A Hebrew Tree Bank Based on Cantillation Marks

What it is about

The Masoretic cantillation marks of the Hebrew Bible act as a punctuation system that splits each verse into nested units. The authors write the rules as a context-free grammar: conjunctive marks join words, disjunctive marks rank the breaks from Rank-1 (Soph Pasuq) down to Rank-5. A CYK parser with separate poetry (Psalms, Proverbs, Job) and prose grammars builds an XML tree bank of every verse. Of 23,213 verses, 23,099 get exactly one tree; the 174 that get several each contain a word with two marks. The authors read the result as a sign that both the grammar and the Masoretic annotation are highly consistent. The grammar uses binary rules, and the paper argues for them over flat ones.

Ideas

• I1. Use the disjunctive rank as a deterministic chunker in the Hebrew reader: the Rank-2 mark (Athnach) gives the main break of a verse. Test whether that break agrees with the main split of the UD root in the tools/hbo output, and report the agreement.
• I2. Binary versus flat bracketing of runs of equal-rank breaks is a real design choice. Mova can run both on the same verses and measure which one matches the parser's constituents better, rather than adopting either by decree.
• I3. The 174 verses with words carrying two marks are a ready list of ambiguous cases. Use them as a test set for the pointed mode of the segmenter and for the cantillation reader.

For Mova

I1 and I2 belong in 2. Cantillation: the accents (te'amim), and I2 also in Syntax: the clause and the sentence as a soft bracketing constraint for gparse on Hebrew. I3 is a test for tools/hbo (pointed mode). The paper's licence is NC, so the verse data is for tests and measurements only, never for a release.

Cite

The paper itself, not this card. BibTeX from the library's record of the work.

@inproceedings{wu2006hebrew,
  title = {{A Hebrew Tree Bank Based on Cantillation Marks}},
  author = {Wu, Andi and Lowery, Kirk},
  booktitle = {Proceedings of the Fifth International Conference on Language Resources and Evaluation (LREC'06)},
  year = {2006},
  url = {https://aclanthology.org/L06-1002/}
}

References

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

1. BFBS (2002). The Masoretes and Punctuation of Bibilcal Hebrew. British & Foreign Bible Society, Machine Assisted Translation Team. find
2. Groves, A & K. Lowery, eds. (2006). The Westminster Hebrew Bible Morphology Database. Philadelphia: Westminster Hebrew Institute. find
3. Jacobson, J.R. (2002). Chanting the Hebrew Bible. Philadelphia: The Jewish Publication Society. find
4. Price, J. (1990). The Syntax of Masoretic Accents in the Hebrew Bible. Lewiston/Queenston/Lampeter: The Edwin Mellen Press. find
5. Richter, H. (2004). Hebrew Cantillation Marks and Their Encoding. http://www.lrz-muenchen.de/~hr/teami. ↗
6. Selkirk, E. (1984). Phonology and Syntax: The Relation between Sound and Structure. Cambridge, MA: The MIT Press. find
7. Wickes, W. (1881). Two Treatises on the Accentuation of the Old Testament. Rev. Reprint by KTAV, New York, 1970. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/lingbulk-2006-wu-hebrew-tree-bank-based-cantillation/