Press Kit · CC BY 4.0

Tri-Covenant Watch — ready-to-cite paragraphs for journalists

7 copy-paste paragraphs · each pairs an international benchmark (EU FRA / UN Women / OECD) with the actual Taiwan government-document count · sources cited · open for reproduction

For journalists: this page is Ctrl+A copyable. Every snippet is length-checked (≤ 80 words), cites ≥ 3 sources (gov / academic / international), and links to the originating raw file for verification.

The platform aggregates 102,728 government-document segments indexed across three NGOs and three years (CEDAW 29,638 / CRC 69,597 / CRPD 3,493). The seven snippets below correspond to 6 cross-covenant structural findings + 1 methodology coda (platform self-audit).

Suggested citation: Tri-Covenant Watch. (2026-05-06). [Title]. cedaw.taiwanmommies.org. CC BY 4.0 — free to reproduce / adapt with attribution.
📌 Snippet 1 · Finding #1+#2 · PI-17

Disabled women's violence — EU 34%, Taiwan gov 0 segments

🇪🇺 EU FRA 34% 🇺🇳 UN Women 2-4× 🇦🇺 2× CRPD docs: 0 segments ~170,000 estimated
The EU Fundamental Rights Agency's 2014 28-country survey found 34% of disabled women had experienced physical or sexual violence in the past 12 months (vs. 19% for non-disabled women); UN Women estimates a 2-4× risk multiplier. Taiwan's estimated affected population is ~170,000 disabled women, yet "disabled women × violence" co-occurs in 0 segments of CRPD government documents (Tri-Covenant Watch, 2026).

🇪🇺 EU FRA 2014 · 🇺🇳 UN Women · 📰 Full case study

📌 Snippet 2 · Finding #3 · PI-12

"Reasonable accommodation" paradox — CRPD's flagship concept appears LEAST in disability documents

CEDAW 100 segs CRC 173 segs CRPD 7 segs 0.2% of 3,493
"Reasonable accommodation" is CRPD Article 2's flagship concept, yet appears in only 7 segments (0.2%) of Taiwan's 3,493 disability government documents — versus 100 in women's docs and 173 in children's. The most plausible explanation: the disability-rights regime in Taiwan still operates as "administrative tasks" rather than "rights discourse" (Tri-Covenant Watch, 2026).

📐 CRPD §2 + GC-2 (2014) · 📰 Full case study · 📊 3-platform frequency

📌 Snippet 3 · Finding #5 · PI-19 + PI-15

Disabled indigenous — the population all three covenants forget

🇦🇺 NDIS 1.4× 🇨🇦 PALS 2× 🇳🇿 Whaikaha 3 covenants total: 1 segment
Australia's NDIS shows indigenous disability prevalence at 1.4× general population; Canada's PALS at 2×; New Zealand established Whaikaha (Ministry of Disabled People) with explicit Māori representation. Taiwan's 102,728 government-document segments yield 1 total segment at the disability×indigenous intersection (CRPD 0 / CRC 1 / CEDAW 0). CEDAW GR-39 §54-58 already mandates such disaggregation (Tri-Covenant Watch, 2026).

🇦🇺 NDIS · 🇨🇦 PALS · 🇳🇿 Whaikaha · 📰 Full case study

📌 Snippet 4 · Finding #4 · PI-18 + CRC PI-14

SOGIESC 100× concentrated in CEDAW — disabled LGBTQI+ erased across all three layers

CEDAW 1,403 CRC 362 CRPD 0 100× concentration
In Taiwan government documents, SOGIESC (sexual orientation / gender identity) discourse is 100× concentrated in the women's framework: CEDAW 1,403 segments, CRC 362, CRPD 0. The women's framework absorbs ~75% of all gender-identity discussion; the disability framework is entirely absent. Cost: disabled LGBTQI+ people are erased across all three protective regimes (Tri-Covenant Watch, 2026).

📐 CRPD §6 + CEDAW GR-28 + CRC GC-22 · 📰 Full case study · 📊 SOGIESC frequency timeline

📌 Snippet 5 · Finding #6 · CRC PI-14 + PI-18

LGBTQI+ children — low across all 3 covenants, "those who should claim, no one claims"

CEDAW 2 segs CRC 1 seg CRPD 0 segs CRC GC-22 (2017) mandates
The UN Committee on the Rights of the Child's GC-22 (2017) §32-34 explicitly requires that children's-rights reports address LGBTQI+ children, but Taiwan's children's-rights government documents contain only 1 segment; across all three covenants, strict intersection yields 3 total segments (CEDAW 2 / CRC 1 / CRPD 0). The structural pattern: those who should claim, no one claims (Tri-Covenant Watch, 2026).

📐 CRC GC-22 (2017) §32-34 · 📰 Full case study

📌 Snippet 6 · Cross-covenant overview · PI-17 + PI-19

3-platform cross-covenant silence — 6 structural findings

102,728 segs 3 platforms 3 NGOs 5 of 6 findings = "0" or single digit
Tri-Covenant Watch's quantitative analysis of 102,728 Taiwan government-document segments identified 6 cross-covenant structural silences: ① disabled women 0; ② DV 1; ③ reasonable accommodation 7 (CRPD); ④ SOGIESC 100×-concentrated in CEDAW, 0 in CRPD; ⑤ disabled indigenous 1 total; ⑥ LGBTQI+ children 3 total. Each maps to a specific UN treaty-body asks under existing covenant obligations (Tri-Covenant Watch, 2026).

📰 6 findings dashboard · 📋 Executive summary 1-pager

📌 Snippet 7 · Methodology coda · Wave 167-171

Self-audit — dead_anchor 88 → 0 (target met 9 months early)

7th indicator baseline 88 target 0 by 2027-Q1 reached 2026-05-06 9 months ahead
Tri-Covenant Watch quantifies government silence — and quantifies its own. The link_audit tool scans all 3 platforms' HTML and classifies broken links into 4 categories. Wave 167 promoted dead_anchor as the monthly tracker's 7th indicator (baseline 50 → target 0 by 2027-Q1); Wave 171 reached zero — 9 months early. Same period: broken_target 3,624 → 10. Lesson: when a quantitative standard is published and watched by readers, self-discipline accelerates (Tri-Covenant Watch, 2026).

📰 Full case study (with W169-171 postscripts) · 🩺 QA dashboard trend

📦 Platform boilerplate (~80 words)

Tri-Covenant Watch is a Taiwan civil-society monitoring platform jointly maintained by AABE (Association for the Advancement of Basic Education) and two partner NGOs, covering Taiwan's implementation of CEDAW (women), CRC (children), and CRPD (disability) — three UN human-rights covenants. Built on full-text indexes of 102,728 government-document segments, it produces cross-covenant quantitative analysis. Updated monthly (auto), with a built-in 7th self-audit indicator (Platform Health). CC BY 4.0 open license.

📞 Interview / further data

Researcher interviews / case inquiries / advanced data (SQLite + JSON API): weall888@gmail.com
Open-source pipeline: github.com/naer-tw/evidence-pipeline
RSS subscription: feed.xml · Atom (monthly trend auto-pushed)
Raw SQLite databases (for research / secondary analysis): each platform's GitHub repo under data/*.db.

All paragraphs released under CC BY 4.0 — free to reproduce / adapt with attribution. Suggested citation:
Tri-Covenant Watch. (2026-05-06). "Press Kit." cedaw.taiwanmommies.org/press-kit-en.html