The hunt for the lost cities of Ptolemy

Abstract

Almost two thousand years ago, Claudius Ptolemy created a world map, identifying the names and coordinates of over 8000 settlements and geographical features. In the British Isles, the locations of many of Ptolemy's cities, such as London and Bath, are known to us today. Yet many still elude us. Most efforts to find and identify these hidden cities, rivers, headlands, and so on have concentrated on the etymology of modern place names, trying to work backwards in time until a match could be found with a name given by Ptolemy. Rather than looking at the names, I am focusing on the other half of Ptolemy's data: the coordinates. Using the data of those cities and landmarks that have already been identified, I am applying a series of best-fit transformations to Ptolemy's map of the British Isles. It is my hope that by ‘correcting’ Ptolemy's map, the validity of previous identifications can be tested and new conjectures can be made about those locations that are still lost. This article was presented as a talk at the Research in Progress meeting of the British Society for the History of Mathematics at The Queen's College, Oxford, 1 March 2008.

Notes

¹ These eight identifications have been made by Stückelberger and Graßhoff, the editors of the most recent Ancient Greek edition of Ptolemy's Geography, which was published in October 2006 after an extensive examination of the surviving Geography manuscripts.

² My guesses are just that and hold no weight. Guesses were made in the following manner: Ptolemy lists a headland and river mouth heading north between the River Boyne and Belfast Lough. Maintaining this order, I have chosen St. John's Point and Strangford Lough as the identification, for no better reason than they were the most prominent of headlands and bodies of water.

³ The rectangles consist of four corner points and four midpoints.

⁴ The rotation of Scotland has been a subject of debate for generations. Though no satisfactory answer has been found, one theory is that Ptolemy adjusted his data so that Scotland would fit below the island of Thule, which was his theoretical northern most habitable land (see arguments in Jones and Keillar 1996).