Answers are grounded in this paper and cite their sources.
Examines economic behaviour in the Eastern Roman Empire, suggesting logistical centralization of capital during turmoil.
This paper reports the first stage of a two-part research programme investigating the spatial organisation of the provincial economy of the Eastern Roman Empire during the Crisis of the Third Century (235–284 CE). Where the Stanford ORBIS geospatial network model reconstructs the formal, state-administered transport system of the Roman world under idealised peacetime conditions, the present study uses empirical numismatic deposition data, principally the spatial distribution of silver and bronze coin hoards recorded in the Coin Hoards of the Roman Empire (CHRE) database, as a proxy for civilian economic behaviour under conditions of systemic military and fiscal collapse. A continuous topographic friction surface derived from Copernicus GLO-30 elevation data is used to generate Minimum Cost Path (MCP) and Least Cost Path (LCP) routing models, isochrone transit-time surfaces, spatial autocorrelation statistics (Global Moran’s I), Density-Based Spatial Clustering (DBSCAN), betweenness-centrality network analysis, and an Agent- Based Model (ABM) of evasion behaviour, applied to a working dataset of 849 localised hoards across Anatolia, the Levant, and the Balkan provinces, bounded chronologically between the accession of Maximinus Thrax (235 CE) and the monetary reforms of Diocletian (284 CE). The paper proceeds dialectically. It first develops, region by region and metric by metric, the observational case for what is termed the Terrestrial Evasion Hypothesis: the proposition that, as the Roman fiscal and military apparatus fractured under Sassanid and Gothic pressure, civilian populations and localised markets withdrew high-purity capital into topographically isolated, high-friction terrain, the Anatolian plateau, the Cappadocian steppe, the Taurus Mountains, and comparable interior massifs, to evade state requisition and the effects of currency debasement. This evidentiary case is built from friction and isolation indices, ceramic and numismatic proxy justification, chronological and metallurgical analysis, isochrone and least-cost-path cartography, spatial autocorrelation testing, network centrality analysis, and mint-to-deposition displacement metrics. The hypothesis is then subjected to a direct empirical test: a predictive Agent-Based Model of the geographic distribution that pure evasion behaviour would generate is compared against the actual, documented archaeological distribution ofhoards. The comparison falsifies the Terrestrial Evasion Hypothesis and the associated Macro-Regional Economic Panic Hypothesis. In their place, the paper proposes and substantiates the Logistical Dependency Hypothesis: that high-intrinsic-value capital during the crisis remained hyper-centralised along the Roman Empire’s own military and maritime logistical corridors, because the physical security these corridors provided outweighed the fiscal cost of continued exposure to state extraction. The paper concludes by identifying the precise point at which this reversal occurs in the data and reserves the full theoretical elaboration of the Logistical Dependency Hypothesis for the second paper in this series. This project does not seek in any way to invalidate the ORBIS project. ORBIS remains the most rigorous available reconstruction of the formal, state-subsidised transport geometry of the Roman world, and the routing and friction architecture used throughout this paper is built to be read alongside it, not against it. Rather, this project seeks to provide a complementary layer of empirical, behaviourally grounded data, describing how the civilian and military economy actually moved and hoarded capital when the conditionsORBIS models as idealised and stable broke down, with the intention of supplementing, rather than superseding, the ORBIS dataset.
No takes yet. Share an insight, caveat, or question.
Damian Noah Dimitrov (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: