The thegn's deed sleeps in the minster's chest

Claim (verbatim)

The great tenth-century diploma was, in the main, a layman's document: the mid-century kings booked land overwhelmingly to named individuals - their ministri, thegns, and household men - not to churches. Yet essentially every surviving text reaches us through an ecclesiastical archive, because a lay beneficiary's diploma survived only if the estate, travelling with its title-deed as English conveyancing required, later came to a church that endured. Two independent filters therefore multiply: the estate had to pass to a church at all, and that church had to be one of the few dozen whose muniments survived. No unbroken lay line transmits a single pre-Conquest diploma - the earliest continuously-kept lay muniments in England begin generations after the Conquest - so the lay charter chest is a wholly lost archive class, glimpsed only in narratives like the Fonthill letter's story of deeds produced, forfeited, and handed on. The surviving corpus is thus doubly unrepresentative: produced largely for laymen, preserved exclusively by churchmen, and thinned at each transfer. The measurable signature is the dissonance itself - a beneficiary population dominated by individuals set against a custodial population that is one hundred percent institutional - and its size fixes the multiplier between the diplomas we have and the diplomas the tenth-century laity actually held.

Prediction clause (verbatim)

Prediction: among Electronic Sawyer royal diplomas dated 925-975 inclusive (excluding writs, wills, and leases), grants to named individual persons will constitute at least 0.55 of the set (primary clause: the individual-beneficiary share at or above 0.55; the verdict follows it), explicitly lay individuals alone at least 0.40, while the custodial archive is an ecclesiastical institution for at least 0.95 of the same set. Disambiguation pinned: the beneficiary is read from the catalogue's headline description of the dispositive act; clergy receiving in a personal capacity count as individuals and are also reported separately; joint or serial beneficiaries count by the first-named recipient. Coverage guard: the test voids if the beneficiary is unidentifiable for more than 10 percent of the set.

Kill-dataset (verbatim)

Kill: the Electronic Sawyer date, beneficiary, and archive fields for the diplomas of 925-975 - a beneficiary-type census crossed against custodial-archive type.

Novelty / leakage triage

anticipated in the literature — this exact test has never been run

That the mid-tenth-century royal diploma was predominantly a layman's grant preserved exclusively through ecclesiastical muniments is stated doctrine - Kelly's essay on lay society and the written word argues exactly this survival filter, with the Fonthill letter as the canonical glimpse of lay deed-keeping, and Keynes's diploma studies note the thegnly beneficiaries of the mid-century kings - but the pinned beneficiary-type census of the 925-975 diplomas (individual share >=0.55, explicitly lay >=0.40, crossed against >=0.95 ecclesiastical custody, with first-named-recipient and personal-capacity rules) has not been run over the catalogue; printed 'most' does not pin these thresholds.

Predictions

No prediction registered yet.