-keeping practice through spelling, handwriting, layout style, etc., which gives us limited access into these unnamed personages who would otherwise disappear in history yet nonetheless played an essential role in émigré community-building.

If the year numbers are the most straightforward part of the data-gathering process, the most confusing piece of data is the Turgenev Library number. It was supposedly assigned by the Library to each individual member, yet approximately half of the cards in our sample (52 out of 103) lack this number; the rest features numbers ranging from 15 to 1381. It remains unknown whether the library assigned a number to everyone but simply did not record it on some of the cards (and recorded these numbers and their associated names in a logbook instead, for example, as was the approach by Sylvia Beach), or whether it was simply an inconsistent method. All numbered cards are dated from pre-1928, and vice versa, which points to a shift in administrative practices, perhaps towards relative disorganization.

Among the numbered cards, the order of the number has little correlation with chronology. For example, card no. 1046 is dated from year 1924, while card no. 1076 is dated from year 1921. It is possible that older numbers that have fallen out of use were recycled and reassigned to new readers, not unlike how Microsoft accounts would be automatically deleted after two years of inactivity. Another possibility is that each reader gets a new card during each active year, while keeping one’s original number. By this logic, the owner of card no. 1046 would have registered at Turgenev for the first time (i.e. paid the deposit) in 1921, not long before the owner of card no. 1076, and continued to be a member in 1924 under a new card.

To prove one hypothesis or the other, we need to confirm two assumptions: 1) that the library indeed issues a new card each year for active members as a convention, 2) that the library keeps the member’s number constant when new cards are issued. Therefore, let us examine the instances of recurrences (i.e. multiple cards belonging to the same individual), to see whether the Turgenev numbers are consistent, and whether the year numbers are consecutive. Within our 103-card sample, there are six pairs of “recurring” cards, each pair belonging to the same reader by name (fig. 2-2). Half of these pairs have the same Turgenev number; the rest have one or more unnumbered cards and thus cannot prove consistency or lack thereof. Most of the pairs are dated from different years, either explicitly confirmed by stamped dates or deduced by content in terms of subscription months and borrowing events. The aforementioned GARF no. 564 card, for example, contains subscription information throughout years 1926-1930, while its pair, no. 565, dates exclusively from 1930, and contains data that is supplementary or consecutive to that of no. 564. The difference of address in many of these pairs will be discussed in a later section.

Despite the limited sample size, we have roughly confirmed the two assumptions above. However, one inconsistency remains to be discussed. GARF no. 188 and 189 represent a peculiar, yet not uncommon case: they originally had the same number, yet one was crossed out and written over with a different number. In fact, in at least 7 out of the 52 numbered cards, mostly dated from 1923-1926, we observe an original Turgenev number (larger than 1,000) stamped and then crossed out, replaced by a new, smaller number written in pencil. Comparing these two sets of numbers, I deduce that sometime between 1923-1926, a systematic reorganization of membership took place, where the library sorted through all the members’ cards, removed the inactive ones, and reassigned these deactivated numbers to the active cards from recent years; thus, no. 1037 took the place of no. 311, no. 1453 became no. 548, etc. Though this revision process might be the time- and labor- consuming, it would have greatly facilitated the library’s internal record-keeping and tracking of active members.

If this is correct, then both of the above hypotheses hold true to explain the lack of correlation between the Turgenev numbers and chronology: the Library both recycled inactive numbers for new members’ use, and kept each individual’s number constant across old and new cards. Further questions could be raised on the specifics of the revision process, if there indeed had been one: What necessitated the revisions – a high proportion of inactive members that made it difficult for the administration to track the active/current membership? How did the Library determine if a user was inactive and to be removed from the system? How often did the Library conduct such revisions – was it a one-time consolidated effort, or spread out throughout a couple years? Why did the Library stop assigning numbers to members altogether after 1928? From the inventory and catalogue data, we already note that 1924 and 1928 are two crucial years: the former being a peak (e.g. publishing the first printed belletristic catalogue) and the latter being the beginning of a decline (e.g. publishing the second and last belletristic catalogue). Perhaps the Library’s membership management faced the same trajectory as its collections management, with 1924 being a time of optimization of administrative practices, and 1928 being the start of a gradual fall into disorganization.

Work Cited:

Jacques Derrida and Eric Prenowitz, "Archive Fever: A Freudian Impression," Diacritics 25, no. 2 (1995).