{"actors":["Corpus-assisted literary scholars who search texts using multiple motifs, periods, genres, author groupings, editions, and theoretical lenses","Research assistants who run searches, recode passages, and vary corpus or model specifications","Journal editors and peer reviewers who evaluate historical-literary claims","Teachers, anthologists, and other downstream users who may treat published interpretations as established"],"affected_objective":"Credibility of corpus-based claims about the prevalence, emergence, or distribution of literary forms before those claims influence publication, canon narratives, or curricular framing.","arm":"COMMON_P1","authority_safety":{"authorized_first_step":"With the research team's consent, conduct a read-only shadow audit of one completed corpus-literary project by reconstructing its attempted comparisons from notes, scripts, and version history; do not alter the manuscript or publication status.","decision_authority":"Researchers control their analyses and claim submissions; journal editors control publication and required status labels; curriculum or anthology bodies retain authority over downstream adoption.","excluded_actions":["Automatically rejecting an interpretation because it emerged exploratorily","Treating multiplicity adjustment as proof of literary importance or causal explanation","Opening a sealed confirmation corpus before the discovery family and confirmation rule are frozen","Changing publication, hiring, reputation, canon, or curriculum decisions during the shadow audit","Requiring access to confidential drafts without contributor consent"],"halt_rollback":"Halt if the ledger exposes confidential drafts, creates disputed attribution of analytic choices, or the confirmation corpus has already been consulted. Revoke access, preserve the original research record, and revert all pilot-only labels without changing underlying scholarship or editorial decisions."},"baseline":"A research team searches one corpus repeatedly across motifs, word lists, genres, periods, authors, editions, passage rules, and theoretical codings, then presents the most coherent pattern in the final article while null searches and alternative specifications remain scattered across notebooks or disappear.","candidate_id":"multiple_testing_discipline__literature_literary_theory__COMMON_P1","causal_chain":["A shared corpus permits many plausible contrasts among motifs, periods, genres, authors, editions, and interpretive codings.","Analysts inspect or revise numerous contrasts and specifications while developing an argument.","Chance variation or specification sensitivity makes at least one pattern appear unusually coherent.","The final article foregrounds that pattern without displaying the opportunity set that produced it.","Editors and readers interpret the selected pattern as if it were the product of a single planned inquiry.","The pattern may then shape literary-historical narratives, teaching, or canon judgments without independent confirmation.","A claim-family ledger restores the attempted opportunity set, applies the selected error-risk rule, labels the result as exploratory or confirmed, and reserves fresh texts for targeted confirmation before consequential use."],"cell_id":"multiple_testing_discipline__literature_literary_theory","consequence":"A specification-dependent coincidence can acquire the status of a stable literary-historical finding, misdirecting later criticism and potentially affecting publication, curricular emphasis, or accounts of a genre or period.","diversity_from_prior_proposals":"The candidate realizes multiplicity through the interaction of interpretive lenses, corpus boundaries, textual editions, and passage-coding decisions, with confirmation performed on sealed literary texts rather than through generic peer review or rerunning the same analysis.","experiment_id":"eoa_inverse_innovation_exp13_second_slot_policy60_20260806","intervention":"Create a Lens-and-Corpus Claim Ledger for corpus-assisted literary criticism. Before promoting a discovered pattern, the team defines the family of eligible motif, author, genre, period, edition, coding, and model comparisons; records every attempted look and material analytic change; selects an error policy that permits exploratory screening but requires a multiplicity-aware rule for promoted discoveries; labels each claim exploratory, adjusted-discovery, confirmed, replicated, or rejected; and tests any publication-leading claim against a sealed set of texts or editions using a frozen coding protocol. The ledger, including null and abandoned paths, accompanies editorial review.","mechanism_mapping":[{"counterfactual_removal":"Without the registry, null searches and abandoned corpus or lens choices can disappear, making the selected pattern seem less searched-for than it was.","mechanism_slug":"claim_registry","role":"Records each claim, corpus boundary, interpretive lens, specification, owner, result, and current evidentiary status."},{"counterfactual_removal":"Without a screening rule, the most attractive result can be promoted solely because it won an unreported search among many candidates.","mechanism_slug":"false_discovery_rate_control","role":"Provides a declared rule for selecting a manageable set of corpus patterns as leads while acknowledging that the selected set may contain false discoveries."},{"counterfactual_removal":"Without reserved texts, the same passages that generated a pattern also appear to confirm it.","mechanism_slug":"holdout_validation","role":"Keeps a sealed set of texts, editions, or passages unavailable during exploration and uses it only after the focal claim and coding rule are frozen."},{"counterfactual_removal":"Without a fixed follow-up, an exploratory pattern may enter the article's thesis or downstream teaching without crossing an explicit evidentiary boundary.","mechanism_slug":"confirmatory_follow_up","role":"Retests the selected pattern prospectively with a frozen contrast, coding manual, exclusions, and decision criterion."},{"counterfactual_removal":"Without robustness disclosure, a claim resting on one favorable tokenization, edition, genre boundary, or theoretical coding can look stable.","mechanism_slug":"multiverse_analysis_report","role":"Shows whether the literary-historical pattern persists across the defensible analytic choices already explored."}],"nearest_rivals":["Preregistration alone can freeze a planned reading but does not account for unregistered exploratory looks or determine the status of discoveries made during search.","Reproducibility protocol can rerun the corpus workflow but can faithfully reproduce a cherry-picked result without calibrating it to the wider claim family.","Close reading can bind an interpretation to textual details but does not by itself account for how many alternative texts, motifs, passages, or lenses were searched.","Robustness or multiverse reporting reveals specification dependence but does not itself reserve fresh evidence or govern promotion from exploratory lead to consequential claim."],"negative_tests":{"intervention_falsifier":"The intervention is undermined if ledger-listed exploratory patterns are promoted at the same rate and with the same editorial language as confirmed claims, or if sealed-text confirmation yields no better survival of promoted claims than the baseline workflow.","problem_falsifier":"The proposed problem is absent if the focal project evaluates one genuinely pre-specified corpus contrast with a fixed coding and analysis path, or if all searched patterns remain explicitly low-stakes prompts for interpretation with no confirmatory, ranking, reputational, curricular, or publication consequence.","risks":["Researchers may omit informal searches, making the recorded family artificially small.","A rigid family definition may suppress legitimate interpretive development or imply that criticism is reducible to statistical testing.","The sealed corpus may differ materially in genre, period, edition quality, or digitization, so failure may reflect transport rather than false discovery.","Formal adjustment may create correction theater while measurement error, confounding, corpus construction, or interpretive ambiguity remains.","The ledger may expose junior collaborators' abandoned ideas or create surveillance and attribution concerns.","Confirmation requirements may privilege large digitized canons and disadvantage scholarship on scarce or marginalized archives."],"strongest_counterevidence":"Many literary interpretations are valued for explanatory coherence, generative reading, or conceptual illumination rather than population-level empirical truth; if the target claim is explicitly of that kind, treating alternative readings as a statistical claim family would mischaracterize the scholarly objective."},"next_evidence_step":"For one consenting completed corpus-literary study, reconstruct at most 30 attempted motif, corpus, subgroup, edition, coding, or model choices from existing records; have two reviewers independently define the claim family; then apply the proposed labels and frozen confirmation rule to one selected claim using a small text subset demonstrably untouched during discovery. Record feasibility, reviewer agreement, missing-search evidence, and whether the claim's status changes; make no publication or curriculum decision from the pilot.","observable_state":"A published or submission-ready literary-historical claim highlights one striking motif or formal pattern, while project files reveal numerous alternative word lists, corpus cuts, author subsets, period boundaries, editions, passage codings, or models whose null and contradictory results are absent from the claim's evidentiary presentation.","prior_art_status":"UNSEARCHED","problem":"Corpus-assisted literary criticism can generate many plausible findings from the same texts by varying motifs, theoretical lenses, author groups, genre boundaries, periods, editions, coding rules, and models. When only the most compelling pattern is presented as a stable discovery, readers cannot see how many opportunities existed for a coincidental or specification-dependent result.","proposal_index":1,"remaining_contrastive_claim":"The distinctive claim is that a corpus-based literary finding should not become publication-leading or decision-relevant merely because it is textually interesting and reproducible on the searched corpus; its evidentiary status must reflect the full family of interpretive and analytic opportunities that produced it, and consequential promotion requires a frozen test on fresh textual evidence.","revision_record":{"claim_changes":["Initial candidate; no prior version or inherited claims."],"conceptual_changes":["Initial mapping of multiple-testing discipline to corpus-assisted literary-historical interpretation."],"evidence_changes":["No external or prior-art evidence consulted; evidentiary assertions are framed as testable hypotheses."],"operational_changes":["Specified a bounded read-only shadow audit, a claim ledger, status labels, and sealed-text confirmation."],"parent_version":null,"progress_targets_addressed":["Defined a concrete many-look literary-theory problem.","Preserved the claim-family, inventory, error-policy, status, confirmation, and discovery-record structure.","Separated authorized pilot activity from editorial and curricular authority.","Included problem and intervention falsifiers plus a bounded first evidence step."]},"schema_version":1,"structural_mapping":[{"archetype_element":"Define the claim family","domain_realization":"Enumerate the motif, trope, form, author, genre, period, edition, passage-selection, coding, and model contrasts that could support the proposed literary-historical conclusion."},{"archetype_element":"Inventory attempted looks","domain_realization":"Log searches, word lists, corpus cuts, exclusions, recodings, theoretical lenses, visualizations, models, and abandoned interpretive paths, including null and contradictory outcomes."},{"archetype_element":"Set the error-risk policy","domain_realization":"Permit broad exploration for generating readings, but require stronger control when a pattern will anchor a publication claim or influence canon and curriculum judgments."},{"archetype_element":"Apply a multiplicity-aware rule","domain_realization":"Use a declared discovery-rate rule for screening, robustness reporting across defensible specifications, and a sealed-text confirmation gate for promoted claims."},{"archetype_element":"Label claim status","domain_realization":"Mark each interpretation as exploratory, adjusted-discovery, confirmed on sealed texts, independently replicated, or rejected in the article and ledger."},{"archetype_element":"Confirm before costly action","domain_realization":"Freeze the selected contrast and coding protocol, then test it on previously untouched texts or editions before using it as an article's central empirical claim or downstream curricular premise."},{"archetype_element":"Preserve the discovery record","domain_realization":"Attach a versioned ledger of selected, null, contradictory, and abandoned analyses to the editorial record so reviewers can reconstruct the true search burden."}],"title":"Lens-and-Corpus Claim Ledger for Literary-Historical Discoveries","version":0}