Retrofitting Reality
You don’t have to win the argument. You just have to distort it enough to shift the narrative.
In my Texas high school, the biology curriculum simply skipped evolution. So naturally, one of the first classes I signed up for in college was evolutionary biology. It did not disappoint. There’s something irresistible about a forbidden idea — especially when it turns out to be a truth that was deliberately withheld.
That experience left me with a lasting sensitivity to how truth gets mistreated in public life. And this week offered a vivid example.
Donald Trump posted an image of himself that most viewers immediately read as Christ-like: hands extended in healing, bathed in divine light, robed, with figures ascending toward something heaven-adjacent. When the backlash came, so did the denial. It wasn’t Jesus, we were told. It was Trump as a doctor, healing the sick.
The explanation didn’t fit the image. But that almost misses the point.
The point is how quickly a claim can be made, challenged, and redefined without ever being held accountable to reality. The conclusion comes first. The explanation gets retrofitted to match. And we’re left arguing about the retrofit instead of the original act.
Which brings us to an actual doctor, and another act of retrofitting. Last week, NIH Director Jay Bhattacharya reportedly halted the release of a publication-ready Morbidity and Mortality Weekly Report (MMWR) on this year’s COVID-19 booster effectiveness. His spokesperson suggested such reviews are routine, citing concerns about the study’s design and methods.
By most accounts, this kind of intervention is far from routine. And the specific concerns raised do not hold up.
The methods used to estimate vaccine effectiveness from real-world data are neither new nor experimental. They are the same approaches epidemiologists have used for decades to evaluate influenza vaccines and countless other public health interventions. What is unusual is political interference at the publication stage of an MMWR, a publication explicitly designed as a rapid, technical communication channel for exactly this kind of data.
And the findings themselves aren’t controversial. Across multiple studies, COVID-19 booster doses have been associated with roughly a 50% reduction in the risk of hospitalization and an 80% reduction in the risk of death, with the largest benefits among older adults. The exact numbers vary by variant and study design. The direction does not.
So why suppress the report?
Bhattacharya’s “concerns about study design” are the administrative-speak version of “it wasn’t Jesus, it was a doctor.” The explanation isn’t meant to convince anyone who knows the field. It’s meant to muddy the conversation enough that we end up debating the explanation instead of the act.
What we’re witnessing isn’t a good-faith debate between competing explanations. It’s something corrosive: the strategic deployment of implausible claims, offered not because anyone expects them to be believed, but because the act of offering them shifts the ground. Suddenly, we’re debating whether Trump meant to invoke Christ, or whether decades of established epidemiologic practice have merit, rather than reckoning with the original acts themselves.
The backwards logic of conclusion first, explanation retrofitted, is how accountability becomes optional. You don’t have to win the argument. You just have to distort it enough to shift the narrative.
That’s a dangerous place for public health. It’s an even more dangerous place for a country.