#  The Marshall IMLS Grant and Resources for Reparative Description 

 





August 07, 2023

 

 

- [ Blog ](/news-categories/blog)
 
 

 

   ![Woman and her daughter standing, with Lorna Marshall squatting next to them, on Fritz Metzger's farm](/sites/g/files/omnuum4921/files/styles/hwp_1_1__720x720_scale/public/peabody/files/128050059.jpg?itok=JOgWKQDV) 

 

 Woman and her daughter standing, with Lorna Marshall squatting next to them, on Fritz Metzger's farm [2001.29.2352](https://collections.peabody.harvard.edu/objects/details/690514?ctx=645de01a027b3ef261a6438cc8f228a3b32a801f&idx=0) Describing archival collections from marginalized and oppressed communities is hard work for archivists. It’s emotional and triggering when we are confronted with racist ideologies baked into the language used in historic materials, especially for those of us who identify with marginalized groups we see in the collections. Reparative description is the practice of deciding how to use language to accurately describe the people and history of a collection without perpetuating harm and how to use language to accurately describe and inform users.

 While there are no clear or correct answers on exactly what terms are best to use in descriptions, there are guidelines for terms not to use. Making decisions about language is iterative, and best done in collaboration with colleagues in and across institutions, and in consultation with source communities. Moreover, it is important to be flexible. A term that is considered accurate and sensitive in 2023 may be considered problematic in 2033.

 The aim of this post is to provide insight into reparative description decisions made by the Peabody Museum’s Marshall Family Archives Digitization Project team and to serve as a resource for cultural heritage stewards embarking on reparative description projects in the future by providing examples, suggestions, and an annotated bibliography.

##  Background

 I started at the Peabody Museum of Archaeology & Ethnology in December 2020 as an Archives Assistant on the Marshall Family Archives Digitization Project. Sponsored in part by the [Institute of Museum and Library Services (IMLS)](https://www.imls.gov), the project is providing online access to over 50,000 negatives, prints, slides, stereoviews, and paper documents. To read more about the IMLS project and the project team members, see [Marshall Family Archives Digitization Project](/blog/marshall-family-archives-digitization-project).

 The material in the collection comes from a series of expeditions the Marshall Family made to the Kalahari region of southwest Africa in the 1950s. Lorna Marshall, a burgeoning anthropologist, conducted field research while Laurence, her husband, organized the expeditions; their son, John Marshall, made documentary films and their daughter, Elizabeth Marshall Thomas, wrote books about her family’s travels. Their work centered on the Indigenous peoples of the regions of what is now Namibia, Botswana, and Angola.

 The collection exemplifies an early era of visual anthropology, a well-meaning effort in salvage ethnography to document the cultures of people whose lives might change because of modernization and neo-colonialism. Although the Marshall Family compensated some of the people they documented, the Family retained control over the images until donating them to the Peabody Museum in 2001.

 Over 50,000 images with extensive original item-level descriptions from the Marshalls posed unique challenges and opportunities for reparative description. They took great care to record their process for research and educational purposes, both for their own publications and for other users. Their descriptions for the images come from journals, diaries, research papers, and photograph catalog books. The bulk of the negatives and slides have item-level descriptions written by Lorna Marshall in photograph catalog books. She entered descriptions based on information gathered from the Ju/’hoansi, Ts’ixa, Herero, Bakgalagadi, and other people the Marshalls documented.

 After reviewing the Marshalls’ descriptions at the outset of the grant project, we saw that the family used 1950s American, and sometimes Afrikaans, terminology rather than the relevant Indigenous language. Moreover, the team determined that many of the descriptions contained outdated and problematic language. The project team had lengthy discussions about what constituted offensive versus reparative language, and what terms should be used. In addition, we consulted contemporary anthropologists with Indigenous contacts, who provided terms in their own language that could be incorporated into the descriptions.

 In this reparative description work, we have made use of the fields available in the Peabody’s collections management system, The Museum System (TMS). TMS, which the Peabody has customized to fit its needs, allows for different flexibilities when it comes to capturing and displaying metadata, more so than a typical archives information management system like ArchivesSpace. Together, the team selected the TMS metadata fields and formats to use that best suited project needs. Examples of this approach include fields such as Display Title (for public use), Transcribed Annotations (for text written directly on the image or its mount), Catalog Transcription (for photograph catalog preservation), Indigenous Term (for translations of Western words), and the Culture/Period Attribute (for recording a person’s ethnicity).

   ![Example of the metadata fields TMS displays](/sites/g/files/omnuum4921/files/styles/hwp_1_1__720x720_scale/public/peabody/files/picture1.png?itok=KUxzzKt1) 

 

 Example of the metadata fields TMS displays in this screenshot from the Peabody Museum's Collections Online [2001.29.261](https://collections.peabody.harvard.edu/objects/details/455734?ctx=e1ad0bc014adcc42524b9b67f09d36322c0907a3&idx=0)##  Suggestions

 1) Deciding What Language Not to Use

 The Marshall team’s first step toward using reparative description was deciding which kinds of language not to use. This included misspellings, mis-genderings, and ethnocentric language, such as terms that interpreted emotional states (like tantrum and frustrated), ones that implied positive or negative value judgements, commentary on people’s physical appearance, and outdated racial terms.

 Terms that are now considered to be outdated and offensive, which appeared often in the photograph catalogs, were flagged. These terms were retained as part of complete transcriptions within the TMS Catalog Transcription field in order to preserve the historical record but were changed in the Display Titles. Examples include terms such as Bushman/Bushmen, negro, and native. Catalog Transcriptions are viewable with a researcher account for the Peabody’s [Collections Online](https://collections.peabody.harvard.edu/collections). In this way, historical language is accessible as a subject of study and for searchability, but database users will not unexpectedly encounter language that may be triggering, especially without context.

 2) Designating Preferred Terms

 After compiling a list of all the terms and phrases not to use for descriptions, a corresponding list of preferred terms was created. This was more complicated than anticipated. We consulted with anthropologists and with community members, and reviewed archival reparative description practices, to make the best-informed decisions. Below are examples of decisions the project team made and the complications that affected the outcome.

 The term “Bushman” provides a useful example of what we contended with. In their photograph catalogs, the Marshall Family used the term “Bushman” or “!Kung” to describe many of the Indigenous people of Southwest Africa at that time. While some scholars and even some Indigenous people still use the term “Bushman,” others consider it to be offensive. “!Kung” is also thought to be outdated and more accurately refers to a language group. The term “San,” which is used in the Smithsonian Institution’s Human Study Film Archives, groups together cultures and ethnicities that are distinct. Thus, whenever possible, we used the specific name for each ethnic group, such as Ju/’hoansi, Gwi, or Naron. In cases where people of multiple ethnicities were together in a photograph, a person-first structure, suggested in [“Archives for Black Lives in Philadelphia: Anti-Racist Description Resources,”](https://archivesforblacklives.files.wordpress.com/2019/10/ardr_final.pdf) was used. If the people in the photograph are of one shared ethnicity or there is only one person depicted, the information is captured in an Attributes field separate from the Display Title.

 The decision to not use “Bushman” or “!Kung” does have repercussions. Because those are the historic terms, they have been used in legacy collections and academic literature, which means researchers rely on those terms when searching. Removing them altogether impacts searchability and access by removing the terms academia routinely uses, making it harder for current researchers to find the material from the Marshall Family Archives. Even the Library of Congress Subject Headings still use “!Kung” in the authority record for Ju/’hoansi. The project team felt it was more important to respect community members’ preference to not use the outdated terms in the Display Title, but mitigated this by including the original description written by the Marshalls in a TMS field called Catalog Transcription, which displays to users with researcher access. Thus, while elimination of the term impacts public-level searching, is a searchable field. This way, users with researcher access can still find the information using the outdated but more commonly used terms by searching on the Description field.

 While the project team decided to capture outdated ethnic terms in the Catalog Transcription field, a different decision was made concerning nicknames. In Ju/’hoan communities, people often have the same name. In Marshall Family journals, research notes, and photograph catalogs, there are nicknames to distinguish between people with the same names. It is unclear if some nicknames were given by Ju/’hoan community members or if the Marshalls made them up, as some of the nicknames embed a negative or positive ethnocentric value. Examples of such nicknames include /Qui Crooked, /Qui Short, or /Qui Navel.

   ![](/sites/g/files/omnuum4921/files/styles/hwp_1_1__720x720_scale/public/peabody/files/171710117.jpg?itok=22uMv-hu) 

 

 "Little ≠Gao" leaning against "/Qui Navel" who is talking to ≠Gao, husband of Bau and /Ghia [2001.29.22172](https://collections.peabody.harvard.edu/objects/details/774062?ctx=48ed877ad93c09b56ea172a21b442fe4cca873d7&idx=0) Removing the nicknames was debated for reasons similar to those for eliminating “Bushman.” Ultimately, we decided that removing them would make writing the Display Titles confusing for team members and confusing for users browsing the collection. Instead, the name and nickname were placed in double quotation marks to show the names were referenced from collection documentation. As it is unclear whether the Marshalls were recording previously established nicknames or providing new ones, we did not want to erase or truncate someone’s nickname without clarity about its origin.

 On the other hand, when the photograph logs contained descriptions of people’s bodies, it was extremely clear these were imposed by the Marshalls, and they were riddled with ethnocentric terms and an often-dehumanizing tone typical of earlier anthropological image descriptions. The project team felt the practice of describing bodies as objects neither aligned with our commitment to reparative description, nor was consistent with the Marshalls’ careful work of recording names and building relationships with the Ju/’hoansi. Therefore, guidelines were established to change anatomical language to more general descriptions and to talk about body positioning without reducing the pictured person to a specimen. For example, “torso” was changed to ≠Toma’s chest, or “side view” was changed to “in profile” or “seen from the side.”

   ![Gao sitting, seen from the side](/sites/g/files/omnuum4921/files/styles/hwp_1_1__720x720_scale/public/peabody/files/171740190.jpg?itok=dtNncYXB) 

 

 **Display Title:** Gao sitting, seen from the side **Catalog Transcription:** Gao 187, the young headman of Band 12 at Gam, seated. / Shows sitting position well [2001.29.25483](https://collections.peabody.harvard.edu/objects/details/778168?ctx=e753ddc6acbe1afbb8e59a0e2ae9173d55e98005&idx=0) This way, people’s humanity remains present in the Display Title, a sentiment that can be applied to all the decisions the project team made.

 3) Documenting

 Because archivists love records, we kept careful track of the decisions made about language use and preferred terms. There are many benefits to this. When questions arise, we can refer to specific language choices; new staff can onboard using the document; and work can easily be shared with colleagues and other institutions. In addition, our own decisions are now part of the history of this collection. Being transparent about the choices we have made allows staff and users, now and in the future, to understand – and evaluate – our rationale.

 The Marshall team created a Marshall Image Cataloging Guide that captures all the decisions made over the past two years, and the examples above come from that document.

##  Annotated Bibliography

 These resources below link to curated bibliographies assembled by other archivists that cover a wide range of topics under the reparative description umbrella.

 Antracoli, Alexis A., Annalise Berdini, Kelly Bolding, Faith Charlton, Amanda Ferrara, Valencia Johnson, and Katy Rawdon. [“Archives for Black Lives in Philadelphia: Anti-Racist Description Resources.”](https://archivesforblacklives.files.wordpress.com/2020/11/ardr_202010.pdf) October 2020: “Archives for Black Lives in Philadelphia: Anti-Racist Description Resources” is a staple in reparative description literature. The “Metadata Recommendations” section is of particular use as it covers the nuances of voice, style, and transparency.

 Lellman, Charlotte G. [“Guidelines for Inclusive and Conscientious Description.”](https://wiki.harvard.edu/confluence/display/hmschommanual/Guidelines%20for%20Inclusive%20and%20Conscientious%20Description) March 2, 2023: This guide created by an archivist at the Center for the History of Medicine at the Countway Library of Harvard Medical School is specific to medical archives which have their own set of legal and ethical considerations for description but are still relevant to the larger discussion of description. The sections about terminology and language are well-considered and include many examples and decision trees.

 Society of American Archivist. [“Inclusive Description.”](https://www2.archivists.org/groups/description-section/inclusive-description) March 24, 2023: This site is updated by the Society of American Archivists (SAA) under the Description Section Steering Committee. There are a variety of resources listed from case studies to presentations to articles on archival descriptive theory.

 Tufts Archival Research Center. ["Additional Reading: Potentially Harmful Language in Archival Description"](https://dca.tufts.edu/about/policies/Additional-Reading-Potentially-Harmful-Language-in-Archival-Description): This list of readings covers several areas such as bias, disability, gender, sexuality, race, and white supremacy.

 Yale Libraries. [“Reparative Archival Description Working Group LibGuide"](https://guides.library.yale.edu/c.php?g=1140330&p=8319098): This LibGuide run by the Reparative Archival Description Working Group is part of the Archival and Manuscript Description Committee at Yale Libraries. Of note in this guide is the “Guiding Principles” tab which expands on the mindset needed to approach this work. The “Recommendations” and “Blog” tabs also have more concrete examples of reparative work.

 At the Peabody Museum, this work falls under our commitment to [Ethical Stewardship](/ethical-stewardship), while the archival community uses terms like reparative or inclusive description. Either way, the goal remains the same: to respect the communities from which collection materials originate and care for their cultural heritage without inflicting further harm. This work cannot be done in a vacuum and hopefully our process will help and inform others.

 Read more about project team’s process: [Rewriting the Past: The Problem of Historic Language in Museum Collections](/blog/rewriting-the-past).

 This project \[MA-245387- OMS-20\] was made possible in part by the Institute of Museum and Library Services.

 [](https://www.imls.gov)

   ![IMLS Logo](/sites/g/files/omnuum4921/files/styles/hwp_1_1__360x360_scale/public/peabody/files/picture2.png?itok=6N0JSEZ0) 

 

 Author: Elise Riley



 

 

 



 

 

 Share on:- [     Facebook ](#)
- [     Twitter ](#)
- [     Linkedin ](#)