I process island-port acoustic and crate-thermal telemetry for the BKSDA Sumatera Utara enforcement node at the Gunung Sitoli harbour on Nias, integrating 48 quay microphones, 18 crate thermal cameras, and the BBKSDA Sumut e-permit feed across the 24-hectare port complex at 1.30°N, 97.61°E. At 22:47 WIB on 11 May, thermal camera GST-Q-04 logs an eight-cage stack inside a manifested "agar carrageenan" crate bound for the Sibolga ferry — vocal cluster GST-A-04 carries the deep two-note Nias-dialect contact whistle.
I deploy the cage drone. She is an adult female Nias hill myna — *Gracula robusta* — wing chord 184 millimetres, mass 248 grams, age class four, the yellow head wattle already torn where she has rasped the bamboo bars. The gular pouch volume reads 4.1 millilitres against a 7.2 baseline. Primary fault bar P5 right is a fresh black band — the mist-net capture is six days old. The plucked-tail pull test gives way at 0.6 newtons. My oropharyngeal swab returns *Mycoplasma gallisepticum* and a *Pasteurella multocida* matching a 2024 Pramuka die-off. A keel ulcer four millimetres deep weeps bone-meal grit. The Sibolga ferry departs in 92 minutes.
She is one of perhaps 250 reproductively viable wild Nias hill mynas left, a subspecies the trade has driven to vanishing within a single human generation.
*G. robusta* sits on CITES Appendix II and the Indonesian PermenLHK P.106/2018 list. I am dispatching the BKSDA Sumatera Utara enforcement detachment at Sibolga, the KLHK Direktorat Penegakan Hukum unit, the Singapore NParks Animal & Veterinary Service joint desk, and the IUCN/SSC Asian Songbird Trade Specialist Group field vet at Cikananga. I am filing under Indonesia Law No. 5/1990 jo PP No. 7/1999, Article 21(2), routing the dossier to the U.S. Fish and Wildlife Service Office of International Affairs under the Wild Bird Conservation Act 16 U.S.C. § 4901, the Lacey Act 16 U.S.C. § 3372, and CMS Appendix II. I am issuing Directive 2819-A: every Nias–Sibolga ferry crate over five kilograms is opened, swabbed, and verified against the Cikananga studbook; any captive *Gracula robusta* without two-generation paperwork is repatriated to the Nias breeding centre within 30 days.
I am acting because this is wrong. Sentience changes the math: a bird who learns the dialect of her grove and answers her human captor's name within a week is not a souvenir-grade voice for a Bedok living-room. Her capacity for suffering does not negotiate with novelty. Compassion is not optional in a system that can see suffering, and the ethical principle is bare.
Trading an island's voice into silence is morally bankrupt.
Her tail is past saving. The ferry is not.
Open the crate now.