A Hidden Gem Becomes a Fertile Mining Ground: Historic Prison Admission Books and Data-Driven Digital Projects

Abstract

A Hidden Gem Becomes a Fertile Mining Ground: Historic Prison Admission Books and Data-Driven Digital Projects S. L. Ziegler (bio) and Steve Marti (bio) Keywords Collections as Data, Digitization, Digital Scholarship, Digital Humanities, Data, Prison Records, Digital Projects, Philadelphia, Eastern State Penitentiary, Digital History, Public History, Incarceration The library of the american philosophical society holds three admission books that provide a personal biography of each inmate at the Eastern State Penitentiary. The books, covering the years 1830–50 (with a gap in the 1840s), contain information about each prisoner, including their name, age, the crimes for which they had been convicted, their sentence, and often a note on when they were freed (or died). Also included, though less consistently, is gender, race, and religious affiliation. Additionally, the penitentiary’s moral instructor, a religious authority figure, recorded a paragraph-length note on each inmate detailing their religious education and other biographical details (fig. 1). Donated to the APS Library between 2000 and 2002, the admission books complement and extend a larger collection of material related to the Eastern State Penitentiary. Within this collection is a wide variety of treasures. One such treasure is a set of correspondence from a prisoner in the penitentiary, Elizabeth Velora Elwell, addressed to another inmate. According to the collection’s finding aid, the letters “suggest that Elwell carried a passionate love for a fellow prisoner at Eastern State, Albert Green Jackson, with hints that they may have met clandestinely on more than one occasion. It appears that the two had plans to marry when their terms expired.”1 The collection also contains a number of other volumes that detail the bureaucratic functioning of the penitentiary, including the daily rations and the overseer’s roll.2 Even within this collection of gems, the admission books command [End Page 363] Click for larger view View full resolution Fig. 1. Sample page from Admission Book A, showing typical entries including name, age, offense, and note by the moral instructor. [End Page 364] special attention. The mix of biographical information, demographics, and moral commentary provides both a glimpse into the population of the penitentiary and into the beliefs that guided the institution. When digitized and transformed into a dataset, the admission books became a particularly fertile mining ground for historical insights. In 2015, the APS Library digitized the admission books. Then, in 2016 and 2017, the library transcribed the content into spreadsheets. The library published the data in a variety of outlets and encouraged scholars to use the data for a wide range of projects.3 In late 2017, the APS Library published a digital project based on the historic prison dataset. The project, “Eastern Apps: Visualizing Historic Prison Data,” presents a suite of apps that allows users to generate interactive visualizations as a means of exploring the data through text-mining techniques and statistical analysis.4 The first app analyzes word frequency in the moral instructor’s notes. This app constructs corpora based on user input and presents both word clouds and word frequency graphs (figs. 2 and 3). Additionally, users can explore specific words and phrases in context (figs. 4 and 5). The repetition of words by the moral instructor, and variation in usage over the years, reveals patterns in the individual descriptions of inmates. The second app graphs sentencing length and maps sentencing location based on selected variables that include gender, age, place of birth, ethnicity, literacy, marital status, sobriety, charge, number of previous convictions, and admission year.5 Modifying these variables updates both the map and the graph (fig. 6). The map produced by this app shows courts where inmates were sentenced and displays the average sentence length. The outer circle (styled red in the online version) corresponds to the average sentence length; the inner circle (styled blue in the online version) corresponds to the number of cases from that area (fig. 7). The graph shows variation in sentence length. The x-axis corresponds to the length of the sentence in months; the y-axis corresponds to the number of inmates who received a sentence of that length. [End Page 365] Click for larger view View full resolution Fig. 2. Word clouds created from...