Subjects - smith-special-collections/sc-documentation GitHub Wiki

Subjects

When creating a new subject entry in ArchivesSpace, you must always select the Source.

  • If the source is a controlled vocabulary, such as Library of Congress Subject Headings, a Getty thesaurus, etc. also add the authority ID. The ID has a different name across vocabularies, but is generally at the top of the entry, with ID or identifier somewhere in the name. For all controlled vocabularies except Homosaurus, this is the full URI. For subjects from Homosaurus, this is the ID only.

  • When adding terms from Getty’s Art & Architecture Thesaurus (AAT), capitalize the first letter of the first word of the term when creating the subject entry in ArchivesSpace.

This page contains two sections on subject analysis; click on the links below to jump to either section.

Introduction to the SSC categorization process

Anti-oppressive subject analysis

Introduction to the SSC categorization process

Goal of this document: 

This is a guide to doing local subject analysis for the Sophia Smith Collection. After reading this, staff should have the information they need to move forward with assigning terms, removing terms, or suggesting new core local terms. There are plans to expand this guide in the future so that it is applicable across Smith College Special Collections three repositories.

The purpose of this resource is to insert a step between the thousands of subjects applied to collections and broad statements about collecting strengths. The categories should allow users to understand what we collect at a glance, and connect to every collection about a certain topic. 

Background Information

Categories fall roughly into two camps: 

  • past collecting priorities, which are larger but not something we are looking to expand right now 

  • current collecting priorities, which tend to have fewer collections, but which will grow in the future.

Assigning categories is an additional step to the subject analysis process; it does not replace more traditional subject analysis.

The main access point for these categories is the Sophia Smith Collection homepage. There are 8 groupings of categories: 

  • Literature and writing

  • The arts

  • Power

  • Social and political movements

  • Social roles

  • Women and religion

  • Identities

  • Work

Each grouping has between 1 and 18 categories, making a total of 58 categories. Except for about 50 outliers, every collection in the SSC has been assigned to at least one of these categories, although most collections have been assigned to more than one. 

  • Categories are selected subjects that describe most of the SSC. They are assigned to collections in ArchivesSpace the same way as any other subject.

  • All the categories can be viewed on the Category list spreadsheet, along with their ArchivesSpace identifier, and if applicable, their scope notes and aliases on the SSC homepage.

Assigning categories

  • A category should describe at least 20% of the content of the collection to be applied. This follows best practice when using LCSH.

  • Identity and social roles categories are different from the other categories, and do not need to describe at least 20% of the collection. 

    • The identity categories are assigned based on how the creator of a collection identified themself. These identities are also assigned to records of organizations that are run by/for a certain identity. The Native American Women's Health Education Resource Center records, for example, are tagged as Native American women. 

    • The social roles categories include Daughters, Female friendship, Girls, Motherhood, Older women, and Significant others. Every individual in the SSC is at least one of these, but these categories are only assigned if the individual's experience in that social role is significant and well documented in their papers.

  • If a new collection does not fit into any of the categories, it should be entered into the  No-category collection tab of the Category list spreadsheet.

  • If you think a new category should be created, you should enter it into the New category proposals tab of the Category list spreadsheet, along with the collections that the proposed category describes. There are some topics the SSC plans to collect in the future that currently represent too few collections to be made into categories, so new categories will certainly be created in the future. For example, collections relating to the full civil rights of trans and gender non-conforming people.

  • If you think a collection is missing a category, or a collection should not have a certain category assigned to it, then ask yourself these questions. 

  • Does the category cover at least 20% of the collection?

  • If an identity category, does the subject of the collection identify as this identity?

  • Is there enough documentation of this category for someone interested in the topic to research this collection?

  • If you have answered these questions and still believe a collection should have or should not have a certain category, then you may in good faith make the change.

Anti-oppressive subject analysis

Background

LCSH does a bad job of classifying marginalized people. If a white woman wanted to research her community's history at a library using LCSH, she would have a far easier time, and feel far more affirmed in the process, than a Latina woman. This amounts to an injustice, and is why anti-oppressive subject analysis is so important, especially at the Sophia Smith Collection, which is meant to honor marginalized history. We don't want our finding aids to undermine our mission.

Smith College Special Collections should use community created thesauri when possible. In cases where we don't have access to those, we should take care to research the terms a community uses. The gold standard in anti-oppressive subject analysis is for people in a community to write the subjects that describe their community. Defining a community from the outside looking in may eliminate information that is important, and not being of the community, the definer wouldn't know it. This is what has happened at the Library of Congress over and over again. When people create subject headings that define their own community, not everyone will be happy, and important context and information may still be lost, but at least the writer made an informed decision.

Guidelines

Consider the following to determine whether a term in  LCSH is insufficient or should be replaced in our records:

  • Is the term outdated, offensive, or dehumanizing?

    • Example of an outdated term: 'Transsexuals.' Transsexual was once a relatively commonly used term, but far fewer people prefer the term today. The term should not be used for anyone who does not explicitly identify with it.

    • Example of an offensive term: 'Prostitutes.' Sex workers' groups have made it clear that 'prostitute' is a demeaning term and that the term sex worker is preferred.

    • Example of a dehumanizing term: 'Frontier and pioneer life -- West (U.S.).' The implications of the terms frontier and pioneer are that people did not live in the West before white colonists arrived. The myth of the empty American frontier denies the full humanness and presence of Native Americans.

  • Is the term not what the people featured in the collection would use?

  • Is the term not what most researchers would use to search for the topic?

In general, be especially thoughtful and considerate when working with subjects describing the marginalized. The farther away from cis, white, straight, middle-class, man a term gets, the worse LCSH will generally do.

Recommended Thesauri

MAIN Revision spreadsheet - for Native American terms

  • If an LCSH term is not in the MAIN spreadsheet, then follow their guidelines

    • For example, the LCSH term Dakota Indians is changed to Dakota by MAIN. I follow this by replacing Mohawk Indians with Mohawk. 

  • As with any other term, look up what the nation/group prefers to call themselves. Instead of Iroquois Indians, use Haudenosaunee. Include a scope note explaining why you use the term you use, and the date you created the subject.

Homosaurus - for LGBT terms or sexual terms

  • Don't use the search function, it's not always reliable. Instead, command-F search for the term you are looking for in their master list (linked). 

  • If the term you are looking for is in Homosaurus, use that.

  • If the term is not in Homosaurus, model the new term on their language. For example, they don't have any terms about students. So to replace the LCSH term "Gay college students" (which was really about queer college students), I used "LGBT college students." 

    • If you have to do this, consider suggesting they add the term you create!

If there is no community thesaurus

  • For most other terms, we are more or less on our own---but if you find a community created thesaurus that can be relied on to have relatively up to date vocabulary, please do add above! 

  • When making a new term not set by a thesaurus, follow these guidelines.

  • Our goal is to use the terms people want to be called by. Even if you think you know, you should do research to find the consensus. Use the best term you find, then write what you found in a scope note. [include scope note here]

To learn more about anti-oppressive subject analysis and the problems with LCSH, here is a reading list.