Uplifting the Lives of the Enslaved through Primary Source Analysis and Data Extraction: Reflections from ARHU SROP 2021 Armani Jackson, Brooklyn Jarrett, Jill Finkel, Kamryn Nelson & Sarah Reeves

cover photo: Robert S. Duncanson, Landscape with Family by Lake, 1850s. image description: landscape oil painting of blueish-yellow sky with grey and yellow clouds, a light blue shore lined with dark green trees. in the water there are two people in a canoe, one is standing and paddling in the back.

Project Abstract and Enslaved.org

The Summer Research Program Opportunity (SROP) 2021 Cohort in the College of Arts and Humanities at the University of Maryland, College Park was partnered with Enslaved.org to research and analyze archival records of enslavement in Maryland.

Enslaved: Peoples of the Historical Slave Trade (Enslaved.org) is an open-source database that explores and reconstructs the narratives of slavery to center the lives of individuals who were enslaved.

Our cohort specifically explored, transcribed, and analyzed the "slave account book" created and kept by Charles Benedict Calvert, circa 1830-1860, which was retrieved from the University of Maryland, College Park.

Intentions of Our Work and Project

Our intentions for our work and this project are to uplift and articulate the lived experiences of enslaved persons, centering their humanity, and empathetically assessing qualitative and quantitative data in order to honor these people and their descendants. Specifically, our intentions are to:

  • Discuss and explore the lives of the enslaved with compassion and tact, bringing their holistic narratives to the forefront of our work and analyses
  • Approach sources from enslavers with caution to not take on oppressive language and assumptions
  • Expose the real history and violence of Charles Benedict Calvert and his legacies specifically at the University of Maryland
  • Document and present data and knowledge that is accessible beyond the restrictions of academia and sustainable for all uses

Language and Vocabulary

There are conscious choices around the language we use in our work. Dominant literature and discussions of enslavement deny personhood and agency to enslaved people, giving enslavers more narrative power and absolving them of their active roles in these violent histories.

“It was very important... to not use language that further dehumanizes people who every system and structure was designed to dehumanize. I think when we hear the word "slave," we think of slavery as being the essence of that person. But if you call someone an enslaved person, then it speaks to a condition. These people were not slaves. Someone chose to force them into the condition of slavery, and that language to me is very important, as is using the word "enslaver" over slave owner because these people didn't have a moral right to own another human being, even though the society allowed it, and I think it needs to be active, that this was an active system of people choosing to treat other human beings as property.” (Nikole Hannah-Jones, 2020)

The purpose of using alternative language when describing the enslaved is to shed light on the correct ways to address people who were and still are ignored and mistreated by the construction of "history." Thus, we use:

  • “Enslaved person” instead of “slave”
  • “Enslaver” instead of “master” or “owner”
  • Example: “Calvert enslaved Robert” instead of “Calvert owned Robert"
“The focus shifts, the onus shifts, and the shame of slavery shifts from the enslaved to the enslavers.” (Bridgette L. Hylton, 2020)

Guiding Questions

  1. How do we take fragmentary evidence, like death certificates and sale receipts, and construct the narrative of a person’s life?
  2. What implications does our work have in confronting, healing, and repairing for this most fundamental American history?
  3. When discussing and writing the histories of enslavement and the American story, particularly while using historical sources written in the voice and perspective of enslavers like Calvert, how can empathetic transcription language and conscientious data work be used to center the stories and humanity of the enslaved?
Robert S. Duncanson, Mountain Pool, 1870. image description: landscape oil painting of cascading water cutting through large bolders. green, yellow, and light brown trees speckle the landscape and surround the waterfall. a small person stands at the end of the large pool of water.

Methodology and Extraction

Early in June, our research cohort received this source with little information of its historical context and no guidance of what exactly to do with it. With no background information on Calvert or the names in this book, we, perhaps like the audience of this presentation, uncovered everything the source had to offer for the first time.

Slave account book of Charles Benedict Calvert, circa 1830-1860. https://hdl.handle.net/1903.1/5750. University of Maryland, College Park Digital Collections. image descriptions: gallery of 17 images with two larger images on the top and bottom and five rows of three smaller images in between. each image is a photocopied page of the slave account book. the pages are light yellow and slightly crinkled with various writings on them, including names of enslaved people, ages, and other numerical values.

Individually, we all created datasheets to begin transcribing and extracting information from the source, which lists the name, age, “valuation”, and sale price of the enslaved persons at Calvert’s several local holdings, circa 1830-1860.

image description: five images of five different datasheets. the datasheets detail the information transcribed from the slave account book. one datasheet is color-coded in green and red, one is pink and blue, one is yellow and white, one is green and white, and one is white with a green heading.

We each approached extracting the data differently...

  • Armani: Focused on creating lists in sets based on gender and color-coded age
  • Brooklyn: Focused on representing data as seen exactly on the page (including markings & unchanged language)
  • Kamryn: Included quantitative information around price and valuation, color-coded categories
  • Sarah: Included age category classifications based on the Enslaved.org categories, language of "assumed gender"
  • Jill: Introduced a column for observational notes

From our individual data sheets, we collectively established consistent language and combined our extractions into one main dataset.

We discussed the intentions we held while working with the data individually and determined which aspects and categories we wanted to maintain, add, or discard. We also discussed how much of our own voices, as the researchers, we wanted to appear in the data.

We decided it was important to explicitly define our chosen terminology for our dataset column headings. This maintains consistency while recording data, as well as describes our intentions for the audience to understand our ongoing choices. Each term is accompanied by examples from the account book to better illustrate our term meanings, as well as to show explicitly how they exist in our source.

image description: text table of fifteen rows and three columns with definitions and descriptions of our field definitions and chosen terminology. the columns from left to right include: Term, Definition, and Example. The rows include, in descending order, Sort Order, Digital Page, Book (L/R), Page Headings, "Nº", "Names", Imputed Gender, Named Place, "Age" Text from Book, Simplified Age, Age Category, "Price"/"Valuation", "Sale", Abbreviations/Words 

In establishing consistent language and deciding which components to include in our combined dataset, we continuously returned to the question of whether or not we should record "imputed gender" classifications. The account book does not have any explicit reference to gender classifications, however, we could imply based on the gendered connotations of most names. Because there was no gender explicitly stated, it qualifies as "imputed" data, meaning it is based on the researcher's voice and assumptions. There were many considerations when determining whether or not we should include this, such as:

  • Adding the category of "imputed gender" has the potential of adding aspects of humanity to the enslaved people in the account book. This is in line with goals we have as a group.
  • Misgendering has the potential of further disrespecting any humanity.
  • Imputing would mean determining gender solely based on names.
  • However, one could argue that the names in the lists conveyed gender because in the time this source was written, given names were often biblical and other references that upheld strict gender binaries.
  • Deciding to gender “Male” and “Female,” without any further acknowledgement may encourage the belief that only two genders exist, or that gender exists at all.
  • None of us have experience in the study of gender for enslaved persons– imputing labels such as “Male” and “Female” may be far from gender as it was understood during enslavement.

Ultimately, we decided to include "imputed gender" because it aided our analyses and visualization of trends and patterns in our data.

Our Final and Combined Dataset:

image description: datasheet of our finalized data extractions containing the names, ages, gender, and valuations of recorded enslaved people. columns from left to right: Sort Order, Digital Page, Book (L/R), Page Headings, "Nº", "Names", Imputed Gender, Named Place, "Age" Text from Book, Simplified Age, Age Category, "Price"/"Valuation", "Sale", Abbreviations/Words
Robert S. Duncanson, Loch Long, 1867. image description: landscape oil painting of a yellow-blue skyline, mountains, and a body of water. the light brown mountains are dotted with green trees. a small house sits on the side of the body of water and two people stand next to a boat that rests on the water.

Data Visualizations

image description: data visualization of the account book. the title reads: "Calvert Account Book: Visual Representation of Enslaved People Categorization". the background is a light yellow and a large grouping of many names in blue, pink, and green appear in the middle of the screen. the bottom-left of the screen contains a graph titled "Name Place to Price/Valuation" that shows nine different bars. the bottom-right contains a graph titled "Age Category to Price Valuation" that shows four different bars.
Robert S. Duncanson, Owl's Head Mountain, 1864. image description: landscape oil painting of a yellowish-blue-pink sky, a large mountain in the center, two smaller mountains in the distance, all surrounding a body of water. the mountains and edges of the water are dotted with green, thin trees and some rocks. a few boats and people are scattered distantly across the water.

Observations and Analysis

After we extracted and finalized our data, we turned our attention to the different markings, symbols, abbreviations, and names that appear throughout the source.

Special Markings

We named this section "special markings" because these markings appeared to us in no specific pattern and were not exclusive to any single or set of records throughout the book.

The first special marking that appears throughout the source is the "x", which appears most often next to the names of enslaved people but also sometimes appeared next to the "N⁰" column, the "age" column, and the "price column."

image description: snapshot of a beige-colored page of the account book with a log numbered 29-33, a corresponding list of the names of enslaved peoples, valuation and sale, all in slight cursive ink. a few "x"s appear in both pen and pencil. "Robert" is written faintly in pencil on the last row. 

The seemingly randomized nature of the "x" marking can lead us to many conclusions. For example, the "x" written next to certain names in the "N⁰" column may have many implications for annotating:

  1. The record of an enslaved person that was caught trying to attempt an escape. Some names have one "x" and others have two "x"’s. It is assumed that if there were three "x"’s next to the name of the enslaved person it could be that they might have been punished or killed for their disobedience.
  2. The status of an enslaved person who is being kept at the listed location under the enslavement of Calvert rather than being sold to another enslaver may be indicated through the absence of an "x" next to their name.
  3. The "x" may also simply represent a log or annotation for counting and tracking throughout the source.
"x" appears in the "N⁰" column and "name" column, also here you can see "do" and other faint writings of what appear to be other names. image description: snapshot of a beige-colored page of the account book with a log numbered 27-35, a corresponding list of the names of enslaved peoples, valuation and sale, all in slight cursive ink.

It is also important to note that some markings are written in pen while others appear in pencil, this is true for certain numerical values and names as well. We infer that something written in pen is more definitive and permanent, whereas pencil indicates non-permanence, the intention to change something, to come back to it.

image description: snapshot of a beige-colored page of the account book with a log numbered 19-26, a corresponding list of the names of enslaved peoples, valuation and sale, all in slight cursive ink. The first name, Henry, has "sold Armfield" written next to it in parentheses, and the following five names have "(do)" written.

Another special marking is "(do)" which appears most often on the last two pages of the source. We assume that this marking means "ditto" because, as seen in the image above, it follows a descriptive marking and repeats until a new descriptor ("sold to Armfield", repeats, ends at "Died").

image description: snapshot of a beige-colored page of the account book with a log numbered 27-35, a corresponding list of the names of enslaved peoples, valuation and sale, all in slight cursive ink. A few of the names have annotations next to them in parentheses, including "Died", "Baker", and "Brown".

Next to the names of some enslaved people appear other names in parentheses. We think these are either the names of other enslavers or property names, possibly indicating a transaction between Calvert and other enslavers.

Familial Breakdowns

We noticed that in the first two pages, people are recorded for the most part in descending age order, with a few outliers. After these pages, there aren’t other apparent ordering patterns, but we do notice that there are times in which the recordings of adult-aged people are followed by child-aged people, enslaved persons with shared last names, and other implications for family connections. Although we can’t fully construct verified and concrete familial connections among the enslaved people named in this source, given time and material restraints, it’s an intention and possible future direction.

  1. The name of an adult-aged enslaved person followed by one or multiple entries of a child-aged enslaved person can infer a familial connection of parent and child(s).
Page 6L is a main example of possible patterns (adult age followed by multiple children) Airy, 38, followed by Kitty: 9, Charles: 3 ½, Francis: 1, Airy:18, Eliza: 11, David: 15, John: 14, Elias: 9. image description: snapshot of a beige-colored page of the account book with the previously listed information in slight cursive ink.
Page 5L, three children (John Hanson:10, Ann Maria: 1, Christy: 9) follow adults Isaac, 40 (who has G.C next to his name) and Caroline, 29.image description: snapshot of a beige-colored page of the account book with the previously listed information in slight cursive ink.

2. An assumed son named Isaac who's labeled as “Tracy’s Son” is indicative that they are the son of another enslaved person named Tracy. There are three Tracy’s documented in the slave account book that can help draw a connection to who is the assumed mother of Isaac.

Page 15R, Negroes to be disposed of, Isaac “Tracy’s Son”. image description: snapshot of a beige-colored page of the account book with the previously listed information in slight cursive ink. "Tracys son" is circled haphazardly. 
Page 2R, Rock Creek Farm, Tracy, age 9, valuation 300. image description: snapshot of a beige-colored page of the account book with the previously listed information in slight cursive ink.
Page 5L, Negroes to be kept, Tracy, Home Place, age 19, valuation 350. image description: snapshot of a beige-colored page of the account book with the previously listed information in slight cursive ink.

It is assumed that the oldest of the enslaved persons named Tracy is the mother of Issac who is the assumed son based on the explicitly written detail.

3. There are three individuals that are assumed to be associated with each other because they share the same last name “Williams.” R. Williams is an assumed enslaver that may possibly enslave a brother and sister: Ephraim is an assumed male and Ruth is an assumed female. Their ages are in close proximity which indicates that they have some relation to each other.

Emphraim Williams, Age 10 and Ruth, Age 6. image description: snapshot of a beige-colored page of the account book with the previously listed information in slight cursive ink.

4. The indication of an initial "G.C." next to six enslaved people leads us to the assumption that this set of individuals are related. If not blood related, these individuals are grouped together under an assumed enslaver with the initials G.C..

Page 5R, Eliza Ann, age 2, valuation 120. image description: snapshot of a beige-colored page of the account book with the previously listed information in slight cursive ink.
Page 5L, Isaac, age 40, valuation 750. image description: snapshot of a beige-colored page of the account book with the previously listed information in slight cursive ink.
5R (Leu, age 16, valuation 700) (Letty, age 25, valuation 350) (William, age 1, valuation 150) (Levi, age 12, valuation 450). image description: snapshot of a beige-colored page of the account book with the previously listed information in slight cursive ink.

Further Connections

Another observation is the duplication of enslaved persons on another page. These duplications show that these individuals have different statuses in reference to where they are located and whether or not they are being kept or disposed of.

Alexander, age 65, valuation 50 appears on page “2L labeled Rock Creek Farm listed 1” and Betty, age 60, valuation 30 appears on page “2L labeled Rock Creek Farm listed 2”. image description: snapshot of a beige-colored page of the account book with the previously listed information in slight cursive pen. the page is titled "Rock Creek Farm" in slight cursive.
Alexander, age 65, valuation 50 reappears “14L labeled Rock Creek and Negroes to be disposed of listed 1” and Betty reappears “14L labeled Rock Creek and Negroes to be disposed of listed 3”. image description: snapshot of a beige-colored page of the account book with the previously stated information in slight curisve pen. the page is titled "Negroes to be disposed of." in slight cursive.

Location Connections and Correlations

image description: white map with hints of light pink and blue. the map is of contemporary Prince George's County.
Left: United States Soil Conservation Service. Prince Georges County, Maryland. Hyattsville, MD: USDA-SCS, 1977. Map. https://www.loc.gov/item/81690002/. Right: Map of Riversdale Demesnes, Rossburg Farm, Situated in Prince George's Co. Md. belonging to Charles B. Calvert, Esq, created by William Sides, land surveyor, 1853. image description: snapshot of section of map above with the location "Riverdale" circled in red.

On the cover of the account book, "Prince George's County" is written. Charles Benedict Calvert inherited Riversdale, land in Prince George's county. Though it is not specified in the account book if any of the enslaved people listed were located in Riversdale, or the farm Calvert owned in Riversdale (Rossburg Farm), this information can help contextualize the location in the account book.

Martenet, Simon J. Martenet's Map of Prince George's County, Maryland. [Baltimore, ?, 1861] Map. https://www.loc.gov/item/2002624036/. image description: gallery of three beige maps. the two maps on the right are snapshots of the bigger map on the left of Prince George's County. the two maps on the right contain three different red circles that point out the names and locations "N.W. Branch," "Chas. B. Calvert," and "Rock Creek."

This map is from 1861 and depicts Prince George's county. On the right, we've displayed images of different locations which have (similar) titles to several of the named places we discovered in the account book. These presented here include, N. W. Branch, Rock Creek, and an area marked Charles B. Calvert (Riversdale). In our account book, we recorded a named place with the title "North-West Farm," and although there isn't a match on this map, "N. W. Branch" is very close to the area marked "Charles B. Calvert," which means this is the area where the recorded North-West Farm existed. Something similar is assumed with Rock Creek.

Simon J. Martenet, Martenet and Bond's Map of Montgomery County, 1865, Library of Congress, MSA SC 1213-1-464. image description: gallery of three light brown maps with many lines and words. the two maps on the right contain five different red circles that point out five names and locations: "R. Williams," "Cabin Branch," "Seneca Creek," "Dry Seneca Cr.", and "Ww. B. Vinson."

This map is a representation of Montgomery County, MD from 1865. Although the contextualizing data we have found so far has addressed Prince George's county and Calvert's involvement there, Montgomery County is not far from PG––making it possible that the locations on this map, which have similar names to some of the named places in the account book, may in fact be those locations.

The similar locations include Cabin Branch, R. Williams (an assumed owner of land, rather than name of land), Seneca, and Vinson. Although we are not certain that these are representations of the specific named places in the account book, the similar names, proximity to each other, and proximity to Prince George's County all add to evidence supporting these locations as the ones written in the account book.

Besides contextualizing the location of our primary source, these maps also provide some help when it comes to making connections and understanding dynamics of the account book. For example, in the book, "R. Williams'" is listed among other named places. With these maps, we can look to observe it to mean R. Williams must have been a landowner, and some of the enslaved people named in the book may have been enslaved by Calvert, however, they were located on R. Williams' land. Although the only evidence we have is the map pictured above and the reference to R. Williams in the account book, the proximity of the name on the map to other locations with names similar to named places in the account books encourages our understandings.

Robert S. Duncanson, A Dream of Italy, 1865. image description: landscape oil painting of a light blue and pink sky, snowcapped mountains in the background, an old castle near a body of water, and stone structures scattered in the right foreground and distance. there are different types of tall green trees on the hills and around the water.

Limitations, Reflections, and Conclusions

Limitations We Encountered in Our Project

  • Time: Although we didn't feel too much of an impact, project ~ 6 weeks long
  • Online setting: Restriction to an online setting caused limitations when it came to collaborating on our project, as well as visiting locations, libraries or archives. However, the online format also made many creative pathways available that would not have been otherwise.
  • Sources reflecting the narratives of the enslavers: As we stated, it was our intention to focus our work on the lives of the enslaved rather than the lives of the enslavers. Work with this focus is difficult when the majority of the sources we found (both supporting sources AND the account book itself) are reflective of the enslavers' narratives.
  • The account book provides a limited amount of data: The book has a lot, we spent a lot of time trying to utilize the aspects that we would often otherwise overlook (blank pages, faint writing, typos, etc), but it also only holds a limited amount of information. This felt particularly difficult in our aim to focus on the names listed, as very little information (in the form of identifiers such as "Valuation," "Age," etc.) was available.
  • Interference in our sources from other researchers: While exploring primary sources, we would often have to consider, when we saw something out of the ordinary, if it was simply that, a mistake by the author, or interference by researchers who interacted with the source before us.
  • Finding a platform that would best tell our story: We explored several different platforms to use to present and display our data, and even found some that we thought fit best, only to later run into technical difficulties.

Conclusions and Implications

From just a singular primary source, we've developed connections, patterns, and more holistic perspectives regarding the lives and experiences of enslaved people in Maryland.

Our hope is to further develop this information as public knowledge for both the descendants of the enslaved to reconnect with their ancestors and construct family lineages and for the public to confront more accurate realities of American history.

While our project is far from complete, it is our deepest intention that this work reconstructs dominant narratives of slavery, progresses the centering of silenced voices in history, and contributes to contemporary dialogues regarding the legacies of institutional violence and long overdue reparations.

Robert S. Duncanson, Landscape with Sheep, n.d. image description: landscape oil painting of a grey-pink-blue-yellow sky filled with clouds. the sun is behind a cloud and sunbeams shine through against distant mountains. the foreground is filled with lush green trees.


Acknowledgments and Thank You

We would like to extend our deepest gratitude to the many people who made this project possible:

  • The Big Ten Academic Alliance and the Andrew W. Mellon Foundation for their generous financial resources that funded our project,
  • The organizers and leaders of the BTAA Summer Research Opportunity Program (SROP), especially Dr. Ralph Bauer, UMD College of Arts and Humanities, and UMD Graduate School,
  • The librarians, archivists, and faculty at the University of Maryland and Maryland State Archives, especially Lae’l Hughes-Watkins and Chris Haley,
  • The researchers and contributors at Enslaved.org, especially Kristina Poznan,
  • Marisol Fila, who supported and guided our work,
  • Betsy Yuen, Patrice Greene, and Ashley Ogwo for being the backbone of our research program,
  • And finally, our mentor and guide, Dr. Daryle Williams!
Created By