Syntax tab
Every token in the corpus carries a signed dependency distance to its head (distance = position of the word minus position of its head; 0 marks the root). The workbench recovers each word’s head from that record and draws the sentence’s dependency tree; the trees are unlabelled, because the corpus stores the attachment but not the relation name.
- Pick a sentence: choose a work, book, and line number; the dependency tree (renderable horizontally, vertically, or as bracketed text) and a token table are drawn for the sentence containing that line (with a selector when two sentences share the line). A manual TSV mode remains for pasting external trees.
- Syntax-Meter Interface Search: because the same tokens carry their metrical record, the tab measures the syntax-metre interface directly from the corpus: where sentence boundaries fall in the verse (line end, the caesura points, the bucolic diaeresis), how many lines are enjambed, how many dependency arcs cross a line boundary, plus head direction and dependency length by part of speech.
1. Pick a sentence from the corpus
Choose a line; the whole sentence containing it is drawn (Homeric sentences often span several lines).
Or paste a tree manually (TSV)
One token per line: id[TAB]form[TAB]lemma[TAB]pos[TAB]head[TAB]deprel[TAB]distance(optional).
2. Sentence structure
Dependency Tree
Corpus-wide measurements of the syntax-metre interface, computed live from the merged record: sentence boundaries located by the metrical position of their final word, enjambment, arcs across the line boundary, and dependency profiles.