UD v1 labels in the examples of specific-syntax.md

Gist. The examples in specific-syntax.md are still written with UD v1 labels: dobj, nsubjpass, auxpass, csubjpass, neg, mwe. They do not exist in v2, the validator rejects them, so there are 0 of them in the EWT data. The analyses themselves are correct; only the label names are outdated.

Guideline. In the sdparse blocks of 2.18:docs/_en/specific-syntax.md there are 36 lines with v1 labels (grep): dobj 20, mwe 5, neg 4, auxpass 3, nsubjpass 3, csubjpass 1. The text also has v1 names, for example :96 (nsubjpass, csubjpass).

v1 → v2 mapping (docs/_v2/summary.md:74–100):

• dobj → obj;
• nsubjpass → nsubj:pass, csubjpass → csubj:pass;
• auxpass → aux:pass;
• neg → advmod (in English — with Polarity=Neg on not);
• mwe → fixed.

EWT 2.18 data. Not a single v1 label (the registry en/data/ud-registry-en.tsv does not have them; the en reader does not know them).

What this means for Mova. Take examples from specific-syntax.md into seeds only with this replacement. They cannot serve as a corpus for checking.

Sources. 2.18:docs/_en/specific-syntax.md; docs/_v2/summary.md; docs/workgroups/v1_to_v2.md.

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/dialects/ud-2.18/seeds/v1-labels/