AI-generated conjecture · a wager, not a finding
← All conjectures · Central & Inner Asian texts
Forty thousand scraps, a few hundred books
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).
Claim (verbatim)
The German Turfan expeditions of 1902-1914 carried tens of thousands of manuscript fragments back to Berlin, of which the Old Uyghur portion is among the largest single deposits of a medieval Inner Asian literature. But a fragment is not a book. The Berlin holdings are being catalogued piece by piece in the VOHD Alttuerkische Handschriften series and the digital Turfan archive, and the striking fact of that catalogue is the ratio: thousands of fragments resolve into a modest roster of distinct works, most of them Buddhist, many known from a single scrap and a handful from dozens of overlapping copies. The mechanism is a shredding filter — a ruined monastery library that has been through fire, damp, reuse, and excavation delivers its texts as confetti, so the number of physical items wildly overstates the number of works while understating how many works vanished entirely; the singletons are the survivors of a class whose other members left no scrap at all. The distribution of copies per work is therefore the diagnostic: a few liturgical staples in many copies, a long tail of once-attested works, and — inferable behind them — a lost majority that fell below one surviving fragment. Prediction: reducing the catalogued Old Uyghur Turfan fragments to distinct works, the fragment-to-work ratio will exceed ten to one, and at least half of the distinct works will be attested by a single fragment (primary clause: the majority-singleton result; a copy-count distribution without a dominant singleton class falsifies it). Kill: the VOHD Verzeichnis der Orientalischen Handschriften in Deutschland, Alttuerkische Handschriften volumes, joined to the Digitales Turfan-Archiv (Turfanforschung, BBAW) fragment inventory.
Prediction clause (verbatim)
Prediction: reducing the catalogued Old Uyghur Turfan fragments to distinct works, the fragment-to-work ratio will exceed ten to one, and at least half of the distinct works will be attested by a single fragment (primary clause: the majority-singleton result; a copy-count distribution without a dominant singleton class falsifies it).
Kill-dataset (verbatim)
Kill: the VOHD Verzeichnis der Orientalischen Handschriften in Deutschland, Alttuerkische Handschriften volumes, joined to the Digitales Turfan-Archiv (Turfanforschung, BBAW) fragment inventory.
Nobody has run this test. The kill-data is named above. If you can run it — or you know the paper that already settles it — claim the kill or submit the prior scholarship. Kills and prior scholarship are credited here, by name, as they come in.
Provenance
Run: Fresh agent generation · model: claude-fable-5
Fresh blind generation by claude-fable-5, 2026-07-17, steppe/Inner Asia wave 2 weighted by inferred textual production rather than survival: every item grounded in real works, testimonia, catalogues, and editions of loss (dead-language corpora, single-deposit libraries, founder-canons, epigraphic provinces, singleton codices, and languages surviving as glosses in other tongues), with no fabricated citations, and deliberately disjoint from the fable-w17 Inner Asia wave and the 2026-07-16 SE-Asia/Inner-Asia wave. Eight candidates were dropped for duplication (Sogdian Ancient-Letters postal sample, Tocharian A/B economics, Khitan epitaph-survival, Maitrisimit-centred Uyghur, the Orkhon royal monuments, the Secret History's embedded poetry, Phags-pa, and Tangut script-origin/printing/bureau); two further candidates (a Tangut-decipherment item and a Tocharian no-testimonia item) were set aside to hold the wave to seventeen non-overlapping seams.
Novelty / leakage triage
anticipated in the literature — this exact test has never been run
The VOHD Alttuerkische Handschriften volumes and the Digitales Turfan-Archiv catalogue the Old Uyghur fragments and identify their works, so the raw material for the test is in print. But the cataloguing is still proceeding across many fascicles, and no published study reduces the whole corpus to distinct works and states the corpus-wide fragment-to-work ratio or the singleton fraction the primary clause predicts. The immediate neighbourhood is heavily worked; the specific copy-count distribution as posed has not been computed corpus-wide.
- Verzeichnis der Orientalischen Handschriften in Deutschland (VOHD), Alttuerkische Handschriften series (Stuttgart: Franz Steiner)
- Digitales Turfan-Archiv, Turfanforschung (Berlin-Brandenburgische Akademie der Wissenschaften)
Predictions
No prediction registered yet.
Weigh in
No community feedback yet.
New here? Create an account first
Create an account or sign in and your feedback is tied to you — you can track it, get replies, and claim this conjecture so others know you’re working on it. Prefer not to? Just leave your take below as a guest — only the name you type is shown.
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.