Convergent Independent Results

Note

Summary release. The body is retained pending a fourth review, scheduled 2027.

Observation

Between ██████ 2022 and ██████ 2024, four groups — none of them this institution, none in contact with the others, working in three countries and at least two unrelated disciplines — reported a result of the same unusual form.

The groups were identified by us retrospectively, in the course of an unrelated literature exercise. None of them appears to be aware of the others. One described their finding as “almost certainly instrumental” and did not pursue it.

The question

Convergence of independent results is normally the strongest evidence available that something is real. The purpose of this note is to argue against over-reading it in this case, because the institution has an obvious motive to over-read it and should say so before someone else does.

Arguments against significance

Shared instrumentation. Two of the four groups used equipment from the same manufacturer, in overlapping generations. A common artefact is not excluded and would be difficult to detect from outside either group.

Shared method lineage. Three of the four cite a common methodological antecedent. Independence of conclusions does not follow from independence of teams when the analysis pipeline has a common ancestor.

Selection. We found these four by looking. We have not established how many groups reported a different result, or no result, over the same period. Absent that denominator, four is not a rate — it is four.

Our own priors. The result’s form is consistent with material already in this Archive. We are therefore the worst-placed party to assess it, and the reviewer who assembled this note had read SEN-14-052 beforehand. That contamination is recorded rather than corrected, because it cannot be corrected.

Arguments for significance

Stated for completeness, and given less weight than they will be given by anyone reading selectively:

• The two groups with unrelated instrumentation and no shared methodological antecedent are, on the evidence available, genuinely independent.
• The result’s form is specific rather than generic, which lowers the probability of coincidental convergence.

Position

The institution’s position is that this is interesting and insufficient.

We are recording it now, at the point of noticing, precisely so that if something further emerges the record shows what we believed before we knew. An account assembled afterwards would be worth much less, and we would not be able to tell the difference.

No programme has been established. No resources beyond the standing review have been assigned. The note exists so that the absence of a conclusion is itself on the record, dated.