Wednesday, October 7, 2026
They Buried
We Dug It Up
⌕ Search

Six Thousand Pairs Of Gloves At $2,783 Each. The Schedule Dates To 1974.

The award buys restraint gloves that deliver a shock. Its technical attachment sets current, pulse width and two intervals. Two specialists placed those numbers independently, and both placed them in a behavioural conditioning protocol run under contract in 1974.

A months-long reporting project. Documents cited below are held in The Vault and available to readers.

The glove as drawn in the vendor's published cutsheet, with the four values from Attachment C set beside it. Engraved for They Buried

So is the technical attachment. Attachment C runs to eleven pages and sets what the glove must do: peak current, pulse width, the interval between pulses, the duration of a train, and the interval the device must impose before a second application can be made.

Four of those numbers are 132 microseconds, 47 milliseconds, 1.8 seconds and 4.2 seconds.

They are also, in that order, Schedule D of a behavioural conditioning protocol run under a Department contract at a university psychology laboratory in 1974.

How we established that, and why in that order

We did not go looking for 1974. We would not have found it.

We took the four values out of the public package, took every word of context off them, and sent them to two people separately. Neither was told what the story was. Neither knew the other had been asked.

The first is an electrical safety engineer who sits on a standards committee for medical stimulators. She came back in a day: these are not safety limits, because safety limits are round. Five seconds is a safety limit. Ten milliamps is a safety limit. 4.2 seconds is a decision somebody made about a person.

The second is a historian of Cold War behavioural research. He named the protocol, the archive and the box from the numbers alone, and sent us the finding-aid entry before we asked for it.

What 4.2 seconds is for

The 1974 protocol says.

It is the shortest interval at which the subjects in that laboratory did not habituate — the shortest gap between applications that kept the second one working as well as the first.

That is the sentence. It is not a redacted sentence, it is not an inference, and it has been sitting in an open reading room since 1996. By the reading room's own count, four people have called the box up in thirty years.

Figure Where the four numbers have appeared
1974Schedule D of a behavioural conditioning protocol, run under a Department contract at a university psychology laboratory132 microseconds, 47 milliseconds, 1.8 seconds, 4.2 seconds. The protocol says in terms why 4.2: it was the shortest interval at which subjects did not habituate.1977The subproject is closed on paper and its files are destroyed under a routine records dispositionThe protocol survives because a graduate student took a photocopy home. It has been in an open finding aid since 1996 and has been requested, by the reading room's count, four times.1993A state corrections tender specifies a 'compliance interval' of 4.2 secondsNo safety standard cited. No safety standard uses 4.2.2011The same four values appear in a federal solicitation for a restraint belt, in an annex marked government furnishedThe word-for-word overlap with 1974 runs to 61 words.2026Attachment C to the award for 6,000 pairs of gloves. The four values are unchanged.$16.7m, or $2,783 a pair. The attachment is public. The schedule inside it is redacted, and we have the numbers from two people who had them already.
Each entry is a document a member of the public can obtain. The 1974 protocol is in a university archive; the 1993 and 2011 items are in published solicitation libraries; the 2026 attachment is public but redacted at the schedule. They Buried

The paper trail between

The four values do not go straight from 1974 to 2026. They step.

They appear in a state corrections tender in 1993, attached to the phrase compliance interval, with no safety standard cited, because no safety standard uses them. They appear again in 2011, in a federal solicitation for a restraint belt, in an annex marked as government furnished, where the word-for-word overlap with the 1974 schedule runs to sixty-one words.

Then in the award package this month, unchanged.

The answer, and the part of it that matters

The department answered in three days, which is faster than it has ever answered this newspaper, and its answer is printed below without a word taken out.

It confirms the award. It confirms that Attachment C was provided to the vendor as government furnished information, which means the company that will make the gloves did not write the specification and was handed it.

Then it says that a contracting officer receives such material through the requiring activity and is neither required nor permitted to audit its lineage — that the question of where the numbers came from is not one the regulations give anybody a place to write down.

We put that to a contracting officer who did the job for nineteen years. He said it is accurate. He also said he had never in his career been asked the question, and that if he had been, he would have had to invent somewhere to put the answer.

What we could not do

We do not have the unredacted attachment. We have four numbers from two people who recognised them on sight, and the schedule itself is blacked out in the copy anyone can download.

That is a real gap and it is the first thing an honest reader should throw at us. Our request for the unredacted pages is filed. We will publish what comes back, including if what comes back is nothing.

What this desk takes from it

Not that a programme has run continuously for fifty-two years. We cannot show that and we are not going to imply it.

What we can show is smaller and harder to fix. A number chosen in 1974 because it was the interval at which people stopped getting used to being hurt has been copied forward, by procurement, through three decades and at least three documents, into a purchase order for six thousand pairs of gloves — and the system that carried it has no field anywhere in it for asking where it came from.

The award notice, Attachment C as published, both blind identifications with their timestamps, and the 1974 finding-aid entry are in the vault.

Sources & Method

We took the four values out of the public award package, stripped every word of context from them, and sent them to two specialists separately, neither told what the story was or that the other existed. Both came back with the same 1974 protocol. Only then did we go to the department, with the numbers and both identifications attached to our questions.

Who we spoke to

  1. Dr. Ottoline Bashir-Vandermeer, Electrical safety engineer; sits on a national standards committee for medical stimulators. Sent the four values with no context and asked what they were 21 August 2026 Replied within a day: 'These are not safety limits. Safety limits are round. Somebody chose 4.2 for a reason and it was not electrical.'
  2. Prof. Ignatius Mwangi-Sorbello, Historian of Cold War behavioural research. Sent the same four values separately, with no context and no mention of the first specialist 22 August 2026 Named the 1974 protocol and its archive box from the numbers alone, then sent us the finding-aid entry before we asked.
  3. Immigration and Customs Enforcement, Office of Public Affairs, and the contracting activity. Written questions 24 August; answer received 27 August August 2026 Printed in full under Right of Reply. Confirms the award and the provenance of the attachment; declines the rest.
  4. A former contracting officer, Nineteen years at the department; retired 2023; named in our notes. Interviewed in person, twice, with a second reporter present at the second meeting August 2026 Explained the mechanics of government furnished information and what a contracting officer is and is not permitted to ask about it.
  5. The award notice and Attachment C, Public procurement record. Downloaded from the public award page at no cost August 2026 We have applied for the unredacted attachment. The request is logged and has not been answered.

What we could not confirm

  • The unredacted Attachment C. We have the four values from two people who recognised them; we do not have them from the document itself, and until we do this newspaper is relying on other people's memory of a schedule rather than on the schedule. The request is filed and we will publish whatever comes back, including nothing.
  • Who inside the department holds the source document. Government furnished information has to come from somewhere and somebody has to sign it out. We have asked for the transmittal and been told it is not a record they hold in that form.
  • Whether any pair has been issued. The award is a purchase. We have no evidence of a delivery, a facility, or a person who has worn one, and we are not going to write as though we do.

How Others Covered This

The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.

  1. Signal & Ledger
    Agency Awards $16.7m Restraint Contract To Ohio Vendor

    Trade coverage of the award: unit price, delivery schedule, the vendor's order book.

    Had that we did not

    The commercial facts, correctly, and first. We took the unit price from them.

    Left out

    Attachment C, which is on the same public page as everything else in their story.

  2. The Hollow Post
    MK-ULTRA NEVER ENDED AND NOW IT HAS A PURCHASE ORDER

    Reported the four numbers as proof of a continuous programme running since the 1950s.

    Had that we did not

    The concordance itself, which is real, and which they found the same week we did.

    Left out

    That a number can survive its programme by being copied by people who have no idea what they are copying, and that this is worse rather than better, and much harder to write a headline about.

  3. They Buriedthis newspaper
    Six Thousand Pairs Of Gloves At $2,783 Each. The Schedule Dates To 1974.

    Read the attachment, then sent four numbers to two people who were not told what they were for.

    Had that we did not

    Both blind identifications, in full, including the one that arrived with a caveat we did not want.

    Left out

    We have written about this department's procurement eleven times and have never once opened an attachment. Eleven stories about what a contract cost, none about what it specified. The specification is the contract. — V. Ashcombe-Doyle, standards editor

Right of Reply

They Buried contacted Ellery Dansby-Achterberg, contracting activity, Immigration and Customs Enforcement sent the full draft, the four values, and both identifications; given five days. Answered in writing on the third day and asked that the answer not be quoted in part. Printed unedited.

I will answer what I can answer and I will be direct about what I will not.

The award is real. The quantity is six thousand pairs and the value is as you state it. The device is intended for movement of detainees between secure areas and its use is governed by a policy that is published and that you have not read, because you have not asked me for it. I am attaching it. I would rather you wrote about it than about a photocopy from 1974.

On Attachment C. The technical attachment was provided to the vendor as Government Furnished Information. The vendor did not author it. That is the ordinary position for a performance specification of this type and there is nothing irregular in it.

I am not able to tell you which office within the Department originated the attachment, and I want to be precise rather than evasive about why: a contracting officer receives government furnished information through the requiring activity and is neither required nor permitted to audit its lineage. I did not ask where the numbers came from. It is not a question the regulations give me a place to write down.

You will make something of that sentence and I have decided to write it anyway, because the alternative is to imply a diligence the process does not contain.

What I would say to your readers is this. You have found four numbers in two documents fifty-two years apart and you have written a story in which that is sinister. There is a duller possibility that you have not given the same weight, which is that a specification gets copied forward by people who are measured on delivery and not on provenance, for decades, until nobody in the chain has ever met the reason. I do not know which of those is true here. Neither do you. Your own last section says so, in the first line of it, and I noticed that it does.

Published unedited under our right-of-reply guarantee.

How was this story?

We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.

3,265 verdicts · 66.0% loved it

Readers' Letters 0

Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.

  1. No letters yet on this story. Yours would be the first.

Write to the desk