Answers are grounded in this paper and cite their sources.
On December 10, 2021, an EF-4 tornado struck downtown Mayfield, Kentucky. Twenty-seven of the thirty-one documented buildings in the listed historic district were subsequently demolished, several structurally viable at the time. Drawing on a building-level dataset, thematic analysis of oral history accounts, and semi-structured interviews with institutional actors, this paper examines what produced that loss at this scale and speed. We argue that tornado damage alone did not determine demolition. Heritage loss in Mayfield was produced by prior erosion of the institutional infrastructure through which preservation could have operated. In that absence, the post-disaster environment was occupied by financial, regulatory, and logistical forces oriented towards clearance, with no countervailing mechanism to interrupt them. Place attachment to the district was strong but did not produce advocacy: twenty of twenty-two participants expressed clear attachment, only three advocated for preservation. Without institutional infrastructure to contest demolition or create time for alternatives, attachment produced grief rather than action, challenging assumptions that attachment to heritage functions as an inherent recovery resource. The case establishes the lower boundary of what post-disaster response can achieve. It identifies pre-event governance capacity as the variable determining which recovery outcomes remain available in small, hazard-exposed jurisdictions with limited administrative capacity.Policy highlightsPre-disaster institutional capacity shaped heritage loss more than damage severity.Without preservation capacity, post-disaster incentives push towards demolition.Place attachment did not become advocacy without institutions to channel it.Heritage planning must anticipate the incentives that fill an institutional gap.Mutual-aid agreements and demolition stays can create decision time at low cost.
No takes yet. Share an insight, caveat, or question.
Maalouf et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: