Empathic Journalism
Journalism for people who don’t trust journalism anymore.
Most reporting asks you to trust it. This asks you to check it. We find the story with the trained empathy of a clinician, then hand you the evidence, archived and fingerprinted, so you can verify every claim without taking our word for any of it.
The problem
In 1976, after Watergate, about 72 percent of Americans told Gallup they trusted the news to report fully and fairly. By 2024 that number had fallen to roughly 32, and lower among the youngest adults. Both figures come from Gallup’s standing trust-in-mass-media series, which publishes its own methodology and can be checked against it. Forty points in fifty years is not a dip in a cycle, and you cannot fix it by being a more careful version of the thing that broke.
The profession’s reflex is to ask for trust more firmly: believe us, we are the credible ones. That request is the disease. So we borrowed a move from cryptography. When a payment system was built for a world that no longer trusted financial middlemen, the design did not make the middlemen more trustworthy. It made them unnecessary. A story you can check for yourself does not need your trust, whereas one that demands trust while refusing to be checked has the logic backwards.
The method, in two beats
Empathic Journalism has two halves, and they have to sound together.
Empathy is the instrument. A clinician is trained to read the gap between what a person says and what a person shows, because that gap is where the real material lives. Turned on a source and on the documents, that reading tells a reporter which claims to chase. The profession filed this under soft skill. It is a detection instrument, and it works like a stethoscope rather than a microphone.
Verification is yours. What the empathy finds, you get to check. Every source is archived where it was read. Every document is fingerprinted, so you can confirm we relied on the same file you are holding. Every claim carries a grade. The reporting tells us where to dig; the protocol lets you check the dig without believing a word we say about ourselves.
The apparatus
Trustless — the protocol
The verification half, made concrete. Before the evidence is gathered, the claim is written as a falsifiable statement with its disproof condition attached. Every cited source is archived and hashed at the moment of reading. Every web source is pushed to a public archive, so a citation points at a frozen copy rather than a link that can rot. The pipeline will not publish a claim whose basis was not preserved. Trustless is outlet-neutral: any journalist can run it.
Receipts — the grade on every claim
Every claim that carries the argument gets one plain word (Established, Supported, Reported, Thin, or Disputed), the receipt it rests on (the source, archived and fingerprinted so you can pull it up yourself), and the one thing that would prove it wrong. Then the whole piece gets a profile: how much of the argument stands on confirmed ground, and which weak claim, if any, the conclusion turns on. You are shown what our certainty is standing on rather than asked to accept it.
The pause — the discipline
We hold a finished piece at least seventy-two hours before it goes out, not to look careful but because the second thoughts arrive in the wait. The house is named for the pause.
The standard
Everything above is enforced, not promised. Work published here declares the EJ-1 standard: a binding right of reply for serious allegations against named subjects, an append-only corrections ledger with no silent edits, and a documented chain of custody from source to sentence. The standard is versioned and public, so you can hold the work to the same rules we do.
Proof
The method is not a posture worn for essays. It produces work that has survived an editor and an independent fact-checker at a regional newsroom, and that publishes its own verification state including when that state is a failure.
An investigation into a regional water authority, published and fact-checked at a regional newsroom, coded 24 board meetings from the minutes and found the board voted on 228 resolutions across two years, was unanimous on 223 of them, split zero times, and authorized more than $871 million, while 156 of those 228 resolutions drew no recorded question from any board member and ten residents spoke in two years. The figures are not assertions. A script in the published package re-derives every one of them from the coded data, and the build fails if any of them drifts.
The reporting here has appeared at PublicSource and the West Hills Gazette.
When a thesis here dies, you see the death certificate with the date on it. A year-long investigation into a cluster of drownings was killed by its own disproof condition when the arithmetic turned against it, and the smaller, truer story underneath, that no audit of the rulings appears in the public record, is the one that survived.
The reader we built this for
You are drawn to the strange, and you are not foolish for it, because some of those threads are real and the institutions meant to settle them gave up the benefit of the doubt. You also want an empirical floor underfoot. You are tired of choosing between two rooms: the conspiracists, who want a world thick with hidden agency where nothing is an accident, and the debunkers, who want a flat world where the strange is always a misunderstanding. Both rooms sell certainty, and this is built for the reader who wants neither, who would rather make sense of a gray world than be handed a flat one.
The water package’s verification gate does not pass, and its page says so at the top rather than the bottom. That is the only reason the rest of this is worth reading.