Verification Policy

The Massachusetts Times · Published by The Massachusetts Times Company · massachusettstimes.us/policies/verification · Effective October 16, 2026

This policy states the standard a claim must meet before The Massachusetts Times publishes it as fact, and what happens to claims that do not meet it. Every sentence here is kept today.

1. The Standard

1.1. Nothing publishes as fact until it is verified. The standard is identical for every work, every class, and every origin: a fact in an opinion piece, a caption, or a machine-compiled table meets the same bar as a fact in the lead story, and the Automation Policy forbids any discount by origin.

1.2. Verification is not a mood of carefulness; it is the acts in this policy, performed before publication, checkable after.

2. Classes of Source

2.1. A first-class source is direct evidence: the named participant speaking of what they did or witnessed, the original document, the raw record, the proceeding itself, our own presence.

2.2. A second-class source is a report of evidence: the secondhand account, another outlet's story, the summary of a document nobody at this desk has read.

2.3. First-class sources establish facts. Second-class sources corroborate, lead us to first-class sources, or publish only as what they are — attributed reports, named and linked under the Attribution Policy, never silently promoted into established fact.

3. The Second Reading

3.1. Before a substantive claim publishes, it is read against its sources in a check separate from the writing of it. The writer's conviction is not the verification; the comparison is.

3.2. The check asks one question of every claim: what, exactly, supports this — and the answer must name a source of the class the claim requires.

4. Verification Dates

4.1. Civic facts — the officeholder, the statute, the rate, the boundary, the standing record a citizen acts on — carry their verification dates in public. A dated fact tells the reader when it was last checked against the world, which is a thing almost no publication will tell them.

4.2. An undated civic fact is a defect in the work that carries it, repairable under the Correction Terms.

5. Withholding Over Guessing

5.1. What cannot be verified is withheld. It is not hedged into print under "reportedly," not laundered through a second-class source, not published with its doubt buried in paragraph nine.

5.2. Where the gap itself matters, the gap is stated honestly — what we sought, what could not be established by publication — because an honest hole in the record is worth more than a confident patch over it.

5.3. Speed never outranks the standard. This institution would rather be second and right than first and wrong, and this policy is where that preference stops being a slogan and becomes a rule.

6. The Held State

6.1. Some claims the evidence raises but does not settle. Those claims are held: kept, marked as contested, and excluded from the record of established fact — neither published as true nor discarded as worthless.

6.2. A held claim may be written about only as what it is — a contested claim, labeled, with the state of the evidence stated — and it is resolved the day the evidence resolves it, in either direction.

6.3. The held state exists because the two easy paths are both lies: publishing the unsettled asserts what the evidence does not, and discarding it pretends the question was never raised. We keep the question, in daylight, at its true weight.

7. After Publication

7.1. Verification does not end at publication. A published fact that fails against new evidence is corrected under the Correction Terms; a work that fails beyond correction is withdrawn under the Archive and Permanence Terms, with its dated notice standing at its address.

7.2. The verification dates of Section 4 are maintained, not abandoned — a civic fact's date advances when the fact is rechecked, and the archive shows the history.

8. What the Reader Is Shown

8.1. The work shows its support: who said what, which documents, what record — attributed at the resolution the Source Terms allow. A reader of ours should be able to see why we believe what we printed.

8.2. Where method matters to the finding — how a count was counted, how a comparison was built — the method is stated in or beside the work.

9. The Record of This Policy

9.1. This policy is kept in dated versions at its address, so the standard any work was held to can always be checked against the work's date.

This policy is one of the Policies this publication keeps.