To better understand the current monument landscape, we assembled a team of monument researchers from around the country. Together, we scoured almost half a million records of historic properties created and maintained by federal, state, local, tribal, institutional, and publicly assembled sources. We reached out to State Historic Preservation Offices and federally recognized Tribal Historic Preservation Offices for their most updated records. Each data source we encountered included publicly accessible digital records about cultural objects in a variety of formats. There were data sets we did not have access to because they are privately held, or they were not currently available digitally. There were data sources we did not use because, upon reviewing them, we found that they were primarily constituent parts of, or redundant with, a larger set; they did not offer key information such as location, which would allow us to condense records via relational databases, or they were in the process of being collected or updated during the period of this study.
In the end, we worked from forty-two data sources incorporating close to 500,000 individual records that included objects commonly referred to as monuments—statues or monoliths constructed with stone or metal, installed or maintained in a public space with the authority of a government agency or institution—as well as nonconventional monument objects like buildings, bridges, streets, historic markers, and place names. Of these forty-two data sources, only one (OpenStreetMap) provides a definition of monument as a guide for contributors to tag entries, with warnings about common misuses.1 Other sources, while not offering a definition of monument, included records that broadly included outdoor sculpture, cultural heritage sites, and other public art and history assets.
A large part of the work of the audit was accessing, converting, parsing, and mapping that data into a single, standardized data set. Our research team designed an algorithm to determine which of these records included conventional monuments. We produced a final study set of 48,178 monuments out of the source materials. While this study set included many idiosyncrasies, omissions, and glitches from its original data sources, we worked to streamline the data by pulling in geographical information (when available), extract names of people and events, review metadata held in records, and look for duplicates in overlapping data sets. We also cross-referenced this data with biographical sources, including Wikidata. Since so many of the monuments in the study set commemorate individuals, we compiled a Top 50 list of individuals for whom we have the most recorded monuments in the US. This list offers a simplified snapshot of the dominant trends in the monument landscape. We used technical approaches to cross-reference the names within the Wikidata category human across multiple categories in each of the forty-two data sources. Public monument records for the Top 50 list of monumental figures were cleaned to remove misidentifications, passing references, and redundant records across and within these data sources.
The research team explored major themes, trends, case studies, and dynamics underpinning the monument landscape. We examined metadata, and contended with the gaps, shortcomings, and blind spots in the data we encountered. We held focus groups with people who study and engage with monuments—scholars, municipal workers, educators, and artists—to help us understand how the data connected to their own experiences of monuments and how to create and support tools for public access.
The findings and data were shared with the Mellon Foundation and then prepared for public release. The National Monument Audit exists as a publication, a website, an ongoing series of essays and programs, and a search interface including open-source code on GitHub.
1. “OpenStreet Map Wiki,” Wikipedia, updated August 4, 2021, https://wiki.openstreetmap.org/wiki/Tag:historic%3Dmonument.