The Institute of Integrative and Interdisciplinary Research has developed a code of practice for the introduction of entities in cosmology.
A code of practice for introducing entities in cosmology. https://doi.org/10.13140/RG.2.2.20426.53446
The code concerns entities known only through their consequences — posited because, if they existed, some observable would follow. This is legitimate and indispensable; the neutrino, the Higgs boson, and gravitational waves were each such entities before they were confirmed. The code does not forbid the unobserved. It asks only that, when an entity is introduced, certain things be exhibited. There are ten norms in four groups, each derived from an elementary proved proposition rather than asserted, and each phrased as an obligation to exhibit something rather than a prohibition on positing:
Structure of the stack of unobserved assumptions
1. Declare the depth of the stack — state how many roles without referents lie between the entity and any observation that bears on it.
2. Do not deepen a stack to rescue it — add no further unobserved posit whose sole function is to shield an existing one from refutation.
Falsifiability
3. Exhibit a discriminating regime — name a conceivable measurement in which the world could say no to this entity, not merely to “too much” of it.
4. State the condition of the entity’s death — name in advance the outcome upon which the entity would be withdrawn.
Discipline of inference
5. Label every load-bearing quantity as measured, declared, or derived — mark whether each is fixed by the data (measured), fixed by choice (declared), or compelled by a proved internal relation from the other inputs (derived), since a conclusion inherits the status of its weakest load-bearing quantity. A quantity counts as derived only if its derivation is exhibited and independently reproducible; a derivation a reader cannot run for himself reverts to declared.
6. Do not read a bound consistent with zero as evidence for the entity — a non-detection constrains abundance conditional on existence; it does not attest existence.
7. Match the loudness of the conclusion to the load-bearing component — claim no more than the component actually doing the work can support.
8. Keep the conditional; refuse the promotion — “if it exists, S follows” is licit; “therefore it exists” is not, absent a diagnostic contact.
Reach
9. Consult adjacent fields before positing — a quantity already measured in a neighboring discipline supplies a referent; do not posit afresh where one exists without saying why.
10. Test the posit at both scales — examine a cosmological posit for its microphysical implications, and a microphysical one for its cosmological implications; one-sided appraisal hides the strongest constraints.
The Institute offers this as an instrument, not an edict. The norms are quantified over every author who posits, so they bind the Institute’s own members exactly as much as anyone else’s; the paper names no program as a model and no paper as a target, illustrating only with historical posits later confirmed or with schematic constructions that have no referent in the current literature. The norms rest on results we would rather see disputed than deferred to: if a rationale fails, the norm should fall with it, and we would welcome being shown where.
The paper:
https://doi.org/10.13140/RG.2.2.20426.53446
Please send your comments, critiques, and reactions to:
Boris Kriger, Lead Investigator
Institute of Integrative and Interdisciplinary Research, Toronto
Institute of Integrative and Interdisciplinary Research 