Labels and symbols: using text on maps to investigate the antiquities on the ordnance survey maps of Great Britain

Valeria Vitale, Katherine McDonough

«Archeologia e Calcolatori» 2024, 35.2, 355-362; doi: 10.19282/ac.35.2.2024.37

Abstract

This paper discusses the application of a machine learning pipeline to automatically digitise text on historical maps and make it searchable, as explored by the ‘Machines Reading Maps’ project. Looking at the Ordnance Survey Maps of Great Britain as a case study, we will suggest ways in which this new kind of open datasets, of both a textual and spatial nature, offers the unprecedented opportunity to study maps at scale, analysing map collections as digital corpora. These new approaches facilitate the use of map as historical sources in humanities research, and their investigation as complex cultural objects that combine heterogeneous knowledge. In particular, we will focus on the uncommonly detailed representation of ancient sites on the Ordnance Survey maps, and how a further layer of information around them is delivered not by the words’ literal meaning but by their appearance. We will propose ways in which this peculiarity could be digitally leveraged to retrace, investigate, and perhaps re-interpret the archaeological information on the Ordnance Survey maps. We will conclude by reflecting on the need for new, more sophisticated workflows that take into account the richness of information delivered by visual clues in words printed on maps.

Figures

Preview