Replication data for: Medical infrastructure, veterans, and mining as potential predictors of the worship of Asclepius in Roman Dacia: A Spatial Quantitative Analysis

This repository serves as supplementary material for the article Medical infrastructure, veterans, and mining as potential predictors of the worship of Asclepius in Roman Dacia: A Spatial Quantitative Analysis. This result is a part of the Socio-Spatial Situatedness of Roman Professions and its Impact on Religion in the Roman Empire: A Formal Modeling Approach project (GM26-21025M) funded by the Czech Science Foundation (GAČR). All csv files are encoded in utf-8 Files Roman cults Asclepius_inscriptions_05_2026.csv: Inscriptions from the Epigraphic Database Heidelberg mentioning the god Asclepius in various noun cases, with the type of inscription attributes votive, dedicatory/building, honorific, and acclamation. Reused for the region of Roman Dacia from Glomb, Tomas, 2021, "Replication data for: The spread of the cult of Asclepius in the context of the Roman army benefited from the presence of physicians: A spatial proximity analysis", https://doi.org/10.18710/279HML, DataverseNO, V2. Asclepius_inscriptions_05_2026.geojson: A geojson file made from Asclepius_inscriptions_05_2026.csv, reprojected to EPSG:3035 ETRS89-extended / LAEA Europe for measurements in meters. Apollo_inscriptions_05_2026.csv: Inscriptions from the Epigraphic Database Heidelberg mentioning the god Apollo in various noun cases, with the type of inscription attributes votive, dedicatory/building, honorific, and acclamation. Reused for the region of Roman Dacia from Glomb, Tomas, 2021, "Replication data for: The spread of the cult of Asclepius in the context of the Roman army benefited from the presence of physicians: A spatial proximity analysis", https://doi.org/10.18710/279HML, DataverseNO, V2. Apollo_inscriptions_05_2026.geojson: A geojson file made from Apollo_inscriptions_05_2026.csv, reprojected to EPSG:3035 ETRS89-extended / LAEA Europe for measurements in meters. Minerva_inscriptions_05_2026.csv: Inscriptions from the Epigraphic Database Heidelberg mentioning the goddess Minerva in various noun cases, with the type of inscription attributes votive, dedicatory/building, honorific, and acclamation. Reused for the region of Roman Dacia from Glomb, Tomas, 2021, "Replication data for: The spread of the cult of Asclepius in the context of the Roman army benefited from the presence of physicians: A spatial proximity analysis", https://doi.org/10.18710/279HML, DataverseNO, V2. Minerva_inscriptions_05_2026.geojson: A geojson file made from Minerva_inscriptions_05_2026.csv, reprojected to EPSG:3035 ETRS89-extended / LAEA Europe for measurements in meters. Jupiter_inscriptions_05_2026.csv: Inscriptions from the Epigraphic Database Heidelberg mentioning the god Jupiter in various noun cases, with the type of inscription attributes votive, dedicatory/building, honorific, and acclamation. Reused for the region of Roman Dacia from Glomb, Tomas, 2021, "Replication data for: The spread of the cult of Asclepius in the context of the Roman army benefited from the presence of physicians: A spatial proximity analysis", https://doi.org/10.18710/279HML, DataverseNO, V2. Jupiter_inscriptions_05_2026.geojson: A geojson file made from Jupiter_inscriptions_05_2026.csv, reprojected to EPSG:3035 ETRS89-extended / LAEA Europe for measurements in meters. Medicine Medicine_Dacia_05_2026.csv: The dataset includes locations where a) medical implements (scalpels, surgical kits) were excavated and reported in literature that was cross-checked for consensus (N=9); b) oculist stamps (collyrium stamps) attesting the presence of eye-specialists (N=2); c) Roman spas were excavated and reported in the literature with epigraphically attested treatments (N=3); the activity of a Roman physician (medicus) was recorded on an inscription (N=1). The dataset (N=15). Medicine_infrastructure_05_2026.geojson: A geojson file made from Medicine_Dacia_05_2026.csv, reprojected to EPSG:3035 ETRS89-extended / LAEA Europe for measurements in meters. Veterans Veterans_votive_inscriptions_05_2026.csv: Inscriptions from the Epigraphic Database Heidelberg mentioning the word veteranus in various noun cases in Roman Dacia, manually checked to ensure they are indeed associated with veterans of the Roman army and not, for example, personal names. This selection was then cleaned to include only selected types of inscriptions such as “votive”, “building/dedicatory”, and “honorific” to approximate activities involving Roman veterans (N=58). Veterans_votive_05_2026.geojson: A geojson file made from Veterans_votive_05_2026.csv, reprojected to EPSG:3035 ETRS89-extended / LAEA Europe for measurements in meters. Mining OXREP_mines_05_2026.csv: The Database of Roman Mines, Version 3.0, from The Oxford Roman Economy Project (Wilson, 2025; OXREP) was used to identify locations in Roman Dacia with attested mines. The database catalogs mines of the Roman Empire; the predominant types are those that produce metals such as gold, silver, copper, and lead. We filtered the data and selected only mines geocoded to Dacia and with the attribute “Roman” as the period of activity (N=118). OXREP_mines_05_2026.geojson: A geojson file made from OXREP_mines_05_2026.csv, reprojected to EPSG:3035 ETRS89-extended / LAEA Europe for measurements in meters. Dacia_Settlements_Miners_05_2026.csv: To estimate the presence of mining communities, the following dataset was created. First, all inscriptions from the province Dacia in EDH were collected. We excluded inscriptions from small, isolated settlements such as villae rusticae or watchtowers, where miners are highly unlikely to have been present. The remaining inscriptions served as a basis for potential mining sites and were subsequently divided into several categories, including forts, vici (villages), urban centers, or mining settlements. A crucial step was determining how to identify with certainty locations with the presence of miners. One of the key sources used to identify Dacian mining areas is the doctoral dissertation Mining under Empire by S. Y. Lim (Lim, 2018), which provides a comprehensive overview of the research. All settlements included in the EDH database that Lim also lists as mining settlements were included in the dataset. Although we initially considered including mining settlements mentioned by Lim that do not have any epigraphical evidence, we ultimately decided to focus only on those with inscriptions to have mutually comparable and consistent datasets. In other words, since the presence of veterans active in writing inscriptions was approximated through epigraphy, we approached the creation of the dataset approximating mining communities using the same principle (as an additional dataset to the OXREP mines). The next step was to confront the epigraphic data with the dataset of Dacian mines from the Oxford Roman Economy Project (OXREP). If the settlement containing epigraphic data was located less than 10km from the nearest metal mine, it was included in the dataset. If it was found within the radius of 20-30km (i.e., a distance one person could travel in a single day), it was marked as “possible” in the dataset. The last factor taken into account is epigraphic material featuring deities specifically associated with mining, such as Terra Mater, Dis Pater, or Pluto. These inscriptions served as supporting evidence that moved settlements from the “possible” to the “true” category when found at a given site. On their own, however, they were rarely used as the main clue. Finally, only locations with the “true” certainty attribute were kept in the dataset, and “possible” locations were excluded (variable title Miners_settlements, N=27). Dacia_Settlements_Miners_05_2026.geojson: A geojson file made from Dacia_Settlements_Miners_05_2026.csv:, reprojected to EPSG:3035 ETRS89-extended / LAEA Europe for measurements in meters. Military Military_personnel_inscriptions_05_2026.csv: Inscriptions with the social status attribute in EDH containing “military personnel” based on military contexts associated with these inscriptions were reused for the region of Roman Dacia from Glomb, Tomas, 2021, "Replication data for: The spread of the cult of Asclepius in the context of the Roman army benefited from the presence of physicians: A spatial proximity analysis", https://doi.org/10.18710/279HML, DataverseNO, V2. These data were further filtered to contain only geocoded inscriptions from the province Dacia based on the province attribute in EDH (N=232). This dataset has sufficient density to approximate Roman military presence in the province. Military_personnel_05_2026.geojson: A geojson file made from Military_personnel_inscriptions_05_2026.csv; reprojected to EPSG:3035 ETRS89-extended / LAEA Europe for measurements in meters. Military_personnel_centroids_05_2026.geojson: Groups of military inscriptions from Military_personnel_05_2026.geojson with the nearest neighbors within 3 kilometers were represented by a single location, the centroid of the cluster (N of centroids = 50). EPSG:3035 ETRS89-extended / LAEA Europe for measurements in meters. Analytical (distances) Military_network_distances.xlsx: An Excel file with the distances in meters from military centroids (Military_personnel_centroids_05_2026.geojson) to their nearest proxy for variables on the Roman road network (Itiner-e). From 50 centroids, 2 are outside the network and not included in the distances. Military_network_distances_05_2026.csv: A csv file with the distances in meters from military centroids (Military_personnel_centroids_05_2026.geojson) to their nearest proxy for variables on the Roman road network for Python script Dacia medicine Zenodo 05 2026.ipynb. Distances_geo.xlsx: An Excel file with the distances in meters from medical infrastructure (Medicine_infrastructure_05_2026.geojson) to their nearest proxy for variables geographically. Includes coordinates, names of the locations, and individual lists with distances to particular variables (HubDist is the attribute with distances). Distances_geo.csv: A csv file with the distances in meters from medical infrastructure (Medicine_infrastructure_05_2026.geojson) to their nearest proxy for variables geographically. Includes coordinates, names of the locations. Distances_geo_python.csv: A csv file with the distances in meters from medical infrastructure (Medicine_infrastructure_05_2026.geojson) to their nearest proxy for variables geographically for Python script Dacia medicine Zenodo 05 2026.ipynb. Distances_roads.xlsx: An Excel file with the distances in meters from medical infrastructure (Medicine_infrastructure_05_2026.geojson) to their nearest proxy for variables on Roman roads (Itiner-e). Includes coordinates, names of the locations, and individual lists with distances to particular variables (length_m is the attribute with distances). Distances_roads.csv: A csv file with the distances in meters from medical infrastructure (Medicine_infrastructure_05_2026.geojson) to their nearest proxy for variables on Roman roads (Itiner-e). Includes coordinates, names of the locations. Distances_roads_python.csv: A csv file with the distances in meters from medical infrastructure (Medicine_infrastructure_05_2026.geojson) to their nearest proxy for variables on Roman roads (Itiner-e) for Python script Dacia medicine Zenodo 05 2026.ipynb. Itiner-e_for_distances shapefiles are reused from de Soto, P., Pažout, A., Brughmans, T., Vahlstrup, P., Auir, A., Bongers, T., Christoffersen, J., Crépy, M., Johansen, M. H., Lewis, J., MANIERE, L., Massa, M., Møller, L. M. H., Redon, B., Renda, G., Şahin, H., Sobotkova, A., Spatzek, A. L., Verhagen, P., & Weissova, B. (2025). A High-Resolution Dataset of Roads of the Roman Empire: Itiner-e static version 2024 (1.3) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.17122148, and further cleaned in Qgis by reprojecting the network into a metric system (CRS: LAEA 3035), splitting the individual roads into 100-meter-long segments, and ensuring that the whole network in Dacia and the surrounding regions is interconnected to ensure precise distance measurements of the shortest possible paths between the data. Analytical (results) Dacia medicine Zenodo 05 2026.ipynb: A Python script for the calculation of "Replication analysis: The impact of medical infrastructure in military contexts on the spread of Asclepius in Dacia" and "The role of veterans and miners in proximity to medical infrastructure". The top part "Replication analysis: The impact of medical infrastructure in military contexts on the spread of Asclepius in Dacia" uses Military_network_distances_05_2026.csv as the input. The lower part "The role of veterans and miners in proximity to medical infrastructure" uses Distances_geo_python.csv or Distances_roads_python.csv as the input. The seed used for the analysis is 20 for reproducibility. correlation_results_military_network_05_2026.csv: The output of the Dacia medicine Zenodo 05 2026.ipynb, The impact of medical infrastructure in military contexts on the spread of Asclepius in Dacia (the upper part). The table lists attributes: var1, var2, r_s, p_standard, perm_p, ci_low, ci_high. Corresponds to Table 1. correlation_results_distances_geo_python_05_2026.csv: The output of the Dacia medicine Zenodo 05 2026.ipynb, The role of veterans and miners in proximity to medical infrastructure (the lower part). The table lists attributes: var1, var2, r_s, ci_low, ci_high, p_standard, perm_p, FDR(perm_p). Corresponds to Table 2. correlation_results_distances_roads_python_05_2026.csv: The output of the Dacia medicine Zenodo 05 2026.ipynb, The role of veterans and miners in proximity to medical infrastructure (the lower part). The table lists attributes: var1, var2, r_s, ci_low, ci_high, p_standard, perm_p, FDR(perm_p). Corresponds to Table 3.