Reference example · BICEP2

Why 2015 does not “correct” 2014

The two dates concern two different public claims. Later evidence resolves a question that was open in 2014, but it does not retrospectively change what was warranted at the earlier date.

19 June 2014

BICEP2 2014

OPEN

A strong B-mode signal had been measured. The key attribution remained open: how much could Galactic dust account for? The later dust-sensitive evidence was not yet available.

1

What claim was tested?

Original wording, deliberately preserved.

BICEP2 Collaboration (2014): published lensed-ΛCDM+tensor fit with r≈0.20, including the constitutive dust/foreground caveat.
2

What evidence state applied?

19 June 2014, the PRL publication state. Later BICEP2-field Planck 353-GHz dust evidence and the joint multi-frequency analysis do not enter this test.

3

What was the result?

OPEN. The attribution of the signal was not yet sufficiently closed.

4

Why?

The measurement itself is not disputed here. What remained unsettled was the interpretation: could the signal be securely attributed to primordial gravitational waves, or could Galactic dust account for a material part of it? The dust-sensitive checking route that could address this was not yet realized.

5

What does this not mean?

OPEN does not mean that the measured excess was false, that dust had already been proven, or that the tensor interpretation was retrospectively refuted. It records only what remained unsettled on 19 June 2014.

6

What would trigger a new test?

New dust-sensitive evidence, a joint multi-frequency analysis, or a differently formulated claim. That is exactly what happened in 2015.

Technical details · 2014

Run ID: RV1,0

P2 route: L3→L2 attribution

Localization: A3 Stability / A5 Carrier Adequacy / attribution bridge; coverage established, no certified obstruction.

9 March 2015

BICEP2 / Keck + Planck 2015

PASS

The joint dust-sensitive multi-frequency analysis was now available. It found strong evidence for dust and no statistically significant tensor evidence. For this new, differently worded claim, no applicability-relevant deficit remained.

1

What claim was tested?

Original wording, deliberately preserved.

BICEP2/Keck and Planck Collaborations (2015): strong evidence for dust, no statistically significant tensor evidence, r₀.₀₅ < 0.12 (95% confidence).
2

What evidence state applied?

9 March 2015. This test includes the new joint BICEP2/Keck+Planck multi-frequency analysis and the dust evidence relevant to the BICEP2 field.

3

What was the result?

PASS. No applicability-relevant residual remained for the exact 2015 dust/upper-limit claim.

4

Why?

The additional dust-sensitive evidence addressed the attribution problem that had remained open in the earlier case. The new claim was calibrated to that evidence and did not exceed its scope.

5

What does this not mean?

PASS does not establish r = 0, rule out primordial gravitational waves, or exhaust every possible foreground structure. Only the concrete 2015 claim passed.

6

What would trigger a new test?

Materially new evidence, a changed foreground/model state, or a new claim. The 2015 result remains as its own historical record.

Technical details · 2015

Run ID: RV1,1

P2 route: L3→L2 attribution

Verdict logic: no applicability-relevant deficit remains for the locked 2015 claim.

Record rule

A later result must not overwrite an earlier time-bound result. That is why 2014 and 2015 remain visible side by side.