One Thousand and One Conjectures

Every synagogue needed Torah scrolls, scrolls wear out fast under liturgical use, and worn scrolls were ritually retired rather than shelved — yet scrolls carry no colophons, so the best-dated manuscript population on earth contains almost none of them, and quantitative histories…

The Masoretes surrounded the biblical text with a lattice of counts and marginal notes — how many times a rare spelling occurs, which word marks a book's midpoint — that modern readers often treat as pious scholasticism. Functionally, it is error-detecting code:…

Marginal masoretic notes are not spread evenly over the biblical text; they thicken around rare spellings, look-alike sequences, and places where one verse could contaminate a near-twin. Copying errors are not spread evenly either — scribes slipped at characteristic traps like repeated…

In many Hebrew Bibles the masora is written as micrography — the notes' tiny letters drawn into carpets, dragons, and geometric borders. This turns a proofreading apparatus into ornament, and ornament obeys the drawing, not the data: lines must fill the shape,…

The Cairo Geniza is routinely treated as an unfiltered mirror of medieval life — whatever paper a community produced supposedly ended up in the chamber. But the institution existed for a reason: texts bearing God's name must not be discarded profanely, and…

The Geniza chamber was emptied in the 1890s with no stratigraphic record, and its contents were dispersed across dozens of libraries — a lost excavation, archaeologically speaking. But the dispersal was not random: batches removed together were sold and donated together, so…

Fustat's Jewish court produced marriage deeds, divorces, sales, and testimonies in quantity, and writing such instruments was a controlled profession — a small circle of court scribes whose hands recur across documents and can be identified. Private letters, by contrast, could be…

In Fatimid chancery practice, respect was spatial: petitions to the mighty left wide blank margins and generous space above the text — deference measured in wasted paper. Jews of Fustat wrote petitions in that idiom, but the conjecture is that the convention…

What a genizah receives is what people had in hand when they cleared their papers: overwhelmingly letters they had received, not the ones they sent, which lay in other people's houses across the Mediterranean — mostly in towns with no surviving genizah.…

Geniza letters are mostly Judaeo-Arabic — Arabic written in Hebrew letters — but they switch into Hebrew in patterned ways: openings, blessings, scriptural tags, condolences. The standard reading treats this as religious reflex; the conjecture is sharper: the density of Hebrew in…

Letter-writers do not invent their openings; they reproduce the current epistolary fashion, and in the medieval Islamicate world that fashion was set by chancery practice, which changed with dynasties and administrations. Jewish letter-writers in Fustat absorbed those conventions through constant contact with…

The Geniza preserves the waste of elementary education: alphabet practice sheets, children's copying exercises, teachers' models — writing produced in the first weeks of instruction, at a rate proportional to how many children entered instruction at all. Literary texts, by contrast, were…

Jewish legal deeds required witnesses to sign, and a signature is an involuntary literacy test: a practiced writer signs fluently, while a man who can barely write draws his name in careful, disconnected strokes — a distinction plainly visible on the page…

A responsum began life as a document: a letter of question sent to an authority, with named parties, places, sums, and local circumstances, and an answer dispatched back. It entered posterity as literature — collected, copied, eventually printed in volumes where the…

The Geniza preserves two kinds of book-lists: inventories of private libraries drawn up for estates, and the working lists of booksellers and their auctions — two censuses of the same book culture taken at different points in a book's life, the shelf…

Geniza book-lists let us count how many books individual medieval households actually owned — a number usually guessed from anecdote. Modern collections and wealth alike tend toward heavy-tailed distributions: many small holders, a few enormous ones. The conjecture is that medieval Jewish…

Jewish communities dated things in two registers: ink — colophons, deeds, letters — and stone, the tombstone epitaph. Both had to choose an era, Anno Mundi or the Seleucid count among others, and eras went in and out of fashion. The conjecture…

Geniza book-lists record actual prices paid for books, and colophons occasionally record scribal fees, giving the raw material for a genuine price series of Hebrew books across centuries — something no narrative source provides. Over those same centuries paper spread and cheapened…

The Black Death killed scribes, patrons, and owners alike, and for most manuscript cultures its effect on book production is inferred rather than measured, because dated books are too few. Hebrew colophons are dense enough in the mid-fourteenth century to see the…

An astronomical table carries a radix — a set of starting positions computed for a chosen epoch date, the zero point from which all its numbers count. The surprising connection is that radix dates are political artifacts, not scientific ones: they cluster…

Sexagesimal digits were written in Arabic abjad letter-numerals, in Hindu-Arabic ciphers, and in various Latin conventions, and each system has its own characteristic confusions — which letter melts into which when a diacritic drops, which cipher flips into which under a tired…

A copied astronomical table inherits every error of its exemplar; a recomputed table refreshes them. Whether a given copy was recomputed or merely transcribed is usually treated as scribal temperament, but the surprising connection is that it is infrastructure: recomputation requires a…

An incipit — the opening words by which a medieval text was cited, catalogued, and sold — behaves like a fossilizing tag, but not all genres fossilize at the same rate. The surprising connection is that astronomical incipits mutate far more slowly…

The great medical handbooks were repeatedly abridged into epitomes, and the epitomes were then stuffed with new matter — this much is familiar. The surprising connection is that the process is a regular two-phase pulse with a stable ratio: first-generation epitomes cut…

Middle English medical writing is routinely described as translated 'from the Latin tradition', as if the whole Latin corpus fed it. The surprising connection is that the vernacular tapped almost exclusively the short-text layer — epitomes, compendia, and extracts — rather than…

The Almagest, the central book of ancient astronomy, rarely travels alone in its manuscripts. The surprising connection is that each translation version travels with its own characteristic convoy of satellite texts — introductions, canons, star-lists — stable enough that a codex's translation…

The standard picture has Hindu-Arabic numerals entering Latin Europe through algorism treatises — books that explain the new arithmetic. The surprising connection is that the zero sign appears earlier and more consistently in Latin astronomical TABLES than in Latin arithmetical texts: the…

Mappae mundi look like pictures copied from pictures, and their genealogies are usually drawn accordingly. The surprising connection is that their place-name errors are textual, not graphic: omissions cluster in runs of names that are adjacent in written geographies but scattered on…

Copyists of world maps worked outward from the ideologically loaded centre, and their attention decayed with radius. The surprising connection is that copying fidelity on mappae mundi is therefore a radial function: error and omission density rises with distance from the map's…

Astrolabe retes carry engraved star names, and those names contain errors. The surprising connection is that the errors match the copyist errors of specific manuscript star-list recensions: engravers worked from written lists at the bench, not from other instruments or from the…

Medieval sundials carry engraved hour-lines and hour labels, and computus manuals — the church's timekeeping textbooks — supplied the vocabulary. The surprising connection is that dial epigraphy tracks textbook succession: when a new computus text achieves dominance in a region's schools, the…

The documentary horoscopes preserved on Greek papyri were computed from astronomical tables, and we can now recompute which ones. The surprising connection is that provincial practitioners used systematically OBSOLETE tables — one to two parameter-generations behind the best contemporary theory — with…

In Greco-Egyptian astronomical papyri, prose and numbers did not switch languages together. The surprising connection is that the tabular matter is the conservative organ: layouts, notational habits, and month-name treatments of the Egyptian (Demotic) tradition persist in tables for generations after the…

Pharmacological recipe collections grew by accretion as owners added what they acquired — this is agreed. The surprising connection is positional: the shared ancestral core of a collection sits at the HEAD, while witness-specific additions cluster at the tail and margins, because…

Our canon of ancient medicine is defined by the great treatises the medieval codex tradition chose to preserve. The surprising connection is that the ground-level papyrus record is dominated by a different textual form entirely: catechistic question-and-answer texts, definition lists, and short…

Provençal and Iberian Hebrew astronomical tables and neighbouring Latin tables drew on the same Andalusi Arabic heritage. The surprising connection is that the Hebrew line functioned as a parameter refrigerator: it preserved Andalusi parameter vintages in working circulation for a century or…

The Toledan Tables were the most copied astronomical dataset of the Latin twelfth and thirteenth centuries, and their witnesses contain identifiable corrupt entries. The surprising connection is that these corruptions form geographic clades: bundles of shared errors mark descent from common exemplars,…

Translation between the scientific languages was not one flow but a ladder of genre-specific speeds. The surprising connection is the ordering: calendrical and computus material crossed a language frontier within roughly a generation, practical table canons within about two, and theoretical astronomy…

A mappa mundi has two inheritable layers: the drawn and written content, and the construction geometry laid down before any ink — compass-hole centres, ruled circles, division angles. The surprising connection is that these layers travel separately: physical workshop templates passed between…

The precision of a copied astronomical table was set not by science but by carpentry: the ruled grid of the page. When a scribe copied a table into a grid ruled for fewer or narrower columns than his exemplar used, the final…

Astrolabe plates and portable dials are engraved for specific latitudes, and it is natural to assume those are the latitudes of the cities where the instruments were made and used. The surprising connection is that they are BOOK latitudes: plate values cluster…

Some scientific content circulated in both prose and mnemonic verse — versified computus, versified algorism, versified regimen. The surprising connection is that the verse versions systematically out-diffused their prose sources into regions of thin book culture, so the verse-to-prose witness ratio of…

The twelfth-century Sicilian translations of Greek astronomy directly from the Greek were textually superior to their Arabic-Latin rivals, yet barely circulated — the fact is known, the shape of their survival is not. The surprising connection is that where the Greek-Latin versions…

A zij — the astronomical handbook built around tables — is modular: syzygy and eclipse tables, planetary equations, a star catalogue, geographical tables. The surprising connection is that the modules cross language frontiers in a canonical order, repeated at every frontier: eclipse…

Medieval books were sold, shelved, inventoried, and censored by their first lines. The surprising connection is that alchemical texts exploited this: they circulated under medical-looking incipits — openings promising waters, oils, and the conservation of the body — at rates far exceeding…

Distances cut into itinerary stones and painted into map legends ought to be independent measurements of the world. The surprising connection is that they share scribal corruptions with the written itinerary tradition: numeral errors born on papyrus and parchment were carved into…

Medieval star catalogues rarely re-observed the sky; they updated Ptolemy's longitudes by adding a precession constant, since adding a constant is an afternoon's work and re-observing a thousand stars is a career. The surprising connection is that the added increments form a…

The uroscopy wheel — the circular diagram of graded flask colors that is medieval medicine's most recognizable image — normally appears embedded in a treatise, and editors assume image and text descended together. The surprising connection is that the wheel travels on…

The university pecia system rented out exemplar quires for piecework copying and is credited with standardizing the scholastic book. The surprising connection is that it moved prose but choked on tables: astronomical codices produced under pecia show split ancestry, their prose affiliating…

Monastic forgery of Anglo-Saxon charters is usually pictured as a slow pious habit; this conjecture joins it instead to legal procedure, the sequence of royal challenges that made written title the winning move in court. The claim is that forgery was a…