Attested: Ptolemy 2,3,8 Σελγοουαι (or Ελγοουαι) with 4 πολεις at Καρβαντοριγον (Raeburnfoot), Ουξελλον (Caerlaverock), Κορδα (Biggar area), and Τριμοντιον (Newstead).
Where: People in the Scottish borders.
Name origin: Possibility (1): R&S followed Jackson (1953:467) in likening the first element to Irish selg ‘hunt’ and its Celtic cognates such as Welsh heliaf ‘to hunt, to pursue’. Irish selg could also mean the target of hunting, which would make it cognate with Scots selkie ‘mermaid’ and OE seolh ‘seal’, German Shelk (in the Nibelungenlied, probably a large stag), and Russian селезень ‘drake’. If this analysis is correct the name need not have been constructed in a distinctively Celtic language, but it might explain the inscription that mentioned carvetior.
(2) As written, the name's most natural parallel is the Greek city of Σελγη, in the south of modern Turkey, whose inhabitants were tough fighters, similar to the Spartans. Maybe the Romans found Scottish border people tough fighters. The closest dictionary word in Greek is ασελγεια ‘licentiousness’, of uncertain origin, but most likely implying that *σελγε meant something like ‘moral strictness’.
(3) PIE *selos ‘swamp’ offers another possibility, though crannog-style dwellings do not seem particularly common in this part of Scotland.
(4) If divided Sel-govae the name would begin like PIE *sel- ‘room, house’, which led to OE sele or sael ‘hall, house, dwelling’, French salle and German Saal ‘large room’, Russian село ‘village’, etc. It would end like cove, related to German Koben ‘stable, shed’. Initial G (the main PIE source of Germanic K) survived in Greek γυπη ‘vulture's nest’ (typically “in caves and on ledges and rock outcrops or caves on steep rock walls”) but changed to ZH in words such as župa found in Croatia and neighbouring countries for what started out meaning ‘cavity, pit’ but evolved to mean something like ‘administrative centre’.
Notes: Analysis (4) is a clear winner because this people's territory contained so many scooped enclosures (Jobey, 1962). This term is used by archaeologists for houses or whole settlements that are partially cut into a hillside, anticipating what modern architects now call earth-sheltered homes, which one can often see being built on TV programs. This name sounds like a a typical Ptolemy comment on the lifestyles of people in a particular area, rather than indicating political, or even ethnic, unity, and has an interesting echo in the way his Σμερται people lived in blackhouse territory. If correct, this analysis raises questions about elements like gob in other early names.
The main arguments against analysis (1) are these. (a) Hunting (of red deer, Irish elks, shelks, reindeer/caribou, or whatever other big game still survived in Roman times) could only have been a leisure activity in areas where most Selgovae lived (mainly by cattle farming), probably capable of producing the same social tensions as fox-hunting today, and unlikely to form the basis for a group name. It could be the viable basis for a subsistence economy only in special cases, such as hunting seals at the coast, where one would want to discuss OE seolh as a parallel some more. (b) There is no agreed etymology for Irish selg. An idea of “set loose the hounds”, based on PIE *selg- ‘to set free, throw away’ (the source of English sulky), has been enthusiastically embraced, but seems contrived, and inferior to a link (suggested by MacBain's dictionary) to PIE *sel- ‘to take, seize’. And semantic evolution from the act of hunting to its quarry, or vice versa, is not agreed. (c) The –ovae ending is not a trivial issue. Similar suffixes form past participles or adjectives. Possible parallels are in Comedovis, Lemovii, and Ληξουβιο, etc. Or perhaps –ovae referred to water, as discussed here, possibly the water's edge, like OE ofer.
Last Edited: 11 September 2017