On September 1, USA TODAY reported that an anonymous Postal Service official warned a new ballot-mail process could “derail the midterm elections” and cause “catastrophic failure.” The official said potentially millions of voters might receive ballots late or not at all.
Those are warnings. They are not measured outcomes.
The disclosure released by Sen. Richard Blumenthal describes the work as “secretive, rushed, chaotic, and fundamentally flawed.” It alleges that if one barcode in a bulk mailing of 10,000 fails to scan, the whole mailing will be rejected and returned.
It also alleges that postal acceptance units will inspect samples of 15, 350 or 400 pieces, depending on the mailing category, with no failed barcode permitted in the sample.
This is admirably easy to understand. One bad scan. Ten thousand ballots go back to the mailer.
It is also not what the public rule says.
The Postal Service’s published rule says noncompliant mailings “will not be accepted and will be returned.” The mailer must correct the errors and submit the mailing again. It says a scan generally takes less than a minute per mailpiece, with the total time depending on how many barcodes are scanned.
The rule does not contain the one-in-10,000 example. It does not publish a zero-error threshold. It does not include the sample sizes alleged in the disclosure. The disclosure acknowledges this.
USPS also declined to publish an anticipated rejection rate.
So the public rule establishes one thing: a noncompliant mailing will not be accepted and will be returned.
It does not establish that one unscannable barcode in a sample makes an entire 10,000-piece mailing noncompliant.
That does not make the whistleblower wrong. Internal instructions can exist without appearing in a public rule. But silence is not confirmation. It is not disproof either. It is silence, an old government product with excellent distribution.
USPS has said the work was being “conducted in a manner consistent with court orders”. The agency also said the disputed pre-acceptance verification procedures were not being performed under a temporary restraining order.
Lauren Bis of the White House called the check “neither complex nor unique,” noting that USPS already handles bulk mail and Intelligent Mail barcodes. Hans von Spakovsky said the allegations were “very general and not specific enough” and presented “only a worst-case scenario.”
These statements address timing and plausibility. They do not provide the missing internal instructions, test records or rejection data. Neither does the disclosure.
One other distinction matters. Blumenthal’s example involving voters who moved or changed names concerns a mismatch in data. That is not the same event as a barcode failing to scan. Both may be relevant. They do not become identical through frequent mention.
The public materials cited so far include no test logs, code records, trouble tickets or observed mass rejection showing the alleged procedure in operation. There is also no published rejection rate showing that the process is harmless.
That leaves three facts:
The Postal Service has a return-and-resubmit rule.
An anonymous official alleges an unpublished sampling method could reject very large ballot mailings after one failed scan.
The available public evidence does not yet show whether that method is operating or how often it would reject mail.
“Catastrophic failure” remains a warning about what could happen. It is not a report of what has happened.
What would settle the matter is ordinary evidence: the internal instruction, a test result and a rejection count.
Elections already receive adjectives by the sack.