Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Enable Blacklight to Discover Translated Works by the Source Language of Translations #4634

Open
minjiec opened this issue Dec 12, 2024 · 1 comment
Labels
feature Updates on existing feature or a new feature request

Comments

@minjiec
Copy link

minjiec commented Dec 12, 2024

User story

Researchers, educators, and translators need to locate translated works, specifically by the source language from which the work is translated.

For example, a researcher may want to survey publications on philosophy translated from German into English in year 2023;
An instructor may want to locate titles (in a specific genre such as fiction or juvenile fiction) translated from Japanese into English, so that they can make a recommended reading list for students who are learning Japanese language and culture;
A translator may want to harvest a list of queer novels translated from French, or a list of poetry translated from Persian, or a list of children's literature about the Holocaust translated from Hebrew, to examine "biases and gaps in the representation of a specific discipline, country or community" (Fricker 2024).

Current barrier

No library discovery system currently supports this kind of queries, even though catalogers have diligently entered the information in bib records, in the form of 041 1# $h (source language of a translated resource). There is no way to search the data.

Reference librarians’ imperfect solutions

In order to assist users’ need for locating translated works, reference librarians occasionally offer search tips. Because none is leveraging the most precise information embedded in 041$h, those methods are not only cumbersome but tend to yield incomplete, outdated, and inaccurate results. For example–

  • Harvard: How do I find all existing translations of a book? - Ask a Librarian

  • NYU: How to Find Translations into and from English - Research Guides at New York University

  • University of Richmond: Identifying Translations

More user stories shared by selectors and CaMS members

The proposed feature would be particularly useful when I answer reference questions about translations from languages I am less familiar with. I may not know the print reference sources of those languages as I would with Chinese, for example. (Martin Heijdra, Director of the East Asian Library)

The UNESCO Index Translationum, which I often relied on for finding translations, is woefully outdated in contents, technical standards, and search functions. I often have students asking me for secondary sources by Chinese scholars in English. A search feature like this would be a great help for finding these. (Joshua Seufert, Chinese Studies Librarian)

I could see a great value in the search function in assisting users with tasks like researching materials in source and target languages, locating bilingual books, or even doing translation studies involving statistics. (Flora Kim, Metadata Operations Specialist)

That feature would be great. I have similar experience with users to Joshua's, especially with undergraduate students from non-EAS departments. I just had a student from a history seminar, who wanted to do "analysis of primary sources," but she really meant English translations of primary sources. (Hyoungbae Lee, Korean Studies Librarian)

This project would be a great help for scholars in comparative literature or translation studies. (Setsuko Noguchi, Japanese Studies Librarian)

Concrete example

Example 1.

A mock-up view:

Advanced search-mock-up0
Figure I. Retrieve all books on the topic of philosophy, translated from German into English and published in 2023.

Example 2. Search bilingual books

The proposed feature will support searches for bilingual books nicely, which may interest language instructors in particular.
Not all bilingual books have the word “bilingual” in their bibliographic records, but give clues in notes like–

Parallel text in English and Chinese.

Texts in English and Latin on facing pages.

Once source language is indexed and made searchable, users can retrieve bilingual books more reliably. Take the example of Figure I above, to narrow down to titles in bilingual versions, users can simply filter the search results further by the language facet “German” (i.e., same as the source language), in order to retrieve titles in both German and English.

Implementation notes, if any

Plan A

  1. For field 041 that has "1" in the first indicator, index the following subfields--
  • $h (original language of the translation)
  • $n (original language of libretto)

as the source language of works in translation.

The language codes in these subfields are the same MARC Code List for Languages as applied in the fixed fields 008

  1. Make "Source language of translations" a drop-down list as one of the limits in Advanced Search

Plan B

Plan B distinguishes between the original language (source language) and the pivot language (aka bridge language, intermediary language, relay language) of a translated work to accommodate more nuanced searches.

  1. For field 041 that has "1" in the first indicator,
    Index $h and $n as the source language of works in translation.
    Index $k as the intermediate language of the translation

  2. Make "Source language of translations" a drop-down list as one of the limits in Advanced Search

  3. Add a checkbox “Also searches intermediary languages”

Advanced search-mock-up2
Figure II. A checkbox for intermediate language
When checked, it searches for translated titles for which German is either the original language or the intermediate language.

A word about the precious real estate of the "Advanced Search" page

I understand that we must limit the number of facets/filters to the smallest essential number, and adding an additional drop-down choice for "source language" would potentially clutter the space. If that would be a concern, one option might be to swap with a less utilized filter or a filter that can afford to be dispensed with due to its availability elsewhere.

For example, I notice "Access" and "Format" have double presence on both the homepage of the catalog and the Advanced Search page. Are these filters in the latter space used heavily? If not, would it suffice to make them available in the homepage/search results page only, thus making room for other features?

Left: Facets on the homepage of the catalog; Right: Advanced Search page

Facet-and-filters
Figure III. Facets and limits

Further explanations of MARC:

The second indicator of 041 does not matter. It can be either blank or 7. In the latter case, $2 would be present.
041 1#
041 17 $2 iso639-3

Reference:

Fricker, Christophe. "Librarians, OCLC and Clarivate: Discover translation!" Linkedin, November 21, 2024, https://www.linkedin.com/pulse/librarians-oclc-clarivate-discover-translation-christophe-fricker-mmf3e/

Proposed by:

  • Minjie Chen
  • Don Thornbury

With input solicited from:

  • Flora M. Kim
  • Hyoungbae Lee
  • Joshua M. Seufert
  • Maria Gorbunova
  • Martin Heijdra
  • Setsuko Noguchi
  • Yang Wang
@minjiec minjiec added the feature Updates on existing feature or a new feature request label Dec 12, 2024
@minjiec
Copy link
Author

minjiec commented Jan 14, 2025

Update 2025-01-14: based on analysis of 041 $h $n $k values, as many as 257k titles (PUL holdings only; not including scsb titles) that contain translations will become retrievable by this new feature.

Here is a breakdown of the source language of translations. Top ten are--

  • English
  • French
  • German
  • Latin
  • Italian
  • Russian
  • Spanish
  • Greek, Ancient (to 1453)
  • Arabic
  • Chinese

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
feature Updates on existing feature or a new feature request
Projects
None yet
Development

No branches or pull requests

1 participant