has been understood as a prehistoric sheiling and also as a barn for hay-fodder. Today
there is no doubt that salr in place names must be seen in a much more ‘aristocratic’
context (Brink 1996 : 255 – 8 ). It was the Old Scandinavian word for a king’s or a
chieftain’s banqueting hall. The few place names containing this word are prominent
places, such as Uppsala in Sweden and Skíringssalr in Norway. A couple of names contain
the god Óðinn’s name: Onsala in Halland, Odensala in Jämtland and the old name of
Huseby in Onsøy (< Óðins-øy) in Østfold, Norway. The element salr is not primarily to be
set in a sacral context, but the denotation ‘hall’ reveals that cultic matters certainly have
been conducted at these places. The word harg, ON ho ̨rgr has a similar background. In
the Icelandic literature it often has the same meaning as hof, hence denoting a cult site or
a cult building. The original meaning of the word was ‘heap of stones’. The assumption
is that from this original meaning a new one, ‘stone altar (on the outside)’, has emerged,
and later on from this ‘cult house’. The word is found as simplex Harg, and in com-
pounds: OSw Oþenshargh, Torshälla (< OSw Þors-hærghe), Skederid (< OSw Skæþ-hargh).
Some originally profane words also obviously occur within a cultic and sacral topo-
nymic context, namely åker ‘arable land’, both as simplex Åker/Aker and in compounds:
Torsåker (< OSw Þors-akir), Onsåker (OSw Oþens-akir), Frösåker (< OSw Frøs-akir)
(Vikstrand 2002 , 2004 ). Similar is the case of vang, ON vangr, in Norway, often found as
the focal farm in the district, by the church. The word eke (< ek-ia) ‘oak grove’ with a
sacral toponymic meaning we find in Onsike (OSw Oþens-eke), Hälke (< OSw Hælgha-eke
‘the holy oak grove’), Alsike (probably < OSw Alhs-eke), and the word böke (< bok-ia)
‘grove of lime trees’ has a similar background. The word lund ‘grove’ had, of course,
originally a profane meaning, but there is no doubt that the word could eventually
appear in a cultic context, not only in compounds (Torslunda, Fröslunda etc.), but also as
a simplex: Lund. An interesting case showing this is the name Oklunda, found in
Östkind’s hundred, in Östergötland. Where the farm Oklunda is situated we have a
runic inscription carved in the rock, saying that the place (during the Viking Age) was a
vi, hence a ‘cult site’, and that cult site must have had the name Oklunda ‘the (cultic)
grove on the yoke’ (referring to the topographical situation) (Gustavson 2003 ; Brink
2003 : 93 – 6 ). In Denmark there are several cult sites containing the word ODa hyllæ
(probably) ‘shelf’, often with the name of the god Óðinn as the qualifier, as in Vonsild
and Onsild (< ODa Othæns-hyllæ) on Jutland.
BIBLIOGRAPHYAnderson (Arngart), O.S. ( 1934 ) The English Hundred-names (Lunds universitets årsskrift. N.F.
Avd. 1 , 30 : 1 ), Lund: no publ.
——( 1939 a) The English Hundred-names. The South-eastern Counties. With a Survey of Elements
Found in Hundred-names and a Chapter on the Origin of the Hundred (Lunds universitets årsskrift.
N.F. Avd. 1 , 37 : 1 ), Lund: Gleerup.
——( 1939 b) The English Hundred-names. The South-western Counties (Lunds universitets årsskrift.
N.F. Avd. 1 , 35 : 5 ), Lund: Gleerup.
Andersson, Th. ( 1965 ) Svenska häradsnamn (Nomina Germanica 14 ), Uppsala: no publ.
——( 1982 ) ‘Hund, hundare och härad från språklig synpunkt’, Bebyggelsehistorisk tidskrift,
4 : 52 – 66.
——( 1992 ) ‘Kultplatsbeteckningar i nordiska ortnamn’, in G. Fellows-Jensen and B. Holmberg
(eds) Sakrale navne (Norna-rapporter 48 ), Uppsala: Norna.
——( 2004 ) ‘Svethiudh. Det svenska rikets kärna’, Namn och bygd, 92 : 5 – 18.
–– chapter 6 : Naming the land––