Research dossier · extract

The Rockefeller money: what the record actually supports

Aleksei Beliankin · checked 5 August 2026 against the IRS Business Master File, the digitised-filings index, and the Rockefeller Archive Center catalogue.
Every claim below carries its class, its evidence, its source, and the observation that would break it. Anything not established is written as not established, including where that is inconvenient.

For thirty years, writing about the Harvard psychiatrist John E. Mack has repeated one figure for Laurance Rockefeller's funding of his work: a million dollars, four years, $250,000 a year. The figure cites a source that says two years. And the foundation any researcher would check first to settle it — the Laurance S. Rockefeller Fund — did not exist until two years after Rockefeller was dead. The number is not merely unverified. The trail that would verify it has been pointing at the wrong door.

What follows is the part of that finding which rests on primary records. It is deliberately narrow. It does not establish a dollar amount, and it does not claim the funding did not happen — the funding is documented. It establishes something more useful to anyone about to write on this: that the obvious diligence produces a false negative from authentic documents, and it names the route that would not.

The finding

CLAIM L13a.1
The Laurance S. Rockefeller Fund that files with the IRS today did not exist when Mack was funded. It cannot hold the 1993 grant.
Fact Primary — IRS Business Master File
Name on fileLaurance S Rockefeller Fund
EIN20-4192256
IRS ruling dateJune 2006
Foundation code4 — private non-operating foundation
Filing years held2006 through 2023, nothing earlier

Laurance Rockefeller died on 11 July 2004. An exemption ruling dated June 2006 is therefore two years posthumous, and thirteen years after the 1993 grant to Mack's Center. The 20- EIN prefix is independently consistent with an issuance in the 2000s.

The consequence is the part worth writing down. Anyone who follows the obvious instinct — pull the Fund's 990-PFs, look for the Mack grant — retrieves a real foundation's real filings, finds no grant, and is entitled to conclude the funding story is unsupported. That conclusion is wrong, and it is wrong on the strength of authentic primary documents.

projects.propublica.org/nonprofits/api/v2/organizations/204192256.json

What would break this
An older, differently-numbered fund of the same name may have existed in the 1990s and been superseded; a same-name entity today does not prove there was no predecessor. So the claim is written narrowly enough to survive that: the entity filing under EIN 20-4192256 cannot be the source of a 1993 payment. Whether an earlier vehicle carried the name is a separate, open question.
CLAIM L13a.2
The recipient is real and its exemption predates the grant — but its returns for the relevant years are not in the online record at all.
Fact Primary — IRS Business Master File
Name on fileCenter For Psychology And Social Change Inc
EIN04-2864296
IRS ruling dateAugust 1991
Status501(c)(3), foundation code 15 — public charity
Filing years held2001–2007, 2009–2012, 2019–2021

Two things are established, and one is closed off. Established: the organisation was tax-exempt before the 1993 grant, and the 04- prefix is a Massachusetts-era EIN consistent with the Cambridge origin. A later address in Boulder, Colorado shows the entity moved, which matters to anyone trying to reach a successor custodian and cautions against assuming Massachusetts records cover its whole life.

Closed off: the digitised filing record begins in 2001. The 1993–1997 Forms 990 — the ones that would report contributions received in the years Rockefeller was giving — are not in it. That is not one vendor's coverage gap; it is the general shape of IRS electronic availability.

projects.propublica.org/nonprofits/api/v2/organizations/42864296.json

What would break this
ProPublica is a mirror, not the IRS. A year absent there is probably absent upstream, but that is an inference, so the claim says only what was measured: not in the online record. The direct test is a Form 4506-A request to the IRS for a named year. It has not been made.

Where the answer actually is

Establishing that the obvious route is a dead end is only half of a useful finding. The other half is naming the route that is not, precisely enough that someone else can walk it.

CLAIM L13.7
The archive holds named folders for the recipient, covering exactly the years in question.
Fact Primary — Rockefeller Archive Center catalogue, 5 Aug 2026

Three catalogue matches for "Center for Psychology and Social Change" resolve to Room 5600 General Files → Science: RG 43 for 1992–1996 — the years that matter — then RG 48 for 1997–2001 and RG 55 for 2001–2006. A separate folder titled "UFO – John Mack," 1992–1995, sits in the same series. Box numbers are not exposed publicly, so a request has to name the series and the folder titles and let the archivists locate the boxes. The Laurance S. Rockefeller papers run to 815 cubic feet, are open for research under the 2009 estate donation terms, and the Center states it holds literary rights and can grant permission to publish.

CLAIM L13.9
A likely booking channel has a name — and that name is not yet evidence.
Inference Contemporaneous press + catalogue holdings

Press from exactly these years describes Laurance Rockefeller's personal giving as flowing through the Fund for the Enhancement of the Human Spirit, and the archive holds Fund folders spanning 1989–2004 including one for 1992–1996. But every one of those citations concerns other gifts — the California Institute of Integral Studies, spiritual-education grants — and none mentions Mack.

Held at inference on purpose
This is the most likely channel and the right thing to ask about by name. It is not a finding, and writing that the Fund paid Mack would be exactly the move this dossier exists to catch. A second hypothesis stays alive alongside it: the Human Potential Foundation folders in the Education series, which is where the pass-through claim would land if it is true.
CLAIM L13.4
Massachusetts looked like a shortcut and is not one.
Fact Negative result

The Center was a Massachusetts charity, and Massachusetts requires an annual Form PC to the Attorney General's Public Charities Division reporting revenue — a filing nobody in the Mack literature appears to have checked. The office still registers the Center. But its portal retains filings for fifteen years, so the 1993–1997 Form PCs exist only in the office's own records and have to be requested directly. Recorded here because a negative result someone else does not have to repeat is worth as much as a positive one.

What this does not establish

No dollar figure. Not one. The commonly cited million dollars over four years remains uncorroborated, the source it is attributed to describes two years, and nothing in the federal record settles it either way. What has changed is that the question is now answerable by a specific request to a specific series of a specific archive, rather than by a filing search that was always going to return nothing.

Method

Four rules, applied to every line above.

  1. Every number carries a source. Anything unsourced is written as missing-evidence rather than softened with a hedge.
  2. Class before prose. A claim is marked fact, reported, inference or speculation before it is written, not after — so the reader is never asked to guess which one they are reading.
  3. Each claim carries its own falsifier. If I cannot state the observation that would break it, I do not yet understand it well enough to publish it.
  4. Negative results are results. A route that dead-ends gets written up with the same care as one that pays, because the next person's time is worth the same as mine.
Need this dossier as a file — to keep, or to forward to somebody who has no reason to open a stranger's website?PDF · 5 pages · 372 KB
Extract from a longer working dossier · full files 03_dossier/L13_rockefeller_money.md and L13a_irs_entity_check.md · Aleksei Beliankin