VCP — Vācaspatyam (1873–1884)

Chapter authored per Decision 29 Tier C. Position 9 in the original atlas chapter ordering — the second Sanskrit-Sanskrit lexicon and the framework's outer limit for the structured-bilingual block model.

Trust Block

• Evidence: CDSL v02 vcp.txt, src/data/cross-dict.json, and linked dictionary-profile/audit notes.
• Limitations: narrative atlas profile; counts summarize committed/generated artifacts and do not replace source edition inspection.
• Validation: checked by npm run build; chapter consistency is tracked in _consistency_audit.
• Owner repo: csl-atlas.
• Next use: use this dictionary profile as context, then open source-linked records or compare the lemma in Reader Lookup.

1. Overview

Vācaspatyam ("the work of Vācaspati", named after the editor's epithet), an encyclopedic Sanskrit-Sanskrit dictionary compiled by Tārānātha Tarkavācaspati, published Calcutta 1873–1884 in seven volumes. VCP is the second indigenous-Indian large-scale Sanskrit-Sanskrit lexicon in CDSL, contemporary with PWG's later volumes and PWK's beginning, and one of the major CDSL Sanskrit-Sanskrit works (50,135 records). VCP is structurally similar to SKD — encyclopedic, monolingual, prose-paragraph entries with inline iti citation — but with a much sparser inline citation density (0.26 iti / record vs SKD's 1.70). In the nine-chapter atlas path, VCP confirms the genre boundary discovered at SKD; the all-dictionary coverage layer extends beyond this chapter path by measuring partial framework fit across every available CDSL v02 dictionary.

2. Genre-bound (the framework does not apply — second confirmation)

VCP confirms the genre boundary that SKD established. The same three preconditions fail in VCP as in SKD:

1. No <lex> tags. Gender + grammatical category encoded inline in Sanskrit.
2. No <ls> tags. Sources cited via inline iti <source> prose.
3. Long encyclopedic prose entries. Mean 494 characters per entry.

The 18-block detector produces the same degenerate output as for SKD: every block-presence column at or near zero, no meaningful profile differentiation.

Block-presence by name:

3. Prose-pattern analysis — sparser iti than SKD

VCP's inline-iti citation density is 0.26 per record — about a sixth of SKD's 1.70. The same iti-formula apparatus is operative (sources named in prose as "… iti "), but VCP's editorial choice was to quote more sparingly. Where SKD entries typically include 1–2 quoted prior-kośa passages, VCP entries are often encyclopedic prose without quotation — Tārānātha's own synthesis, with named-source attribution reserved for direct quotations. This is a different editorial philosophy within the same genre:

SKD is the quotation-rich kośa (the prior tradition speaks through the entries); VCP is the synthesis-rich kośa (the editor speaks, citing only where direct quotation matters). Both are valid genre-internal variants — neither maps to the European-bilingual structured-block framework.

4. Why the framework changes here — boundary, not endpoint

This chapter makes the framework's scope explicit. The 18-block apparatus developed for MW in PAPER.md §3 was designed to analyse structured bilingual dictionaries — works where every source citation occupies a discrete tagged slot and every entry decomposes into a kernel + enrichment block-economy. SKD and VCP do not have this structure; they are encyclopedic Sanskrit-Sanskrit kośa works whose source-discipline lives in prose. The 18-block detector applied to them returns a degenerate "everything is zero" reading — not because the dictionaries are deficient, but because they are a different kind of artefact.

PAPER.md §8 names this explicitly:

"The block apparatus is therefore genre-bound to structured bilingual dictionaries — a useful limit on the framework's scope."

And PAPER.md §10 further:

"Sanskrit-Sanskrit lexica … require a different microstructural framework, oriented to inline-iti citation rather than tagged sources."

The original chapter sequence therefore reaches a boundary at VCP, but the atlas itself should not stop there. The next layer is an all-dictionary coverage and size inventory: first measure which blocks partially transfer to every CDSL v02 dictionary, then design a companion microanalysis for inline-prose citation, with iti as its primary structural unit and paragraph-flow as its block unit. That extension would include SKD, VCP, and the indigenous-kośa tradition the four kosha repos (ARMH, ABCH, ACPH, ACSJ) represent.

5. Lineage statement

VCP occupies the synthesis-rich position in the indigenous Sanskrit kośa tradition. Tārānātha Tarkavācaspati (active 1855–1885) was a Calcutta-based Sanskrit grammarian and lexicographer; his Vācaspatyam draws on the same prior-kośa sources as SKD (Amarakośa, Hemacandra, Medinīkośa, Halāyudha) but with a more synthetic editorial voice — quoting less, explaining more. VCP is contemporary with PWG's later volumes (1873–1884 vs PWG's last volume 1875) but the two enterprises did not draw on each other: PWG cited prior indigenous kośa sources directly; VCP did not engage with the European tradition. The two run as parallel monuments of 19th-century Sanskrit-lexicography on opposite sides of the genre boundary.

The four kosha repos that resolve MW's <ls>L.</ls> hedge are prior to both SKD and VCP in the indigenous tradition; SKD and VCP synthesised them rather than introducing new material. The full lineage chain of the indigenous tradition lies upstream of this atlas.

6. Cross-references — divergence/convergence with adjacent chapters

7. Decisions log

• Boundary chapter in the original nine-dictionary path — cross-references now point onward to all-dictionary coverage rather than treating VCP as the atlas's last possible node.
• Tier C template per Decision 29 §29.2.
• Two-monograph comparison with SKD structures §3: SKD vs VCP shows two editorial philosophies within the genre-bound kośa tradition (quotation-rich vs synthesis-rich).
• Explicit framework-limit statement in §4 — the atlas's purpose in including SKD and VCP is to demarcate the framework's outer scope, not to extend the framework to cover them.
• Phase-5 future work flagged in §4 as the next research project (all-dictionary coverage first, then a different microstructural tool for inline-iti citation).
• Effect-size threshold (D19): all numerical comparisons (SKD 1.70 vs VCP 0.26, both vs MW 0) are large effects.

8. Data dictionary + reproducibility manifest

• DATA_DICTIONARY.md — VCP/DATA_DICTIONARY.md (notes absence of <lex>/<ls>)
• Source file: csl-orig/v02/vcp/vcp.txt
• Block-detector script: figures/scripts/export_data.py (same degenerate output as SKD)
• Cross-dict aggregate JSON: src/data/cross-dict.json (includes VCP metadata; VCP does not use tagged <ls> citation blocks)
• License: CC-BY-SA-4.0

See also (tools + epilogue)

• Cross-dictionary comparison — VCP shown alongside SKD and the structured bilingual dicts for the nine-chapter comparison
• All-dictionary coverage — VCP shown inside the full CDSL v02 coverage and size inventory
• VCP#31 — docs-pass tracking issue
• Boundary of the nine-chapter path. Continue to the coverage tool for the full-dictionary inventory.

Source: CDSL vcp.txt 2026-05-23 · MWS docs-pass commit reflects audit pipeline as of 2026-05-28 · CC-BY-SA-4.0