Modes of Organization
For seven decades the articles were organized in one singular order: by group, by primary subject code, and then in reverse chronological order. There were no indexes to provide access by date, source, or subject across groups. In fact, there was no practical way to determine the number of articles altogether, or in any particular category, although it was possible to estimate the amount of material for each ethnic group. The Newberry digitization project provides access to the articles primarily through the categories of the original editors. These are often not the categories that scholars, librarians, editors, or curators would use today, but they are constitutive of the collection, and are essential to understanding it and making good use of it.
Twenty-two ethnic groups make up the Survey as we know it. The bulk of the work focused on ten groups: Czech, which 1930s editors called "Bohemian" for the purposes of categorization, Danish, German, Greek, Jewish, Lithuanian, Norwegian, Polish, Russian, and Swedish. Some groups include a relatively small number of articles, particularly Albanian, Serbian, Slovene, Chinese, and Filipino. The Slovene articles are not included in the current digital collection for reasons of availability of scanned images.
The categories of the digital edition are those of the original Survey editors. They were willing to use their own judgment in applying their ethnic categorizations flexibly to accommodate linguistic, national, and religious identities, and they occasionally adapted their editing practices to adjust to the limitations of the categories they imposed. One of the most notable examples of this is the category of "Spanish," which described Spanish-language sources from communities based in the Americas rather than the nation of Spain. A majority of the articles in the Spanish section are qualified as "Mexican," though there are scattered instances of other secondary qualifications, such as "Costa Rican," "Columbian," "Cuban," and even "Brazilian." Similarly in the Serbian section there are a couple of articles qualified as "Montenegrin." The 1930s editors were neither consistent nor thorough in applying these secondary terms even according to their own understanding of them, and so the present digital edition does not include these terms in the interface for browsing.
The subject codes are fundamental to the organization of the Press Survey. The categories and terms are unique to the project, and were developed under the supervision of Bessie Louise Pierce, a historian at the University of Chicago who was conducting extensive newspaper research for what became her three-volume History of Chicago (1937-1957). The translators and editors who worked on the project in the 1930s consulted a manual that listed the codes and explained what they were intended to mean, and how they should guide the selection of articles for the Survey. The present digital resource includes a digital transcription of this manual, and subject codes listed at the end of each article include links to the relevant paragraphs in the code manual.
The subject scheme is arranged hierarchically with five top-level groupings: I. Attitudes, II. Contributions and Activities, III. Assimilation, IV. Representative Individuals, and V. Miscellaneous Characteristics. The first few of these are elaborated in several layers of subcategories. Only the lowest levels of the hierarchy produced terms that were applied to individual articles. That is, the first code applied toa individual articles is I A 1 a, "Attitudes / Education / Secular / Elementary, Higher (High School and College)." Roman numeral I, "Attitudes," was not applied to single articles on its own. Roman numeral IV, "Representative Individuals," has no subcategories.
WPA editors assigned each article to a primary ethnic group with a primary subject code, and this has been for decades the most fundamental mode of organization of the Survey articles. Editors in many cases assigned additional subject codes, apparently intending them to be indexed and cross-referenced. Sometimes the editors assigned secondary codes in the context of other groups than the primary group assigned to the article. There are examples of this among the Scandinavian groups in particular.
In this digital edition, the filtering and browsing tools at the left show parenthetical counts of articles in each subject code within the constraints of other active filters. These counts reflect only primary subject code assignments. The full set of results returned, however, includes instances where any selected codes appear as secondary codes. The parenthetical counts at left thus offer an opportunity for some preliminary analysis of a set of results in the aggregate, while respecting the special status of the primary codes, without giving up the ability to find articles based on their secondary codes.
The Foreign Language Press Survey is primarily a collection of translations of newspaper articles that were originally published in Chicago in languages other than English. Editors in the 1930s would assign a translator a particular set of newspapers to work through. Translators proposed articles for inclusion based on the selection standards of the project. The Survey did not translate every article in an issue. They might select just one, or none. They did not include complete runs of every newspaper, nor where they comprehensive in including the full range of papers published at any particular time. It is perhaps impossible to know from the Survey itself what they omitted, or why. Researchers looking for particular newspapers or particular events may be disappointed, and may need to be cautious on how they interpret what cannot be found, as well as what can be.
On the other hand, Press Survey editors made their selections on the basis of a positive vision of what they hoped would be of use to future researchers. As a result, they included materials that go beyond the expected foreign-language newspaper sources. They transcribed articles from English-language papers that offered information on ethnic communities. They consulted books, sociological journals, scrapbooks, yearbooks, event programs, and publications and record books of ethnic societies of various kinds. They also occasionally transcribed correspondence of ethnic community leaders. Finally, they conducted interviews with individuals, and typed their notes in the same form as their newspaper transcriptions.
The original Press Survey sheets were never organized by source, and so the editors never fully confronted the demands of standardizing styles and references. The current digital edition has sought to facilitate browsing by consolidating references to newspaper names, generally adopting the style used most frequently by the editors in the 1930s. A few new categories of source have been introduced in order to consolidate many of the non-newspaper sources. These appear in the source list in square brackets, and include association documents, correspondence, information attributed to Press Survey project staff, interviews, and what the original project itself termed "miscellaneous material," often manuscript records and ephemera relating to ethnic associations held in the private collections of community members.