📄 Royal Names in Old English Poetry

Summary

Malone argues that many royal names in Old English verse began as descriptive epithets for a ruler. Poets then reused these epithets as personal names for invented kings, often the legendary founders of dynasties. He opens by rejecting the reading, repeated by Robert Ramsay in the first issue of Names, that the Scyldings were named after their shields. In his account, the poets created Scyld as an idealized founder and gave him the epithet 'shield, protector' as his name. The dynasty was first named after its real founder, Healfdene. Later the plural Scyldingas was stretched to mean the Danes in general.

He then works through names drawn mainly from Beowulf, Widsith and royal genealogies. For each, he weighs its etymology and its uses as a common noun or adjective. He sorts the material into three groups that differ in how strong the evidence is. The study is meant as illustrative and preliminary. It also warns that many compound names carry no real meaning, so each name needs its own historical and contextual study.

Key points

• Scylding develops in three stages: first 'descendant of Scyld', then the Danish dynasty, then, in poetry only, the Danes as a whole (a part standing for the whole).
• Weapon and armour words serve as royal epithets and become founder names:
  • Scyld means 'shield'.
  • Helm means 'helmet' and is linked to the Helmings and the Wulfings.
  • Brond means 'sword'; it appears in the West Saxon and Bernician pedigrees and is linked to the Brondings.
  • brand is applied to Hrothgar at Beowulf line 1020.
• Other names read as epithets for invented kings include:
  • Freawine ('lord and friend') and Wig ('good at fighting').
  • Breca ('breaker' of rings or of enemy ranks) and Wod ('furious').
  • Wald ('ruler'), Gifica ('the generous one'), Henden, Hun, Meaca, Hagena and Holen.
• Attila means 'little father' in Gothic, a name the subject Ostrogoths gave the Hunnish king. Names ending in -ric, -wald and -mund are often epithetic in meaning, though rarely attested as actual epithets.
• The three groups are:
  1. Words attested both as epithets and as names, e.g. Helm, Freawine.
  2. Words found only as names but easily read as epithets, and belonging to legend, e.g. Breca, Herela.
  3. Ordinary Germanic names, e.g. Theodric, Sigemund.

Interesting ideas

• I1. Malone grades every claim by evidence level: attested in both uses, plausible only from etymology plus a legendary context, or ordinary usage.
• I2. He traces meaning through explicit stages ('descendant of X' → dynasty → whole people) instead of jumping straight to the final sense.
• I3. Many compound names are not compositional (Beadohild is not literally 'war-war'), so splitting a name into elements does not reliably give its meaning.
• I4. He drops his earlier identification of Beadeca with Totila because the Widsith passage concerns the fourth century while Totila lived in the sixth.
• I5. He insists that every context of a name be compared and weighed before reaching any conclusion.

For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

• I1 (world): Knowledge-graph facts derived from names or etymology could carry a similar three-level provenance tag, so answers can say how solid a fact is.
• I2 (frames): Metonymy in coreference (a ruling house standing for a nation) could be resolved by stored, explainable chains from part to whole.
• I3 (tokens): The lemmatizer and morphology should not assign word-like semantics to proper-name parts unless the name is separately marked as meaningful.
• I4 (world): When identifying entities, dates should be checked for consistency, and the system should refuse a merge when the time periods clash.
• I5 (learn): Silver labels for names could require agreement across several occurrences in context before being accepted, with single-context cases serving as a negative control.