About This Data

For this exhibit we have focused on three main 'themes' of analysis - location, gender, and time. Of course, there are many other ways to analyze the data and we encourage you to do so. To this end, we have uploaded the dataset to our website for you to download and try yourselves.


Dig into the Data: What will you find?

There are several ways to find, download, remix, and edit the dataset used in this gallery.


We're curious to know what you find! Please contact us with any questions, feedback, or results using or analyzing the data at digitalprojects@amphilsoc.org.

Where has this data come from?

Over a two year period, staff at the American Philosophical Society digitized and transcribed the bound volume of records creating a large dataset. This data can be visualized in many ways. Each map, graph, and diagram can be used in data stories that tell us about those who were indentured.

This dataset was created from a volume of over 800 pages that records information pertaining to individuals entering contracts of indentured servitude in Philadelphia from 1771-1773. Each entry contains details about the person to be indentured, including their name, country of origin, length of contract, and amount of debt owed. The records not only list the name of the person, but contain details on their profession and on the terms of the indenture. Although the volume is described as the records of German immigrants, there are other indentures included, such as that of John Slour, a free African American, records of those arriving from Ireland, and of young Philadelphians entering indentured contracts. The volume had been on loan to the City Archives until 1987. During that time, approximately twenty pages went missing. Otherwise, the volume appears to be complete and contains over 5,139 records. Having all of these records in spreadsheet form means that instead of reading over six hundred pages of handwritten text, we can now analyze two years’ worth of records quickly and by different measures.


Is there anything the data doesn’t show?

Data are not objective. When we create data sets, we make many decisions about what to include, what to leave out, how to categorize words or locations and more. We view historic data through a contemporary lens. For example, the records taken at the time of indenture did not capture the gender of the individual entering into the contract. Therefore, to build our visualizations, we have manually assessed, by name, whether each person is likely to be female or male. However, this means that we are only able to consider gender in binary terms and not across its full spectrum. Transparency about these choices is of the utmost importance to this, and all, open data projects.

The records of indentured servitude and apprenticeship, when viewed as data, create distance from the practice that they describe. Entering into such a contract likely meant a difficult life for the indentured person or apprentice. This volume contains snapshots of critical points in the lives of thousands of individuals. While examining the visualizations produced from this dataset allows us to highlight unique individual stories, it can also hide many others that become lost in numbers and statistics.

By looking at the dataset as a whole, we see only the big picture. It can be easy to forget about the details and thus the people behind the records. Contracts of indenture and apprentice tell us about a critical part of their lives, but not the whole story. Reducing human experience to data can obscure the emotional and full lives of each individual. For example, how did families feel when they were forced to sign differing contracts that would split them by household or location? How did new mothers complete their contracts with newborns to care for?

We have tried to show both views of this data by presenting individual stories alongside visualizations. However, there is much lost in the space between the small and large scale. It is also important to remember that when we build visualizations, we have feelings about the data we are presenting and might be predisposed to present them in a way that is sympathetic to these. This is another reason why it is important to be transparent about how and why we build and design visualizations.


About data transparency at APS

The American Philosophical Society Library & Museum maintains an active Open Data Initiative. Under this initiative, the Center for Digital Scholarship (1) identifies content in the APS Library that is conducive to being reconfigured as structured datasets; (2) encourages the use and reuse of the data by opening the data to all, and by easing access to the data.

Recognizing that the processes of data collection, organization, interpretation, and visualization have the potential to enact, reify, or reinforce cultural and structural harms, biases, and inequalities, the CDS endorses a policy of data transparency. Due to the quantitative demands of computational analysis methods, even well-intentioned and careful data work may misrepresent the nuance, complexity, and uncertainty inherent in human lives and experiences. A responsibly humanistic approach to data therefore requires transparency around method and labor, as well as a critical assessment of the data’s uncertainties, biases, and limitations.


Contributors and Process

  • Phase One: June 2017-March 2019
    Process: Digitization, transcription, and basic data structuring by Benjamin Weinstein (Washington College Explore America Intern) and Cynthia Heider.
    Deliverables: Publication of “Digitizing Indenture Records,” “Indenture Mining: Making Pre-Industrial Tradeswomen Visible (Part I),” and “Indenture Mining (Part II)” on the American Philosophical Society blog.
  • Phase Two: March-April 2019
    Process: Data restructuring and refinement in OpenRefine for more effective computational analysis by Cynthia Heider.
    Changes to data: Standardization of “Bound As” column value to reflect “Apprentice,” “Servant,” or blank as indicated in the original indenture; splitting of “Name” field
    Deliverables: Finalized dataset submitted to MEAD and made available via GitHub and the APS Digital Library. Publication of “Datafied Redemptioners/Redeeming the Data: What’s New at the APS Center for Digital Scholarship” on the American Philosophical Society blog.
  • Phase Three: May-July 2019
    Process: Online exhibition planning, text creation, data augmentation, data visualization and mapping using Tableau by Nicôle Meehan (APS Digital Humanities Fellow). Contextual research by Nicôle Meehan and Cynthia Heider.
    Changes to data: Addition of geo-coordinates to allow analysis of departure/arrival points; tentative categorization of gender based upon name analysis and use of pronouns in the original document
    Deliverables: Publication of dataset on MEAD.
  • Phase Four: August-December 2019
    Process: Website design and construction by Bayard Miller using HTML5Up and StorymapJS with contributions by Cynthia Heider. Drafting of data transparency policy by Bayard Miller and Cynthia Heider.
  • Phase Five: March-August 2020
    Process: Additional text written and edited by Bayard Miller and Cynthia Heider. Home page redesign and visualization annotation done by Cynthia Heider.