From 62.0 births per 1,000 women of childbearing age in 2016 to 53.8 in 2024. That is a 13% drop in nine years, and the single interruption was a small rebound in 2021 after the first pandemic year. It did not last. Every year since has set a new low.

This is the kind of number that gets filed under “interesting demographic trend” and then ignored, because nothing appears to happen. No one is hurt today. The schools are still open. The pension checks still clear.

That is exactly the problem. The consequences of a birth rate arrive twenty and forty and sixty years after the births don’t. Which means the damage from the last nine years is already locked in — and the countries that started falling before we did are living in our future right now.

First, look at where the births went

The headline number hides the more interesting story. American births aren’t simply falling. They’re moving in different directions - but they all point to the same thing: decline.

Every age group under 35 is down, and the younger the group, the harder the fall — teen births down 40%, ages 18-19 down 37%, ages 20-24 down 24%. Every age group over 35 is up. Births to women 45 and older rose 22%.

The dividing line runs straight through the mid-thirties, and it has not moved in nine years.

Some of this is a genuine social good and nobody should pretend otherwise. A 40% collapse in births to 15-to-17-year-olds is a win by many standards. But the same shift has a second effect that doesn’t make the press release: a birth delayed is frequently a birth that never happens. Push the average first birth far enough into the thirties and biology starts taking a cut of the intention. The chart isn’t a picture of Americans choosing childlessness. It’s a picture of Americans choosing later, over and over, until later closes.

And the fall is not evenly shared.

Native American fertility fell 25% and Asian fertility 24% in nine years — the steepest drops of any group. Black fertility fell 20%, White 12%, Hispanic 9%. The Hispanic rate, long the demographic cushion under American population projections, is now 64.3 and dropping.

Now look at Greece

Here is the part that makes this urgent rather than academic. We do not have to speculate about what comes next. Europe started down this road decades ahead of us, and the bills are arriving on schedule.

Greece spends 14% of its entire economy on pensions — and it was 16.4% in 2021, the highest share in the EU, before reform trimmed it. That money comes from workers. Today roughly five Greek workers support two retirees. By 2050 it’s four workers supporting three. The old-age dependency ratio nearly doubles, from 39 to 74.

Now here’s the line item nobody reads. The European Commission’s own 2024 Ageing Report projects that total age-related spending rises only modestly — from 25.1% to 26.5% of GDP by 2070. Manageable! Except look under the hood at how that number stays flat. The dependency ratio alone pushes pension costs up 6.2 points of GDP. The forecast only balances because the same model assumes the benefit each retiree actually receives falls (−2.9), that coverage narrows (−1.3), and that people work longer (−1.1).

The reassuring projection is reassuring because it has already priced in a smaller check. Europe has not solved this. Europe has scheduled the cut and declined to announce it.

It slows the economy, and we can measure by how much

Maestas, Mullen and Powell, in the American Economic Journal: Macroeconomics (2023), used variation across US states from 1980 to 2010 and found that every 10% rise in the share of the population over 60 cuts GDP per capita by 5.5%.

The mechanism is the surprise. Only a third of that came from fewer people working. Two-thirds was productivity — each remaining worker producing less. Aging doesn’t merely shrink your workforce. It appears to slow down the one you still have.

Their estimate is that aging shaved 0.3 percentage points a year off American per-capita growth between 1980 and 2010. Compounded across a generation, that is the difference between a rich country and a much richer one.

Charles Jones takes it further in the American Economic Review — “The End of Economic Growth?” — and his argument is uncomfortable and clean. Modern growth models run on people discovering ideas. Fewer people, fewer ideas. If population shrinks permanently, you don’t get slower growth. You get what he calls the “empty planet” outcome: living standards stagnate for a population that keeps dwindling.

That is the top journal in the field, not a pamphlet.

Old societies stop starting things

This one I’d never seen quantified until now. Liang, Wang and Lazear in the Journal of Political Economy found that a one-standard-deviation drop in a country’s median age raises new business formation by 2.5 percentage points — roughly 40% of the average rate.

And the detail that should stop you: older societies have lower entrepreneurship at every age, including among their young. A 30-year-old in an old country starts fewer companies than a 30-year-old in a young one.

The authors’ proposed mechanism is blockage. When older workers hold the senior positions, younger workers never acquire the skills and responsibility that let them strike out on their own. An aging society doesn’t just have fewer founders because it has fewer young people. It produces less capable young people.

The map empties before the country does

Depopulation is not a gradual, even thinning. It’s a doom loop in specific places: the school closes, the bus route ends, the clinic shuts, the last young family leaves.

• Greece closed more than 700 schools in 2025. Student numbers are down about 150,000 since 2018. A Greek school is suspended below 15 students and permanently closed after three years under the line.

• Greece peaked at 10.82 million people in 2010 and was at 10.43 million by 2023 — 400,000 people gone in thirteen years.

• Italy’s fertility rate hit 1.18 in 2024, the lowest in the EU. Italy now counts roughly 1,000 comuni fantasma — ghost towns — and as many as 6,000 abandoned hamlets.

• Spain’s countryside has lost 28% of its population in fifty years. Five rural regions covering 53% of Spanish land hold 15% of Spanish people. The España Vaciada — “emptied Spain” — movement put 100,000 protesters in Madrid in 2019 and forced the creation of a government ministry for depopulation.

A government ministry for depopulation. Sit with that.

And then there’s the army

EU births fell below 4 million in 2022, the first time since 1960. That is the recruiting pool, and it is already fixed for the next eighteen years. You cannot appropriate your way out of it.

Germany reports 20,000 military billets sitting entirely vacant, with its own parliamentary commissioner warning that the armed forces are aging and shrinking simultaneously. In 2026 Germany began screening every 18-year-old man for military suitability — a step back toward the conscription it had abandoned. France, Croatia and Latvia are moving the same way.

The EU defense commissioner says Europe needs to be able to move 100,000 troops quickly. Finding them is now the binding constraint. Not the budget. The people.

The bill that comes forty years out

Demographers track something called kinlessness — adults who reach old age with no living spouse and no children. Among Americans 55 and older, roughly 7-8% of White adults and 10-11% of Black adults were kinless as of 2015, projected to reach 8-10% and 12-13% by 2040.

Unpaid family care is the invisible foundation of elder care in every Western country. It never appears in a budget because nobody invoices for it. The kinless have less wealth, worse health, and less support, and somebody pays for that.

Low fertility today is a long-term-care bill forty years out, and it lands on a working-age population that is smaller for the same reason.

Where I’ll concede the argument

There is a serious counterargument and it deserves a fair hearing, not a strawman.

Acemoglu and Restrepo find no negative relationship between aging and GDP growth across countries. Their explanation: aging societies automate. Robot adoption rises precisely where the middle-aged workforce is shrinking, and productivity rises fastest in the industries most open to automation. Scarce labor makes labor-saving technology pay.

But notice what automation cannot do. A robot does not pay into a pay-as-you-go pension system. It does not serve in an infantry battalion. It does not keep a rural school open, and it will not sit with you when you’re 86 and kinless.

Yet (I guess).

Who pays the pensions, who staffs the nursing home, who fills the army, and who lives in the village. Those four questions are the case, and automation answers none of them.

And money alone doesn’t buy babies

Before anyone reaches for the obvious fix: Hungary spends around 5% of GDP on family policy, among the highest in the world and at a scale no other Western country has attempted.

Its fertility rate went from 1.23 in 2011 to about 1.56 in 2024. That is real movement and pronatalists are entitled to it. It is also, after fourteen years and enormous cost, nowhere near the 2.1 replacement level. Demographers add a further caution: some of that gain is timing rather than quantity. The period fertility rate rises when women who postponed births catch up, which flatters the number without necessarily producing more children per woman over a lifetime.

The honest read is that pronatalist spending moves the needle and does not come close to solving the problem.

The part that should actually frighten you

Wolfgang Lutz and colleagues proposed the “low fertility trap” — three self-reinforcing loops. Demographic: fewer girls born now means fewer potential mothers later, so births keep falling even at a steady birth rate. Sociological: people who grow up with few siblings want fewer children themselves, so the preference shifts, not just the behavior. Economic: rising material expectations plus falling relative income for young adults pushes childbearing later, and later becomes never.

Every one of those loops feeds the next one.

And underneath all of it sits the fact that makes this different from every other policy problem: population momentum runs in reverse.

Because the large cohorts are already past childbearing age, decline continues for decades even if the fertility rate recovered to replacement tomorrow. There is no fast lever. There is no emergency appropriation. There is no election that fixes it.

The 2040s are already set. We are only now getting around to reading the terms.

Charts: CDC / National Center for Health Statistics, Data Query System — “Birth and fertility rates, by age group, race, and Hispanic origin of mother: United States,” from Health, United States. Underlying data: National Vital Statistics System, Birth File. Note that the US figures above are the general fertility rate — births per 1,000 women aged 15-44 — while the Italian and Hungarian figures are total fertility rate, births per woman. Different units, same direction.

Sources: Maestas, Mullen & Powell, AEJ: Macroeconomics (2023) · Jones, American Economic Review 112(11) (2022) · Liang, Wang & Lazear, Journal of Political Economy 126(S1) (2018) · Lutz, Skirbekk & Testa, Vienna Yearbook of Population Research (2006) · Acemoglu & Restrepo, NBER WP 24421 · European Commission 2024 Ageing Report and Greece country fiche · ECB Economic Bulletin, May 2024 · Eurostat · Verdery et al., Journals of Gerontology: Series B 74(8) · AEI and N-IUSSP on Hungary.

His path since is unusual enough to be worth stating precisely, because people keep getting it wrong.

He was appointed director of the FDA’s Center for Biologics Evaluation and Research — CBER, the division that regulates vaccines and gene therapies — in May 2025, replacing Peter Marks. On July 29, 2025, under sustained political attack and in the middle of a fight over a Sarepta gene therapy, he abruptly resigned. Two weeks later he was back, reinstated at the FDA’s request after a White House review of the campaign against him. He left again at the end of April 2026; Commissioner Makary characterized it as the end of a planned one-year leave from his UCSF faculty position rather than a resignation under fire.

Critic, then insider, then out, then in, then out. He has now sat down with Reason, and he is telling all the truths.

▸ EMBED THE YOUTUBE VIDEO HERE —

The thing a former CBER director is not supposed to say

The interviewer raises the question of whether any child has died from the COVID vaccine. Prasad does not hedge:

“Absolutely, is it the case that kids died of myocarditis from the COVID-19 vaccine? Yes, absolutely it is the case.“

Then he does something more useful than the assertion — he explains why the official coding says otherwise, and why that coding is misleading.

For some vaccines, he notes, you can find physical proof. Give MMR and a child develops a granuloma, you can biopsy it and find vaccine product. Give the varicella vaccine and a child dies, you can sample the cerebrospinal fluid and find the vaccine strain.

Myocarditis leaves no such fingerprint. There is no assay that pulls the vaccine out of the heart muscle. So under a standard demanding that kind of certainty, the deaths get coded “possible” or “probable” and never “definitive” — not because the evidence is weak, but because the test does not exist.

His point, and it is the sharpest thing in the interview:

“Just because the death isn’t coded as certain and coded as probable or possible, that doesn’t negate the fact that it is attributable. In fact, the standard across all of drug development is possible and probable are attributable.“

That is the whole argument in two sentences. Every other drug in America is regulated on the basis that “possible” and “probable” adverse events count. For this one product, a different and unmeetable standard was applied, and the resulting zero was reported as though it were a finding.

He follows it with an analogy about base rates that is hard to argue with. Send a million people down one stair. Most walk it off. A thousand twist an ankle. Ten split their heads open. One dies. Not because the stair is dangerous — because a million is a large number. Myocarditis is a known side effect. The arithmetic follows.

And then the part that is actually news

The interviewer notes the published figures: zero childhood deaths definitively linked to the vaccine, five classified as possible and two as probable — seven total — and observes that this is fewer than the ten that had circulated earlier.

Prasad’s answer:

“They changed the report.”

“I had a draft report with 10. They changed it after the media publicity of my letter.“

Asked who changed it, he says: the people who did the report.

We covered the ten. In November 2025 we published FDA Finally Admits It: At Least 10 Healthy Kids Likely Died from the Covid Vaccine. That number then quietly became seven.

Prasad is now saying, on camera and on the record, that the revision followed press attention to his letter.

To be careful: this is his account. He is a party to the dispute and an interested one. We have not seen the draft, and “they changed it after the publicity” describes a sequence, not necessarily a cause. But the man saying it ran the center that produced the report, and the specific claim — a draft of ten, a published seven, a revision after coverage — is checkable by anyone with access to the drafts. Somebody should go get them.

Fauci

Asked directly whether Fauci committed a crime, Prasad says yes, and locates it in the gain-of-function funding:

“I think that was a crime for which he’s been pardoned. I think it’s not just my assessment. I think that’s what Joe Biden’s autopen thought… when he signed, or whoever signed, a pardon — that he committed a crime. That’s why he has a pardon that goes back that far.“

The reasoning is worth following. The pardon reaches back to 2014. Prasad’s argument is that the scope of the pardon is itself evidence: you do not indemnify someone that far back for conduct nobody believed was criminal.

On the legacy question, he doesn’t reach for nuance:

“I think very poorly. I think this is where the media is completely enraptured by the guy.”

And then the bill of particulars — the man who told you to put a cloth mask on a two-year-old; the six-foot rule that he later conceded was, in Prasad’s paraphrase, a number they made up; the school closures, including a call to Bill de Blasio urging him to close New York’s schools, recorded in Fauci’s own diary.

On origins, Prasad points at the same diary. Fauci writes of being brought into a SCIF and shown intercepted Chinese communications that made the CIA suspicious of a laboratory incident. And on the Huanan market — the origin Fauci publicly endorsed — the diary reportedly concedes that at best it was an amplifying event, not the source.

“I suspect that many intelligence people do in fact know the answer, and that answer is probably lab leak.”

The part nobody will quote, and the most important

Everything above will be clipped and shared. This won’t be, and it should be.

Prasad went inside. He got the job, lost it, got it back, and served out a year running the division that regulates every vaccine in America. Asked what public health could become, he describes something modest — honest, proportionate, humble, bipartisan.

Then he describes what it is:

“Public health is far from these things. Public health could actually aspire to be bipartisan. Public health is not. It’s becoming an increasingly entrenched left-wing discipline run by left-wing ideologues, reaching more and more absurd conclusions.”

Asked whether any of that changed during his tenure: “Not much at all.”

And then the sentence that should stop you:

“Changing an entrenched organization with tens of thousands of people is very difficult… I don’t think a real reform is possible. I suspect it’ll splinter into different sort of public health groups.”

That is not a critic on the outside throwing rocks. That is a man who was handed actual regulatory authority over American biologics, used it, and came out the other side saying the institution cannot be fixed — only fragmented and routed around.

For those of us who spent four years arguing that these agencies needed reform, that is a genuinely uncomfortable conclusion, and we’re not going to pretend otherwise.

The framing worth stealing

One more, because it dismantles the dumbest argument in this entire debate in about fifteen words.

People tell him vaccines save lives. His answer:

“Drugs save lives. The right drug given to the right person at the right time is life-saving. But the wrong drug given to the wrong person at the wrong time can be harmful. Vaccines are the same way.“

You cannot say anything meaningful about “vaccines” as a category, he argues, any more than you can about “Tylenol, aspirin and Plavix all in the same breath.” They are pharmaceutical products. Each one has an indication, a population, a dose, a risk profile, and a benefit that depends entirely on who’s receiving it.

That is not an anti-vaccine position. It is the position every other class of drug in America is already regulated under. The only reason it sounds radical here is that for five years we were told the category was exempt.

Interview: Reason, with former FDA CBER director Vinay Prasad.

Here is the first line of a message sent on August 8, 2021 by Michael Osterholm — Director of the Center for Infectious Disease Research and Policy, and a member of Biden’s COVID-19 Advisory Board — addressed to “Jeff, Vivek, Tony and Rochelle”:

“I’m sending this via text messaging to avoid any FOIA issues.”

That is the opening sentence. The stated purpose of the channel, written down in the channel itself.

The recipients, per Herridge’s reporting: Jeff Zients, then the White House COVID coordinator and later Biden’s chief of staff. Vivek Murthy, Surgeon General. Anthony Fauci — confirmed. Rochelle Walensky, CDC Director.

That is the entire command structure of the American pandemic response, receiving a message whose author announced in his first line that he was routing it around the federal records law.

What he had to say that was worth hiding

Read the rest and the choice of channel starts to look less like habit and more like judgment.

Osterholm tells them he has been consistent since April 2020 about “the limited benefit that cloth face coverings play in reducing risk.” He tells them N95s are what actually work against aerosol transmission. And then he says this about the agency one of his recipients was running:

“none of the studies that CDC uses to support its statement as to the significant protection of face cloth coverings stand up to scientific scrutiny.”

He adds that CDC “was very late to the game recognizing that aerosols play a significant role in the transmission picture.”

He closes by predicting what history will conclude:

“I am certain one day that one of the take-away findings of this pandemic was the constant USG emphasis on masking while at the same time providing minimal guidance to the public what effective masking means.”

Be precise about what Osterholm was arguing, because it matters and because getting it wrong would be the easy way to lose this. He was not against masking. He says plainly, “I am all for ‘masking’ but we must be promoting the use of N-95s.” His complaint is that the government was telling Americans to do a thing he did not believe worked, instead of the thing he did.

That distinction doesn’t soften the story. It sharpens it. The dispute inside the room was never whether cloth face coverings did much. It was what to say in public about it.

And note the trigger. A Newsmax reporter had asked Jen Psaki about Osterholm’s own comments on cloth masks. His response was to reassure the task force — over a channel he’d chosen to defeat FOIA — that the reporter had mischaracterized him, while restating the substance of exactly what the reporter had reported.

He said it in public too. We wrote it down at the time.

None of the science here was secret. Osterholm’s views were on the record, and this newsletter covered them as they happened.

In February 2021 we wrote up Expert versus Expert — Osterholm, then a top health adviser to the incoming president, flatly contradicting Fauci’s double-masking recommendation: “I do not support the idea of two masks.” He explained that what matters is fit and filtration, comparing the seal to swimming goggles.

In October 2021 we ran The most important 5 mins you can spend today, on Osterholm going on CNN and dismantling the claim that masks had eliminated the flu.

So the record is not that a lone dissenter was ignored. It is that the man saying cloth masks don’t do much was inside the tent, said so publicly, said so privately to the four people running the response — and the guidance did not change.

And then they masked the children anyway

Here is the part that should end careers.

August 8, 2021 was three weeks before the school year. Over the following months, districts across the country mandated cloth face coverings on children — including preschoolers — and in many places kept them there deep into 2022.

The people issuing and defending that guidance had in hand, in writing, from one of their own advisers: cloth coverings offer very limited protection, and the CDC’s own supporting studies do not withstand scrutiny.

We said so at the time, repeatedly, and were treated as cranks for it:

• Masking Children is an Ineffective Policy and Not Supported by Research — September 2021, weeks after this text was sent

• Infographic: Making Children Mask is Misguided Health Policy — January 2022

• Mask Mandates and Trauma — December 2021

• Masking Children is Insane and Harmful — July 2022

• Debunking the Mask Mandate: A Systematic Review of COVID-19 Child Masking — December 2023

• New Study! Masking Kids → Unproven Benefits & Potential Harms — October 2024

Even the CDC eventually conceded the product was inadequate — we covered that in CDC: Ditch that Cloth Mask! in January 2022, five months after Osterholm’s text.

And there’s a wrinkle nobody in that group text addressed. Osterholm’s answer was N95s. N95s are not a serious option for small children — they are not fit-tested for them, and the respirators that work for an industrial hygienist do not work for a four-year-old. Which means the honest conclusion available to those four officials in August 2021 was that for young children there was no effective masking option on the table at all.

They did not tell parents that. They mandated the thing their own adviser said didn’t work.

Two days ago we published what that decision may have cost: speech-language IEPs at the highest rate ever recorded, and the states that masked longest carrying the most young children in special education, with the backlog and lost-classroom-time explanations both tested and both failing.

The Morens plea, two days earlier

The timing here is not a coincidence so much as a pattern coming due.

On August 18, 2026 — one day before Herridge published — David Morens, senior adviser in the NIAID Office of the Director from 2006 to 2022 and Fauci’s longtime lieutenant, pleaded guilty in federal court to conspiracy relating to a scheme to evade the Freedom of Information Act and the Federal Records Act.

Per the Justice Department, Morens conspired with Peter Daszak of EcoHealth Alliance — identified in the plea as “Co-Conspirator 1” — to destroy or conceal records after NIH terminated EcoHealth’s bat coronavirus grant. Four other felony counts were dropped as part of the deal. He faces up to five years and a $250,000 fine, with sentencing set for November 12, 2026 before Judge Paula Xinis in Greenbelt, Maryland.

Herridge notes the statute that makes this more than an ethics problem: 18 U.S.C. § 2071 makes it a crime to willfully conceal, remove, or destroy federal records.

So assemble the sequence:

February 2020 — Fauci to NIH Director Francis Collins: please delete this e-mail after you read it. August 2021 — Osterholm to the four most senior officials in the response: I’m sending this via text messaging to avoid any FOIA issues. 2021–22 — Morens, later by his own guilty plea, conspiring to keep records out of FOIA.

Three different people. Three different agencies and roles. The same instinct, expressed in nearly the same words.

That is not a few bad actors. That is a working culture, in which the reflex on encountering a fact the public might not like was to make sure the public could not obtain the record of it.

The question Herridge asks, which is the right one

Her post ends where the story now has to go:

How widespread, and how many separate channels were used to avoid FOIA?

We know of at least three now, across three organizations, spanning at least two administrations. Every one surfaced by accident — through a subpoena, a lawsuit, a criminal referral, or a senator with jurisdiction and patience.

None of it surfaced because the system worked. It surfaced because the system was forced.

Which raises the obvious question about everything that was never subpoenaed, never litigated, and never referred.

Reporting by Catherine Herridge, August 19, 2026, from records obtained by Sen. Chuck Grassley via HHS. Osterholm message dated August 8, 2021. Morens plea reported August 18, 2026; details from the U.S. Attorney’s Office for the District of Maryland. Prior RationalGround coverage linked throughout.

We never believed it, and we said so at the time — loudly enough that it cost some of us our accounts and a few of us our jobs.

That wasn’t a hunch. It came from four years of hearing the same thing from the same people: from the parents in this community, from the teachers who wrote to us, from speech-language pathologists who described waiting lists they had never seen in twenty-year careers, from the preschool directors who kept saying something is different about this cohort. Anecdote after anecdote after anecdote, from every state, all pointing the same direction, all dismissed as anecdote.

We now have the administrative record. Not a survey. Not a model. The federal count of every child in every state, every year.

It says the people writing to us were right.

The flattest line in the dataset broke

Of everything in the federal special-education file, the steadiest series was speech. For eight consecutive years — 2012-13 through 2019-20 — the rate at which American children were placed on a speech-language IEP sat between 16.74 and 17.24 per 1,000. No trend. No drift. A ruler.

It is now 18.20 per 1,000. The highest in the thirteen years of data that exist. About 5% above where that flat line pointed.

In children rather than rates: 1,352,054 on speech-language IEPs in 2019-20 became 1,454,259 by 2024-25. A 7.6% increase — while the child population was shrinking.

Now the control that turns this from a curiosity into a finding.

All of special education fell during the closure year, and fell harder than speech did — down 5.4%. And it has since climbed back to precisely its own trend: +0.8%. Dead on. Whatever disruption the closures caused to identifying children with disabilities generally, the system absorbed it and returned to baseline.

Speech didn’t return to baseline. Speech overshot and stayed there.

That gap — between the category that recovered and the category that didn’t — is the whole story.

And it tracks how long a state kept masks on children

Here is where it stops being just a national curiosity.

Using the Oxford COVID-19 Government Response Tracker, each state was scored on how many days it required masks statewide between September 2020 and June 2022 — counting only genuine statewide mandates, not one county’s rule. The spread is enormous: zero days in Florida, Georgia, Montana, Oklahoma, South Dakota and Tennessee. 668 days — the entire window — in New Jersey, Rhode Island and Washington.

Then: how far does each state’s special-education enrollment now sit above or below the trend that state itself was on before 2020? Every state is measured against its own history. No cross-state comparison of who identifies children more aggressively. Each state is its own control.

For preschoolers — ages 3 and 4:

+2.21 percentage points of excess enrollment for every 100 days of statewide mask mandate. (t = 3.58)

The 20 states that masked 400+ days now average +5.0% above their own pre-2020 trend. The 12 states that masked 100 days or fewer average −9.2% — still below theirs.

A fourteen-point gap.

And it does not break when you push on it. Drop any single state and the coefficient stays between +1.95 and +2.51. Remove the four highest and four lowest states and it holds at +1.61. A bootstrap puts the 95% range at +1.12 to +3.27 — nowhere near zero.

It is not a backlog

The first thing a defender of these policies will say is that long-masking states also closed schools longer, missed more evaluations, and are simply working through a queue.

That was tested. It fails.

How far a state’s preschool enrollment fell during 2020–22 has no relationship whatsoever to how long it masked. +0.24 points per 100 days, t = 0.60. Statistical noise.

Both groups fell into the same hole. Long-maskers dropped 15.3%. Short-maskers dropped 18.3% — further. Then:

Fell 2020–22 Where they are now Masked 400+ days −15.3% +5.0% Masked ≤100 days −18.3% −9.2%

Same hole. Opposite climb.

If this were a backlog clearing, the states that fell furthest would rebound hardest. The correlation runs +0.56 — the wrong sign entirely. The states that fell least came back most.

One more nail: nationally, preschool special education is still cumulatively below its pre-2020 trend. Nobody has finished catching up. The long-masking states are not overshooting a queue. They are on a different road.

It is not missed school either

The second defense: it wasn’t the masks, it was the lost classroom time.

So that was measured properly — not estimated. The COVID-19 School Data Hub recorded what actually happened in 14,921 school districts across 48 states: the share of 2020-21 each district spent in person, hybrid, or fully virtual, weighted by enrollment. Florida lost 3% of the year. Texas lost 11%. California lost 82%. Hawaii lost 83%.

Put both variables in the same model. They’re correlated — r = +0.47 — but nowhere near identical, so the model can separate them.

For preschoolers, it isn’t close:

Mask days Lost classroom time Ages 3–4, special education +2.22 (t = 2.90) −0.15 (t = −0.19)

Adding lost classroom time to the model improves the fit by nothing. The explained variance is 0.191 with it and 0.191 without it.

The leading alternative explanation contributes literally zero.

Mask duration correlates with state politics. State politics also predicts school funding, special-education eligibility rules, and how aggressively districts evaluate children. We cannot rule that out, but we feel confident producing these charts as proxy data for what we already knew: masking kids was a bad idea.

One more number, from the American Speech-Language-Hearing Association: 69% of speech-language pathologists reported increased referrals since 2020. They didn’t need our regression. They were living in it.

We should never have had to prove this. But here it is.

All figures are from federal administrative records, reproducible from public sources. Special education counts: U.S. Department of Education IDEA Section 618 child count, Part B (ages 3–21) and Part C (birth–2), 2012-13 through 2024-25, restricted to the 55 reporting units present in every year. Population denominators: U.S. Census Bureau state estimates by single year of age. Mask policy: Oxford COVID-19 Government Response Tracker, indicator H6, statewide-general days only. Learning modality: COVID-19 School Data Hub, 14,921 districts across 48 states. The ages 3–4 measure covers all disabilities — the federal file has never published the speech category by single year of age. IEP counts are a record of what school systems did, not a direct measure of children; that is the nature of the source and we present it as such.

It moved into the peer-reviewed literature.

Three papers, four years, one Springer journal. Together they say something that until recently you could not say in public without being throttled for it: that the science underneath the COVID vaccination program was not rigorous, and that the machinery built to catch exactly that kind of failure did not catch it.

Start with the dose

In September 2022, four researchers — David Dunn, Richard Gilson, Sheena McCormack and Laura McCoy — did something obvious that nobody had done. They went back and read the early-phase dose-finding trials for the nine COVID-19 vaccines approved by the World Health Organization, plus one more that showed partial efficacy.

Dose-finding is the unglamorous part of drug development. Before you run the big efficacy trial, you test several dose levels in small groups to work out which one gives you the best balance of immune response and side effects. Then you carry that dose forward, and eventually a regulator authorises it, and eventually people get it in their arms.

Here is what the review found.

The number of dose levels actually evaluated ranged from one to seven depending on the vaccine. The number of subjects per dose level ranged from 15 to 190. At the low end, that is fifteen people.

Higher doses did tend to produce better immune responses — but the difference between the top dose and the one below it was typically less than 1.6-fold on both binding antibody concentration and neutralising antibody titre. Meanwhile reactogenicity — the fevers, the aches, the lost day — rose with dose, as it always does.

And the finding that should have been a scandal on the day it published:

All of the trials had at least one important design limitation: few doses evaluated, large gaps between adjacent doses, or an inadequate sample size.

All of them.

Their conclusion, in careful academic register:

it is “uncertain whether the single dose taken into clinical efficacy trials, and subsequently authorised by regulatory agencies, was optimal.”

And more pointedly: “the recommended doses for some vaccines appear to be unnecessarily high.“

Over twelve billion doses have been administered worldwide. The amount in each of them was selected by comparing, in some cases, a single tested level — sometimes in groups of fifteen — with gaps between dose levels too wide to distinguish between them.

It means nobody established that it was right, and the people who eventually checked found the evidence base underneath one of the largest medical interventions in human history was thin enough to fit in a paragraph.

One honest note, which we’d rather make ourselves: the Dunn review is a preprint. It has not been through peer review.

Then the bioethicists picked it up

In December 2023, the Journal of Bioethical Inquiry published a paper by five academics who are about as far from the fringe as this subject gets: Wendy Lipworth and Ian Kerridge at Macquarie, Cameron Stewart at Sydney Law School, Diego Silva at Sydney Health Ethics, and Ross Upshur at the University of Toronto’s Dalla Lana School of Public Health.

Their subject was what happens to science when you speed it up. Their conclusion was that the rush to produce and apply pandemic research “created opportunities for unscrupulous actors” and — the phrase to remember —

“compromised the methodological, peer review, and advisory processes that would usually identify sub-standard research and prevent compromised clinical or policy-level decisions.”

To wit: the systems that exist to catch bad science before it reaches your arm were themselves degraded by the emergency. Compromised. Bypassed.

They give examples. The modelling studies used as the basis for policy on isolation, vaccination and treatment “were biased and used poor controls, inappropriate inclusion and exclusion criteria, and poor statistical analysis.” Studies of the antiviral molnupiravir overstated the benefit.

And on vaccines specifically, this:

“some early phase studies to determine vaccine doses were heterogeneous and underpowered”

The citation on that sentence is Dunn et al.

So the preprint we flagged thirty seconds ago is carried into the peer-reviewed literature by five senior bioethicists who evidently read it and agreed with it. That is how a preprint is supposed to earn its keep, and it is worth more than a citation count.

We’re going to be careful here, because the distinction matters and someone will check. Lipworth and colleagues did not publish a general verdict that COVID vaccine science was poor and biased. Their “biased and poor controls” language is about the modelling behind policy. Their vaccine-specific claim is narrower and sharper: the dose-finding was underpowered. Anyone quoting them as having condemned vaccine science wholesale is overreaching, and we’re not going to do it.

What happened in between

Between the 2023 paper and this week, we published a fair amount of what those compromised advisory processes looked like from the inside.

A CDC that drafted a national Health Alert Network advisory about myocarditis in teenage boys in May 2021, and never sent it — downgraded to a webpage after the FDA objected, with the line telling doctors to keep those kids off the field deleted along the way.

Three officials privately agreeing in January 2021 that fever after the second dose “theoretically could be associated with miscarriage in the 1st trimester” — while the CDC was sending Facebook and Twitter decks with “infertility” on the misinformation list, and following up to check the posts had come down.

A 2021 FOIA release in which the deliberative-process exemption blacked out every scientific judgment in the February 2020 origins thread and released the flight schedules.

And this month, an mRNA flu vaccine approved for the population that supplies three-quarters of flu deaths on antibody levels rather than efficacy, against a comparator the FDA itself conceded in writing “is not the preferred standard of care in this age stratum.”

Lipworth et al. described a mechanism. We have been documenting it running.

And then yesterday

On August 11, 2026, the Journal of Bioethical Inquiry published “Lack of Rigour on COVID-19 Vaccines” by Raphael Lataster of the University of Sydney.

The argument is a direct reply to the 2023 paper. Lipworth and colleagues hypothesised that sped-up science had produced a multitude of mistakes. Lataster examines the more recent evidence and concludes they were right — and that the mistakes ran through the vaccines’ development, recommendation, media coverage, and mandating.

Four domains. Every one of them is a place where this newsletter has receipts.

Now the caveats, because we would rather list them than have them listed for us. Lataster is a single author advancing an argument, not a systematic review or a consensus statement. He also wrote the Substack post that brought this to wider attention, which means the “experts say” framing circulating today is partly an author describing his own work. None of that makes him wrong. All of it is worth knowing before you cite him.

Why this matters more than another document dump

Because of where it is.

For five years the standard move against anyone raising these questions was not to answer them. It was to place the person outside the boundary of legitimate discussion — a conspiracy theorist, an anti-vaxxer, a spreader of misinformation. That boundary was enforced, as we have documented at length, with the direct participation of federal agencies working through private platforms.

The boundary does not hold when the argument is inside a Springer journal, made by professors of bioethics at Macquarie and Toronto, resting on a review of the dose-finding literature by researchers at University College London and the MRC.

You cannot deplatform the Journal of Bioethical Inquiry.

The dose was never properly established. The processes that should have caught that were compromised by speed. And the peer-reviewed literature now says both of those things out loud.

It took four years, and it should not have taken a Senate subpoena, a FOIA lawsuit and a change of administration to get here. But here we are.

Sources: Dunn D, Gilson R, McCormack S, McCoy L, “Dose of approved COVID-19 vaccines is based on weak evidence: a review of early-phase, dose-finding trials,” medRxiv, 22 September 2022 (preprint), doi:10.1101/2022.09.20.22276701. Lipworth W, Kerridge I, Stewart C, Silva D, Upshur R, “The Fragility of Scientific Rigour and Integrity in ‘Sped up Science’,” Journal of Bioethical Inquiry 2023;20(4):607–616, doi:10.1007/s11673-023-10289-w. Lataster R, “Lack of Rigour on COVID-19 Vaccines,” Journal of Bioethical Inquiry, online ahead of print 11 August 2026, doi:10.1007/s11673-025-10548-y. Prior RationalGround reporting as linked.

Three officials, in writing, treating a reproductive risk as a real question worth raising.

Now here is what was happening to Americans who raised the same question in public.

The deck

The lawsuit I filed against Facebook and other government entities elicited some interesting exhibits during discovery.

The CDC produced a slide deck for the trust-and-safety teams at Facebook and Twitter. It is on CDC letterhead, carries the CDC 24/7 logo, and is titled “COVID Vaccine Misinformation: Hot Topics.” It is dated May 14, 2021. Two decks with that title appear in the exhibits.

Inside: a formal agenda of misinformation topics, a set of “Example posts” marked as what the platforms should be on the lookout for, and a logistics slide naming a CDC point of contact for the recurring meetings.

The slides are marked “NOT FOR FURTHER DISTRIBUTION.”

The meetings had a name. They were called “Be On the Lookout” meetings — BOLO. The first misinformation deck went to Facebook in December 2020, a month before the text messages. Formal BOLO meetings ran May 14, May 28, and June 18, 2021.

This is not a leak, an allegation, or an inference. It is the case record in Hart v. Facebook, and the deposition of the CDC’s Chief of Digital Media.

The loop, in one sentence

March 31, 2021. The CDC to Facebook:

“It looks like the post from last week’s deck about infertility and side effects have all been removed. Were those re-evaluated by the moderation team or taken down for another reason?”

Read that again slowly, because it is the entire argument in a single sentence.

Posts about infertility and vaccine side effects were in a CDC deck. The posts came off the platform. The CDC followed up to ask why they came off.

Flag, removal, confirmation. The government checking its own work.

And in the same exchange, a second question — about local news coverage of deaths after vaccination:

“One of the main themes we’re seeing and from the CrowdTangle report is local news coverage of deaths after receiving the vaccine. What’s the approach for adding labels to those stories?“

Not fringe accounts. Not anonymous cranks. Local news coverage.

A note on “Census,” which appears throughout the record: it does not mean the Census Bureau as you’d assume. It refers to a CDC-linked communications team working the same channel.

Four categories. Two of them were reproductive health and the government’s own data.

Carol Crawford, the CDC’s Chief of Digital Media, sent Twitter a spreadsheet of four example misinformation categories — at Twitter’s request, so the platform could see what would come up in the BOLO meetings:

1. Vaccines aren’t FDA approved

2. Fraudulent cures

3. VAERS data taken out of context

4. Infertility

We’ll be fair about the first two. Fraudulent cures are actual fraud, and so are forged vaccination cards, which appear elsewhere in the same workflow. Nobody sensible objects to a platform removing a scam.

The other two are a different animal.

VAERS is the government’s own adverse-event reporting system. It exists so that the public can report and examine suspected harms. Citizens discussing the contents of a federal safety database became a category on a federal misinformation list.

And infertility is not a product claim. It is a safety question about reproductive health — the identical territory the three most senior health officials in the country were privately texting about ten weeks earlier.

There is significant evidence that the Covid vax can impact fertility.

Crawford’s testimony on the fourth category: “I’m assuming this was people claiming that getting the vaccines led to infertility.”

Assuming. About a category she supplied.

The themes that carried “potential risks”

From the monitoring reports flowing to the CDC, here are the themes singled out as carrying “potential risks”:

• Reports of healthcare workers refusing the vaccine — with a Forbes article cited

• Posts about alleged vaccine-related deaths

• News and reports of severe vaccine reactions

Nurses declining a shot is a fact about the world. It either happened or it didn’t. It is not a claim that can be true or false — it is an event, reported in Forbes.

It was flagged as a risk.

Then the standard moved

Facebook sent the CDC lists of claims and asked the agency to assess, for each one, whether it was false “and if believed this claim could contribute to vaccine refusals.”

Crawford noticed the change herself: “this added the whole ‘could this contribute to vaccine refusals’ element.”

Understand what happened there. The test was no longer is this false. The test became is this false, and would believing it cause someone to decline a vaccine.

Those are entirely different questions, and only one of them is about truth. Under the second test, a true statement that discourages vaccination is a problem to be managed.

The same message describes a new Instagram penalty for accounts repeatedly posting content that violated COVID policy: they “may now lose the ability to be tagged or mentioned or may see pop-ups asking if they’d like to delete certain posts.”

Not removal. Erasure from the conversation, and nudges to delete your own words.

A claim the CDC admitted it could not debunk

On a claim about COVID survival rates, the CDC’s own internal note reads:

“Not able to debunk now, inconclusive. Scientists would be hesitant to attach a correct number to the survival rates.”

The note adds that the claim is “tied to the VAERS issue.”

The agency could not say the claim was false. It said so, in writing, to itself.

The claim stayed in the misinformation workflow anyway.

Pre-clearing a product that didn’t exist yet

Facebook, to the CDC, on paediatric claims:

“We understand the FDA is considering giving emergency use authorization for COVID-19 vaccine for children under five in coming weeks. We are considering how our existing policy on COVID-19 vaccines should apply to claims about children 6 months to 4 years once the vaccine is approved for use. Can you please assess for each claim whether it is false for children in this age range and if believed, likely to contribute…”

A private company asked a federal agency to determine in advance what would be false about a medical product that had not yet been authorised for the population in question.

The safety data did not exist yet. The misinformation policy did.

Facebook later told the CDC that once authorisation came through for children, it “immediately updated our policies globally to remove false claims about the COVID-19 vaccine for children, e.g., the COVID vaccine is not safe for kids.”

What a federal appeals court made of this

The Fifth Circuit reviewed the same programme in Missouri v. Biden:

“The CDC’s relationship with the platforms began by defining—in ‘Be On the Lookout’ meetings—what was (and was not) ‘misinformation’ for the platforms… the CDC was entangled in the platforms’ decision-making processes.“

Entangled. A court’s word, not ours.

And the finding that disposes of any remaining argument that this only ever touched falsehoods — a platform committed to a policy change:

“focused on reducing the virality of anti-vaccine content even when it does not contain actionable misinformation.“

Even when it does not contain actionable misinformation. Content that broke no rule, throttled because of what it was about.

The counter-argument, in her own words

Crawford’s position throughout her deposition is that the CDC was sharing scientific information and asking questions, not directing removals:

“I don’t think we were asking them to add labels… We were asking them what their approach for labels were.”

That is her testimony and we’re not going to paraphrase her into an admission she never made. She says the platforms invited the CDC to flag major themes, and she says she was answering questions rather than issuing instructions.

Put her denial next to the March 31 email and decide for yourself. An agency that is merely curious about approaches does not follow up to confirm that specific posts came down.

What this cost, and why it still matters

Here is our position, stated plainly.

The jury is genuinely still out on what these vaccines did to pregnant women. We do not claim they caused miscarriages. We claim the risk was never zero — the officials running the response said as much to each other in writing — and that the benefit to healthy pregnant women, who were at low absolute risk from COVID, was oversold with a confidence the data never supported.

But that is not the scandal. The science can be argued. The scandal is that it couldn’t be.

While Fauci, Walensky and Murthy were privately agreeing that fever after dose two could be linked to first-trimester miscarriage, their own agency was sending private companies decks with “infertility” on the list, checking whether those posts had come down, asking about labels on local news coverage of deaths, and flagging nurses who declined the shot as a risk theme.

The conversation they were permitted to have with each other was the conversation Americans were not permitted to have with each other.

That asymmetry is the whole of it. Not a disagreement about virology — a government that reserved a question for itself and policed the public out of it. And it is precisely why the question is still unresolved five years later: the process that would have surfaced an answer was throttled while the answer was still forming.

One more date. On April 1, 2021, HHS entered a $15 million advertising agreement with Facebook. Nick Clegg told Surgeon General Murthy that Facebook had removed over 20 million pieces of COVID and vaccine content.

And on July 13, 2021 — after the ad money, after the BOLO meetings — this newsletter’s founder was suspended from Facebook over a graphic about masking children. That suspension is why Hart v. Facebook exists, and why these documents are public at all.

We were told we were spreading misinformation.

The people who told us that were, at the same time, texting each other about miscarriage risk.

Sources: the case record in Hart v. Facebook — 32 documents, 1,354 pages, including the November 15, 2022 deposition of Carol Crawford, CDC Chief of Digital Media, and the First Amended Complaint with exhibits. Every image above is a page from the court record with its Excerpts of Record stamp visible. Appellate quotations are from the Fifth Circuit’s opinion in Missouri v. Biden. Originals at the Liberty Justice Center’s case page. Text-message quotations are from the August 10, 2026 release by Senators Ron Johnson and Rand Paul.

The first batch is about pregnant women.

January 25, 2021

A group chat between three people: Fauci, Surgeon General Vivek Murthy, and CDC Director Rochelle Walensky. Murthy wants to know what to tell pregnant women.

Walensky answers first, and her answer is worth keeping in mind for everything that follows — her own description of the evidence base is that “this is all a data free” zone. CDC had stood up a pregnancy registry. Some 15,000 pregnant women were enrolled in V-safe.

Then, at 4:56 PM, Fauci raises the perception problem:

“There is this misperception out there that we need to deal with that mRNA somehow can ‘get into your genes’. When the misperception gets extended to an early developing fetus, the anxiety heightens”

An hour and forty-nine minutes later, he writes again.

“I asked around a bit more and another issue came up that you need to be aware of. Since many people have significant cytokines storm and fever after the 2nd dose, this theoretically could be associated with miscarriage in the 1st trimester“

This is the message Johnson led his release with, and rightly. But the part that has been overlooked in the coverage so far is what happened next.

Two minutes later, Walensky: “Definitely a good point, esp after dose two.”

Forty minutes after that, Murthy: “This is really helpful, thanks… That’s a really good point re the cytokine storm post second dose.“

The Director of the CDC and the Surgeon General of the United States both agreed, in writing, that the concern was legitimate. This was not one official musing privately. It was the three most visible public health officials in America reaching a shared conclusion on a Monday evening.

The next afternoon, WHO said the quiet part out loud

At 4:34 PM on January 26, Murthy brings news:

“Tony, Rochelle, did you see this announcement today from WHO saying they advise against pregnant women taking the Moderna vaccine bc of lack of data? It has led to a flurry of questions. What is your take? I imagine we will all get asked about this.”

Twenty-one minutes later, Fauci sends the message that has been quoted everywhere today:

“I would say that in any vaccine in which data are lacking in pregnant women one must weigh the potential risks against the benefits. COVID-19 can be a serious disease especially in pregnancy. In the vaccine trials, pregnant women were not allowed but following the EUA, a large number (more than 10,000) pregnant women have been vaccinated. No issues have arisen. The FDA and the ACIP and the ACOG are all permissive for pregnant women to have a choice. Animal tox studies did not raise any red flags.“

Sit with the sequence. The World Health Organization advises against a product in pregnancy for lack of data. Twenty-one minutes later, the position that would be repeated publicly for the next year is composed in a text message.

And read the message itself. In the same breath as “no red flags,” he concedes that pregnant women “were not allowed” in the trials. Both things are in the same paragraph.

Murthy’s reaction to the WHO advisory is the tell:

“That’s a strong statement to make and potentially quite damaging to public confidence among pregnant women.“

Not is WHO right. Whether it would damage confidence.

He had done this properly once before

In 2009, during H1N1, the same agency under the same director ran dedicated pregnancy trials. From Fauci’s own diary that November:

“NIAID is the only one giving good news si[n]ce our pregnancy trials and pediatric trials are giving good data – only 1 dose of 15 ugs for pregnant women is required for robust response…”

Pregnant women got their own trials and their own dose-finding data. He was proud of it.

In 2020 they were excluded from the trials entirely — and he was told so, in writing, before a single dose shipped. A National Academies draft circulating in his own inbox that September:

“it is concerning that most, if not all, of the current Phase II/III trials exclude pregnant women… OWS, NIH, and CDC should include assessment of vaccine efficacy, effectiveness, and safety among pregnant women in their clinical development and post-marketing surveillance plans.“

The same point appeared twice in his daily news digest, via a STAT piece headlined “Will Covid-19 vaccines be safe for children and pregnant women? The data, so far, are lacking.”

Nobody had to discover this problem later. It was flagged in advance, in his own reading pile.

February 8: the co-inventor writes

Nine days after the “no red flags” text — and five days after Fauci repeated the phrase on camera — Drew Weissman, co-inventor of the nucleoside-modified mRNA platform, emailed him directly:

“using a dose 100-fold higher than current vaccines and given IV, the mRNA-LNPs go to the placenta and fetus, measured by translation of the mRNA and cause an immune response in the amniotic fluid. In non-pregnant female mice, the uterus and ovaries take up mRNA-LNPs and translate the mRNA. We are repeating these studies with human dose levels… and are waiting to release the findings until these studies are completed”

We are quoting the whole sentence deliberately, including the conditions, because they matter and because the first reply to any excerpt will be to point them out: this was 100 times the human dose, given intravenously. The vaccine is given intramuscularly at 1×. Anyone telling you this email proves the vaccine crosses the placenta in humans is overreading it, and we’re not going to do that.

Fauci replied the same night, at 9:12 PM:

“Thanks for sending this. Looks like you have touched all the bases of informing people of your results… Congratulations on the Rosenstiel Award!“

(I wonder if he was a bit jealous of that award!)

March 1: the version at human dose

Three weeks later Weissman came back, having run it the way the vaccine is actually given — human dose levels, intramuscular, in pregnant mice.

“We still have some concern over the data and are talking to all of the appropriate people.”

This is the sentence that matters, and it’s the one that survives the dose-and-route objection.

Fauci forwarded it within the hour to Vaccine Research Center director John Mascola, copying Cliff Lane: “Let us discuss.”

The internal answer came back the next day from the VRC’s Karin Bok. She reported no concerns — and in doing so, recorded what the expert panel had already decided:

“The SMEs on that call (especially obgyns) were aware that mRNA might reach the fetus. The only discussion was around the possibility of that mRNA to integrate into the nuclear DNA. And the second main concern was the fever caused by the vaccine, especially during the first trimester, but they went around that by recommending the administration of acetaminophen after vaccination.“

Now put that beside a message from five weeks earlier.

On January 26, Walensky had written: “ACIP meeting tomorrow. Shall I flag it for them?“

So the chain runs end to end, across two document sets obtained by two different senators on two separate tracks. January 25: Fauci raises fever-driven first-trimester miscarriage risk. Walensky and Murthy agree it’s a real concern. January 26: Walensky offers to put it in front of ACIP. March 2: the record shows the panel’s answer to that exact concern.

Take Tylenol.

We want to be fair about Bok’s email, because it is also the strongest rebuttal in the file and it sits in the same PDF. Her position is that mRNA reaching the placenta is an expected finding in mice, that the obstetricians on the panel knew it, and that they weighed it. That is a defensible answer. The story is not that nobody considered the question. The story is what they decided once they had, and how little of that deliberation reached the women being asked to decide.

Then the language hardened

April 21, 2021 — the Shimabukuro V-safe paper in NEJM reports a 12.6% spontaneous-abortion rate (104 of 827) and becomes the reassurance everyone cites. Worth knowing about that denominator: some 700 of those 827 women were vaccinated in the third trimester, past the window in which a spontaneous abortion under 20 weeks could be counted at all. The rate was computed against a group that largely could not have experienced the outcome being measured.

August 11, 2021 — CDC’s Sascha Ellington: “We are not seeing a signal of safety concerns of the vaccine in pregnancy.”

August 30, 2021 — Fauci: “there’s no indication whatsoever that there’s any increase of any adverse issues in a pregnant woman who was vaccinated.”

September 29, 2021 — Walensky issues a health alert. CDC now “strongly recommends” vaccination before or during pregnancy.

February 1, 2022 — Murthy, the man who asked the original question thirteen months earlier: “It’s all the more important for you to get vaccinated if you are pregnant or if you are trying to become pregnant… anytime during your pregnancy is actually fine.“

The private uncertainty of January 2021 never made it into the public record. The public language moved in exactly one direction, and it wasn’t toward the caveats.

What this is

It is not proof that the vaccine caused miscarriages. Nothing here shows that, and we’re not going to claim it.

What the documents show is narrower, and it is enough:

The three officials directing America’s pandemic response privately identified a specific, biologically plausible risk to first-trimester pregnancies. They agreed with each other that it was a real concern. They knew pregnant women had been excluded from the trials, because it had been flagged in writing months earlier. When a foreign regulator said so out loud, the response was composed in twenty-one minutes and worried about public confidence. The concern went to an expert panel and came back answered with an over-the-counter fever reducer. And over the following year the public messaging grew steadily more absolute, ending with a Surgeon General telling women trying to conceive that any point in pregnancy is fine.

Pregnant women were never told any of this was under discussion. They were told there were no red flags.

There are 34,000 more messages on that phone.

Sources: text messages released 10 August 2026 by Senators Ron Johnson and Rand Paul (4 pp, rendered here at 200 dpi from the official PDF); the Weissman correspondence from Senator Paul’s 24 July 2026 award-records release, pp. 55–60; NIH FOIA case 57534 for the National Academies draft and news digests; Fauci’s H1N1-era diary from the Senate release of 24 July 2026; Shimabukuro et al., NEJM, 21 April 2021. Public statements are quoted with their dates from the senators’ compiled timeline.

You see a black box. The agency cites an exemption code. That’s the end of it. You can suspect, you can complain, and you have no way to check.

Not this time. In four cases we now have both versions of the same page — the redacted release, and the same document later produced through litigation, a court case, or a congressional in-camera review. You can lay them side by side and read what the black box was covering.

In every single case, the withheld material is not an internal proposal, not a negotiating position, not a private phone number.

It is a scientist saying where he thought the virus came from. The redaction reason is obvious — CYA.

What the exemptions are actually for

Three FOIA exemptions matter here.

(b)(5) — deliberative process. Protects the internal back-and-forth of government before a decision is made, so officials can argue freely.

(b)(4) — trade secrets. Protects confidential commercial and financial information.

(b)(6) — personal privacy. Email addresses, phone numbers, personal details.

This is mostly about (b)(5) and (b)(4).

Case one: the pleasantry survived

On February 1 and 2, 2020, Anthony Fauci, NIH Director Francis Collins, NIH’s Lawrence Tabak and Jeremy Farrar of the Wellcome Trust ran an urgent email thread with the world’s leading virologists about whether SARS-CoV-2 had come out of a laboratory. Subject line: “Teleconference.”

The 2021 FOIA release contains that thread. It is redacted 21 separate times under (b)(5) — deliberative process. That one thread accounts for 5% of every (b)(5) redaction across the entire 3,234-page release.

The unredacted version came out later through litigation. Start with a single email, because it is the whole story in one paragraph.

Collins writes to Farrar at 12:03 on February 2. The released version gives you the courtesy line — thanks for forwarding these reflections from Mike and Bob — and then a black box.

What the black box covered, in the very same paragraph:

“I hadn’t given much consideration to the idea of lab-based evolution by tissue-culture passage, but that is worth including on the list of options. Waiting a month sounds like a really bad idea.”

The Director of the National Institutes of Health, saying laboratory evolution belongs on the list of options. Withheld as a deliberative process. The thank-you was released.

It happens again in the same thread. In an email that morning, Collins’s flight schedule was released — “available any time today except 3:15 – 5:45 pm EST (on a plane).” The sentence above it was not:

“Though the arguments from Ron Fouchier and Christian Drosten are presented with more forcefulness than necessary, I am coming around to the view that a natural origin is more likely. But I share your view that a swift convening of experts in a confidence-inspiring framework… is needed, or the voices of conspiracy will quickly dominate.”

His plane time was public. His assessment of the origin of the pandemic was a state secret. Why? CYA.

What else was under the boxes

The same thread, same exemption:

Jeremy Farrar, head of the Wellcome Trust, putting a number on it:

“On a spectrum if 0 is nature and 100 is release — I am honestly at 50!”

Andrew Rambaut, the evolutionary biologist:

“From a (natural) evolutionary point of view the only thing here that strikes me as unusual is the furin cleavage site. It strongly suggests to me that we are missing something important in the origin of this virus.”

Michael Farzan, who discovered the SARS receptor, on how it might have happened accidentally — passaging SARS-like coronaviruses on human cell lines under BSL-2 conditions — and his own odds:

“I am 70:30 or 60:40.”

Robert Garry, who two months later would co-author Proximal Origin, the paper that publicly declared a lab origin implausible:

“I really can’t think of a plausible natural scenario where you get from the bat virus or one very similar to it to nCoV where you insert exactly 4 amino acids, 12 nucleotides, that all have to be added at the exact same time to gain this function… Of course in the lab it would be easy to generate the perfect 12 base insert that you wanted.”

And Ron Fouchier, arguing for shutting the discussion down — with a reason that is worth reading twice, because it isn’t a scientific one:

“further debate about such accusations would unnecessarily distract top researchers from their active duties and do unnecessary harm to science in general and science in China in particular.”

Incredible they thought they could hide this under redactions.

The tell

Here is what did survive the redactions across the whole thread. Every substantive released sentence:

“Do you have a minute for a quick call?” — “they need to decide today in my view.” — “Sorry that I took so long to weigh in.” — “Really appreciate us thinking through the options.” — “See thoughts overnight from others.” — “I’m available any time today except 3:15 – 5:45 pm EST (on a plane).” — “This is a very complex issue.” — “I suggest we don’t get into a further scientific discussion here, but wait for that group to be established.”

Scheduling. Pleasantries. Throat-clearing.

Not one scientific judgment survived.

A reviewer working carelessly produces a mess — some science through, some logistics blacked out. What’s here is the opposite of a mess. It is clean, and it is consistent, and it sorted the material by exactly one criterion.

The small absurdity that makes the point better than any argument:

Fauci explains that he was slow to reply because he’d been on conference calls. The kind of meeting he had been sitting in was withheld as deliberative process.

Case two: fourteen words of “trade secret”

Two days later, the same men were still at it. This one wasn’t withheld under deliberative process. It was withheld under (b)(4) — trade secrets and confidential commercial information.

The released version is one word and a black bar. “Yes,”

The full sentence:

“Yes, I’d be interested in the proposal of accidental lab passage in animals (which ones?).”

Fourteen words. There is no product in that sentence. No contractor, no pricing, no formula, no commercial party of any kind. It is the Director of the NIH asking which animals a laboratory accident might have involved.

It was withheld under the exemption for trade secrets.

The message he was replying to was blacked out too — Farrar writing that “’Engineered’ probably not. Remains very real possib[ility] of accidental lab passage in animals,” noting Eddie Holmes was “60:40 lab side,” and adding: “I remain 50:50.”

Case three: the report that was blacked out whole

In October 2017 — two years before the pandemic — a NIAID official named Dr. Ping Chen filed a report on biosafety at the Wuhan Institute of Virology. It contained lines like this one:

“It is clear to me by talking to the technician that certainly there is a need for training support.”

When HHS released Chen’s report under FOIA, it redacted the entire document. Every word, citing privacy and deliberative process.

Senator Ron Johnson’s staff were eventually permitted to read the unredacted version in camera — in a room at HHS headquarters, no copies allowed. They transcribed it by hand and published both versions side by side in September 2023.

Johnson’s conclusion:

“In the public FOIA document, HHS redacted Dr. Chen’s entire report claiming that it contains privacy and deliberative information. Following my staff’s in camera review of the report, it seems apparent that the only reason that HHS redacted this information was to hide the report’s contents from the American people.“

Two details sharpen it. Chen had been named specifically in a June 2021 records request from five senators — HHS knew precisely what was being asked for. And it was still withholding her report two years after that.

Case four: the vocabulary that vanished

This one needs no side-by-side comparison. It’s a word count.

Across all 3,234 pages of the 2021 release, against the 208 pages of the later litigation release:

Term 2021 release (3,234 pp) Litigation release (208 pp) furin 0 38 passage / passaging 0 37 glycan 1 34 outbred 0 5 inbred 0 4

Zero. In three thousand pages.

The obvious objection is that these scientists must simply have been outside the scope of the request. They were not. They are all over the 2021 release: Fouchier appears 8 times, Garry 7, Rambaut 9, Marion Koopmans 12, Patrick Vallance 12, Farrar 82, Kristian Andersen 18.

The people are there. Their science is not.

The single surviving “glycans” is worth quoting for what surrounds it — Farrar to Fauci and Collins, February 5, 2020, on the WHO group that would investigate the origins:

“They have asked for names to sit on that Group - please do send any names… With names to be put forward into the Group from us and pressure on this group from your and our teams next week… they will add further comments on the glycans.”

The body that would investigate the origin question was being staffed from this circle, and the paper that would publicly settle it was being drafted — in the same email.

Case five: the number that ends the legal argument

The 2021 release invokes deliberative process 139 times and trade secrets 121 times.

NIH FOIA case 57534 — 7,386 pages of the same official’s email, released in 2026 — invokes deliberative process zero times.

Same official. Same kind of records. Twice the page count.

One release treated his internal correspondence as legally protected. The other withheld not a single line on those grounds.

The exemption was never a legal necessity. It was a choice, and a different administration made the opposite one.

What this proves, and what it doesn’t

We’re going to be careful here, because the case is strong enough that overstating it would be the only way to lose.

Documented and checkable by anyone: the (b)(5) redactions in the February 2020 thread covered scientific opinion while leaving scheduling intact. The trade-secrets exemption was applied to a sentence containing no commercial information. HHS blacked out Chen’s report in full. The word counts are the word counts. FOIA 57534 uses no (b)(5) at all.

Inference, and we’ll label it: that this reflects intent rather than an overzealous reviewer. It is a strong inference — line-by-line consistency across 21 redactions in one thread is difficult to produce by accident — but it is an inference. Senator Johnson states it as his conclusion. We’re comfortable quoting him.

What it does not prove: that COVID came from a lab. Nothing above establishes that. What it establishes is narrower and, frankly, worse: serious scientists privately considered a laboratory origin plausible in February 2020, and the American public was not permitted to watch them do it.

The exemption for deliberative process exists so that officials can think out loud without fear. It was used here to make sure nobody found out that they had.

Not a national security interest. Not a commercial secret. Not privacy.

Cover for the people who would have to answer questions.

Sources: the 2021 NIH FOIA release of Anthony Fauci’s email (3,234 pp), the later litigation release of the same February 2020 correspondence (208 pp), Senator Ron Johnson’s September 2023 publication of Dr. Ping Chen’s 2017 report alongside the redacted version, and NIH FOIA case 57534 (7,386 pp, released August 2026). Comparison images show the same email from two releases; the grey bars remaining in the unredacted versions are ordinary (b)(6) privacy redactions applied in both. Nothing here was unmasked or reverse-engineered — the unredacted pages come from later official releases. Every quotation is citable to a Bates-numbered page.

Let’s get the disclosure out of the way first, because it isn’t a disclosure so much as the reason this newsletter exists.

The origins of Rational Ground are tied directly to Dr. Scott Atlas’s efforts at the White House. We formed a pro bono data team and for roughly six months provided nearly daily support to Dr. Atlas, helping him right the ship. We were proud of that work then and we’re proud to say so now.

So read what follows knowing exactly where we sit. Then read the emails, which say what they say regardless.

The document set

On August 3, 2026, the Independent Institute’s Phillip W. Magness obtained and released NIH FOIA case 57534 — seven volumes, 7,386 pages of Anthony Fauci’s NIH email.

The important thing about this release isn’t the page count. It’s what isn’t blacked out.

Across all 7,386 pages there are zero (b)(5) withholdings — the deliberative-process exemption, the one agencies use to hide internal discussion. The 2021 Leopold release had 414 of them. What remains here is routine privacy redaction.

For the first time, the internal conversation is legible.

August: the part that cuts against us

We’re going to start with the material that complicates our own story, because a piece that skips it isn’t worth reading.

In late August 2020, Fauci’s engagement with Atlas’s scientific argument was real. Atlas had advanced the position that cross-reactive T-cell immunity from prior common-cold coronavirus exposure meant the herd immunity threshold was far lower than assumed. Fauci didn’t wave it away. He went and asked.

On August 25 he wrote to Deborah Birx summarizing what Shane Crotty and Alex Sette — two genuinely first-rate immunologists at Scripps — had to say about it, and NIAID began scoping actual studies to test the question, including expanded longitudinal cohorts.

In the same period he wrote something that, read in 2026, is almost poignant:

“As I mentioned to you last week, we should sit down with Scott and explicitly spell out our differences since they will become public very soon. We have to be able to say that we gave him the courtesy to know that we disagree.“

That is a man behaving properly.

The same email contains the other August tell. Learning that the White House was bringing in outside physicians, Fauci wrote to Birx:

“The White House is sending a strong signal by having Scott Atlas, along with Dr. Jay Bhattacharya, Dr. Joseph Ladapo, Dr. Martin Kulldorff, and Dr. Cody Meissner up there. As Butch Cassidy said to the Sundance Kid — ‘Who are those guys?’ They are stacking the deck against you.”

Readers of this newsletter will recognize most of those names. In August 2020 they were, to the man running the response, an unfamiliar threat.

October 6: the removal demand

Six weeks later the courtesy is gone.

A Business Insider reporter emailed Fauci for comment on an interview she’d conducted with Atlas. Fauci forwarded it to Marc Short, the Vice President’s chief of staff, under the subject line “PLEASE READ!”

“I am not planning to respond publicly since that would be beneath my dignity. However, a line has been crossed here am [sic] I strongly suggest that Dr. Atlas be removed from the Task Force and quite frankly from the White House itself.“

Then, after disputing Atlas’s characterizations of him, comes the part that deserves to be read twice:

“what he has been preaching about public health measures has put the POTUS and the White House staff at risk, resulting in the infection of our President and several White House staff and other government officials. As I mentioned, I am not going to address this publicly, but it must be addressed internally.”

Fauci told the Vice President’s chief of staff that Scott Atlas was responsible for infecting the President of the United States. Trump had been discharged from Walter Reed five days earlier.

And in the same breath he states the operating principle out loud: nothing in public, everything internally.

October 8: “crazy Scott Atlas”

Two days later, on Atlas’s television bookings:

“I just pinged Alyssa in an almost annoyed way asking what the heck happened to the list that Courtney sent down, when she is allowing crazy Scott Atlas to be on FOX. I think [I] got her attention since she texted right back and said that she will follow-up with Courtney.”

The same day, Fauci forwarded Francis Collins an article attacking the Great Barrington Declaration, and separately circulated a piece calling focused protection a deadly delusion. This is the week the GBD was signed, and the internal response was immediate.

October 14: “it’s him or us”

On the morning of October 14, a group of veteran AIDS activists and academics — Peter Staley, Gregg Gonsalves of Yale, Mark Harrington, David Barr — emailed Fauci and Birx under the subject line “This is nuts.”

Staley’s message is blunt:

“Isn’t it time for drawing a red-line regarding Atlas? The last thing the WH wants right now is seeing headlines saying Fauci and Birx resigned from the task force in protest. Why is this crazy man with no infectious disease experience cock-blocking all your access to the president... It’s time for you both to tell Pence, ‘it’s him or us.’“

Fauci replied that afternoon:

“I am trying to get this jerk shut down by working from within the White House. They promise they will handle it and nothing happens. I will keep trying.”

Birx answered the group too:

“I am just glad it’s in the open who has been undermining Tony and I in the WH and who was manipulating the data. We are pushing back very hard.”

The same night: Kushner

At 10:05 p.m. Eastern that evening, Fauci wrote to Birx:

“I had a long talk with Jared tonight about Scott Atlas. I really laid it on him, but he still does not completely understand how dangerous this guy is. He is still in the realm of ‘differences of opinion’. You have got to speak with Jared and confirm what I said that this guy is killing people. I likened him [to] the AIDS denialists from the 1980s.“

Two things about that email.

First, the comparison. For Anthony Fauci, whose career was forged in the AIDS crisis, calling someone an AIDS denialist is the most severe charge he has. He deployed it against a Stanford professor of radiology over a disagreement about pandemic policy — in writing, to a colleague, asking her to repeat it to the President’s son-in-law.

Second, the corroboration. Fauci’s personal diary, released separately by Senator Rand Paul’s committee, records a conversation with Kushner about Atlas that same night. Two independently obtained document sets, two different FOIA tracks, the same evening.

Only one person here is being a jerk, and it ain’t Scott.

October 16: “Please speak out for me”

Two days later, Fauci wrote to Birx that he would miss the Task Force meeting — a routine colonoscopy, “not for CNN consumption ☺” — and left instructions:

“Over the past week I have come out very strongly publicly against the ‘Great Barrington Declaration’... Scott Atlas may try to counter this. Please speak out for me.“

He then previewed Atlas’s expected arguments about post-COVID syndromes and cardiac MRI findings and asked Birx again: “Please speak out when he tries to play these things down.“

By October 18:

“Atlas did not go to the TF meeting, but he is present at the daily ‘huddle’ with the Vice President, Jared, and the other senior staff where he spouts his nonsense. He is driving Deb crazy.”

Birx has recently come out and thrown Dr. Fauci under the bus — but her hands are NOT clean here.

November: the closing argument

November 1, in Fauci’s own words: “that fool Scott Atlas.”

November 11, after a Politico piece on the Atlas–Fauci feud, Fauci wrote to White House communications director Alyssa Farah: “Do you folks have any control over this guy?”

Farah’s reply is the quiet confirmation that the campaign had been running for months:

“I Will immediately raise with Dr Atlas. This is unacceptable. He goes rogue and it’s wildly unhelpful. I’m so sorry you have to deal with this. I’ve advocated for his dismissal for months to no avail.“

By November 17, Fauci was tracking attendance — noting Atlas “was not physically present at the last meeting on November 9,” and that the Vice President had not called on him.

I remember that fateful day. Scott had headed back to California for Thanksgiving. President Trump had lost the election, and everything felt a bit uncertain as Scott’s tenure came to an end. I remember that day Mike Pence brought Dr. Burks and Dr. Fauci back onto the press platform, where they hadn’t been in quite a while. I texted Scott, and he was relieved that President Trump wasn’t there, but I could see the writing on the wall. Our political nightmare had just begun. 2021 and 2022 would prove fateful, as all of us know.

Scott Atlas resigned on December 1, 2020.

What this is, and what it isn’t

Fifteen weeks. From “we should sit down with Scott” and “give him the courtesy” to “this guy is killing people,” with a formal removal demand to the Vice President’s office in between.

A friend of ours online put it best: you can hear the line forming behind every one of these emails. Who will rid me of this meddlesome doctor?

Now the fair part, because we’ve been on the receiving end of unfair and we don’t intend to return the favor.

Nothing here is illegal. A federal official is entitled to think a colleague is wrong, to say so internally, and to argue for personnel changes. Fauci’s August engagement with the T-cell question was genuine — he commissioned real work from real immunologists rather than dismissing it. He also, by his own repeated statement, declined to attack Atlas publicly.

But that last point is the story, not the defense. “I am not going to address this publicly, but it must be addressed internally.” The public saw a scientific disagreement between two advisers, conducted with restraint. The private record shows a sustained fifteen-week effort to get one of them fired — routed through the Vice President’s chief of staff, the White House communications director, the President’s son-in-law, and a group of outside activists urging an ultimatum.

Americans were told there was a debate happening inside the White House. There was. They just weren’t told that one side was working to end it by removing the other man from the building.

We knew Scott Atlas. We worked for him for free because we believed the data supported what he was saying about focused protection, about schools, about the costs nobody was counting. We took a great deal of abuse for it.

Reading these emails six years on, we feel something close to vindicated. Mostly we feel like the argument was never allowed to be an argument.

Be sure to pick up Scott’s excellent book recounting his fight against Dr. Fauci

Source: NIH FOIA case 57534, seven volumes, 7,386 pages, obtained by Phillip W. Magness of the Independent Institute and released August 3, 2026. Quotations are verbatim from the released pages; OCR artifacts have been corrected and original typos preserved and marked [sic]. Every quotation above is citable to a specific volume, PDF page and Bates number. Corroborating diary material comes from the separate release by the Senate Homeland Security and Governmental Affairs Committee, July 24, 2026.

That number is real. It is also not efficacy, and the difference matters more than anything else in this story.

We read the full document set — the approval letter, the prescribing information, FDA’s 90-page briefing document to its advisory committee, and CDC’s own burden tables going back eight seasons. What follows is almost entirely FDA quoting itself. The agency approved this product while writing down, in detail, the reasons to be careful about it.

Start with what wasn’t tested

There was no placebo group.

Study P304 randomized roughly 40,300 people to either MFLUSIVA or Fluarix — an ordinary standard-dose flu shot. Every efficacy number you will read about this vaccine is a comparison between two vaccines.

That design choice is (historically) defensible. Once an effective vaccine exists for a population, running a placebo arm raises real ethical problems, and regulators routinely accept active-comparator trials for exactly that reason. We’re not going to pretend otherwise.

But it has a consequence that gets lost in the press release: nobody has ever measured this shot against no shot at all. “26.6% more effective” means 26.6% better than the thing it was tested against. It is a margin, not a measure.

Number one: the margin is 0.73 percentage points

In the trial, 2.04% of the MFLUSIVA group caught PCR-confirmed influenza-like illness. 2.77% of the Fluarix group did. That gap — 411 cases versus 557 — is the entire finding.

In absolute terms it is 0.73 percentage points. Which means roughly 137 people have to switch from the ordinary flu shot to the mRNA shot to prevent one case of lab-confirmed flu-like illness over one season.

And note what happened in the comparator arm during a season CDC classified as high severity: 97.2% of people who got the ordinary, cheap, widely available flu shot did not get PCR-confirmed flu-like illness anyway.

The endpoint is also softer than most readers will assume. “Protocol-defined ILI” required one systemic symptom — which could be tiredness or a headache — plus one respiratory symptom such as a cough, plus a positive PCR test. It is not hospitalization. It is not death.

Number two: flu kills the old, and the old got the weaker approval

Influenza is overwhelmingly a disease that kills elderly people. We pulled CDC’s burden estimates for eight seasons. Adults 65 and older account for between 68% and 87% of all U.S. influenza deaths — every single season. FDA’s own briefing document puts it at roughly 75%.

That is precisely the group FDA did not approve on demonstrated efficacy.

Adults 50 to 64 got a traditional approval. Adults 65 and older got an accelerated approval based on antibody levels — hemagglutination inhibition titers and seroconversion rates — drawn from a different study, in a different season, using a different formulation than the product being licensed, bridged after the fact.

To be precise, because this is the sentence people will try to twist: seniors are approved. They are approved on a weaker evidentiary standard than everyone else.

FDA has required Moderna to run a confirmatory trial to find out whether the thing actually works better in that population. It begins August 31, 2026. The final report is due May 31, 2030. :/

For the group that supplies roughly three-quarters of America’s flu deaths, the answer arrives in four years.

Number three: FDA says the comparator was the wrong shot

This is the part that should end the “breakthrough” framing, and it is FDA’s language, not ours:

“For adults 65 yoa and older, high-dose (HD-IIV4), recombinant (RIV), and adjuvanted (aIIV4) influenza vaccines are preferentially recommended by the Advisory Committee on Immunization Practices (ACIP) over SD formulations. Accordingly, the SD comparator used in P304 is not the preferred standard of care in this age stratum.”

And:

the rVE “does not directly establish the magnitude of benefit relative to HD, RIV or adjuvanted vaccines — the vaccines most adults 65 yoa and older would otherwise receive.”

CDC tells people over 65 to get the high-dose shot. Moderna’s trial tested against the standard-dose shot.

Moderna beat the shot that seniors are told not to get.

FDA even cites ICH E10 — the international guideline holding that a control group should reflect actual standard of care — and then notes that it didn’t here.

Number four: the severe-outcome claim is twelve hospitalizations

You will see “47.9% reduction in serious outcomes” quoted. Here is what sits underneath it, across roughly 40,300 people:

Twenty-two participants versus forty-two sought a higher level of care. Six versus twelve went to an emergency room. Four versus eight were hospitalized.

Twelve hospitalizations, in total, in a trial of forty thousand people.

FDA is explicit that this is not a finding to lean on: relative efficacy “could not be calculated for hospitalizations and ER visits individually due to limited case counts,” and the study “was not powered to evaluate healthcare outcomes.”

No deaths in either arm were attributed to influenza.

Where the effect holds, and where it vanishes

Break the result apart and the picture gets thinner.

In participants with high-risk medical conditions — the people who actually need protection — rVE was 22.3%. In healthy participants it was 32.1%. FDA’s own reading: the direction of that difference “suggests that rVE in the healthier study population may overestimate effectiveness in the general U.S. target population.”

In adults 75 and older, the oldest group actually enrolled, the result was not statistically significant: 44 cases versus 59, rVE 25.3%, confidence interval running from −10.4% to 49.5%.

Against influenza B, also not significant: 25 cases versus 35, confidence interval −18.5% to 57.5%.

And there is a specific problem with resting the senior approval on antibody titers. From the approved label: for B/Victoria, the correlation between antibody levels and actual illness “was not statistically significant; therefore, effectiveness against this strain cannot be reliably inferred based on HAI titers.” The surrogate endpoint doesn’t hold for that strain — and the surrogate is the whole basis of the 65+ approval.

The trial studied healthier people than the ones flu kills

Immunocompromised patients: excluded. Moderately or severely frail participants: under 1%. Nursing home or assisted-living residence: not even recorded. Median age: 64. Only 11.6% were 75 or older.

Fifty-seven percent of trial participants had at least one high-risk condition, against an estimated 78–93% in the real-world target population.

FDA’s conclusion, again in its own words: these exclusions limit “direct applicability of the efficacy data to a substantial portion of the intended patient population,” and the gap “is significant because these populations face the highest absolute risk of severe influenza-related complications.”

The trade nobody puts in the press release

Grade 3 systemic reactions — the kind that stop your day — occurred in 5.5% of MFLUSIVA recipients versus 0.9% on the ordinary shot. Any systemic reaction: 58.0% versus 32.4%. Any local reaction: 67.5% versus 32.1%.

That excess works out to roughly one additional day-stopping reaction for every 22 people vaccinated, against one case of illness prevented for every 137.

Roughly six people knocked out of their day for every one case of flu-like illness avoided.

The fair counter, which belongs in print: those reactions had a median duration of about two days and resolved on their own. Influenza can be considerably worse than two bad days. But the trade is real, and it is not what “26.6% more effective” conveys to someone rolling up a sleeve.

What we are not claiming

Rational Ground’s brand is receipts, so here are the ones that cut the other way.

This was not rushed. FDA’s advisory committee voted 9–0 that the benefits outweigh the risks, for both age groups. FDA initially refused even to file the application. The review took eight months from submission to approval.

The death imbalance has an explanation. Across the pooled Phase 3 safety database of nearly 72,000 people, deaths of unspecified cause ran 23 in the MFLUSIVA arm versus 9 in comparators. That looks alarming, and FDA worked through it in detail: deaths within 28 days of vaccination were 3 versus 2, most occurred months later with no temporal clustering, and nearly all involved heavy pre-existing illness. FDA concluded none were vaccine-related. We think that assessment is reasonable. We would also note that no autopsies were performed on any of the mRNA-1010 deaths, so the conclusion rests on record review.

The safety database can’t answer the rare-event question either way. FDA states plainly that it “is insufficient to detect rare adverse events, including myocarditis, Guillain-Barré syndrome, and other neurological events of interest.”

And this may genuinely be a better flu shot. CDC’s own presentation to the same committee put existing vaccine effectiveness in adults 65+ at 31–37% against unvaccinated this past season — and noted that was “lower than recent previous influenza seasons.” Layer a 26.6% relative improvement on top and you land somewhere near 49–52%. That would be a real gain. It is also an arithmetic projection, not a measured result, because no trial has ever compared this product to no vaccine at all.

but… it’s still mRNA

Our concern with mRNA as a platform has never rested on the spike protein alone — it rests on what else is in the vial. Kevin McKernan, who ran R&D for the Human Genome Project's sequencing effort at MIT, found residual plasmid DNA in Pfizer and Moderna vials at levels far beyond regulatory limits, along with an SV40 promoter-enhancer sequence in the Pfizer product — an element used in gene therapy to drive DNA toward the nucleus, and one that regulators were never told about. That work has now cleared peer review.

Speicher, Rose and McKernan, writing in Autoimmunity in December 2025, tested 32 vials from 16 lots of documented Ontario provenance and measured total DNA at 36 to 153 times the regulatory limit for Pfizer and 112 to 627 times for Moderna. The most damning part is methodological. Regulators certify these products using qPCR, which counts only short, specific, amplifiable fragments — and by qPCR every Moderna vial passed. By fluorometry, which measures all the DNA actually present, not one did. The limit wasn't breached in secret; it was measured with an instrument that couldn't see it. What that DNA does inside a human being remains genuinely unresolved, and we won't pretend otherwise. But "we never looked properly" is not the same finding as "there's nothing there" — and after four years, the burden belongs to the people who picked the assay.

So, for myself and my family — we’ll pass on this shot, thanks!

Every figure above is drawn from FDA’s approval letter (STN BL 125869/0, August 5, 2026), the approved MFLUSIVA prescribing information, FDA’s briefing document to the Vaccines and Related Biological Products Advisory Committee for its June 18, 2026 meeting, Moderna’s briefing document, CDC’s presentation to that same committee, and CDC’s published influenza burden estimates by age group. One gap worth disclosing: CDC’s by-age burden table for the 2024-25 season — the trial’s own season — was not retrievable; CDC’s live site blocks automated access and the archived snapshot predates the season’s end. The eight complete seasons shown make the point without it.

Matt Walsh, on August 5, to an audience of more than a million: the rise in grocery prices over the past six years is a mathematical reality, it hasn’t stopped or reversed, and conservatives had better come up with an answer that isn’t “learn to make cheap burritos” or they will get crushed. He also made a sharper political point — that almost no Republican was arguing the affordability crisis was fake during the Biden years, which makes the sudden discovery of statistical nuance look exactly like what it is.

Jeremy Kauffman, a bit later: as far as he can tell it simply isn’t true that food has gotten more expensive in real dollars in the United States. Why is everyone so convinced it has? Where’s the evidence? He posted a chart to back it up — food-at-home CPI divided by overall CPI, running back to 1985, wobbling in a narrow band the whole way. From 1985 to 2025, grocery prices rose 200.4% while all prices rose 199.2%. Groceries, in real terms, went up four-tenths of one percent in forty years.

Both men are arguing in good faith. Both have real data. And the reason this argument never resolves is that they are answering two different questions and both getting the right answer.

Here’s the referee.

First, the thing nobody disputes

A basket of goods that cost $100 in January 2019 cost $103.90 the day Joe Biden was inaugurated. It cost $126.20 the day Donald Trump returned. As of June 2026, it costs $132.70.

That line essentially never goes down. June 2026 delivered the first monthly decline since April 2020 — down 0.4%, driven almost entirely by a 9.7% one-month collapse in gasoline after the spring’s Iran-war oil spike faded. One month, after six years.

This is the part people mean when they say prices went up and never came back. They’re right. Disinflation is not deflation. When a 9.1% inflation rate falls to 3%, prices are still rising — just more slowly, from a level that never returned.

Kauffman’s ground, and it’s solid

The strongest possible version of the “it’s flat” argument isn’t a price index at all. It’s work-time: how many minutes does the average private-sector worker have to labor to buy a thing? No index, no deflator, no chained-dollar methodology to argue about. Price divided by hourly wage.

Run every item that way from January 2019 to June 2026 and groceries as a whole cost 2.3% less work time than they did before COVID. Bacon is down 13%. Chicken breast is exactly flat. Eggs — after the entire avian-flu round trip — are up one percent.

Kauffman is right. In aggregate, measured the cleanest way available, the American grocery basket has roughly kept pace with the American paycheck. Zoom out to his forty-year window and the case is stronger still: by USDA’s accounting, food’s share of household income fell for half a century.

Anyone claiming groceries have become categorically unaffordable relative to wages is, on the aggregate numbers, wrong.

Walsh’s ground, and it’s also solid

Now look at the same chart again, at the top instead of the bottom.

Coffee costs 64% more work time. Gasoline costs 34% more — 5.0 minutes a gallon in 2019, 6.7 minutes now. Ground beef costs 32% more work time: 8.3 minutes a pound then, 10.9 minutes now. Milk is up 9%, electricity 8%, bread 4%.

That’s the same dataset producing the opposite headline, and neither number is wrong. The average is flat because bacon and chicken and eggs pull down while beef and gas and coffee pull up.

Nobody buys the average. Households buy specific things, notice specific prices, and anchor hardest on the items they purchase most often and see most clearly — the gallon of gas on the illuminated sign, the pound of ground beef, the morning coffee. Those are precisely the items that got worse.

And there was a genuine squeeze, not a perceived one. Food-at-home prices have risen 33.2% since January 2019 against wages up 36.4% — but the path matters more than the endpoint. The food line ran above the wage line from 2021 into 2023. For roughly two years, groceries outran paychecks. Measured from January 2021, food still costs about 1.4% more work time today than it did then.

USDA’s own numbers say Americans spent 11.3% of household income on food in 2022 — the highest share since 1991. That is not a vibe. That is the statistic Kauffman’s forty-year trend line smooths away.

The whole argument is a fight about start dates

Ask whether wages are keeping up with prices and the honest answer is: compared to when?

• Against January 2019, the average hour of work buys 3.6% more.

• Against January 2021, it buys 0.7% less.

• Against January 2025, it buys 0.7% more.

Three defensible answers, three different starting lines, one dataset. Pick 2019 and you get Kauffman’s story. Pick 2021 and you get Walsh’s. Neither is cherry-picking in bad faith; they’re just standing in different places.

Kauffman’s own chart concedes this if you read the table under it. Over 1985–2025 real grocery prices rose 0.4%. But over February 2020 to June 2026 they rose 2.1%, and over June 2021 to June 2026, 1.7%. His long trend is real and his recent numbers point the other way, which is precisely the shape of the disagreement.

The full tab

Since January 2025: coffee +35% — a record $9.72 a pound in April 2026, driven by Brazilian and Vietnamese droughts plus tariffs. Gasoline +31%, from $3.21 to $4.20, with a $4.65 peak in May 2026 during the Iran conflict. Ground beef +23% to a record $6.90 a pound on the tightest cattle supply in decades. Electricity +11%, a record 19.8 cents per kilowatt-hour. Milk +7%. Chicken +5%.

Down: eggs −57%, from $4.95 to $2.14.

Be careful with that last one, because it’s the number you’ll hear most. It’s real — and it’s an avian-flu round trip. Eggs spiked to $6.23 in March 2025 and came back down. Measured against 2019, eggs are still up 38%.

Which is the pattern for everything. Against 2019: gas +83%, beef +80%, milk +48%, electricity +47%, bread +42%, eggs +38%, chicken +37%, bacon +19%.

Nothing on that list is back to 2019 prices. Not one item.

The scoreboard, presented straight

Inauguration month to inauguration month:

• Trump’s first term: +7.7% over four years — 1.9% a year.

• Biden: +21.4% over four years — 5.0% a year.

• Trump’s second term so far: +5.1% over seventeen months — about 3.6% a year annualized.

Better than Biden. Worse than Trump I. That’s the receipt, and we’re publishing it in that order because the alternative is doing exactly what Walsh warned about.

The obligatory caveat, stated once: presidents don’t set prices. Avian flu set the egg price. Cattle herd cycles set the beef price. Brazilian weather set the coffee price. A war near the Strait of Hormuz set the May gasoline price. The Federal Reserve matters more than any of them. Attributing a CPI print to whoever occupies the White House is bad economics regardless of who benefits from it that month.

But the scoreboard is the scoreboard, and voters are going to keep it whether or not economists approve.

The verdict

The honest statement is narrower than either side’s tweet, and here it is:

Aggregate food has roughly kept pace with wages across the full 2019–2026 window. Specific staples have not. The 2021–23 squeeze was real. And the recovery since is recent, uneven, and completely invisible on the price tag.

That last clause is the one that decides the political argument.

Even where wages have caught up, the nominal price is permanent. Ground beef that cost $3.80 costs $6.83, and it will cost $6.83 forever, no matter what happens to your paycheck. The wage recovery arrives as a number on a pay stub you see twice a month. The price increase arrives every single time you stand at a register. One of those is salient and one of them isn’t, and no amount of correct math changes which is which.

So Kauffman’s question — where’s the evidence? — has an answer: it’s in chart six, at the top of the list, in the items people actually watch. And Walsh’s warning has an answer too: the aggregate data really does show groceries roughly tracking wages, and pretending otherwise is its own kind of dishonesty.

But Walsh is right about the politics in a way that transcends the statistics. Telling people that their lived experience is a measurement artifact is a losing argument even when the measurement is correct. It was a losing argument when the Biden administration made it in 2022 with “Bidenomics,” and it will be a losing argument for anyone who makes it in 2026. The winning move is to say the true thing: some prices genuinely outran you, the squeeze was real, and here is what we intend to do about the ones still climbing.

Sources: BLS Consumer Price Index (CPI-U, 12-month change, not seasonally adjusted) and BLS average price data for all shelf prices; average hourly earnings of private employees; USDA ERS food expenditure series; FRED series including CUUR0000SAF11 for food at home. Data retrieved August 4–5, 2026, running through June 2026. Coffee work-time is measured from January 2021 because BLS did not publish an average price for it in January 2019. October 2025 CPI was never published because of the government shutdown; our charts show the gap rather than interpolating across it.

Every number above is explorable. Hover any point, open any “view data” table, and check the months yourself — the full interactive price page is here →

On Wednesday, July 29, Anthony Fauci sat before the Senate Homeland Security and Governmental Affairs Committee — subpoenaed, sworn, and silent. Within minutes of the gavel he read a statement accusing Chairman Rand Paul of running a prosecution in search of a crime, and then delivered the line he would repeat, by reporters’ count, more than four dozen times in the first two hours:

“On the advice of counsel, I respectfully decline to answer based upon my rights under the Fifth Amendment.”

Fauci has testified before Congress more than 250 times across nearly four decades. This was the first time he took the Fifth.

Two weeks earlier, on July 15, Jay Bhattacharya sat down with Marissa Streit for an hour on PragerU’s Real Talk. No subpoena. No counsel. No carve-outs. He is the sitting Director of the National Institutes of Health and Acting Director of the CDC — the most legally exposed chair in American public health — and he answered every question put to him, including the ones with his own agencies’ fingerprints on them.

Put the two transcripts side by side and you have the whole story of the last six years in one frame.

What silence sounds like

Let’s be precise about what Fauci declined to discuss, because the list is the indictment.

He would not answer about the origins of SARS-CoV-2. He would not answer about gain-of-function research. He would not answer whether he misled Congress about either. And — this is the part worth sitting with — he would not answer a procedural question about documents.

Other exchanges were more personal and frustrating but damning to Dr. Fauci.

Paul’s predicate for the subpoena was two emails. On February 2, 2020 — one day after the now-infamous call that produced “Proximal Origins” — Fauci wrote to then-NIH Director Francis Collins about getting ahead of “the science and narrative” of the virus’s origins, and closed with: “Please delete this e-mail after you read it.” On July 20, 2020, he sent his chief of staff a message disparaging Paul, with the same instruction attached.

In 2024, under oath, Fauci denied deleting records or obstructing anything.

Paul’s response to the Biden pardon, which blankets Fauci from 2014 through January 2025, was the sharpest thing said in the room: a pardon “does not convert a misleading statement to a truthful one.” He closed with a promise — “There will be repercussions to your refusal to testify today.”

Set aside whether you think a preemptive pardon was justified. Notice the structure it created. The pardon removed the legal jeopardy for everything already done. The Fifth Amendment now removes the obligation to describe it. Between the two, the entire record of how American pandemic policy was made has been walled off from the only body empowered to examine it.

That is not a legal technicality. That is an accountability vacuum with a constitutional roof over it.

What openness sounds like

Now the other transcript.

Bhattacharya opens the PragerU interview by describing 2020 as the year his career stopped mattering to him — he watched children locked out of school, he opposed the mandates, he got canceled for it. “I thought my career was over,” he says. “And now I’m the director of the National Institutes of Health.”

He was blacklisted on Twitter at the government’s request. He co-authored the Great Barrington Declaration, the document Francis Collins famously demanded a “quick and devastating published takedown” of on October 8, 2020 — a takedown organized from inside the very building Bhattacharya now runs.

And here is what he says, in public, unprompted, about the institution he inherited:

“Most scientists and people in public health — if they’re honest — recognize the colossal failures that happened of science and public health during that era.”

He names the damage without hedging: kids out of school for two and three years and still behind, the economic hangover, and — a phrase no NIH director has ever uttered — “all the people that suffered from vaccine injuries that are unacknowledged.”

On the mandate-era claim that started the whole censorship apparatus, he is blunter than any of his critics predicted he would dare be as a sitting director:

“Does the COVID vaccine stop you from getting and spreading COVID? I can see and you can see that it doesn’t — but it’s a peer-reviewed journal that says that it does.”

Asked directly about reports of a 235-word screening list at NIH, he doesn’t spin it, doesn’t refer it to comms, doesn’t say the report is under review:

“I am deeply against banned word lists. I don’t believe in them. I will not have my institution embrace them.”

He explains that grant-screening tools use keywords and that a human reads what they flag — and that in many cases the reviewer discovers the DEI language was camouflage, tacked on by researchers who had learned that “in order to get a grant, you have to pretend to be doing DEI and then you’d secretly be doing good science underneath.”

Read that twice. The Director of the NIH just described his own agency’s recent grant regime as a system that taught American scientists to lie in order to work.

The reversal is structural, not rhetorical

Anyone can hold a press conference about restoring trust. What separates this from the usual Washington penance tour is that Bhattacharya is dismantling the specific machinery that produced the failure.

The replication crisis. He puts the number on the table: “60% of the time, 70% of the time, the published biomedical research literature is not reliable.” Methods aren’t described clearly, datasets aren’t available, tissue samples aren’t available. So NIH is now funding replication work as first-class science, standing up an NIH journal that publishes replication studies with the peer reviews attached in public, rebuilding literature search so a paper displays every study that tested the same idea instead of the prestige of its masthead, and building metrics that reward sharing data and code.

The safety trap in peer review. He and a Stanford colleague indexed every new word and phrase in the biomedical literature since 1940 to date how old the ideas in a given paper are. The finding: in the 1980s, NIH-funded researchers published ideas zero to two years old. By the 2000s and 2010s, seven to eight years old. The agency had quietly optimized for will this work over will this disrupt — partly because you had to already hold a large NIH grant to sit on a study section, which means the incumbent reviews the insurgent. The fix, which he concedes has a boring name, is the uniform funding strategy: institute directors are now pushed to read the full review rather than rank by a composite score that mostly predicts publishability. “I fundamentally don’t care if a project gets a paper in a journal,” he says. “I care if the project disrupts science.”

Epistemology. This is the deepest cut, and it’s the one that indicts the Fauci-Collins era most directly without ever naming it. Bhattacharya reaches for Galileo — the first scientific revolution wasn’t about telescopes, it was about stripping ecclesiastical authorities of the power to rule on physical reality. Then:

“Why is it that a relatively small number of people, in a very non-transparent way, get to decide for all of society what’s true or false about physical reality?”

That is a description of 2020 and 2021 written as a philosophy lecture. The Proximal Origins paper was an ecclesiastical ruling. The GBD takedown was an ecclesiastical ruling. The blacklist was enforcement.

Mandates. On childhood vaccines he refuses both scripts. He defends MMR and DPT as essential and says so plainly. He also says every medicine has a benefit and a harm, that we owe something to the people harmed, that the COVID shot isn’t necessary for most kids, and that the schedule has been shifted toward a European model with Secretary Kennedy. Then the sentence that is the entire argument of the last six years compressed into fourteen words:

“It’s not going to be a universal take-it-or-else. You have no trust.”

The same chair, six years apart

Here is the juxtaposition, stated as plainly as it deserves.

The men who built the policy will not describe it — one is pardoned and silent, the other authored the takedown email and has never sat for a hearing like Wednesday’s at all. The man they tried to erase now holds the chair, and cannot stop describing it: the failures, the injuries, the censorship, the rot in peer review, the flatlined life expectancy, his own agency’s capture.

One era is protected by counsel. The other is conducted on camera, for an hour, with the tape running.

Bhattacharya was asked what success looks like in twenty years. Not a Nobel, he said. Not a citation count. Reverse the longevity flatline. People living longer, healthier, more fulfilling lives. “That’s the measure.”

It is a modest answer from a man who was told his career was over for saying children should be in school.

Fauci had four dozen chances Wednesday to say something equally modest. He took the Fifth instead.

Quotes from Dr. Bhattacharya are drawn from “Dr. Jay Bhattacharya on Depoliticizing the NIH and Ushering in a New Era of Scientific Innovation,” Real Talk with Marissa Streit, PragerU, July 15, 2026, lightly cleaned of verbal stumbles. Hearing quotes and details are from contemporaneous reporting on the July 29, 2026 Senate Homeland Security and Governmental Affairs Committee hearing (STAT, TIME, Boston Globe, Axios) and from Chairman Paul’s own committee release on the February 2 and July 20, 2020 emails.

JUST. TURN. AROUND. FAUCI!

(he had to consult his lawyers)

Excerpts from Prologue & Chapter 2

“Lest Satan should get an advantage of us: for we are not ignorant of his devices.”

- 2 Corinthians 2:11 (KJV)

Hell keeps minutes.

This will surprise those who imagine the place as fire and shrieking, all fiercely carnivirous, all chaos, no administration. The letters in this book should cure that misconseption. Whatever else those years were, and they were many things, including a real plague with real dead, they were also, below, a project: scoped, staffed, minuted, and reviewed. Great evil is seldom loud. It is procedural. It files away everything, and you never noticed.

How this file surfaced I will not fully explain, partly to protect what must still be protected, mostly because I cannot for fear of retribution. I can say that the letters arrived the way everything true about those years arrived: late, out of order, in pieces, and through the very procedures that had been built to bury such things. Call it a leak if you like. Call it, as the lawyers say, production assets. Somewhere in the middle of the correspondence, the senior devil assures his nephew that his own records are kept in the one archive that has never seen the light of day. You are holding the reply. This are the pantheon fruits of our FOIA request.

The file is incomplete, these type of files always are, and where the gaps fall, I have left them; the silences of devils are instructive too. The books, we are promised, get opened in the end. Regard this as an early draft, a forgotten manuscript.

What you hold, then, is the working correspondence of two devils. The elder, Screwtape, has seen plagues enough to find them boring. The younger, talented, careerist, anxious for commendation, signs himself Dr. F., Undersecretary for Respiratory Affairs. Readers who come hunting a particular face behind the initial will leave unsatisfied, and should. Hell assigns specific desks with like-minded biographies. The Undersecretary is a post, a titular honorarium and these denoted pedastals outlive their occupants. Or more likely, their occupants have stepped off.

Notice, as you read, what the two of them almost never discuss: the virus. It drifts through these pages as weather, a thing that happened to be happening. Their professional interest was never the disease, but what fear it could instill. Their focus is always how the virus could be spun into action, what the virus could be taught to do, to love, to truth, to worship, to children, to memory, to the face across the table.

Their patient was never the sick man. It was you.

Two notes on method, because the book asks something of its reader.

First: devils on the record praise what should appall them and dread what should be cherished, so everything here arrives upside down and must be read in a mirror. When the correspondents gloat, someone was being broken. When they grow uneasy and sense the event horizons: quiet houses, mothers doing arithmetic, men who frame the letters that ended their careers, any two witnesses comparing notes, you have found the exact coordinates where the light was getting in. Their contempt is a map. I know no better index of what actually mattered in those years than the short list of things these two feared.

Second: do not cite these letters as history. Devils are indifferent to facts and say so with pride; the figures they spout to one another prove nothing and were never meant to. But a devil reporting to a devil about the work cannot afford to lie about the work; Hell audits, and so the correspondence is unreliable about every number and unimpeachable about every untoward intention. Cite nothing. Recognize everything.

And you will recognize it. That is the last thing to say, and the reason the book exists. The famous rooms are mostly offstage here. The files return, instead, to kitchens, thresholds, classrooms, group chats, parish halls, nursing-home windows, the seating of dinner tables, because that, by the devils’ own accounting, is where the era was actually decided, and theirs is the one accounting in this matter I have come to trust. Few readers will get far without meeting a letter they wrote, or received, or chose, gently, with sorrowful wording, not to send. I met several of mine. I finished anyway, and I mention it only because the devils, late in the file, name the last weapon they keep for the day when all the others have been blunted: the tiredness of readers.

So read it rested. Take it slowly; it was done to you slowly. And when you have finished, do the one thing this correspondence fears from its first page to its last: say what you saw, plainly, out loud, to someone near.

It was always the nearness they were after.

Chapter 2 - The Mask: Symbol of Submission

“False face must hide what the false heart doth know.”

- William Shakespeare, Macbeth, Act I, Scene VII

Letter VI - Dr. F. to Screwtape

My Dear Uncle,

A boy in my sector is learning to talk this year. Unfortunately for him, the human mouths he would mimic are covered, his teacher’s, the grocer’s, the better part of his mother’s day, and so he is doing his best with eyebrows. I open with this small impact as a touchpoint of the season’s inventory, and the season, Uncle, is this: the free citizens of the West have covered their faces. I ask you to sit with that sentence a moment, as I have. No conqueror ever managed it. Mohammed came close but we triumphed in covering the male counterparts as well! You said the mask deserved a letter of its own; it has since earned a volume.

Their litany to date has been: a frightened spring, a bulbous chart, a quiet fortnight renewing itself, the cadence your last letter described, and then this: the faces of the West withdrawn from circulation, like a currency recalled.

Understand what we have and have not issued. A temporary measure is simply a tool with a date on it. This has no date, but it has acquired a meaning: it is a confession of virtue, worn in public, the first garment in living memory that announces the wearer’s morality at fifty yards. Lockdown was a intangible policy of shut doors; a man could pace his floor for weeks and resent it privately. The mask is a tangible policy of the person. It rides the breath, tugs at the ears, dampens against the lip. It is a policy they can feel against their skin, and that discomfort is now rehearsed hourly.

First, I must also report a complication which I believe we have successfully converted into an asset. Our opening messages, you will recall, discouraged the flimsy cloth, not needed by the well, reserved only for the wards. Then the doctrine reversed, and the article became sacred. I braced for damage; hypocrisy is the one charge the laity can prosecute without counsel. The damage did not come. The reversal was absorbed, retailed to the public as the science evolving. Absorbed, Uncle, is the word I adore here, because absorption is the product. To obey a stable rule, a man needs only obedience. To obey a rule that has reversed, he must surrender his memory of the rule before, must agree, in some back room of his head, that what was said was never quite said. Practice this twice and the faculty that compares yesterday with today learns to ask for instructions before it speaks tomorrow. You will remember my anxieties regarding comparison. We are now manufacturing the antidote, and the public wears it on its face.

The technical disputes, weaves, droplets, inches, aerosolization (especially that!), continue, and I leave them undisturbed, for the cloth’s office was never chiefly respiratory. The face was always the problem, Uncle. The face is where they are specific. A crowd can be managed; a face insists on being someone. And the mouth carries half their grammar: the smile, the frown, the half-laugh, the tremble before tears, embarrassment, tenderness. Cover it and you do not merely filter the breath. You abridge the person. To cover the face is to efface the self, and we have arranged for the effacement to be self-administered, daily, with elastic.

The yields are everywhere. An old man at the pharmacy asks the clerk to repeat herself, twice, then pretends he has heard and returns home without his task fulfilled; he had been reading lips for years without knowing it, and goes home having purchased silence, which he calls not making a fuss. A widow at a rationed funeral receives condolences from a row of anxious foreheads. You can still smile with your eyes, they keep telling one another. Eyes do something, Uncle. From a stranger at six feet, it reads mostly as watching. I’m told a man’s untoward imagination is heightened because he cannot verify attractiveness in the face, so his gaze wanders elsewhere and his shame aggregates hourly for the lust-adjacent moment we have manufactured. After all, even the leper has a size 1 waist.

And the sorting, instantaneous, universal, unpaid. Covered: safe, obedient, kind. Bare: reckless, selfish, suspect. No warrant, no public tribunal; a man’s character is now adjudicated at the threshold, by glance, in under a second. We have never fielded a faster court decision. Fear, meanwhile, has entered the inventory of the pocket: wallet, keys, phone... mask, four mnemonic beats, performed without thought. The first three items are designed for a free world. The fourth is against it. The hand now remembers fear on the way out of the house.

Enforcement costs us nothing. A woman in a bakery lowered her mask to make herself understood to the man I mentioned above, and was corrected by a stranger in the queue before any official could have been summoned. The stranger held no office and required none; the office has been decentralized with innate instructions. Whole populations now serve as unpaid minor officials, and they enforce this mechanism with an energy no salary could buy, because they are paid in standing, its a wonder anyone sleeps at night with virtue signals beaming brightly from every pair of eyes peaking above the dutiful masked compliant population. A further yield I had not projected: the anxious wear it with relief. For some of them the face was always an exposure, being seen the heaviest tax of their social day. We have issued them a sanctioned veil and named the hiding responsibility. They are not merely compliant, Uncle. They are grateful.

And I have found a household, a particular couple, and I mean to measure the season by their actions; I am keeping their file on my own desk. A husband and a wife, comfortable, a good house on a quiet street, one child, two cars. I have entered them in the registry as the Volunteers, because that is the whole of their art: no one ever has to ask them. The new constabulary are paid in standing, as I reported, and standing is a wage; withhold it, and they resign. These two enforce our mandates unbidden, at their own expense, and thank us for the opportunity.

Mark what we were given to work with, Uncle, because it was nothing but virtue. He is a protector by birth, he stands between his family and the world and is proud to stand there. She is conscientious to the point of restlessness; she cannot let a rule lie unobserved or a targeted task go undone. In a duller decade the two of them would have died quietly, somewhat admired but mostly unremarkable. The era found me the seam in each to make them powerful, if only for this season. His penchant to protect I turned a few degrees outward, until it became a browbeating patrol. Her conscience I turned the same few degrees, until it became a busybody’s sanctioned watch. By June, he was stopping strangers on the footpath to account for their nearness; it is the cloth that licenses the stopping, Uncle; a bare face at forty yards has become a thing one citizen may take up with another, like smoke seen over a neighbor’s fence. By September she had learned the number one rings to report a gathering, and had rung it, over a christening, the returned home and slept the sleep of the just. They have not felt themselves change. They tell each other, nightly, across the kitchen table, that they have simply begun, at last, to care.

For the archive, I commend the doorway of the restaurant, where the liturgy is at its most refined. The patron masks to cross the threshold, walks ten feet behind a hostess, sits, and unmasks for ninety minutes of talk and chewing among strangers. Should he rise, the cloth quickly returns for the twelve yards to the lavatory and the twelve yards back. The virus, in this cosmology, is a creature of exquisite manners: ferocious in vestibules and corridors, harmless wherever food is served; it hunts the standing and respects the seated. No one believes this. That is the achievement. They perform it anyway, for the hostess, for the room, for their own membership, and every performance teaches the muscle that notices absurdity to keep its findings to itself. A man who will costume himself for the walk to the lavatory has accepted the essential terms: his perception is negotiable; his standing is not. The fogged spectacles alone repay this effort. I keep a small file of these moments just to laugh.

You will ask for outcome figures. Of course, I decline to keep any. A tool is judged worthy by outcomes; a ritual is deemed worthy by meaning alone; and I have moved the visage article from the first category to the second. The formula of consecration is two words: for others. I did not coin it; I only priced it. It converts fear into generosity at scale. The frightened man is no longer hiding, he is giving. Conformity is rebranded compassion; suspicion, prudence; and fear’s old appetite for company becomes a work of mercy. We have made the mask mean love, Uncle, and a man who will quarrel all day with a regulation will, frankly, avoid a quarrel with love.

Now the anxieties, filed per protocol. The cloth sits unevenly. The clerk wears it ten hours; the man she serves, ten minutes. The young talking boy I mentioned looks up at his teacher and finds only eyes above fabric, and has begun reading foreheads for temperment. The old, as noted, are missing more than sound. And in my surveillance of parents I record the first appearances of a word that concerns me. Not uncomfortable. Not inconvenient. Indignity. A tired mother, watching her son mask for the third recess of the day, has stopped asking whether it is comfortable and begun asking whether it is small sacrifice or not. Discomfort, Uncle, is a sensation, and sensations fatigue. Indignity is a claim, and claims can and do organize.

My question for doctrine: how do I keep a billion small losses filed under inconvenience?

Your devoted nephew,

Dr. F.

Undersecretary for Respiratory Affairs

P.S. They have begun manufacturing the article in patterns, florals, team colors, seasonal prints. I confess the holly print disquieted even me. An emergency with a product line is an emergency that has decided to stay.

Letter VII - Screwtape to Dr. F.

My Dear Dr. F.,

Your report gave me an entire afternoon of satisfaction. However, I warn both of us: distrust that sensation. Satisfaction is the posture in which our kind is most often found asleep at the rail. You have built something genuinely fine, and you have understood, by my count, rather less than half of it.

First: you were right to leave the technical quarrels running, and you must never let them settle. Every hour the humans spend disputing weaves and droplets is an hour spent agreeing on the premise underneath both positions, that the question is material. It is not. If the cloth stopped everything, or nothing, our position would be identical, provided the article continues to mean what you have made it mean. The deeper work was never filtration. It is formation.

Second, your question, how to keep the losses filed under inconvenience, has most of its answer already in your inventory; you have only to see what you are holding. The one fatal misstep would be to let the article be experienced as oppression. Oppression is a known stimulant. It has summoned more courage out of that species than we have ever managed to put back, and martyrs have been cut from cheaper cloth than this. No. The mask must remain what you have made it: a token of care. Which means the language must stay tender, all the way down. A small burden, reluctantly asked. For the vulnerable. Children are resilient. You can still smile with your eyes. Keep the entire vocabulary in the nursery register, and file every spiritual loss, recognition, expression, the freedom of one face toward another, under inconvenience. The filing is the whole mechanism, nephew: it makes the loss unspeakable by making it minor. A man may rise against an indignity; no man rises against an inconvenience without becoming, publicly, the smallest person in the room. Your tired mother is not dangerous so long as she cannot say the word indignity aloud without hearing how it sounds next to it’s just a piece of cloth. And the citizens she might have recruited will go on enforcing tenderly, for cruelty is most durable when it is experienced by the cruel as compassion.

Your Volunteers, next. Please nephew, strike the label before it sets. It flatters you, not them, and a flattered tempter mislabels his own assets. These two are not your discovery, nephew; they are the oldest stock in the house. Every plague century mints the identical pair, and I keep a drawer of them going back to the lazarettos: the Matron, who holds the street under observation from behind her curtain and keeps the record of who called on whom, and the Marshal, who plants himself at the gate and rules on who may pass. The faces are reissued each century; the office is eternal. In Milan they carried the health board’s staff and hunted the untori I told you of. Yours carry an application on a telephone and an inexhaustible supply of conviction.

Here is the instruction that will keep your pride from spoiling them. They are not being made cruel, cruelty is the amateur’s aim, and it spoils, for a cruel man may weary of himself or be shamed out of it. They are being made righteous, which keeps. The righteous cannot be shamed; to them shame is merely the wicked resisting the good. Your Marshal takes no pleasure in stopping the stranger, he believes he is saving him, and the man’s resentment only proves the peril. Your Matron did not relish her call; she wept over it, and the weeping is the bolt thrown home. A woman who has suffered to inform on her neighbor will defend to her grave the necessity that made her suffer, for if it was not necessary, what is she? Keep them in that grief, nephew. It is the most commodious cell we have ever furnished, and the door locks only from the inside.

Now the face, since you have at last begun to see it. The Enemy committed an extravagance there which we have never forgiven. No two faces issued alike, and then the claim that He knows each one by heart. The face is where the species defeats our redux. A man can hate a class for thirty years without missing a meal; show him a single face from inside that class, at the wrong moment, in the wrong light, and the whole account comes under audit. Shame keeps its office in the face. So does courage. Mercy is transacted there almost exclusively, the face is where one soul discovers that another is looking back. Your bureaucrats survive by never being looked back at; an order covering a million children is permitted only because no single obeying child conveys their confusion and marked sadness, except through the mask . So mark what your cloth has actually done. It has not blinded anyone. It has made the street a procession of categories, covered, bare; safe, suspect, where there was a procession of persons. A category cannot be absorbed at sight. That is not a side effect. That is the work. And it extends our oldest inversion to point-blank range: a man may now stand a yard from his neighbor and still withhold his face. Absence, worn. See that it goes on meaning devotion.

Your next assignment is saturation. Move the article from emergency to scenery. I want it in the advertisements, on the photographs, where it filters nothing, which is precisely its value there, for meaning is the only work an image can do. On the mannequins. On the school posters with their round, smiling letters. On the clergy at their livestreams; on celebrities at their ceremonies; in the children’s cartoons; in the airline videos with their serene, half-erased crews. Patterns, team colors, your holly print. The goal is not that the mask be required. Requirements can be repealed. The goal is that its absence come to feel indecent, the bare face in public reading as a kind of undress. Law compels the body, nephew. Indecency instructs the soul.

As for the contradictions your restaurant liturgy displays so handsomely: stop apologizing for them in your filings. You still account absurdity a cost. Properly managed, absurdity is not a weakness in the regime; it is one of its tests. Any rule that makes sense can be obeyed in mere agreement. The doorway ritual asks for something better: that a man override the report of his own eyes in favor of the meaning assigned by authority, and then, this is the kernel, that he not say what he noticed. Each performance is a small referendum on which shall be master, his perception or his standing, and we want the question rehearsed daily until it no longer feels like a question.

Which brings me to the man you did not mention, and he is the one I think about: the quiet one who has noticed. Not the shouters, the shouters manage themselves, and a fine or two will keep them shrill and unpersuasive. I mean the decent man at the table, doing silent arithmetic about the corridor and the chair. Your entire task with him is the administration of his loneliness. He must never learn how many of him there are. Keep him unsure whether his observations are intelligent or merely unkind; that is the hinge, for a decent man will suppress a true thought indefinitely if he suspects the thought is cruel. Give him nothing so vivifying as persecution. Certainly, make your preparations for censorship, that will come soon enough. But be cautious. He will eventually give pause at the office in a candid moment of truth. He will indeed notice the dinner invitation that stops coming. Fines make martyrs. Pauses make members.

The children require their own paragraph, and here your reporting tends sentimental in the wrong direction: you cite their adaptability as a success. Of course they adapt. Children adapt to anything; that was never in question and was never the prize. Adaptation is not evidence that nothing was bent. Adaptation is the bending. The sapling tied to the stake does not complain either. I read mention another report of a boy who chose the dinosaur print and felt lucky, let him feel lucky; his mother, watching him feel lucky, is the one who goes quiet, and her you will file under inconvenience as instructed. What the boy is learning underneath the print is the catechism we could never have taught openly: that the other person is, first, a risk. Approach is the exception; distance, the rule; the stranger, a hazard until cleared. A child who learns strangers as hazards before he learns them as faces will furnish us with business for decades. On the harm itself: publish nothing, claim nothing, and forbid the eager among your staff to gloat in trade journals. Let it arrive late and diffuse, an anecdote here, a specialist’s worried report there, a grief with no date on it. Harm that arrives all at once produces inquiries. Harm that arrives gradually produces essays.

And now lift your eyes from the cloth, because the cloth was never the objective. Today it is fabric over the mouth. Tomorrow it will be a credential in the pocket; after that, some other token, the form will change and the changing itself is useful, since every new token re-runs the training from the beginning. What we are installing is the reflex beneath all tokens: I prove my goodness by displaying public obedience. Say that sentence slowly until you hear everything inside it. A creature that proves its goodness by display has moved its conscience out of doors, relocated it to the audience, and a conscience approval kept by the audience is an approval we can edit. The artifact is not the achievement. The reflex, the reaction, the retribution for disobedience is the achievement.

Two cautions in closing. Do not press humiliation openly. We are not after conquest, which is loud and breeds resistance; we are after participation, which is quiet and breeds itself. The ideal subject is not the man ordered into the mask but the man who feels the stranger approaching on the path and reaches up, unbidden, to draw the cloth over his nose, a small bow performed for no audience but his own conscience, which now reports to us. The safest mask is the one adjusted by the wearer before anyone asks. Once the gesture is self-administered, the enforcement has moved indoors, where no repeal can reach it.

And when the cloth finally comes off, seasons end, nephew; I have outlived thousands, do not mourn the prop. If you have worked properly, the loss will by then be stored where grief cannot take inventory. The man will not miss the open faces of strangers; he will have stopped expecting them. The child will not mourn what the street withheld from him; he never learned its name. Your talking boy will carry the era at the edges of his sentences for sixty years, and no listener will ever learn what she is hearing. She will be hearing the cloth, nephew: a language keeps what a childhood pays. What cannot be imagined cannot be mourned, and what cannot be mourned cannot be reclaimed. So let the masks come off on schedule, with ceremony if they like. Your assignment is the face beneath: see that it emerges trained, guarded at the mouth, careful at the eyes, a face that has learned to behave like a mask.

Your affectionate uncle,

Screwtape

Everyone remembers the 2020 argument about an “October surprise” vaccine — the worry that Donald Trump would force an emergency authorization out of the FDA in time to help him win. What we didn’t have, until now, was Fauci’s own contemporaneous account of what he did about it.

His diary has it. Across five months, he tracks the pressure campaign, names the people applying it, and records the moment he picked up the phone and told the FDA official who would have to sign the authorization that he had his back if he refused.

Who’s who in these entries

Peter Marks -Director of the FDA center that reviews vaccines. The authorization would go through him.

Stephen Hahn - FDA Commissioner, Marks’s boss. Fauci repeatedly worries he will “cave.”

Mark Meadows - White House Chief of Staff.

Alex Azar - Secretary of Health and Human Services.

Larry Horowitz - Works closely with House Speaker Nancy Pelosi. He is Fauci’s source, again and again.

Francis Collins & Cliff Lane - NIH Director and Fauci’s deputy at NIAID. They make the call to Marks with him.

June 20, 2020 “I may need to blow the whistle on her”

It starts earlier than you’d think, and it starts with Deborah Birx — not the White House political staff.

Fauci writes that he is “getting progressively disillusioned” with the White House coronavirus response coordinator. His complaint is that she wants to skip the clinical trials and move straight to emergency authorizations:

Deb has ststed [sic] that we should noit [sic] do clinical trials since “I want to see every infected person in a nursing home get plasma early on and steroids later. Forget the trial, we need to do this on EUAs. She says the same thing about vaccine trials. Clearly, she is trying to provide Trump with his October surprise. I may need to blow the whistle on her.Diary, June 20, 2020 — released package p. 212

Note what this is: five months before the election, Fauci has already framed the acceleration of vaccine approval as an election play, and has already contemplated going public to stop it.

Late July 2020 - The tip-off from the Speaker’s office

By late July, Fauci is recording the same worry about Meadows. On July 26 he watches the chief of staff on television and writes that it is “his way of backing out of a failure and trying to get an ‘October Surprise’… Heads up for him trying to push an early approval of vaccine.”

The next day, July 27, Fauci is in the Oval Office briefing Trump on therapeutics and vaccines. His note is almost sympathetic: Trump “accepted that the vaccine might not be ready before the election, but he was not happy about it.”

Then, on July 31, comes the entry that changes the picture. Fauci’s information does not come from inside the administration he serves. It comes from Nancy Pelosi’s office:

Important call form [sic] Larry Horowitz who works closely with nancy Pelosi. Apparently, There was a meeting in the WH with Steve Mnuchin, Mark Meadows, Nancy Pelosi and Chuck Schummer. Steve and Mark told Nancy and Chuck that AstraZeneca will have vaccine in September and that it will be given an EUA… Apparently, Peter Marks does not know anything about this and Larry just informed him. Bottom line is that they are going to pressure Steve Han [sic] (who likely will cave in in the idea of an EUA) to issue one even though Peter has serius [sic] issues with this. An EUA will likely get in the way of completing the RCTs that are going on. Clearly. This is politically driven and will be a ploy for Trump to get re-elected.Diary, July 31, 2020 — released package p. 267

Read that middle sentence again. The FDA’s top vaccine regulator did not know the White House was planning to authorize a vaccine in September. He found out because a Pelosi aide told him. The political positioning here on both sides is crazy.

August 7, 2020 - The phone call

A week later, Fauci records what he did with the information. This is the entry the whole story turns on:

Call from Larry Horowitz (works with Nancy Pelosi)… He had warned me a few days ago about Meadows trying to pressure Hahn and Marks to grant UA [sic — EUA] (or even BLA) prematurely to vaccines (likely AZ) in time for the election. I called Peter Marks a couple of days ago (together with Collins and Lane) and explained that we would back him if he holds the line. He called the AZ package “a hot mess” and definitely not in shape for an EUADiary, August 7, 2020 — released package p. 274

Three of the most senior scientists in the federal government — the NIAID director, the NIH director, and Fauci’s own deputy — telephoned the regulator who would sign the authorization and told him they would back him if he held out.

Fauci’s justification is on the page and it is (he claims) a scientific one. Marks called the AstraZeneca submission “a hot mess.” An early authorization, Fauci had written a week earlier, “will likely get in the way of completing the RCTs that are going on.” Whether you find that persuasive is the whole argument. But the intervention itself is no longer in dispute. He wrote it down.

The pattern repeats through August. On the 8th, Horowitz flags that an FDA official needs support against a hydroxychloroquine authorization being pushed by Peter Navarro — and that Hahn told her to hold her decision so he could “speak with Meadows.”

On the 24th, Pfizer CEO Albert Bourla calls Fauci, worried AstraZeneca will get an authorization before the other companies finish their trials. On the 26th, after the FDA authorizes convalescent plasma, Fauci writes that Hahn “likely caved to Meadows. Who is looking for good news before the election.”

October 29, 2020 - Five days out

With the election five days away, the fight moves to monoclonal antibody drugs from Lilly and Regeneron. Fauci flags the entry himself, in red ink, in his own journal:

Some potentially nefarious stuff going on. The FDA career scientists are hesitant to grant EUAs to Lilly’s and Regeneron’s mAb based on the current data. Alex Azar is really pushing them and furious because he thinks FDA is “slow-walking” the process so as not to have it out before the election in 5 days. Clearly, Alex wants to please the POTUS so that they can say that they have a breakthrough… right before the election… We encouraged Francis to support the career scientists at FDA since they do not want to rush this through. They need more time to gather important information since what they have now is rather weak. Diary, October 29, 2020 — released package p. 432

Same shape as August: the administration pushes to get something announced before the vote, and Fauci works the other way, encouraging the NIH director to stiffen the FDA’s resistance.

November 21, 2020 - The accusation lands

Trump lost. Pfizer announced its 95% efficacy result on November 9 — six days after Election Day. And a little over two weeks later, the same back-channel that had been feeding Fauci information all summer delivered the other side’s verdict:

Got call from Larry Horowitz tonight telling me that Peter Marks told him that Trump called up Steve Hahn today and told him that Alex Azar said that Steve lost the POTUS the election by holding up the news about the Pfizer vaccine. Trump was cursing Steve on the phone and then hung up. Steve did not know if he was going to get fired and told this to Peter who called up Larry. Still not sure what is going to happen. Diary, November 21, 2020 — released package p. 488

There is one more entry worth noting. On November 9, the day Pfizer’s result was announced, Fauci wrote that he was “surprised that there was not even a hint in Pence’s voice that declaring the vaccine successful a few days AFTER the election was a conspiracy of some sort. I have not even heard any of that from the Oval Office.”

He was waiting for the accusation. It arrived twelve days later, and it was aimed at Stephen Hahn.

What the record does establish, in his own hand: Fauci personally intervened with the FDA’s top vaccine regulator to encourage him to resist a pre-election authorization, and his intelligence on White House vaccine plans ran repeatedly through the office of the Speaker of the House — on July 31, August 7, August 8, and again in November.

THE MERCH WAVE

March 2020

01—09

01 Fauci donuts - Donuts Delite sold thousands; copycats followed nationwide.

02 T-shirts with his face - Commercially produced and sold across the country.

03 Prayer candles - Votive candles cast him as a haloed saint.

04 Socks, mugs, pillows, sweatshirts, magnets, bottle openers - A full secondary merch wave, all unlicensed.

05 ‘Honk for Dr. Fauci’ bumper stickers

06 ‘Color Me Fauci’ coloring book - Crayon-in outlines of his face.

07 Instagram Live with Steph Curry - Obama and Bieber replied with emojis.

08 FaceTime interview with Trevor Noah

09 A song dedicated to him on San Francisco radio - Curtis Mayfield’s ‘Keep on Pushing.’

March 2020

“For better or worse, I am becoming an international celebrity.”

March 2020

“CRAZY, CRAZY – donuts are being named after me; T shirts have my picture on them and are being sold; San Francisco radio is dedicating a song to me – Curtis Mayfield singing “Keep on Pushing”; coloring pictures with crayons for my picture, face time with Trevor Noah; Instagram with Steph Curry”

April 2020

10 Brad Pitt plays him on SNL - Pitt thanked him ‘in a very serious tone’ when the sketch ended.

11 Sammie Award - Outstanding Federal Employee of the Year.

12 Official Fauci bobblehead - Raised $100,000 for healthcare-worker charities.

13 ‘Facepalm’ and ‘red tie’ bobblehead variants - $200,000 across all versions.

14 ‘Fauci Pouchy’ cocktail bags - Cocktails-to-go in transparent pouches bearing his image.

15 Lingua Franca cashmere sweater - A high-end knitwear brand’s Fauci edition.

16 Portrait sells for $15,000 - Jeremy Penn painting; proceeds to charity.

17 Petition: name him Sexiest Man Alive - 10,000+ signatures sent to People.

18 Video games using his name and likeness

19 Organized fan clubs - With membership and their own merchandise.

20 News stories about grandmothers with crushes on him

21 Restaurant dishes named after him

April 2020 · original spelling [sic]

“It is not hyoperbole to say that today I am the most famous and talked about person in the country and one of the most recognizable peson in the world. Countless profiles on me in multiple media. I cannot even read all the things that are written abuyt me since I am so busy….seriously.”

April 2020

“The ‘swag’ around me is getting really ridiculous… Geeeez!! I do not like this one bit.”

April 2020

“Continued crazy stuff: Petition to name me ‘Sexiest Man Alive’; video games named after me; Fauci Bobblehead; dishes in restaurants named after me, and on and on.”

April 2020

“The ridiculous attention to me continues; fan clubs, Swags, am constantly on TV”

Faucoi remarks about the fuss arrive in the first two months… and hen the diary goes quiet about the attention while the ledger keeps growing for another two and a half years.

THE BRAND EXPANDS

May–June 2020

22 Linguine with white clam sauce - A Long Island clam bar named its signature dish after him.

23 - A racehorse named Fauci runs at Belmont Park - Went off as the 4-to-5 betting favorite.

23 THE AWARD SEASON

October–November 2020

24 Time 100 — ranked #1 in Leaders - Top of Time’s Leaders category.

25 - Leonardo da Vinci Award A celebrity tribute reel accompanied the honor.

December 2020

26. People Person of the Year - Alongside Clooney, Gomez, and King.

27 Kate McKinnon begins a recurring SNL impersonation

28 Eleanor Roosevelt Prize - Qmerican Bar Association.

29 Sicilian of the Year Award - I Love Sicilia magazine.

30 Mother Teresa Award 0 Harmony Foundation.

“Extraordinary enthusiastic reception… totally mobbed for selfies.”

During his time as NIAID Director, Dr. Anthony Fauci kept meticulous daily journals. He recorded conversations, decisions, and his thoughts on the actions of politicians, colleagues, and the press — with near-daily entries throughout the COVID pandemic. His diary is now public, released by the U.S. Senate Committee on Homeland Security and Governmental Affairs under Chairman Rand Paul.

It runs to hundreds of pages. And what it shows is a gap between two parallel stories: the one Fauci told the American public, and the one he wrote down for himself.

Night One: The Alarm Went Straight to Him

On January 31, 2020, Dr. Jeremy Farrar — director of the Wellcome Trust — got off a routine call with Fauci about the emerging Wuhan outbreak, then immediately patched Fauci back into a second call. On the line: Dr. Kristian Andersen, an evolutionary biologist at Scripps Research, and Dr. Eddie Holmes from Sydney. They were not trading internet rumors.

Fauci’s diary records what they were actually discussing — the furin cleavage site of the SARS-CoV-2 genome. This structural feature had no close parallel in known bat coronaviruses. The scientists on the call raised the possibility that it could not have occurred naturally and may have been deliberately inserted. They discussed whether the virus had escaped from a laboratory, and whether the release was accidental or intentional — with accidental release considered the more likely scenario if it was a lab origin at all.

“They bring up the possibility that this could have been deliberately inserted and either accidentally released or deliberately released by a crazy person in the lab, the former being the most likely.”— Fauci’s diary, Jan. 31, 2020

This happened before most Americans knew there was an origins debate. Fauci was in the room — not as a passive bystander, but as the person Farrar called for advice. His diary records that he immediately suggested getting a larger group of qualified virologists together and convening the next day. He then called HHS officials, consulted Secretary Azar, and brought in Francis Collins.

The Feb. 1 Call Was Not a Consensus

The next morning, February 1, 2020, Fauci and Francis Collins convened an emergency teleconference. Participants included Collins, Farrar, Patrick Vallance, Andersen, Holmes, Ron Fouchier, Christian Drosten, Robert Garry, and others.

This call has been described, and its aftermath used to suggest, that the scientific community quickly reviewed the evidence and concluded the virus was not lab-made. Fauci’s own diary says something different.

“There was not total agreement about the likelihood of deliberate insertion. Ron Fouchier said he was sure that this could occur naturally… The rest felt that deliberate insertion was possible and given the fact that Dr. Zheng-Li Shi at the University of Wuhan has been working for years in GOF in coronaviruses… we could not let this go.”— Fauci’s diary, Feb. 1, 2020

Fouchier and Drosten argued for natural origin. “The rest” — Fauci’s phrase — thought deliberate insertion was possible. Fauci wrote: “We could not let this go.” The room was divided. The Proximal Origin paper, which publicly declared that “no laboratory-based scenario was plausible,” was submitted four days later.

The Public Story Hardened. The Evidence Hadn’t.

On February 6, 2020, the government completed its first formal review of the origins question. According to Rand Paul’s committee summary, it found no clear evidence of malfeasance — but explicitly concluded the issue still needed investigation.

By April 17, 2020, Fauci was publicly telling the country the virus was “totally consistent” with a natural species jump. By May, he was dismissing the lab-leak hypothesis as a “circular argument.”

The diary captures a widening gap: an internal review that said keep investigating, and a public-facing Fauci increasingly confident that the question was settled.

When Questions Came Back, the Fight Moved to Messaging

By May 2021, congressional investigators had begun asking about the NIH’s role in funding research at the Wuhan Institute of Virology. Fauci’s diary records that HHS public-affairs officials resisted putting out a proactive NIH statement with direct responses to those congressional inquiries.

Fauci’s own private assessment of that silence is striking:

“If the government stayed quiet, it would look like ‘we’re covering things up.’”— Fauci’s diary, May 2021

That is Fauci’s private read of what NIH’s communications strategy looked like from the outside. It is either a warning he was trying to prevent — or a description of exactly what was happening.

He Pushed Back Against the FBI’s Lab-Leak Conclusion

On July 7, 2021, Fauci met with Beth Cameron and Maher Bitar — the NSC’s senior director for intelligence programs — inside the EEOB Sensitive Compartmented Information Facility (SCIF). The FBI, by this point, had assessed that COVID-19 most likely originated in a laboratory.

Fauci’s diary is blunt about his view of that conclusion:

“It became clear that the FBI has no idea what they’re talking about since they are convinced that COVID originated in a laboratory leak.”— Fauci’s diary, July 7, 2021

He then urged NSC officials to work closely with Andersen, Holmes, Garry, and Rambaut — the same circle of scientists at the center of the early origins calls, and the co-authors of Proximal Origin. Fauci was not just privately skeptical of the FBI’s conclusion. He was actively steering intelligence officials toward the same scientific allies who had shaped the public narrative.

The Ally Who Made Him Feel Better

Around May 2021 — the same month Fauci warned that staying quiet would look like a cover-up — he had a private conversation with Kristian Andersen. This is the same Andersen who, in January 2020, had privately written that the lab-escape hypothesis was “so friggin’ likely to have happened.”

A year later, Andersen was privately reassuring Fauci. According to the diary, Andersen was “very emphatic” that the work done at the Wuhan Institute under the NIH grant was not gain-of-function research, and confirmed that chimeric viruses made under that grant were “not even close sequence-wise” to SARS-CoV-2.

“I feel much better now that he is in total support of what I have been saying.”— Fauci’s diary, May 2021

The same scientist. A year apart. First privately alarmed. Then privately reassuring. What changed in between was the publication of Proximal Origin — the paper Andersen co-authored that declared the lab origin implausible.

Scott Atlas: The Dissenter Fauci Couldn’t Dismiss on Science Alone

In August 2020, President Trump brought in Dr. Scott Atlas — a radiologist and health policy fellow at Stanford’s Hoover Institution — as a White House special assistant. You recall that the team here at Rational Ground was part of the support group of data analysts that Dr. Atlas relied on for his data analysis while he was at the White House. We are proud to play a particular role in this thorn in Fauci’s side

Atlas argued that schools could and should reopen, that children did not drive community spread, and that lockdowns caused more harm than they prevented.

Fauci’s diary entry on Atlas’s arrival frames the dispute in political and credentialing terms, not scientific ones:

“Trump has brought in a Dr. Scott Atlas, a radiation oncologist from Stanford Hoover Institute. He is a FOX News Commentator and Trump has seen him on FOX. He is against everything we are saying… He obviously does not believe in community spread. This is going to be a real problem.”— Fauci’s diary, August 12, 2020

The conflict intensified through the fall. In late October, Fauci gave a Washington Post interview in which he, by his own diary account, “strongly blasted Scott Atlas.” The White House pushed back. Alyssa Farah, a senior White House aide, texted Fauci at 7 AM the next morning:

“I am so sorry that fool Scott Atlas keeps attacking you. I’m even more sorry I’m not in a better position of influence to see him out the door. I’m trying to make people see he’s a disaster.”— Alyssa Farah, quoted in Fauci’s diary

Atlas resigned in December 2020. Fauci’s diary entry on the resignation is telling:

“Scott Atlas resigns! I imagine that is because everything that is happening now with the surges has finally proven him to be incorrect on virtually everything that he has said. The one area where he was correct was in saying that school children are not the drivers of the outbreak.”— Fauci’s diary, December 1, 2020

Atlas was publicly dismissed, mocked in the press, and branded a fringe figure. Fauci’s private diary concedes that Atlas was right about the thing parents cared about most — whether keeping kids out of school made any difference.

MORE TO COME!

For years, Ralph Baric was the quiet center of the COVID-origins story.

He was the coronavirus engineer at the University of North Carolina whose handwritten note on the 2018 DEFUSE proposal warned that American researchers would “likely freak out” over the biosafety of what was planned. He was the “B” in the story we published — the man who actually knew how to build these viruses. And for six years, he mostly didn’t talk.

On April 10, 2026, that changed. Baric sat for a voluntary transcribed interview with the staff of Senator Rand Paul’s Homeland Security Committee. On July 23, Paul released the transcript. Read in his own words, it is the most candid account we have yet from anyone who was actually at the workbench.

Here is what he said.

“That’s a gain-of-function experiment. There’s no other way to think about it.”

Start with the word everyone spent years fighting over: gain-of-function.

DARPA’s PREEMPT program — the solicitation that DEFUSE was written to answer — asked researchers to learn what makes a virus jump between species “down to the nucleotide level.” Investigators asked Baric what that meant. He didn’t hedge:

“That’s a gain-of-function experiment. There’s no other way to think about it… when I read that, that’s exactly what they’re asking for.”

Then they turned to DEFUSE itself — the proposal Baric submitted alongside Peter Daszak of EcoHealth Alliance and Zhengli Shi of the Wuhan Institute of Virology. The plan described finding a bat coronavirus, and, if the science warranted, dropping a furin cleavage site into a viral backbone. The furin cleavage site is the exact feature of SARS-CoV-2 that first made serious scientists look sideways at a natural origin — because its closest relatives don’t have one.

Who would have performed that insertion? Baric was asked directly.

“That part would be done by me.”

Let that land. The single most suspicious feature of the pandemic virus was, on paper, a job assigned to a lab in North Carolina — by the lab’s own principal investigator, under oath-grade questioning.

DEFUSE was never funded; DARPA rejected it. No one has shown that SARS-CoV-2 is the DEFUSE construct. But the thing skeptics were mocked for even describing — a plan to engineer a furin site into a bat coronavirus, with a named American scientist to do the cutting — Baric just confirmed in plain English.

The experiment he buried — and what it does to “Proximal Origin”

This is the part that should reset the entire debate.

The Proximal Origin of SARS-CoV-2 — the March 2020 paper used to shut the lab-leak question down — rested on a deceptively simple argument. If someone had engineered this virus, the authors reasoned, they would have built it to bind human cells as tightly as possible, the “ideal” configuration. The real virus doesn’t bind in that ideal way. Therefore, they concluded, no one designed it.

Baric ran an experiment, years earlier, that quietly demolishes that logic.

He explained that a biochemist colleague believed the tightest possible grip on the human receptor would be “the most optimal and most dangerous form of virus.” Baric disagreed — and tested it:

“If they react too well, that interaction can’t come apart… the virus, as it tries to release, gets caught up in what’s called a dominant negative effect, and it will prevent virus replication… I tested that hypothesis and showed that I was right, that the super-binding things were just as deleterious as the really poor-binding ones.”

In plain terms: grip too tightly and the virus gets stuck and can’t spread. So a knowledgeable designer would deliberately avoid the “ideal” binding that Proximal Origin assumed an engineer would use. The very feature the paper held up as proof of natural origin is exactly what a competent engineer might choose on purpose.

And then the tell:

“I subsequently told NIH and destroyed those viruses. I have never told anyone, and I would recommend that this committee not release that.”

He reported it to NIH. He destroyed the materials. He’d told no one. The finding that undercuts the field’s central natural-origin argument sat unpublished — and the man who ran it asked Senate investigators to keep it quiet.

The ghost on the February 1 call

Regular readers know the February 1, 2020 teleconference — Fauci, Francis Collins, Jeremy Farrar, and the authors of Proximal Origin, one day after Kristian Andersen privately called a lab origin “so friggin’ likely.”

Baric confirmed he was on that call. But he appears on no invitation list. He couldn’t say who invited him:

“I couldn’t find how I knew about the meeting.”

Two participants have since told the Committee they didn’t even know he was on the line. And Baric gave conflicting accounts of whether he spoke — first saying he stayed silent so as not to “change the direction of the conversation,” then saying “I think so… we went around and introduced everybody,” and that he was “surprised that they didn’t know I was there.”

The world’s foremost coronavirus engineer was on the call that shaped the official origin narrative — and can’t explain how he got there.

The rest of the ledger

Three more admissions, briefly, because they matter:

• The mouse model. Baric served more than a decade as a paid member of the intelligence community’s Biological Science Expert Group. He confirmed sending his U.S. taxpayer-funded humanized-mouse model to Shi at the Wuhan lab. When China later tried to sell that model back to American scientists, he said: “What kind of crap is this? This is not ethical behavior.” He framed the sharing as compelled — “I’m required to share” — and told Congress that if they don’t like it, “you all should write regulations that prevent that.”

• Five years of silence on USAID money. In November 2015, Daszak sent an urgent email asking that USAID PREDICT and NIAID funding be acknowledged on Baric’s paper with Shi. The acknowledgment wasn’t added until 2020 — five years later. That delay sits awkwardly against then-USAID Administrator Samantha Power’s 2022 testimony that the agency had “no evidence” it funded, and had “not authorized,” gain-of-function research.

• NIH lost its own homework. When NIH officials met with Baric in February 2020 and asked for his research records — the ones NIH supposedly reviewed years earlier — Baric says they “didn’t know where the documents were.” And DEFUSE, he testified, never came up in his conversation with Fauci.

Read it honestly

None of this is a confession that Baric built SARS-CoV-2. DEFUSE was never funded. A plan to insert a furin site is not proof that this furin site came from that plan, and Baric plainly believes his own work was responsible and defensible. He is a hostile-but-voluntary witness, questioned by a committee with a thesis.

But strip the spin from both directions and what remains is remarkable enough. The engineer who would have done the furin-site work says the program was gain-of-function, “no other way to think about it.” He ran and buried an experiment that turns the marquee argument of Proximal Origin inside out. He was on the secret call and can’t say why. And the federal agencies that were supposed to be watching couldn’t find their own records.

For years, saying any of this out loud got you called a conspiracy theorist. The furin-site designer just said it himself — into a court reporter’s transcript.

(Senator Paul separately released private Slack messages from the Proximal Origin authors the same week; their contents reinforce the same gap between what these scientists said in public and in private. We’ll have more on those.)

About the source. Quotations are drawn from the official transcript of the April 10, 2026 voluntary transcribed interview of Dr. Ralph Baric before the U.S. Senate Committee on Homeland Security and Governmental Affairs, released July 23, 2026 by Chairman Rand Paul, and from the Committee’s summary of key findings. The interview and companion Slack release were first publicized by Senator Paul.

In early February 2021, Peter Daszak was on his way home.

You might recall our article from the other day with the video talking about Daszak.

The president of EcoHealth Alliance had just spent weeks in China as part of the World Health Organization’s team investigating the origins of COVID-19 — the mission that took him, however briefly, to the Wuhan Institute of Virology, the lab his own organization had helped fund with U.S. taxpayer money. His flight was booked: Qatar Airways 701, Doha to New York’s JFK, landing February 5 at 2:25 p.m.

Waiting for him, on paper, was U.S. Customs and Border Protection.

According to documents released this month by Senator Rand Paul — chairman of the Senate Homeland Security and Governmental Affairs Committee — and first reported by Bill Melugin, officers at CBP’s National Targeting Center had built a formal targeting event around Daszak’s arrival and prepared a Tactical Terrorism Response Team examination. They had a plan, and they had questions.

Then, days before he landed, the plan died. Not because anyone at CBP changed their mind — because another agency intervened.

Why they wanted him

CBP’s interest in Daszak wasn’t a hunch. The internal referral, dated January 11, 2021, called him “an extremely high person of interest” and recommended “a thorough inspection either outbound/inbound.” Officers noted a specific worry: that Daszak “could hand carry and/or mail samples back from China.”

The targeting package laid out why. It cited Daszak’s role as principal investigator on NIH grants tied to bat-coronavirus research — including grant R01AI110964, “Understanding the Risk of Bat Coronavirus Emergence” — and his 15-year collaboration with Zhengli Shi, the Wuhan virologist known as “Bat Woman.” It flagged his co-authorship, with Shi and UNC’s Ralph Baric, of the 2013 paper reporting a bat SARS-like virus that could infect human cells, and the 2015 Nature piece on “engineered bat virus” research. And in the officers’ own blunt phrasing:

“Much of the funding DASZAK receives is put towards gain of function research.”

The package even noted that NIH had asked Daszak to retrieve the original virus type-specimen used by the Wuhan lab and return it to the United States — a request Daszak reportedly called “outrageous” and did not appear to intend to honor.

What they planned to ask

This is the part that reads like a checklist someone actually wanted answered. CBP’s recommended questions for the exam included:

• Why was Daszak chosen for the WHO investigative team?

• Where in China did he go, and where did he stay?

• Did the team visit the Wuhan Institute of Virology? What areas did they inspect, and whom did they meet?

• Did the team visit the Chinese CDC?

• Did Daszak collect any biological samples during the trip? Is he traveling with any? Did he mail any samples back to the U.S.? Did he take any materials with him to China?

• Has Daszak maintained contact with anyone at the WIV — and what do they discuss?

The exam was also slated to “include review of media” — that is, a search of his phones and devices, with the search level to be set at the port of entry.

In other words: this was not a wave-through. Customs had assembled the framework to question a returning WHO investigator about whether he was physically carrying biological material across the U.S. border, and to look at what was on his phone.

The intervention

It never happened. And the documents are unusually clear about why.

On February 2, a National Targeting Center message rescinded the inspection request, noting the “subject may be of interest to the FBI” and instructing colleagues to “please disregard request for inspection.” Then, on February 4 — the day before Daszak’s flight — a CBP Deputy Chief Officer serving as the agency’s Joint Terrorism Task Force liaison sent the line that anchors the whole release:

“Please be advised FBI New York requested we do not stop the subject below.”

The final disposition record puts it even more plainly:

“FBI requested not to stop subject. No further NTC action.”

Daszak landed at JFK. As far as these records show, no one from CBP questioned him, and no one searched his devices or his luggage.

Read it honestly

Here is what the documents establish, and here is what they don’t — because the gap matters.

They establish that career Customs officers, on their own initiative, identified Daszak as a high-priority subject, built a counterterrorism-style inspection around his return, wanted to ask whether he was moving biological samples, and were called off by the FBI’s New York field office at the last minute.

They do not establish why the FBI made that call. And there’s an innocent reading worth stating out loud: when a Joint Terrorism Task Force asks CBP to “not stop” someone, it is sometimes precisely because the Bureau has its own active interest in that person and doesn’t want a border stop to tip them off or interfere. A stand-down order is not automatically a cover-up; it can be the opposite. But the track record during the Biden years is not great for calming our suspicions.

The redactions hide the names, the office details, and the FBI’s stated reasoning — so the single most important question here, why, is exactly the thing blacked out.

What’s not in dispute is the shape of it: in February 2021, one arm of the federal government wanted to search Peter Daszak at the border, and another arm told it to stand down. For a figure already at the center of the COVID-origins conflict-of-interest story — the man who funneled NIH money to Wuhan, secretly organized the Lancet letter branding lab-leak a “conspiracy theory,” and then sat on the WHO team sent to investigate the very lab he funded — that is a document set worth having in the public record.

The files are public now. The one line still hidden is the one that would explain everything.

About the source. Drawn from a package of internal CBP records released in July 2026 by Senator Rand Paul, Chairman of the U.S. Senate Committee on Homeland Security and Governmental Affairs, and reported by Bill Melugin. The records are marked Law Enforcement Sensitive / For Official Use Only and were released in redacted form; blacked-out specifics — including names and the FBI’s stated rationale — are left blank here. Quotations are taken from the released text.