Information that might be useful as you manage your data.
The goal of this project is simple: to identify where contributors to Poetry and The Little Review lived and to add that information to the data the MJP has already collected for these two Chicago-based journals. With this new residential information, we can map the geographical distribution of contributors (and their contributions) and also track the relative importance of different geographical areas, like Chicago, for either journal.
Despite its simplicity, realizing this goal has involved a great deal of work—because there were over a thousand contributors to track down, because residence information was often not readily available, and because I wound up making numerous adjustments to the data as I worked out what should count as a “contributor,” a “contribution,” and a “residence,” which all become more complicated upon closer examination. The following are the main tasks that I undertook, more or less at the same time:
After I settled upon a single list of contributors for each journal and assigned to each contributor one or more residences (in files 1, 2, and 3), I reformatted the residence information in these files, and associated each contributor with a single primary residence, in order to make it easier to manipulate the data in spreadsheet form (in files 4 and 5). I also added the expanded residence and name authority information to the MJP’s “every-contributor” dataset for each journal, so users could relate all of this residence information to all of the data we’ve previously recorded about each contributor’s contributions in these journals (in files 7 and 8). Finally, I tested and proofed all eight data files while I created the materials that appear in the Data Overview and Data Visualization sections on this site.
What follows is a description of each task that I performed and the protocols that I followed to make the new data as reliable, consistent, and functional as possible.
Working with the MJP’s catalogue records for Poetry and the Little Review, I made a series of adjustments to the names and number of contributors, as well as the number of their contributions, in order to come up with a more accurate list of contributors for this project.
One small adjustment was to throw out some contributors whose items did not meet the standard of a contribution for this project. Here, I mostly I eliminated 30 or so authors from the Little Review who had written about eighty short quotes and aphorisms that were published in the early volumes of the magazine to fill empty space at the bottom of a page. Although it’s helpful to have these authors and their quotes show up in author searches of the MJP database, they mostly express the editors’ interests and values, and are not submissions by the authors who originally composed them. I did not throw out, however, more substantial items by authors who were not alive at the time of publication—and thus, like the authors of the quotes, not conscious or deliberate contributors to either journal. The most common instance of this are several series of ancient poems translated from Chinese or Italian, in both journals, where I’ve recognized both the poets and their modern translators as contributors. One implication of this decision is that most contributors from China in the magazine datasets had been dead for many centuries when the magazines were published.
A much larger adjustment that I made to the contributors list involved reconciling various expressions of a contributor’s name in either journal and also attempting to identify by name the authors who published under initials or pseudonyms. When the MJP creates a catalogue record for a contributor, its policy has always been to reproduce whatever name appears in the journal, replicating any variation or even mistake in the name rather than inferring the correct name or actual individual represented by that signifier. The drawback to this policy is that a single contributor may appear under numerous names and variants, both within a journal and between journals, which increases the number of unique contributors while decreasing the number of contributions attributed to an individual. For example, both Poetry and the Little Review followed the common convention of using authors’ full names to sign their literary contributions (like poems or stories) but only their initials to sign some of their lesser journalistic contributions (like book reviews). When a person contributed both items to an issue, which happened often, it appears that there are two contributors with one contribution each, instead of one contributor with two contributions.
The ideal solution to this problem at the MJP would be to create an authorities index that identifies all the different names used for and by the same contributor, and then add the contributor’s authorized name to the catalogue records for each of the author’s contributions. Without such a resource at my disposal, I built an ad hoc authorities list for this project, reconciling as many name variants as I could and selecting just one name as the contributor’s “authorized name” while subordinating all other name variants to it. More specifically:
Protocols for choosing an authorized name
There is not much at stake in choosing an authorized name in this project, since all that matters is that each contributor is represented by a unique identifier. But in assigning authorized names, I generally opted for the more commonly used name variant, unless the less used variant was more informative. In a very few instances, I replaced initials or an alias used in the magazine with the author’s true name (e.g., “J. Q.” with “John Quinn”) simply because the actual name helped establish the author’s identity and residence and not because it decreased the overall number of contributors.
Procedure for decoding initials and aliases and for reconciling name variants
To identify contributors who are listed solely by initials, I first reviewed my alphabetized list of contributors (derived from the MJP’s “every contributor” dataset) and checked to see if the initials correspond to any of the named contributors. When they did, I investigated the connection and reconciled the initialed contributor with the author’s full name if there was enough evidence, either from within or beyond the journal, to support their common identity: e.g., if an initialled contributor published about the same time or in the same issues as a contributor with a matching full name, just as “G. S.” and “George Soule” published items in the same issue of the Little Review. If I could not achieve a level of confidence about a contributor’s identity, I simply left the initials unreconciled.
To identify the authors behind aliases, I mostly consulted external evidence like memoirs. Although I decoded some aliases this way, there remain 51 contributors to the Little Review whom I could not identify by residence, primarily because so many inscrutible aliases populate its pages.
Most of my adjustments to the contributors list for either journal involved reconciling an author’s name variants and thus decreasing the overall number of contributors on the list; but in some few instances I instead grew the list by differientiating among authors who shared the same name. In particular, I wound up counting separately all contributors named Anonymous (and Anon), since it’s unlikely that any of them are the same person (though to be sure, I also checked the issues for overlap amongst them). To prevent these contributors from being conflated in spreadsheets and pivot tables, I decided to assign each a number: e.g., Anonymous (1), Anonymous (2), and so on. But I also investigated and counted separately other authors sharing the same signifier: e.g., two letter writers in the Little Review who go by the initials “A. S.,” since they come from two different cities (New York and Pittsburgh) with five years separating their contributions.
Finally, after I reconciled name variants within each journal, I reconciled any variation in how authors who contributed to both journals were named, generally using for Little Review contributors the name or spelling that I had authorized for that author’s contributions to Poetry.
The journals themselves were my two main sources of information about where their contributors lived: the Notes section at the back of each issue of Poetry, which typically includes residence information about each poet published in the issue, and the “Reader Critic” or letters department of the Little Review, which identifies most letter writers by location. I transcribed without question this information from either journal, unless I had some clear reason to doubt it.
Although this residence information from the journals provided the foundation for both master lists (files 1 and 2), it left many contributors still unidentified. In Poetry, book reviewers and other non-poetry contributors are conspicuously left out of the Notes section, and the Little Review lacks a corresponding section on Contributors Notes, so residence information about even primary contributors is largely misisng from the journal. However, I did a quick “sweep” of every issue of both journals to collect any information about contributor locations that was included but not foregrounded for readers: e.g., in the early issues of the Little Review, I often identified contributors as coming from Chicago if their review of a play or art exhibit in Chicago showed that the writers had first-hand experience of that event (though even here I tried to corroborate that information with independent evidence of the writers’ whereabouts).
Besides the journals themselves, other sources that I turned to included archival materials from the Newberry library’s collection, in particular two address books owned by Eunice Tietjens (a Poetry editor and contributor to both journals), one from 1913 and the other undated but possibly from the 1930s. Though I found a handful of usable addresses in these sources, the time it took to process them was ultimately not warranted by the information they yielded.
So the second most useful and efficient source of contributor residence information for this project was simply the internet. Online digitized books yielded some good information without too much effort, while Wikipedia surprisingly offered a lot of information as well about even quite minor authors. The drawback to internet searching was establishing reliable residence information for contributors during the span of years when they contributed to these two journals. To locate information from that short span of time, I turned to ancestry sites (e.g., FamilySearch), which gave me access to a host of government documents, including birth certificates, travel logs, and census information for 1910 and 1920. The drawback to the sites I used is that they focused on U.S. citizens, and often a contributor’s name would be shared by 10-20 different people—making this kind of resource useful for only those contributors with distinctive names.
Because I had 1125 contributors to track down, I could afford to do only a minimum amount of research on any one contributor. So the information I collected is limited by these time constraints. Undoubtedly, more in depth research will yield more specific and reliable residence information.
Protocols for intepreting residence
Even if I were able to collect accurate information about where contributors lived during the time of their contributions (which may be epitomized by a return address written on their submissions), I would still need to figure out how to define a “residence,” which can be fluid and open to interpretation. Some contributors, for instance, were travelling when their contributions appeared in the journals, while others moved from one residence to another over the course of their contributions, and still others had two or more residences at the same time. So I established some loose rules for what counts as a residence.
I’ve mostly construed residence as the location where contributors were living (not just staying) at the time of their contributions. Complications to this definition include:
In this project, I’ve sought to distinguish residence (where someone lives) from nationality (where someone was born, where someone comes from), and I’ve privileged residence over nationality whenever possible in the data I’ve collected. This distinction is especially important for cataloguing contributors who lived abroad, which was common during the modernist era. Accordingly, James Joyce, William Butler Yeats, and Padraic Colum were all “Irish,” but I’ve listed them as residing in “Zurich,” “London,” and “New York,” respectively, since this is where each lived (or had a permanent, stable residence) at the time of their contributions.
However, if I had strong doubts about where contributors resided during the time of their contributions, I opted instead for the author’s national identity, since identifying where someone comes from is more informative than offering nothing about them and also more accurate than offering an unsubstantiated residence. In my data files, nationality is likely standing in for residence when the country of residence for a contributor is listed without a corresponding city.
For authors whose contributions are posthumous, I made no attempt to identify when the work that appears in the magazine was composed and where the author lived at that time; rather, I simply recorded the authors’ nationality or the location where they resided through their most productive period.
Finally, if I didn’t uncover residence information for authors whose contributions were translated into English from a foreign language, I sometimes inferred that the author came from the country commmonly associated with that language; in these instances, “translated from the Italian of” = Italy. That’s hardly a reliable expression of residence, but it’s still better than nothing.
Other decisions I made for collecting and organizing residence information
Dealing with multiple locations
When my sources indicated that a contributor lived at two or more residences at the same time, I included all the residences, serially, but selected one as primary, based on the available information; marked that residence as (1); and listed that residence first, to ensure all primary residences show up when the spreadsheet is sorted by residence.
When a contributor with multiple contributions moved to one or more new residences over the course of making those contributions, I also listed these additional residences, more or less serially (from the left to right), but again privileged one over the others, marked it with a (1), and then placed it first among the residences listed for that contributor. To decide which residence would count as primary over time:
Because residence can have different meanings, and because I didn’t want to err in being either too generous or too stingy in deciding what it means to come from Chicago, I decided to construe “coming from Chicago” in two ways, one more strict than the other.
I counted contributors as “Chicagoans” (and entered “yes” for them in column 3/C) if they met any of the following criteria: