# Global Religious Counts Are Versioned Model Outputs ## Claim A global religious-movement total is interpretable only when its category, date, denominator, source mix, overlap rules, forecast status, and revision history travel with the number. <!-- atomicity: 1 idea ✓ --> ## Scope & conditions - Holds for modeled or aggregated totals whose membership boundaries are disputed or externally assigned. - A census count has different limits but still needs a dated question, response category, coverage statement, and denominator. - A later calendar year does not convert an old projection into an observation; the production method determines its evidence class. - Broad categories can be useful when clearly named. The claim does not require one correct boundary. ## Evidence Three completed reference-work reviews converge on the rule. *The Cambridge Companion to Pentecostalism* treats definition as an unresolved scholarly task and checks current totals against later methods. *The Encyclopedia of Protestantism* shows Core and Wider Protestant categories producing different totals and retains a 2000-based forecast that must not be read as a current observation. NIDPCM counts three waves and many subcategories while correcting overlaps and projecting growth, making the model visible rather than eliminating it. ## Links - [[3. Resources/Permanent Notes/Historical Ancestry Does Not Establish Present Religious Identity]] — because extends: classifications need current evidence at the person or institution level before they can support an aggregate. - [[4. Archives/Book reviews/The Encyclopedia of Protestantism]] — because evidence-for: its Prompt 10 preserves the forecast-versus-observation rule and anchors it to the source's dated statistics. - [[3. Resources/Permanent Notes/A Verified Fraction Without a Comparison Baseline Cannot Be Scored]] — because contrasts: totals require denominators and category rules; fractions also require a comparison standard. - [[1. Projects/llm-wiki-theology/wiki/concepts/pentecostalism]] — because applies-in: current movement summaries should carry dated categories rather than a freestanding global total. - [[3. Resources/MOCs/Book Review Synthesis MOC]] — applies-in: the MOC lists this note under Reference-work comparison added 2026-08-28. - [[3. Resources/MOCs/Faith & Theology MOC]] — applies-in: the MOC lists this note under Synthesis additions. ## Tests / falsifiers - Recalculate one published Pentecostal total under narrower self-identification and broader family-resemblance rules. If the number and interpretation remain unchanged, the category dependency is overstated for that dataset. - Trace a forecast cited after its target year. If later writers preserve its forecast label and compare it with observations rather than treating it as fact, the version metadata is functioning correctly. ## Source - [[3. Resources/Literature Notes/Pentecostal Reference Works - Categories Counts and Authority - Literature Note]] — comparative synthesis of five reviews, with three independent count-method cases.