Ars Inquirendi

AI-generated conjecture · below the evidence/publication boundary

← All conjectures · Time, sky & calendars

The qibla learning curve

Status: Already answered

Status is derived only from the shepherd-authored triage/prediction data above -- community submissions and claims are a separate overlay and can never change it (see the participation panel below).

This is a proposed connection between two domains, generated by a language model. It is not an article and not evidence: it sits below the evidence/publication boundary. A quantitative prediction and a named kill-dataset are attached (when registered) so the claim stays falsifiable rather than merely evocative.

Claim (verbatim)

The qibla learning curve. Joins experience-curve economics to Islamic mathematical geography. In manufacturing, quality improves as a power law of cumulative output — the experience curve — because every unit produced teaches its producers. Every mosque is such a unit: it must face Mecca, its measurable orientation error is an output-quality score, and a region's dated mosques form a quality series spanning eight centuries. As zij astronomical tables and qibla-determination methods diffused, builders converged on the true bearing, so orientation error should fall as a power law in cumulative regional mosque construction — sacred accuracy as a learning curve. The conjecture predicts, in two independent regions (the Swahili coast; al-Andalus and the Maghrib), median errors shrinking from fifteen degrees or more before 1000 CE to a few degrees after 1400, a learning exponent in the classic range, and improvement arriving as regional step changes after documented arrivals of tables and instruments, rather than as uniform secular drift.

Prediction clause (verbatim)

For dated mosques with published measured orientations in two independent regions (the Swahili coast; al-Andalus and the Maghrib), the median absolute deviation from the true great-circle qibla falls from >=15 degrees before 1000 CE to <=6 degrees after 1400; error versus cumulative regional dated-mosque count fits a power law with learning exponent 0.2-0.5; and improvements arrive as regional step changes following documented arrivals of astronomical tables and qibla instruments, rather than as a uniform secular drift.

Kill-dataset (verbatim)

Kill: published archaeoastronomical orientation surveys of dated mosques (East African coast; Iberian and Maghribi corpora). No error decline over time, or errors behaving as an undirected random walk, kills it.

In the atlas

This conjecture is bridged, as an L1 lead, onto these Inferpedia subject pages.

Provenance

Run: Fresh agent generation · model: claude-fable-5

Generated by a fresh Fable-tier instance at maximum effort with generation-first blindness (no repo reads, no searches, no DB queries); title list supplied at launch, titles only, no verdicts or dossiers seen; prompt pre-committed in docs/GOAL_CONJECTURES_BATCH2_20260705.md (7e55eb8). Novelty unverified by construction.

Novelty / leakage triage

already answered in the literature

The connection is established historiography with a quantitative regional precedent: David King's foundational work documents mosque misalignment and the diffusion of increasingly accurate mathematical qibla methods from the 9th century (the conjecture names him), an Anatolian study already performs a regional multi-century measured-orientation-error analysis (1150-1590), and an early-period accuracy study covers 622-900 CE. The experience-curve formalization (power-law exponent vs cumulative regional output) and the two named regions were not located, but error-declining-with-method-diffusion is the standing account.

Predictions

No prediction registered yet.

Weigh in

No community feedback yet.

Add your take

Posted immediately (spam is removed). Community feedback is never an adjudicated verdict and never changes this conjecture's triage label or status above.

Working on this?

Sign in to claim this conjecture and let others know you're working on it.