The use of the Septuagint in Afrikaans Bible translations

What it is about

The paper compares three Afrikaans Old Testament translations (1933/53, 1983 and 2020) on one question: where do they depart from the Masoretic Text in favour of the ancient versions such as the Septuagint? The method is a qualitative comparison of verse-level choices, illustrated with cases from 1 Samuel (for example 1:24, 3:13 with the scribal tiqqun, 6:18 and 10:19) and from Ezekiel. Counted by instance, the 1983 translation refers to the ancient versions more often than the 1933/53 pair does. The 2020 translation footnotes every deviation from the Masoretic Text, and even some verses it translates from it when the ancient versions have important variant readings. The paper's main finding is that the ancient versions were often used to normalise a text where the Masoretic reading presents a serious problem or ambiguity, and that the normalisation was mostly unannounced in the older translations. The counts are case counts, without statistics.

Ideas

• I1. Treat a reading chosen from another witness as an author delta: where a translation picks a different reading for the same source words, log it as an override with its verse id. Count the overrides per translator and per reading source, so that the delta learner sees the same kind of evidence as the footnotes.
• I2. Use the footnotes as a labelled list of expected overrides. The 2020 translation marks each deviation, so its list can serve as a held-out check of the delta learner. It is a check on the learner, not a training source.
• I3. Keep two candidate concepts for an ambiguous verse rather than one, and let the chosen author decide between them. This addresses reader ambiguity (open problems 1 and 5), and the choice is then an explicit, countable delta.

For Mova

The paper's cases are the kind of author delta that tools/translate learns, here for Hebrew with ancient witnesses. The Hebrew side is read through tools/macula and tools/hbo. The paper gives no measure to copy, so only the logging and the held-out check are proposed. Summary only is needed.

Topics: Machine translation

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.4102/hts.v76i4.6137

@article{rooy2020use,
  title = {{The use of the Septuagint in Afrikaans Bible translations}},
  author = {Rooy, Herculaas F. van},
  journal = {HTS Teologiese Studies/Theological Studies 76},
  year = {2020},
  doi = {10.4102/hts.v76i4.6137},
  url = {https://doi.org/10.4102/hts.v76i4.6137}
}

References

2 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Allen, L.C., 1994, Ezekiel 1–19 (WBC), Word, Dallas, TX. find
2. Barthélemy, D., 1982, Critique textuelle de l’Ancien Testament, 1, Josué, Juges, Ruth, Samuel, Rois, Chroniques, Esdras, Néhémie, Esther (OBO 50/1), Universitätsverlag, Freiburg/ Vandenhoeck & Ruprecht, Göttingen. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/rooy-2020-use-septuagint-afrikaans-bible-translations/