Peder Gammeltoft
https://orcid.org/0000-0002-6996-9616 
University of Bergen

https://doi.org/10.4467/K7501.45/22.23.18056

As with other research, an important part of place-name research is how to localize one’s object of study. Traditionally, place-name localization has been indirect, stated in relation to in which administrative unit the place name was situated. With the so-called spatial turn, geolocation and place-name geodata have started to play an even more significant role in digital onomastics.
A tacit premise of geolocation is that it is feature oriented, regardless of whether it is point, multi-point, line or polygon oriented. This is principally at odds with the nature of place names, which have an in-built multi-referentiality. No one has focused on this theoretical problem so far, although, to the advanced user of geolocation-oriented digital onomastics, this is a constantly recurring problem.
The present paper explores this problem and proposes a solution to it by introducing the notion of the Unique Place-Name Concept to geolocated place-name databases. This addition will have the added bonus of enabling quick comparisons among multiple features with the same name origin and thus strengthening place-name standardization, making it easier to utilize geodata onomastically, and preventing the doubling of data, among other benefits.

Keywords digital onomastics, place-name databases, conceptual modelling, name-centric model