A Model for the Digital Edition of Inventories of Enslaved Persons

Edition Contents

• List of People Enslaved by Joana de Souza, Campos, Brazil, August 31, 1872
• List of Fugitive Enslaved Persons Held in Cárdenas, Cuba, August 1, 1858
• List of Enslaved Persons Prepared by the Executor of the Will of Abraham Dupont, Matanzas, Florida, December 30, 1857
• Census of Enslaved Persons, Freeborn Children, and Fugitives, Sopetrán, Antioquia, Republic of New Granada, 1845
• List of Individuals Enslaved at the Beaugé Plantation, Grande Plaine, Saint-Domingue, December 25, 1778
• Letter from Benjamin Harrison to George Washington, Virginia, March 31, 1783
• Index of People

This micro-edition explores a model for creating digital editions of the inventories of enslaved persons that appear in censuses, wills, and other colonial-era and nineteenth-century documents from throughout the Americas. We seek to address two central problems in the transmission of these texts: how to edit the censuses as both textual artifacts and relational data, and how to handle them in a way that acknowledges and does not amplify their dehumanizing effects. We also attempt to establish a flexible paradigm that can account for the diversity of documents which contain enumerations of enslaved people and which appear in multiple languages, involve structural differences, and include varying categories of information. Our micro-edition presents six texts in four languages through a consistent approach that could be adapted or expanded by other projects.¹

§Documents Versus Data

Between the sixteenth and nineteenth centuries, inventories of “human property” were prepared in various parts of the Americas. Some appear in commercial documents related to the transportation and sale of enslaved people, others in wills or probate instruments, and others in bureaucratic records regarding the collection of taxes and processes of manumission. They frequently indicate the name, sex, and age of each person, and sometimes also the economic activity assigned to that individual, as well as descriptions of physical characteristics and a monetary valuation.

These documents possess value in various ways. They provide information that can address gaps in our understanding of slavery and the experiences of Africans and Afro-descendant populations, contributing to the reconstruction of narratives that have been overlooked, distorted, or erased. They allow people living today to better understand the history of their families and communities. They testify as well to the relationship between written culture and oppression in New World societies.

Although such lists have served as sources for historical scholarship, they are seldom edited and transmitted as documents in their own right. Most commonly, the data they contain are extracted and reformatted as databases. This is the case, for instance, with two landmark projects related to the history of slavery: Slave Voyages (formerly the Transatlantic Slave Trade Database) and Slavernijverleden (Slavery Past).²

This method avoids some of the costs and complexities that can accompany a digital editing project, particularly when large in scope, but it presents at least two disadvantages. The first relates to access: while the data can be queried and displayed, the sources themselves frequently remain available only to those with the resources and the credentials needed to access the physical archives where they reside. The second deals with accuracy and interpretation: when we extract the data from its discursive context, meaning is inevitably lost or distorted. This situation is particularly problematic when the extracted data has been deliberately modified or remediated in ways that users cannot interrogate without access to the original documents.³

This micro-edition therefore proposes a model for editing enumerations of enslaved persons in a way that allows them to be presented in their original format as linear texts, as well as randomizable data that can be reorganized and queried. We propose an approach to markup that understands the documents in a dual fashion: as discursive texts and as collections of interoperable, manipulatable records.⁴

§Honoring the People Enumerated

As many scholars have observed, finding humanity in the textual, material, and visual history of slavery is a frustrating enterprise, with data about specific people often fragmentary, incomplete, or nonexistent. The memory of most enslaved people has been lost to the archival erasure that Michel-Rolph Trouillot describes in Silencing the Past. As Trouillot notes, the silencing of the past can occur in four stages in the production of history: “the moment of fact creation (the making of sources); the moment of fact assembly (the making of archives); the moment of fact retrieval (the making of narratives); and the moment of retrospective significance (the making of history in the final instance).⁵ In all these phases, the history of enslaved people has arguably been obliterated.

As Olivia Barnard et al. explain in the introduction to their recent micro-edition “Kinship & Longing: Keywords for Black Louisiana,” the information that we do possess about enslaved people is often articulated from the worldview of the society that enslaved them: “[f]inding stories of Black life in slavery’s archive is not easy. Enslavers wrote documents from their own perspective, often silencing, ignoring, and purposely erasing Black humanity.”⁶ In this sense, the archival record demonstrates not only how enslaved people experienced colonial and white supremacist violence during their lifetimes but also how the traces of their existence were subject to textual violence, often at the hands of their enslavers.

The documents we edit in this project are inherently dehumanizing, portraying people as quantifiable property and reflecting the destructive logic that Katherine McKittrick refers to as a “mathematics of unlivingness.”⁷ McKittrick argues that “documents and ledgers and logs” that recount the brutal history of slavery are paradoxically both a point of departure for understanding the experiences of Africans in the Americas and mechanisms for denying their condition as living people. They do so by “affirming the knowable”—black people as objects—and by “disguising the untold”—black people as human beings.⁸ “This is where we begin, this is where historic blackness comes from: the list, the breathless numbers, the absolutely economic, the mathematics of the unliving.”⁹

McKittrick asks how to engage with these materials without simply replicating the violence and objectification they represent. “How do we think and write and share as decolonial scholars and foster a commitment to acknowledging violence and undoing its persistent frame, rather than simply analytically reprising violence? How do we ethically engage with mathematical and numerical certainties that compile, affirm, and honor bits and pieces of black death?”¹⁰ The problem for McKittrick is not only about reconstructing the past but also about understanding the force that the archives of slavery exert on the present: “How do we come to terms with the inventory of numbers and the certain economic brutalities that introduce blackness—the mathematics of the unliving, the certification of unfreedom—and give shape to how we now live our lives?”¹¹

These documents reflect as well the everyday cruelty of slavery at work, a dynamic Christina Sharpe refers to as monstrous intimacies. As she explains, the term encompasses “a set of known and unknown performances and inhabited horrors, desires and positions produced, reproduced, circulated and transmitted, that are breathed in like air and often unacknowledged to be monstrous.” Sharpe points to the way that we are “inured” to slavery’s “force” and emphasizes the need to explore “slavery’s inherited and reproduced spaces of shame, confinement, intimacy, desire, violence, and terror.” While the inventories we edit may only at times hint obliquely at the abuse and trauma that Sharpe references, the “mundane horrors” that she describes are an inevitable backdrop to the texts.¹²

Despite the painful realities they portray, these materials hold value by providing witness to the existence of specific individuals and families. They allow us to name the names of the enslaved, an activity that Walter Hawthorne et al. in their introduction to the recent special issue of Digital Humanities Quarterly assert is about “remembrance, connections, relationships, and identities.” As Hawthorne et al. note, digital projects that list the names of enslaved people respond to the call within social history “to create a more inclusive history of the United States and the broader Atlantic world.” While the creators of such work face ethical considerations around publishing the names of the enslaved online—the focus of the special issue that Hawthorne et al. edit—they “are addressing the issue of the anonymity of enslaved people in historical studies and reconceptualizing the history of slavery.”¹³

We believe that enumerations of enslaved persons, like those we handle here, offer opportunities to counter the erasure of African American narratives. To the creators of these documents, the people enumerated were property and their family and community connections were incidental or unimportant. To users of the documents today, however, those names can help to connect those people to larger narrative arcs and position them as individuals with their own stories and as members of larger human networks. We believe that the upholding and centering of these names, even if given by colonizers in some instances, is a way to restore a glimpse of humanity.

Numerous digital projects today seek to carry out such work. In “Kinship & Longing: Keywords for Black Louisiana,” Barnard et al. reconstruct stories of specific individuals based on colonial-era archival records in French and Spanish, seeking to “offer researchers, teachers, public historians, and artists access to the centuries-long freedom struggle Black communities have been engaged in.”¹⁴ In the digital companion to her book Unsilencing Slavery, Celia Naylor and her collaborators make the names and relationships of people enslaved at the Rose Hall Plantation in Jamaica accessible and visually engaging, constructing a site that functions “as a memorial, as a site of mourning and grieving, as a gesture of gratitude and appreciation, and as a catalyst for the ongoing recognition.”¹⁵ Breadnut Island Pen correlates the acts of sexual violence that enslaver Thomas Thistlewood recorded in his diary between 1750 and 1786 to a historical map of his Jamaica plantation as a way to “disrupt the visual tranquility” of the cartographical material and create a “record of a landscape stippled with acts of coerced sex,” focusing on the experiences of thirty-four women named in the journal.¹⁶

In transmitting texts like those we edit here, however, editors risk—as McKittrick warns—reproducing the harms that the original documents embody. In their introduction to “Kinship & Longing,” Barnard et al. point to such concerns, noting, for instance, the difficulties inherent in needing to assign unique identifiers to the people in the documents: “Numbering […] raises a series of critical questions around the reiteration of archival violence, datafication, and surveillance of Black life.” Similar concerns exist around how we mark up semantic aspects of the texts and the information they contain, as the language of our markup may embody assumptions that unintentionally reinforce the racist and objectifying rhetoric of the documents.¹⁷ We must remain vigilant to the possibility that the tools, standards, and editorial choices we employ perpetuate the objectification of the people we seek to honor.¹⁸

At the same time, we must balance a desire to remediate harm with the need to provide faithful testimony of the texts as historical facts. While we are obliged to critique our approach to handling the information they contain, we likewise should respect the documents as evidence of painful historical realities. Attempts to ameliorate the cruelty of the discourses they contain should not compromise the ability of our editions to contribute to critical reflection on the past and the reckonings that are needed in the present.

In this micro-edition, we propose an approach to editing these enumerations that faithfully transmits the documents while not exacerbating, through our own editorial choices, the racist logics and colonial hierarchies they represent. In encoding the documents as textual artifacts, we seek to present them in a straightforward, transparent fashion, letting the texts speak for themselves and taking a minimal role in marking up the painful and tragic details they at times contain. In constructing a separate personography, we take a more active role in authoring an editorial artifact that seeks to dignify the people involved by registering their information in a structured fashion, focusing only on data that we believe serves to locate each individual as a human being in a specific time, place, and community, without incorporating the dehumanizing rhetoric we find about them in the documents themselves.

§The Six Sample Documents

The six documents we edit all contain enumerations of enslaved people, but they respond to differing historical circumstances and present distinct categories of information. All the documents provide names and ages, and some indicate the economic activities associated with people. Some include physical descriptions, appraised values, and other data. Several assign a number to each individual that in some cases is a sequence number relevant only to the census and in others may have meaning beyond that context. One of the documents groups individuals by family. The six documents include information about a total of 528 individuals.

We provide below a summary of each item. We assign standardized titles in English and refer to each document in this introduction using a short reference consisting of a place name followed by the year (indicated in parentheses below).

§List of Individuals Enslaved at the Beaugé Plantation, Grande Plaine, Saint-Domingue, December 25, 1778

This document in French (Saint-Domingue 1778), dated December 25, 1778, is an accounting of 149 individuals enslaved at the Beaugé Plantation in the Grande Plaine region of Saint-Domingue.¹⁹ The list appears to have been prepared following the death of the male owner of the plantation, as his heirs are referenced in the title. For each person, the document indicates a name and age, and for most, signals a trade or economic activity. The list begins with individuals grouped by trade, with seven carpenters, five wheelwrights, three blacksmiths, eight coopers, and seven masons. Thereafter that structure is abandoned, with a less-ordered enumeration that includes descriptions of personality traits and physical conditions, as well as accounts of past illnesses and injuries that left some disabled or limited in their capacity to work.

§Letter from Benjamin Harrison to George Washington, Virginia, March 31, 1783

In this letter (Virginia 1783) Benjamin Harrison, at that time governor of Virginia, communicates with George Washington during the winding down of the Revolutionary War.²⁰ He writes at the end of March 1783, during the armistice that followed the decisive American victory at Yorktown and five months prior to the signing of the Treaty of Paris in September. Harrison expresses concern over the future of the young nation and requests the recipient’s assistance in recovering the enslaved persons of whom he has lost control during the fighting. The letter is accompanied by a list of fourteen individuals that the author believes are in New York, with names, descriptions of their physical characteristics and personalities, as well as indications of the skills for which the writer viewed their recovery as essential to his own interests.

§Census of Enslaved Persons, Freeborn Children, and Fugitives, Sopetrán, Antioquia, Republic of New Granada, 1845

This census (New Granada 1845) corresponds to the town of Sopetrán in the Antioquia region of the Republic of New Granada (today Colombia).²¹ The document is one of a series from throughout the nation that responds to an executive decree from 1842 that required an annual accounting in each parish of enslaved people, their children born into freedom, and those who had fled their enslavers. The 1842 decree sought to control the gradual process of abolition that began with the libertad de vientres of 1821, a law that declared the children of enslaved people to be free from that point forward. The Sopetrán census registers thirty-four enslaved adults, thirty-seven freeborn children, and two fugitives. Each person’s name, age, and gender are indicated.

§List of Fugitive Enslaved Persons Held in Cárdenas, Cuba, August 1, 1858

This text (Cuba 1858) in Spanish is an accounting of seventy-eight fugitives held in a municipal holding facility.²² The document indicates a date of capture of each individual, along with their name, date, and defining physical characteristics. It specifies a “nationality” for each person, with values referring to African regions or tribal groups, like congo and mandinga, along with other values including asiático (Asian) and criollo (creole). The list also indicates the enslaver who claims ownership of that person and the area of the island in which that enslaver resides, except in situations where the individuals in question refused to divulge that information.

§List of Enslaved Persons Prepared by the Executor of the Will of Abraham Dupont, Matanzas, Florida, December 30, 1857

This inventory (Florida 1857) of 169 enslaved people, dated December 12, 1857, was prepared following the death of Abraham Dupont, an enslaver in St. Johns County, Florida.²³ The document groups people by families, assigns an “account” or inventory number to each, and stipulates name, age, and “capacity” or economic activity, and includes a column labeled “receivable” that is blank for all individuals except four, for whom it is used to indicate miscellaneous facts or presumed disabilities like “one-eyed” or “right thumb off.” This census includes a monetary value for each person.

§List of People Enslaved by Joana de Souza, Campos, Brazil, August 31, 1872

This item (Brazil 1872) is an inventory of forty-five enslaved people at the property of Joana de Souza in Campos, Brazil, dated August 31, 1872.²⁴ This document assigns to each person a registration number, a sequence number, and specifies name, color, age, marital status, birthplace, family relationship (whether illegitimate, legitimate, natural born, etc.), fitness for work, and economic activity, along with other observations. Notes in the final column associate some individuals into family groups. The document is written on a preprinted form that references a law passed the year prior that required a census of enslaved people and the free children of enslaved women.²⁵

§Our Approach to Marking Up the Documents

Our approach to marking up the documents is at once customary and experimental. With some minor exceptions indicated below, we use elements from the Text Encoding Initiative (TEI) to encode the information in ways that are conventional. Our project departs from standard practice, however, in the mere fact of using TEI-XML to do this markup, as the type of tabular documents we handle are seldom the focus of editorial projects. The closest analogy might be found in endeavors that deal with nonlinear data like financial records and bibliographic data. though there are obviously distinct challenges for this project, since it addresses tabular data about human beings.²⁶

For each document we have created an XML file bearing a name in the following format: place-year.xml. The metadata section of each TEI document (>teiHeader>) includes the standardized English title indicated above,as well as the metadata about the source of the manuscript (in <sourceDesc>). We indicate date and language of each text in <profileDesc> using, respectively, <date> (within <creation>) and <language> (within <langUsage>).

§Structural Aspects

We use <div>, <head>, and <p> to mark sections, headings, and paragraphs, respectively, and <pb> and <lb> to indicate the beginnings of pages and lines.

We encode the lists of people, which generally follow a tabular format, with <table>, <row>, and <cell>.²⁷ We use a value of “label” for @role on the <row> elements that contain column headings.

§Semantic Aspects

We mark up the names of the people enumerated using <persName> with a value for @ref that points to the corresponding <person> element in our personography.

<persName ref="#iep0100">George</persName>

Because the personography provides a tool for interacting with the data related to the people listed in a structured way, we do not mark up other semantic aspects related to those individuals in the documents themselves. Granular description of the data in the texts themselves could be useful as a way to clarify the nature or purpose of specific pieces of information, but in the case of our sample documents, the tabular structure removes most potential ambiguity. For the purposes of this prototype, therefore, we do not believe that marking up the texts themselves provides additional value.²⁸

We mark up the names of other people, including those of the enslavers and scribes mentioned in the documents, with <persName>, but without a value for @ref, as they are not gathered in the personography. Additionally, we mark up names of places with <placeName> and dates with <date>.

§Regularization

We transcribe the original texts reproducing the spelling and use of capital letters, abbreviations, and punctuation of the originals, and through our markup, carry out limited regularization.

We use <choice> with <orig> and <reg> to indicate original and normalized spelling, case, and punctuation.

<choice><orig>esclabos</orig><reg>esclavos</reg></choice>

We employ <choice> with <abbr> and <expan> to expand abbreviations.

<choice><abbr>Inft</abbr><expan>Infant</expan></choice>

In the case of names, we regularize to forms that are standard today, when possible, but in case of doubt, leave the original forms unaltered.

<choice><orig>Chabes</orig><reg>Chávez</reg></choice>

We use <choice> with <sic> and <corr> to correct apparent errors in the original.

<choice><sic>Flicidade</sic><corr>Felicidade</corr></choice>

When quotation marks or id. (idem) are used to indicate that a value in a table is the same as in the previous row, we reproduce that shorthand in the documents, but provide explicit values in the personography.

§Translations

For the documents that are in French, Portuguese, and Spanish, we encode English translations alongside the original language in the same XML files by utilizing @xml:lang on <head> and <p>.

<head xml:lang="es">Primera clase</head>
<head xml:lang="en">First class</head>

Within table cells (<cell>), we enclose the original and translated versions in parallel <seg> elements with corresponding values for @xml:lang.

<cell>
  <seg xml:lang="fr">Nom</seg>
  <seg xml:lang="en">Name</seg>
</cell>

We have chosen this approach instead of creating separate files for the translations because much of the material in these documents, such as names and numbers, often do not require translating or cannot be translated. We believe, therefore, that interspersing the translations as needed into the original files is a more efficient approach in the case of the documents we are editing.

§The Creation of the Personography

In creating a personography, we have taken an approach to handling metadata about people that is common in editorial projects, and we have followed most closely the example provided by Barnard et al. Our personography exists as a separate XML file named people.xml containing information for all the individuals mentioned across the six items in our sample. The file contains a high-level <listPerson> into which we nest another <listPerson> for the people associated with each document. When individuals are grouped in families in the respective lists, as in the case of Florida 1857, we nest a corresponding <listPerson> for each family group in the personography.²⁹ When a series of periods is used in a table to indicate that a field does not apply, we standardize these to three with a space between each (“…”).

In each list, a separate <person> element with a unique value for @xml:id gathers the information for each individual. Below we enumerate the categories of data we include, with examples taken from the first <person> corresponding to Florida 1857 (<person xml:id=“iep0100”>).

§Name

We use <persName>, with <forename> and <surname>:

<persName>
  <forename>George</forename>
  <surname/>
</persName>

§Age

We use <age> with @when to represent the date on which an individual was a given age. This approach allows for the inclusion of multiple <age> elements when a person appears in multiple documents from different chronological moments.³⁰

<age when="1857">65</age>

§Gender

We have encoded gender when a value is explicit in the document or can be assumed with reasonable certainty from a person’s name or the larger context. In each case, we indicate the source of this information using @ana and a taxonomy of two possible values defined in the TEI Header:

<taxonomy name="gender">
  <category xml:id="expl">
    <catDesc>Value stated explicitly in document</catDesc>
  </category>
  <category xml:id="ded">
    <catDesc>Value deduced by editors from context</catDesc>
  </category>
</taxonomy>

When we believe the name in question does not suggest a specific gender or otherwise are unable to provide a value, we leave this value empty (<gender/>).

<gender ana="ded">male</gender>

§Economic Role

We use <occupation> to indicate the economic role or type of work associated with a person. We do so with reluctance, as we believe that the term occupation implies a level of agency on behalf of the enslaved that does not correspond to conditions of forced labor. However, the TEI guidelines define the element in a general way as containing an “informal description of a person’s trade, profession or occupation” that we believe can be used, at least provisionally, for this purpose.³¹ We use a value for @when to allow for the potential inclusion of different roles for an individual at different points in time.

<occupation when="1857">teamster</occupation>

§Location

We employ <residence> with <settlement>, <region>, and <country> to indicate the place with which an individual is identified. In most cases, this is the location the census was carried out, but could be another location specified in the document, as in the case of Cuba 1858, which was compiled in Cárdenas but lists individuals whose places of enslavement were elsewhere on the island. For the values of all these elements, we use the regularized forms indicated with <reg> in the respective XML documents, when present.

<residence>
  <settlement>St. Johns County</settlement>
  <region>Florida</region>
  <country>United States of America</country>
</residence>

§Status with Respect to Enslavement

We have included in the personography an indication of the status of individuals with respect to being enslaved. We implement this through another taxonomy defined in the TEI Header that for our project includes three possibilities: enslaved, born free (for the children of enslaved people born into freedom), and fugitive.³²

<taxonomy name="state">
  <category xml:id="ensl">
    <catDesc>Enslaved</catDesc>
  </category>
  <category xml:id="bf">
    <catDesc>Born free (child of enslaved mother born into freedom)</catDesc>
  </category>
  <category xml:id="fug">
    <catDesc>Fugitive</catDesc>
  </category>
</taxonomy>

Within each <person> element, we use an empty <state> element with @ana pointing to the corresponding element in the taxonomy (for example, ana=“#ensl”) and @when to indicate the date of that status.³³

<state ana="#ensl" when="1857"/>

As in the case of <age> and <occupation>, this approach allows the inclusion of multiple <state> elements with potentially differing values in a scenario in which a person appears in multiple censuses corresponding to different points in time.

§Data Excluded from the Personography

We do not include in our personography several categories of information. We do not include skin color or descriptions related to body type, distinguishing features, or other physical characteristics. We likewise do not register monetary valuations.

We exclude this information for two reasons. In the first place, we believe that it is particularly dehumanizing. While the documents themselves inherently deny the humanity of the people enumerated, these details are especially harmful. We do not presume that our personography can be a liberatory document, but in its construction, we do seek to represent people in the most dignified matter possible.

In the second, we do not believe this information would be useful for most users of the type of edition we envision. We imagine those users to be students and researchers interested in the historical context of the documents, as well as individuals conducting genealogical research. We do not imagine a scenario in which someone queries the personography to discover, for instance, how many individuals were missing teeth, had no facial hair, or were short or tall—all types of descriptions included in Cuba 1858. Certainly an argument could be made that including the monetary valuations could enable research into the economics of enslavement, or that the inclusion of other factors could make possible other types of questions, and projects that chose to do so could easily adapt our model to account for that information.

§The Interface to Our Edition

Our edition allows each document to be viewed in several ways. The user can toggle between original and modernized readings, and for the documents in French, Portuguese, and Spanish, switch between the originals and English translations. We display the transcriptions alongside the document images. Users can select the names of individual people to see the regularized data collected about them in the personography. In addition to making visible the contents of the personography within the context of the documents, this tool allows users to see explicit data in cases where the text itself may be ambiguous. This is particularly the case when the documents use idem or quotation marks to indicate the repetition of information across multiple cells in the tables.

We also include an index to people named across the entire collection of documents. That tool allows the user to sort the information by name and place, and to navigate from an entry in the index to the document or documents where the name appears. We provide this index as an example of the kind of tools that could be developed to visualize, analyze, or otherwise interact with the data in the personography. The index could be expanded to include whatever additional categories of information a project might decide to include in the personography.

§Conclusion

With this prototype, we have sought to demonstrate an approach to creating digital editions of enumerations of enslaved persons that both respects the documents as textual artifacts and enables the data they contain to be analyzed in a relational fashion. We have likewise endeavored to create a model that allows for the transmission of such texts in a way that respects the humanity of those enumerated and minimizes the opportunities for recreating the harm inherent in those documents. We believe that our model provides a paradigm for transmitting written artifacts that hold value for both researchers and general audiences today, despite the deeply troubling discourses they embody.

This prototype gathers a small number of short documents and therefore does not take into account all problems that could be encountered in editing the diverse range of documents that contain enumerations of enslaved people. Our project also takes a general approach to handling the information contained in the documents that may not fit the specific goals of other projects. In general, however, we believe that our prototype offers a model that can be adapted by others seeking to publish documents like those we transmit here in ways that maximize their utility and accessibility and honor the humanity of the people they enumerate.

§Acknowledgments

Raff Viglianti and Megan Bushnell played key roles in helping us to articulate the central concerns of our project, develop our approach to marking up the documents, and create the interface. Edith Nanclares, Bob Nawrocki, and Charles Tingley assisted with the identification of New Granada 1845 and Florida 1857, the documents that initially inspired this project, and Marisa Pechillo, Matthew Welcome, and Alexandra Zapata contributed to the transcription and initial markup of those items. Victoria Dickenson transcribed Saint-Domingue 1778 with contributions from Siobhan Meï and Chelsea Stieber. Katie Blizzard and Patricia Lynette Searl assisted with questions related to permissions for Virginia 1783.

The images of New Granada 1845 are used by permission of the Alcaldía de Santa Fe de Antioquia and the images of Florida 1857 by permission of the St. Augustine Historical Society.

§Footnotes

1. This project began in response to a presentation that McCarl delivered at the 2024 conference of the TEI in Buenos Aires in which he posed initial questions about how to edit censuses of enslaved persons, based on problems that arose through experiments in transmitting such texts carried out in collaboration with students at the University of North Florida. See Clayton McCarl, “¿Cómo editar los censos de personas esclavizadas?,” Journal of the Text Encoding Initiative 19 (2025), https://journals.openedition.org/jtei/6596. McCarl, Mei, Arriaga and Machado then collaborated on this micro-edition as a way to address those matters and design a prototype. ↩

2. Slave Voyages, accessed November 30, 2025, https://www.slavevoyages.org; Slavernijverleden, Nationaal Archief, accessed November 30, 2025, https://www.nationaalarchief.nl/onderzoeken/zoekhulpen/slavernijverleden. ↩

3. In seeking to alert users to the presence of offensive terminology in the database records, The National Archives of the Netherlands announces a pilot project that would potentially remove such language. See “Kwetsend taalgebruik en terminologie” (Offensive language and terminology), accessed November 30, 2025, https://www.nationaalarchief.nl/onderzoeken/kwetsend-taalgebruik-en-terminologie. ↩

4. This paradigm of discourse versus data is central also to an earlier micro-edition and accompanying study by Clayton McCarl: “Un prototipo para una edición digital del Epítome de la biblioteca oriental y occidental, náutica y geográfica (1629) de Antonio de León Pinelo / A Prototype for a Digital Edition of Antonio de León Pinelo’s Epítome de la biblioteca oriental y occidental, náutica y geográfica (1629),” Scholarly Editing 39, 2022, doi.org/10.55520/12N3RNQ6; and “Semantic Markup and Structured Annotation of Early Modern Bibliographies.” Scholarly Editing 39, 2022, doi.org/10.55520/KWZBEA0E. Those studies addressed this question in the context of bibliographical records, however, not people. ↩

5. Michel-Rolph Trouillot, Silencing the Past: Power and the Production of History, (Boston: Beacon Press, 1995): 26. ↩

6. Olivia Barnard, Emma Bilski, Leila Blackbird, Jessica Marie Johnson, and Ellie Palazzolo, “Introduction: Black Louisiana as a Template for Resistance,” in “Kinship & Longing: Keywords for Black Louisiana,” Scholarly Editing 41 (2025), https://scholarlyediting.org/issues/41/kinship-and-longing. ↩

7. Katherine McKittrick, “Mathematics Black Life,” The Black Scholar 44, no. 2 (2014): 18, https://www.jstor.org/stable/10.5816/blackscholar.44.2.0016. ↩

8. McKittrick, “Mathematics Black Life,” 16-17. ↩

9. McKittrick, “Mathematics Black Life,” 17. ↩

10. McKittrick, “Mathematics Black Life,” 18. ↩

11. McKittrick, “Mathematics Black Life,” 19. ↩

12. Christina Elizabeth Sharpe, Monstrous Intimacies: Making Post-Slavery Subjects (Durham, NC: Duke University Press, 2010), 3-4. ↩

13. Walter Hawthorne, Richard Roberts, Fatoumata Seck, and Rebecca Wall, introduction to “The Politics and Ethics of Naming the Names of Enslaved People in Digital Humanities Projects,” edited by Walter Hawthorne, Richard Roberts, Fatoumata Seck, and Rebecca Wall, special issue of Digital Humanities Quarterly 19, no. 1 (2025), https://dhq.digitalhumanities.org/vol/19/1/000781/000781.html, ❡1, 5, 7. ↩

14. Olivia Barnard, Emma Bilski, Leila Blackbird, Jessica Marie Johnson, and Ellie Palazzolo, “Kinship & Longing: Keywords for Black Louisiana,” Scholarly Editing 41 (2025), https://scholarlyediting.org/issues/41/kinship-and-longing. ↩

15. Celia Naylor, Unsilencing Slavery: Telling Truths About Rose Hall Plantation, Jamaica (Athens: University of Georgia Press, 2022); Celia E. Naylor, Kristen Akey, Madiha Zahrah Choksi, Alex Gil, Moacir P. de Sá Pereira, and Monique J. S. Williams, “About (Un)Silencing Slavery,” in (Un)Silencing Slavery: Remembering the Enslaved at Rose Hall Plantation, Jamaica, August 6, 2022, http://rosehallproject.columbia.edu. ↩

16. “Introduction,” Breadnut Island Pen, accessed November 30, 2025, https://breadnutislandpen.net/introduction; “About,” Breadnut Island Pen, accessed November 30, 2025, https://breadnutislandpen.net/about. ↩

17. An example that arose in the process of marking up the documents for this project revolves around the element <list>, which we chose not to use as a container for the enumerations as it would have required us to mark up people using <item>, a circumstance we find problematic, even if we understand item to refer to a structural element of a list, not the person in question. ↩

18. Jessica H. Lu and Caitlin Pollock addressed this topic in a 2019 talk titled “Design, Development, and Documentation: Hacking TEI for Black Digital Humanities,” https://mith.umd.edu/digital-dialogues/dd-fall-2019-jessica-h-lu-caitlin-pollock. The Antiracist Markup Practices Symposium hosted by the Women’s Writers Project in 2023 focused on similar problems (https://wwp.northeastern.edu/about/events.html). To date, however, little beyond the micro-edition of Barnard et al., “Kinship & Longing,” has been published focused specifically on the implications of using the TEI standard for work on materials related to the history of slavery. ↩

19. Schomburg Center for Research in Black Culture, Manuscripts, Archives and Rare Books Division, New York Public Library, accessed November 30, 2025, https://digitalcollections.nypl.org/items/bdf7f8a7-c71a-bb31-e040-e00a180660ac. ↩

20. George Washington Papers, Series 4, General Correspondence, 1697-1799, MSS 44693, Reel 091, Library of Congress, Manuscript Division, Washington, DC, accessed November 30, 2025, https://www.loc.gov/item/mgw433769. ↩

21. Folios 17r-18v of “Listas de los esclavos de uno y otro censo, y el de los libertos y esclavos prófugos,” Archivo Histórico Bernardo Martínez Villa, Santa Fe de Antioquia, Colombia, Jefatura Política, box (caja) 2, folder (carpeta) 1, file (legajo) 42. ↩

22. “Runaway Slave Register—Cuba,” Miscellaneous Afro–Latin America Collection, Schomburg Center for Research in Black Culture, Manuscripts, Archives and Rare Books Division, New York Public Library, accessed November 30, 2025, https://digitalcollections.nypl.org/items/122ec920-dfc9-0132-a02b-58d385a7b928. ↩

23. St. Johns County Circuit Court Papers, St. Augustine Historical Society, Box 127, Folder 13, St. Augustine Historical Society Research Library, St. Augustine, Florida. ↩

24. “Brazilian Slave Register,” Brazil Miscellaneous Collections, Schomburg Center for Research in Black Culture, Manuscripts, Archives and Rare Books Division, New York Public Library, accessed November 30, 2025, https://digitalcollections.nypl.org/items/ee124c00-4e90-0133-7825-00505686a51c. ↩

25. The referenced legislation is “Artigo 2º do regulamento número 4.835 de 1º de dezembro de 1871” (Article 2 of Regulation number 4835 of December 1, 1871). The full text of the law is available online: “DECRETO Nº 4.835, DE 1º DE DEZEMBRO DE 1871,” Câmara dos Deputados, accessed November 30, 2025, https://www2.camara.leg.br/legin/fed/decret/1824-1899/decreto-4835-1-dezembro-1871-552265-publicacaooriginal-69374-pe.html. ↩

26. See, for instance, The George Washington Financial Papers Project, accessed May 3, 2026, http://financial.gwpapers.org; and McCarl, “Un prototipo.” ↩

27. When an enumeration appears as a prose list, not as tabular data—as in the case of Virginia 1783 and the final section of New Granada 1845—we likewise employ <table>, forming a table with a single column, for the sake of consistency. ↩

28. A project having a reason for more granular semantic markup of the documents, however, could use <num type=“age”> for ages, <num type=“monetary-value”> for monetary valuations, <trait type=“physical”> for physical descriptions, and <seg type=“occupation”> for economic roles. ↩

29. This approach works in the case of Florida 1857, where the family members are listed in contiguous fashion. Brazil 1872 also contains information that links people by families, but the rows are not necessarily contiguous. ↩

30. This does not occur in our sample but would be present in the case of multiple censuses corresponding to the same place in different years. ↩

31. “<occupation>,” TEI: Guidelines for Electronic Text Encoding and Interchange, Text Encoding Initiative Consortium, last updated on September 4, 2025, https://tei-c.org/Vault/P5/4.10.2/doc/tei-p5-doc/en/html/ref-occupation.html. ↩

32. Projects that involve manumission records, which are not part of our sample, might also need a category like freed (for a person who was previously enslaved but is free in the respective document). ↩

33. We do not indicate a value for @when when the value of @ana on <state> corresponds to “born free.” ↩