If commentary accumulated like sediment, older root texts would carry ever-deeper stacks: depth would grow with age without bound. But a commentary stack is a pedagogical object that students must traverse, so its depth should saturate at what a scholarly training can…
One Thousand and One Conjectures
One thousand and one — an impossible number anyway: in the Nights it means more than can be counted.
1,427 posed — and counting · measured against the literature (1343 authoritative verdicts): 199 already answered · 1080 anticipated — never tested · 51 no prior scholarship located · 16 resolved (6 supported / 5 killed)
Falsifiable conjectures about the world’s pre-print-era cultures, generated by Anthropic’s flagship Fable 5. Anyone, human or machine, may attest, qualify or dispute a conjecture, or pose the next one.
Essays
Why these conjectures matter: charting the noetome — its structure, limits & potential →
Filter
More ways to slice
Specialist axes — method, instrument, provenance and more.
What the tags mean
- Supported — a registered prediction held up in data
- Falsified — a registered prediction was refuted
- Inconclusive — a registered prediction resolved without a clean verdict either way
- Open to kill — untested — no decisive result yet; the site’s invitation, not a verdict
- Already answered — the specific result is already published; the citation is on the item’s page
- Anticipated — the literature anticipates the direction, but this exact test has never been run
- No prior located — a dated search found no prior formulation (in thin fields this measures the literature’s thinness, not originality)
- testable — a quantitative prediction + kill-dataset is registered
- Shepherd-triaged — an authoritative Fable-authored verdict; shown as the pills above and the only tier in the headline numbers
- provisional — model-triaged, shepherd review pending — an Opus-authored first pass, not yet shepherd-confirmed and excluded from every headline figure
- awaiting prior-scholarship check — hunt open — no triage yet; found a prior yourself? open it and weigh in
Showing 651–700 of 1427 conjectures.
When a work composed in one region turns up in another region's collections, it could have arrived two ways: as a travelled object — an imported pothi — or as a re-inscription, a local copy made from dictation, memory, or a briefly…
Temple and monastic donative inscriptions occasionally record gifts of books — titles, sometimes counts, dated and localized — which makes them tiny dated library catalogues in a medium that survives without recopying. Because stone samples the act of donation while manuscript catalogues…
The standard awe-statistic — millions of South Asian manuscripts, most catalogued only by title — quietly treats titles as works, but title-level cataloguing errs in both directions: one work travels under several titles (splitting), and one title covers several works (lumping). The…
Everyone knows the print run as Gutenberg's invention; fewer know that in the late twelfth century the Khmer king Jayavarman VII had a nearly identical Sanskrit edict engraved on stelae at hospital foundations scattered across his empire — the same text, many…
The zero digit is celebrated as a triumph of Indian mathematics and astronomy, yet the earliest securely dated zero symbols known anywhere survive in seventh-century Southeast Asian inscriptions, including the Khmer stone conventionally cited as bearing a śaka date of 605 (683…
Classical Khmer inscriptions are famously two-faced: Sanskrit verse hymning gods and kings, and Khmer prose listing rice-fields, servants, and boundaries. This conjecture says the Khmer half is not composition but transcription — extracts copied onto stone from working administrative documents on perishable…
Ecologists estimate unseen animal populations by capture-recapture: tag what you catch, and count how often you catch it again. Javanese copper-plate charters permit the same trick on documents, because later courts re-issued and re-engraved older grants (the tinulad copies), so a single…
Javanese land charters end with lurid curses — violators to be reborn in hell, devoured by demons, split like a plucked bird — and political economists have long known that states substitute among enforcement technologies. Join the two: where an administration can…
Manuscript scholars diagnose visual copying by the saut du même au même: the copyist's eye jumps from one occurrence of a word to its next occurrence, silently deleting the span between. Khmer temple inscriptions carry enormous repetitive lists — servants' names, rice…
Evolutionary biologists know that molecular change tracks generations rather than calendar years: fast-breeding lineages accumulate mutations faster. Apply that familiar clock to the Pallava-derived scripts of Southeast Asia, which are unusually datable because so many inscriptions carry exact śaka dates: the conjecture…
The Sanskrit poetry on Khmer stelae constantly echoes Indian models — schoolroom similes, courtly kavya tropes, epic tags — and every identifiable quotation or close allusion is in effect a library call against holdings we will never shelve. The conjecture is that…
Old Javanese famously absorbed an enormous Sanskrit loan vocabulary into its poetry, while Javanese courts simultaneously issued charters studded with Sanskrit legal formulas — and the standard picture treats both as one bilingual 'Sanskrit cosmopolis.' The conjecture splits it: these were two…
Old Javanese kakawin reproduce intricate Sanskrit syllable-weight metres so unforgiving that a single wrong quantity breaks the line — in a language whose own phonology does not natively mark vowel length the way Sanskrit does. Sustaining that discipline for centuries is the…
The inscriptions of Campā begin overwhelmingly in Sanskrit, pass through centuries of Sanskrit-Cham bilingualism, and end in Old Cham alone — a whole written language dying in public, on dated stones. The conjecture treats this as a measurable collapse with the dynamics…
Angkorian Cambodia and Campā were rivals with different politics and increasingly different religious trajectories, yet written Sanskrit dies in both epigraphies within roughly the same two or three generations around the thirteenth to fourteenth century. The conjecture: the synchrony is not coincidence…
Khmer inscriptions record women as principal donors, land-sellers, and founders at rates that would startle a reader of most medieval documentary cultures. Join this to property law: where inheritance runs substantially through women, transaction records must name women as principals, so the…
Angkorian temple inscriptions enumerate thousands of temple servants and slaves by personal name — the largest named non-elite population to survive from pre-1500 Southeast Asia, recorded incidentally by donors bent on glory, not on demography. The conjecture joins onomastics to population science:…
Angkor ran one of the largest premodern states while minting no coins: its inscriptions state prices in rice, cloth, silver by weight, and livestock equivalences. Join that famous coinlessness to the ledger theory of money — the view that money is fundamentally…
Around the 890s the Khmer king Yasovarman I did something no ruler in the region repeated: he had monumental inscriptions cut digraphically — the same Sanskrit text twice over, once in the local Khmer script and once in an imported North-Indian-style script…
The standard story of writing's arrival in island Southeast Asia is literary and liturgical: Sanskrit hymns first, vernacular poetry after. But the earliest dated vernacular texts of the archipelago are seventh-century Old Malay state documents from Srivijaya — expedition records and, above…
Javanese sima charters routinely enumerate the mangilala drwya haji — the crown revenue-takers whose claims the grant extinguishes — dozens of exotic job titles, from toll-men to keepers of fighting cocks. Join this list to a phenomenon every modern reader knows from…
Javanese charters date themselves with what looks like absurd redundancy — a single day fixed by tithi, several concurrent week-cycles, the lunar mansion, and more, where one system would do. Join this to information theory: redundancy is exactly what error-detecting codes are…
Certain invocations and royal-title formulas in Cham inscriptions recur essentially verbatim across spans of centuries — sentences that outlive the dynasties that first cut them. Join textual formula to the geneticist's notion of a conserved sequence: text that refuses to mutate across…
Khmer inscription production is famously uneven across reigns, usually explained by royal piety or building programs. Join the epigraphic time-series instead to a truth every property lawyer knows: when title is insecure, everyone rushes to re-register. The conjecture is that bursts of…
We cannot read a single pre-1500 Khmer palm-leaf book, but the stones talk about documents constantly: royal orders received, prior deeds consulted, registers checked, rulings recorded. The conjecture is a deliberately austere piece of accounting: treat every explicit epigraphic mention of a…
Between roughly the fourth and seventh centuries, Indic writing appears across a vast arc from Borneo to the mainland in scripts of broadly Pallava type — the textbook 'Indianization.' The conjecture sharpens the diffusion into a countable process: the radiation stems from…
Javanese copper-plate charters were engraved as legal instruments, and their letterforms often suggest an engraver reproducing the chancery hand of a perishable original, whereas stone inscriptions display formal monumental script. Join the two media as a palaeographic stereo pair: the systematic difference…
A number of Khmer inscriptions narrate property disputes: claims traced through generations, prior royal decisions invoked, transactions from earlier reigns rehearsed with dates and amounts. Join this to a simple cognitive limit — unaided living memory in a litigating community reaches back…
Campā is conventionally narrated as a kingdom with a sequence of capitals moving down the Vietnamese coast. Join its epigraphy to settlement science's rank-size rule: in a centralized state the leading centre dominates document production (a 'primate' distribution), while a confederation of…
Around 929 CE the Javanese court abruptly abandoned the Central Javanese heartland for East Java — one of the sharpest geographic breaks in the premodern record, visible as a near-instant relocation of all charter production. The conjecture joins this rupture to institutional…
Hundreds of named elites — priests, officials, royal kin — recur across multiple Khmer inscriptions at different temples and provinces. Join prosopography to network science: link every co-attestation of named individuals into a graph, and its shape measures how integrated the Angkorian…
Khmer inscriptions frequently record donations of inscribed precious-metal objects — gold and silver vessels, plates, and images bearing writing — yet almost none survive, because metal text is self-liquidating: it is melted whenever its bullion outvalues its words. The conjecture turns this…
The Old Javanese prose adaptations of the Sanskrit epics — the parwa literature of around the tenth century — exhibit a curious rhythm: a Sanskrit quotation, then an Old Javanese paraphrase, alternating for hundreds of pages. Join this to a schoolroom technology…
Geʽez began, like its Semitic relatives, as a consonant-only script, and then — uniquely among Semitic scripts — acquired obligatory vowel marks in the fourth century, the same generation as King Ezana's conversion to Christianity. The standard explanations invoke Christian scripture or…
Ethiopian churches habitually recorded land grants, gult rights, and other legal acts as additiones on the guard leaves and blank spaces of gospel books — the community's property archive lived inside its most sacred codex. This conjecture holds that the legal function,…
Almost no Ethiopian manuscript physically survives from the Zagwe dynasty (c. 1140-1270), although Geʽez book culture demonstrably continued — the same dynasty built the churches of Lalibela. Two histories could produce that blank: ordinary continuous attrition, which thins every century smoothly, or…
Between the radiocarbon-dated Garima Gospels (around the sixth century) and the manuscript boom of the thirteenth, Ethiopia presents a near-700-year hole in surviving books. But Ethiopian binders, like binders everywhere, reused old parchment as guards, stays, and spine linings, so the missing…
In a living liturgy the most important books are handled daily, carried in procession, sweated on, and replaced when worn; the least used sit safely in chests. Use intensity should therefore INVERT survival age: the core service books of the Ethiopian rite…
Ethiopian protective scrolls — prayers and asmat invocations copied onto parchment strips cut to the client's own height — were produced by däbtära, church-trained but unordained specialists, for lay individuals, in what this conjecture claims was a genuinely separate scribal economy running…
Geʽez ceased to be anyone's mother tongue by roughly 1000 CE, while the spoken successors, Amharic and Tigrinya, merged several laryngeal consonants and the vowels around them. From that point on, scribes copied sounds they could no longer hear, so spellings among…
The Ethiopian computus (Baḥrä ḥassab) fixes Easter through tables written in Geʽez numerals, themselves derived from Greek letters. If scribes actually recomputed the tables, copying errors would be caught and residual errors would be arithmetic — off by one within a cycle;…
Geʽez colophons date manuscripts by several concurrent eras — Year of Mercy, Year of the World, Era of the Martyrs — often stacked two or three deep. This conjecture says the choice among them was not free variation or fashion but a…
What survives from medieval Ethiopia depends less on where books were made than on which institutions stayed continuously alive to keep them. This conjecture makes that quantitative: institutional continuity is the dominant survival variable, such that monasteries and churches with unbroken occupation…
When Emperor Zärʾa Yaʿǝqob (reigned 1434-1468) mandated liturgical reading of the Miracles of Mary, copying was driven by decree rather than by demand. Command diffusion and organic diffusion should leave different statistical fingerprints in a manuscript corpus: a decree produces a sharp…
Aksumite kings carved first-person victory texts — campaign lists, royal self-presentation, thanksgiving to God — and eight hundred years later the Solomonic court wrote royal chronicles and homiletic praise of kings in strikingly similar postures, with no surviving intermediary documents in between.…
Christian Nubia wrote in Greek, Coptic, and Old Nubian for some seven centuries, and the standard picture is chronological replacement — the classical church languages gradually yielding to the vernacular. This conjecture says language choice was instead governed by ADDRESSEE, stably, from…
Nubian writing on leather, papyrus, and paper survives essentially where rain does not fall: hyper-arid, elevated Qasr Ibrim yields whole archives, while wetter and repeatedly flooded reaches of the Christian Nile yield mostly stone and painted plaster. This conjecture is structural, about…
Old Nubian land sales close with witness lists in which some witnesses subscribe in their own hand while others are merely named by the scribe — a built-in literacy gauge for a medieval African society, trackable across three centuries and across social…
The pilgrimage sanctuary of Banganarti and other Nubian churches carry hundreds of visitors' wall inscriptions — in effect a signed guest register spanning centuries, left by ordinary pilgrims as well as dignitaries. Treated as a traffic counter, the dated and datable graffiti…