Horses and Housewares: Indexing George Washington’s Financial Papers

TOPICS: Digital Humanities, Financial Papers, Student’s Perspective

by Reese Fulgenzi, Undergraduate Student Worker
June 14, 2019

Homepage of the George Washington Financial Papers Project website.

One of The Washington Papers’ central tenets, producing “accessible transcriptions of historical documents,”1 is perhaps best demonstrated by the project’s work on George Washington’s financial papers. The zeroing in on the Washingtons’ lives that the financial papers provides is incredible; small details are captured and preserved, down to the exact day that sundries were purchased or that employees were paid. Even beyond the numbers, the language and phrasing of these documents provide a glimpse into the world of colonial Virginia.

I was elated to begin work on the financial papers in January of this year because of the richness of the documents. They seemed to be filled with even more information than Washington’s letters and other records. Financial documents, however, are matched in their wealth of content by their complexity. Hundreds of expenses occupy single books, with each entry later re-recorded in journals and ledgers. The documents are filled with abbreviations and shorthand—not to mention imprecise subtotals and, of course, 18th-century cursive. Some of these challenges are resolved fairly easily with transcription or annotation, while others, such as the categorization of items mentioned, require more careful work.

Coming from a science background, I was not unfamiliar with the goal of accessibility. Citizen science, scientific democratization, and other similar efforts demand that information be accessible both physically and intellectually. The Financial Papers Project resembles these scientific efforts, providing researchers and the general public with freely accessible, annotated transcriptions of the materials. But upon further study, I found that the project’s goal of accessibility was more nuanced than simply making Washington’s financial papers available and legible. For instance, some elements of the documents, like abbreviations and shorthand, were edited only for standardization. While directly transcribing “do” [ditto] or “&c” [etc.] sacrifices accessibility, it maintains the authenticity of the materials, preserving the author’s intention and the document’s historical context.

My work on the George Washington Financial Papers Project began with a transcribed pocket book (also known as a waste book), a document that records general and ad hoc expenses, ranging from clothes and jewelry to hay and cords of wood. In addition to verifying the prepared transcription of the waste book, my work involved the recording of metadata, including line numbers, annotations, explanations, and classification of content. The identification of such information allows for a searchable online database: users can query all mentions of services, slaves, loans, livestock, and more.

Working with project editors Jennifer Stertzer and Erica Cavanaugh, I helped tag and sort unique items, occupations, and individuals. Hitting the right degree of atomization—descriptive and distinct enough to be useful, but also broad enough to be searchable—required careful thinking and comparison to other entries. The taxonomic categories of agriculture, food, household, clothing, occupations, and individuals were created, and tagging was performed manually, followed by revisions. (The distinction between agriculture and food, for example, was not made initially).

How to condense tags was similarly demanding: “ivory thimble” versus “thimble,” “oznabrigs” versus “cloth,” or “hickory” versus “wood.” How much information is lost when tags are simplified, and how is searchability or accessibility improved? For the documents with which I was working, I erred on the side of sorting items into larger collections. No information is entirely lost by more generous sorting, as each tag links back to the original transcription. Using broader tags, however, allows for more questions to be answered by pointing users to a larger collection of potentially useful material. As more financial documents are tagged and a broader database is developed, the possibilities of greater quantitative analysis increase immensely.

Tagging individuals was less ambiguous than objects but was not without its own challenges. Sometimes (as with any historical document), it was impossible to identify individuals referenced only by their first names or occupations. Other illegible names could be inferred from context or from reviewing Washington’s correspondence. And some names, primarily those of family members, could be narrowed down based on their surname of “Washington” or “Custis.” Unfortunately, not all individuals were identifiable, but such is the burden of the editor. Some notes on purchases were also illegible; after all, a waste book is not the same as a formal household ledger.

The George Washington Financial Papers Project’s goal of providing accessible and accurate transcriptions that were easily searchable, was achieved. This work certainly broadened my experiences, swapping out spreadsheets of university geology collections and genomic assays for historical finances, and I am excited by the shared goals of such projects. Democratization of information and improved accessibility are laudable aims in any discipline, and I look forward to seeing the future unfold for digital humanities initiatives, through which more and more material is made available to all.


1. See The Washington Papers’ About page: