📄 Rewriting Anglo-Saxon: Notes on the Presence of Old English in Contemporary Literature

Summary

Fernando Galván's essay (University of La Laguna, SELIM 2) challenges the view that Old English literature is dead and can be dropped from university curricula. It is literary criticism, not empirical work. Its frame is intertextuality: every text reuses earlier words, yet each new context changes what they mean. Borges's "Pierre Menard" serves as the model case, with Gérard Genette's typology of transtextuality as the analytic tool.

The author first notes that Old English devices never fully vanished after the Norman Conquest. Following John Speirs, he points to alliteration and compounding surviving in later English poets. He then examines twentieth-century rewritings by Pound, Auden, Yourcenar and Borges. Pound's version of The Seafarer is defended as an attempt to recreate the original's sound and effect rather than its dictionary sense. Borges's Spanish poems on Hengist and on the Brunanburh poet merge his own voice with the Anglo-Saxon one.

The conclusion is that creative rewriting, as opposed to purely philological reconstruction, keeps Old English alive for modern readers.

Key points

• Genette's Palimpsestes (1982) supplies five kinds of transtextuality: intertextuality, paratextuality, metatextuality, hypertextuality and architextuality.
• Pound's Seafarer has known lexical errors: wrecan given as "reckon", Lifge mid englum as "'mid the English", and byrig faeriath turned into berries. The author argues these were deliberate trade-offs for sound.
• Michael Alexander calls the demand for a lexically exact prose version an "archaeological fallacy". In his view, Pound's phonetic simulacrum may come closer to the original audience's experience.
• The motivation is Valentine Cunningham's 1991 TLS call to end compulsory Anglo-Saxon in the Oxford English course. Hopkins's 1882 letter praising Anglo-Saxon purity is also cited.
• The Borges poems analysed include "Hengist quiere hombres (449 A.D.)", "A un poeta sajón" and "Lo perdido". The essay finds paratextuality and metatextuality dominant in the Hengist poem.

Interesting ideas

• I1. The Menard argument says an identical string can carry different meanings when the author, era or reader changes. Interpretation therefore depends on context, not only on surface form.
• I2. Genette's five-way typology names distinct kinds of relation between texts, such as quoting, commenting, imitating, framing and genre membership, instead of lumping them together as "influence".
• I3. Pound's translation sacrificed lexical accuracy for sound and effect, while critics judged it only by dictionary meaning. One output can thus be strong on one axis and weak on another.
• I4. Several of Pound's slips (wrecan → "reckon", englum → "English") come from mapping a word onto a similar-sounding modern one. This is the classic false-friend trap.
• I5. The "archaeological fallacy" warns that a literal reconstruction is not automatically closer to the original experience.

For Mova

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

• I1 (frames): Pragmatic readings such as irony or implicature should take discourse and speaker context as explicit inputs, so the same sentence can legitimately receive different frames.
• I2 (world): Cross-text links in the knowledge graph could be typed with a small closed set like this, keeping provenance and reuse relations explainable.
• I3 (back): Evaluate generation and translation on separate axes, such as propositional fidelity and stylistic or prosodic fidelity, rather than one blended score.
• I4 (tokens): Lemma and gloss lookup for archaic or foreign material could flag sound-alike candidates as suspect, so these false-friend matches are caught before reaching the tree.
• I5 (learn): Treat literal paraphrase as a negative control in evaluation, checking whether a model's "faithful" output actually preserves what readers need rather than just word-level overlap.