Experimenter's Regress¶
The appraisal loop in which a contested result needs a competent experiment, while competence is judged by producing the still-unknown correct result.
Core Idea¶
Experimenter's regress arises when a new phenomenon has no agreed correct experimental result. A result is trusted only if a competent experiment produced it, but experimenters may judge competence by whether the procedure produced the very result still in dispute. A positive and a null test can each be rejected as poorly performed; repeating the experiment may reproduce the same appraisal question.[^collins]
Harry Collins developed the pattern from early gravitational-radiation disputes. It diagnoses a problem of evidence appraisal, not the physical truth of either side or the impossibility of all replication. Independent calibration and other checks can help when their relevance to the target is independently warranted.[collins][calibration]
Scope of Application¶
In Collins's Weber-era case, positive resonant-bar reports and negative replications were weighed alongside contested judgments about detector competence. Collins and Pinch also discuss the early cold-fusion controversy: reported excess heat and negative replications raised questions about measurement and replication practice. Pinch explicitly applies the regress concept to that second case; doing so does not endorse the physical effect.[collins][ref-fe232e0733ec][^ref-f192bffc4f86]
The concept applies to other frontier disputes only if one can show both directions: results accepted because the experiment is deemed competent, and competence inferred from producing the expected result. Mere disagreement or a failed replication is not enough.[^collins]
Clarity¶
Keep the reported output, the performance of the apparatus and the existence of the target effect separate. The regress appears when the latter two are used to certify each other without independent support. It does not imply dishonesty, confirmation bias or that a claim is false.[^collins]
A known test signal may offer an external performance check, but a surrogate's similarity to the novel signal itself needs argument. Collins notes this for electrostatic-pulse calibration of gravity-wave detectors; Allan Franklin's original work articulates epistemic strategies for rational experimental appraisal. Neither blanket social-only closure nor automatic calibration success follows from the pattern.[calibration][ref-4e21f37d057a]
Manages Complexity¶
The entry turns a diffuse dispute into a dependency question: Which evidence supports the outcome, and which evidence supports the competence of the experiment that produced it? If the answers rely on each other, another replication may inherit rather than settle the dependence.[^collins]
This diagnostic cannot replace technical examination of controls, sensitivity or error. Its useful consequence is to locate where a genuinely independent competence test would change the evidence situation.[calibration][ref-4e21f37d057a]
Abstract Reasoning¶
For each positive or null result, ask why that experiment counts as well executed. If the answer is only that it produced what the evaluator already believes is correct, the result cannot independently settle that belief. Apply the same test to the next attempted replication.[^collins]
Then identify candidate independent standards and test their relevance. A calibration can break one dependence only if its signal tests the capability that the contested claim requires; otherwise the adequacy of the surrogate may become a new point of dispute.[calibration][ref-4e21f37d057a]
Knowledge Transfer¶
Resonant-bar detection and electrochemical calorimetry have different physical objects, but the same appraisal roles can recur: contested outcome, skilled procedure, competence judged through outcome, outcome judged through competence, and replication whose status remains open. What transfers is this epistemic structure, not the experiments' physics.[collins][ref-fe232e0733ec][^ref-f192bffc4f86]
Live Infinite Regress is related when replications repeatedly generate the same competence demand, but its unbounded-chain requirement makes it an unsafe strict parent for the core two-way loop. The entry is staged unparented. Calibration is a possible exit, not a prerequisite; Circular Reasoning is related but concerns argumentative support rather than experimental skill. This entry remains domain-specific to contested empirical practice.
[^collins]: Harry Collins, author's account of the 1975 “Seven Sexes” paper and “Gravitational Waves and Scientific Realism”, author version, pp. 1–2. [^ref-fe232e0733ec]: Harry Collins and Trevor Pinch, The Golem, chapter “The sun in a test tube: the story of cold fusion.” [^ref-f192bffc4f86]: Trevor Pinch, “Cold Fusion and the Sociology of Scientific Knowledge”, Technical Communication Quarterly 3 (1994), abstract. [^calibration]: Harry Collins, author's note on Changing Order, chapter 4, electrostatic calibration and replication caveat. [^ref-4e21f37d057a]: Allan Franklin, “The Epistemology of Experiment”, in The Neglect of Experiment (1986), ch. 6, author summary; 1994 original reply.
Neighborhood in Abstraction Space¶
Experimenter's Regress sits in a sparse region of the domain-specific corpus (66th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Strategic Decision Biases & Mechanisms (29 abstractions)
Nearest neighbors
- Conditionality principle — 0.84
- Educational Assessment — 0.84
- Outcome bias — 0.84
- Dual-Character Concept — 0.84
- Cooperative Pulling Paradigm — 0.84
Computed from structural-signature embeddings · 2026-10-08