📄 Lutz Edzard (ed.), Semitic and Afroasiatic: Challenges and Opportunities

Summary

This is not an NLP paper. It is Alessandro Bausi's short review of Lutz Edzard's edited volume Semitic and Afroasiatic: Challenges and Opportunities (Harrassowitz, 2012). The book's goal is to help Semitists see their own data within the wider Afroasiatic family. It does this through compact grammatical sketches of the non-Semitic branches: Egyptian, Berber, Cushitic, Chadic and, tentatively, Omotic. Each sketch comes with annotated sample texts, interlinear translations and a short lexicon section. Semitic gets no sketch of its own and appears only for contrast. The volume also stays neutral on how Afroasiatic should be classified or how it developed.

Bausi calls the concept innovative, although its format follows Bergsträsser's older introduction to Semitic languages. He lists concrete problems: wrong Ge'ez verb patterns, inconsistent transcription, typos and a Coptic spelling error. His broader criticism is that the book pays too little attention to drift, convergence and internal development. He still concludes that a pan-Afroasiatic view is valuable, above all for Ethio-Semitic specialists, and singles out Appleyard's Cushitic chapter for praise.

Key points

• The volume appeared as Porta Linguarum Orientalium, Neue Serie 24 (2012, 414 pp.). Its authors are:
  • Edzard (introduction),
  • Ruth Kramer (Egyptian),
  • Mohamed Elmedlaoui (Berber),
  • David Appleyard (Cushitic: Beja, Bilin, Oromo, Somali, Iraqw), with Kjell Magne Yri on Sidaama,
  • Herrmann Jungraithmayr (Chadic: Kulere, Mubi, Mushere, Tangale),
  • Rolf Theil (Omotic), with Binyam Sisay Mendisu on Koorete.
• The interlinear annotation uses abbreviations aligned with the Leipzig glossing rules.
• Bausi argues that the Amerindian example given for root-plus-template morphology really resembles Indo-European vowel alternation, as in Greek leip-/lip-/loip-.
• He traces the Tigrinya genitive particle nay- to the noun nəway ("goods, property"), following Praetorius. This counters claims that the particle has an Egyptian or Cushitic origin.
• The Ethiopian area is unusual in containing three Afroasiatic families (Semitic, Cushitic, Omotic) as well as non-Afroasiatic languages.

Interesting ideas

• I1. Every grammar sketch is paired with real sample texts glossed morpheme by morpheme under one shared abbreviation standard, so readers can check claims against data.
• I2. The reviewer treats mixed transcription conventions within one book as a real defect, because readers cannot tell which system applies.
• I3. Root-and-template patterns are shown to resemble ordinary vowel alternation in other families, so they need not be treated as exotic.
• I4. Before positing a foreign origin for a structure, Bausi checks whether internal development explains it, citing a typology where possession nouns turn into genitive markers.
• I5. Diverging hypotheses are laid out side by side without the volume forcing a verdict.

For Mova

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

• I1 (tokens): The morphology layer of crate en could emit Leipzig-style gloss lines for each word, giving a readable trace of how lemma and features were derived.
• I2 (tokens): Fix one normalization scheme for spelling variants and transliterations before tagging, and flag input that mixes schemes.
• I3 (tokens): English strong verbs (sing/sang/sung) could be lemmatized as a stem plus a vowel template rather than as memorized exception lists, which also makes the analysis explainable.
• I4 (world): Possessive constructions could map consistently onto the holdings layer, since possession marking often grows out of words for "property."
• I5 (learn): When witness parsers disagree, keep the competing analyses with their sources instead of silently picking one, and resolve them later with evidence.