📄 Anglo-Saxon Onomastics in the Old English Andreas

Summary

Robinson examines a textual puzzle in the Old English poem Andreas. At lines 1189–90 the saint addresses the Devil as deofles stræl ("devil's dart"), and the Blickling prose homily on St. Andrew likewise uses stræl. The Greek source instead names the Devil Belial and then explains a byname (Amael, or Samael in one variant) as meaning "blind." The English versions therefore both drop the source's name-gloss and replace the name with a word for "arrow."

Earlier editors explained stræl differently. Zupitza blamed a mistranslation in a lost Latin intermediary, and Brooks suggested it meant a dart-armed demon. Robinson argues that it was a deliberate etymological substitution that links Belial to Greek belos ("dart"). His main support comes from a speech the poet adds later, found in neither the Greek nor the Latin versions, in which the devils name spear and arrow as their weapons against Andreas. He also places the reading within a wider Anglo-Saxon habit of playing on name meanings.

The result recasts an apparent scribal blemish as part of the poet's design. Robinson admits that the exact source text is unknown, so his conclusions remain tentative.

Key points

• The passage under study is Andreas 1189–90, cited from Kenneth R. Brooks' 1961 Oxford edition. The parallel prose is in The Blickling Homilies, edited by R. Morris (1880).
• Several meanings of Belial circulated in the period. Glossaries gave pestilens, Bede gave absque jugo, and Jerome and Bede gave filius praevaricationis, apostata, caeca angustia and caecum lumen. Modern Hebraists read the name as "uselessness."
• The second deviation from the sources is at lines 1330–32. There the poet invents a devil's speech that names gar (spear) and earh (arrow) as the weapons to be used.
• Robinson cites parallel onomastic play elsewhere:
  • In "The Avenging of the Savior," a converted prefect is renamed Titus, glossed as arfæst ("pious").
  • In Ælfric's St. Vincent, the verb oferswiðan ("conquer") occurs ten times, echoing vincere.
  • In Andreas itself, Mambre is linked to perspicuus and Dauid to fortis manu.
• Medieval etymologists often drew name meanings from more than one language. This makes a Greek-based reading of a Hebrew name plausible.

Interesting ideas

• I1. Robinson treats an odd translation choice as intentional only after finding a second, independent deviation that points the same way; one anomaly is weak, two aligned anomalies suggest design.
• I2. The paper notes a double discrepancy against the source: an omitted gloss and a substituted word, which are explained together rather than separately.
• I3. A metaphorical epithet ("devil's dart") stands for a named entity (Belial) and still refers to him.
• I4. Name meanings can foreshadow narrative roles, as with Vincent the victor and Datianus the one who gives up.
• I5. A speculative reading is backed by a simple count: the verb meaning "conquer" recurs ten times, mostly around one character.

For Mova

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

• I1 (learn): Before a self-induced repair rule rewrites a parse that disagrees with silver data, the system could check whether other independent signals in the same text support the "deviant" reading.
• I2 (back): When translating through the meta-world, Mova could compare source and output graphs for omissions and substitutions jointly, and flag pairs where one plausibly compensates for the other.
• I3 (frames): Coreference could allow figurative noun phrases built from a known attribute of an entity to link to that entity, keeping the epithet-to-attribute link as the explanation.
• I4 (world): For fables and tales, a small lexicon of meaningful names could serve as weak, overridable priors on which character wins, loses, or deceives.
• I5 (answer): When answering theme questions about a story, Mova could cite how often a verb class attaches to a character as transparent, checkable evidence.