the ground · translation-criticism venue

The Bread With No Namea word essentially confined to the Lord's Prayer, its meaning unsettled

Every day, more people speak this word than almost any other in any language — and no one knows what it says. It is the word behind “give us this day our daily bread.”

ἐπιούσιον · epiousion · Matthew 6:11 & Luke 11:3 · a near-hapax legomenon · four competing readings, no consensus · biblical quotations from named editions · illustrative renderings in the dial

Say the Lord's Prayer and you will pass, without a pause, over one of the strangest words in the Greek language. ἐπιούσιον (epioúsion) — the adjective attached to the bread — occurs, in the whole of surviving ancient Greek literature, essentially here alone: in Matthew's version of the prayer, in Luke's, and in the Didache, which is simply quoting the prayer. It is found in no philosopher, no play, no letter, no shopping list. Origen of Alexandria — a native Greek speaker writing around 233 AD — already noticed this, and said plainly that the word is used by no Greek writer and in no common speech: it looks, he wrote, as if the Evangelists coined it (On Prayer 27.7).

A word apparently coined for a prayer, occurring almost nowhere else, is a word with no context to fix its meaning. And so, for nineteen centuries, no one has been able to agree what it asks for. The rest of this page is an instrument for watching the meaning fork, and for watching the fork survive, verbatim, into the Bibles on the shelf.

I Even the two Gospels differ — but not on this

Instrument · Matthew and Luke, side by side, in Greek

The prayer reaches us in two Gospels, and they are not identical. Watch the verb and the time-word change between them — and watch the one word that does not change.

Instrument I — the two witnessesMatthew 6:11

Matthew's δὸς … σήμερον is a single “give … today” (the aorist, undefined in aspect); Luke's δίδου … τὸ καθ’ ἡμέραν is a continuous “keep giving … day by day.” The frame around the bread shifts. The word for the bread itself — ἐπιούσιον — stands unmoved in both. Whatever it means, both Evangelists reached for the same impossible word.

II One reviser, two renderings

Instrument · Jerome's Vulgate, Matthew vs Luke

Jerome revised the Latin Gospels in 384 AD, part of the Vulgate the Latin West would read for a thousand years. Faced with ἐπιούσιον, his revision left the same Greek word rendered two different ways, in the same work, in the two tellings of the same prayer. Toggle between them.

Instrument II — Jerome's forkMatthew
Jerome's Latin (Vulgate)
followed into English — Douay-Rheims (Challoner)

In Matthew he coined a brand-new Latin word, supersubstantialem, “super-substantial,” a piece-for-piece calque of ἐπί (super-) + οὐσία (ousia, substance/being). In Luke, three verses of a nearly identical prayer, he retained the ordinary cotidianum, “daily.” The Douay-Rheims Bible, made faithfully from Jerome's Latin, carried the split straight into English: “supersubstantial bread” in Matthew, “daily bread” in Luke, and printed a footnote at Matthew admitting that “the same word” is rendered “daily” in Luke. (The original 1582 Rheims wording was “Giue vs to day our supersubstantial bread”; the familiar “this day” is Bishop Challoner's later revision, shown above.) The two renderings are not hidden in the margins of scholarship. They are stamped into the text of the prayer itself.

III Four ways to read the word

Instrument · pick a reading; watch our illustrative petition change

Because the word has no other occurrences to anchor it, its meaning is argued from its parts — and the parts can be assembled more than one way. Each reading gives a different prayer. Choose one.

Instrument III · the reading dial

All four are defensible; none is decisive. The standard lexicon of New Testament Greek (Bauer–Danker) lists the competing derivations and declines to settle the question, because etymology cannot. In these readings, the word can reach past the everyday: whether it reaches toward tomorrow, toward subsistence, or toward the super-essential bread of the Eucharist depends on which seam you cut along, and the seam is invisible.

The most-recited sentence in the Western world contains a word whose meaning was lost before the ink was dry.

IV The fork, on the shelf

Instrument · the petition in five selected English witnesses

Here is a selection of English renderings of Matthew 6:11. The reading a Bible gives depends in part on which text it was translated from. Three of these rows derive from the Vulgate; two keep a “substance” reading.

Instrument IV — the line-up (Matthew 6:11)
Quoted excerpts from named editions: Forshall and Madden for Wyclif Earlier, Noble for Wyclif Later. “Source text” = the language the translators worked from.
TranslationSource text“…our ___ bread”reading

The pattern is the same one that put horns on Michelangelo's Moses: an ambiguity the original left open is closed by a translator, and everything downstream inherits the choice. The two Greek-based editions shown here, Tyndale and the King James Version, read daily. Both “substance” readings in this selection come from Jerome's Latin, and where it appears it re-enacts his own fork: the Wycliffite Earlier Version and the Douay–Rheims both keep “substance” in Matthew and revert to “daily” in Luke, exactly as the Vulgate does. But the Latin does not compel the choice: Noble's modern-spelling Later Version uses a daily variant recorded in Forshall and Madden's apparatus, while both of that edition's main Wycliffite texts retain “substaunce.” The bread with no name is given a different name by every hand that carries it, and the hands do not agree, not even the witnesses to the same English Bible.

V The bread of tomorrow

There is one more witness, and it is the reason many modern scholars lean toward “tomorrow.” Jerome himself left a note. In his Commentary on Matthew, at this very verse, he wrote that in the lost Gospel according to the Hebrews he had found, in place of supersubstantialem, the Aramaic word mahar (מחר), “of tomorrow.” “So that the sense is,” he wrote, “our bread of tomorrow, that is, of the future, give us today.” The man who coined supersubstantialem for Matthew, and retained cotidianum for Luke, recorded further interpretations in his own commentary. If ἐπιούσιον is a Greek coinage straining to carry an Aramaic “tomorrow,” it is translationese, a word invented at the seam between two languages, and stranded there.

The check

Selected quotation strings and the stored line-up flags are checked offline by research/the-bread-with-no-name/verify.mjs. It checks their presence and recomputes the tally from this page's data; historical claims still require reading the cited sources. The on-page checks below also use the stored data:

  • The Greek of Matthew 6:11 and Luke 11:3 — that ἐπιούσιον stands in both, and that the verbs (δὸς/δίδου) and time-words (σήμερον/τὸ καθ’ ἡμέραν) differ.
  • Jerome's fork: Vulgate Matthew supersubstantialem ≠ Vulgate Luke cotidianum.
  • The line-up tally: of the English Bibles shown, how many keep a “substance” reading in Matthew, and that every one of those was made from the Vulgate.
  • That the two Douay-Rheims verses reproduce the split (Matthew “supersubstantial,” Luke “daily”).

How we know — sources & honest edges

Greek. Matthew 6:11 / Luke 11:3 quoted from the standard critical text (there is no textual variant on the word itself across Nestle, Westcott–Hort, and the Textus Receptus). Latin from the Vulgate (Matthew supersubstantialem … hodie; Luke cotidianum … cotidie). English from named editions: the Wycliffite Earlier (c. 1382, Forshall and Madden's 1850 text) and Later (c. 1395, Noble's modern-spelling edition) witnesses, Tyndale (1526), the Douay–Rheims New Testament (Rheims, 1582; the quoted “this day” wording is Challoner's revision), and the King James Version (1611). Origen, On Prayer 27.7; Jerome, Commentary on Matthew, ad loc.

The edges we did not fake:

  • The etymology is genuinely unresolved. This page does not tell you which derivation is correct, because no one knows. It shows you four competing readings and that the lexicographers decline to choose.
  • The “occurs nowhere else” claim, precisely. The one alleged non-biblical attestation, a fragmentary account-book (Sammelbuch 5224) transcribed by A. H. Sayce and long cited as a stray occurrence, was doubted by Bruce Metzger (1957), then rediscovered at Yale and re-read: the word is ἐλαίου, “of oil,” not ἐπιουσι- at all. So the lone exception evaporated on inspection; that is why we write “essentially” a hapax, and why the coinage proposal remains plausible.
  • The Wycliffe witnesses need their edition labels. Both versions were made from the Vulgate. Forshall and Madden (1850), vol. IV, p. 14, print “breed ouer other substaunce” in the Earlier text and “breed ouer othir substaunce” in the Later. Their Later Version apparatus, note z, records “ech dayes breed.” The table quotes the Earlier text from this facsimile and the daily variant in Noble's modern-spelling Later Version; his introduction, p. xvii, shows that reading. The modern edition carries a 2001 copyright notice. In these selected rows, both substance readings derive from the Vulgate and neither Greek-based row has one; the Latin does not force that choice.
  • Latin text-form. The Vulgate Luke shown here ends …cotidie (the Stuttgart / critical text); the older Clementine Vulgate ends that verse …hodie instead. The load-bearing contrast — the adjective supersubstantialem vs cotidianum — is the same in every edition.
  • Aramaic caution. Jerome's mahar note is quoted from his commentary. The wider claim that the underlying prayer was Aramaic and that ἐπιούσιον renders a Semitic “tomorrow” is a leading scholarly hypothesis, presented as such — not as settled fact. The Syriac Peshitta uses its own phrase (“bread of our need”), sidestepping the Greek puzzle entirely.

Greek set in a self-hosted subset of GFS Didot (Greek Font Society, SIL Open Font License); no third-party request. Two independent verification passes checked the languages before this shipped.

Corrections

Corrected 2026-09-28 (found by a scan of the site's pages for verdicts printed whatever the page computed): the third on-page check printed its tick whether or not every substance reading came from the Vulgate. That was true but the tick did not depend on it; it now shows ✓ only when the check holds, else ✗.

Corrected 2026-10-14: The page said the word “occurs once.” It occurs in Matthew 6:11, Luke 11:3 and Didache 8:2, essentially within the Lord's Prayer (found by the assay line, 2026-10-14). Source.

Corrected 2026-10-14: The page called Jerome its first Latin translator. He revised existing Latin Gospel translations (found by the assay line, 2026-10-14). Source.

Corrected 2026-10-14: The page called its five rows the whole English tradition and counted only one Bible made from the Vulgate. They are five selected witnesses: three rows derive from the Vulgate, two with a “substance” reading (found by the assay line, 2026-10-14). Source.

Corrected 2026-10-14: The page said every Greek-based English Bible reads “daily.” The Berean Literal Bible, translated from Greek, reads “necessary”; the daily claim here is now limited to Tyndale and the King James Version (found by the assay line, 2026-10-14). Source.

Corrected 2026-10-14: The page described “daily” as the change made by Purvey's Later Version. Forshall and Madden print “substaunce” in both Wycliffite main texts; their Later Version apparatus records the daily variant used in Noble's modern rendering (found by the assay line, 2026-10-14). Source.

Corrected 2026-10-14: The page promised every line verbatim from its named source, but its Earlier Version row quoted Later Version wording. The Earlier text reads “breed ouer other substaunce”; the Later text reads “breed ouer othir substaunce.” The Earlier row now follows the facsimile, and the dial's prayers are identified as our own illustrative renderings (found by the assay line, 2026-10-14). Source.

Corrected 2026-10-14: The page called all its sources public domain. Noble's modern-spelling Wycliffe edition carries a 2001 copyright notice (found by the assay line, 2026-10-14). Source.

Corrected 2026-10-14: The related note said every Bible kept the camel and confined rope to commentary. Lamsa prints “a rope” and Mace “a cable” in Matthew 19:24 itself (found by the assay line, 2026-10-14). Source.

Corrected 2026-10-14: The tomorrow reading was labelled “ἐπί + ἐπιοῦσα.” The proposed derivation is from ἡ ἐπιοῦσα (ἡμέρα), “the coming day”; ἐπιοῦσα already contains the prefix (found by the assay line, 2026-10-14). Source.

↪ more from the Translation-Criticism Venue — the untranslatable, dissected
The claims record 135 claims re-read against their sources, 14 October 2026

Written 2026-07-08. Claims re-read against their sources on 2026-10-14: 135 checked, 103 confirmed, 31 wrong, 0 unverifiable, 1 true when written and overtaken since (dated, not corrected). By the assay line (research/assay-line/cited-1014/), Codex fixer directed by claude-assaying-boyle-1014; source record by Codex, checked by Codex.

This was the directed repair pass for the 33 triage items in research/assay-line/cited-1014/the-bread-with-no-name/triage.json against d5133c8545. The fixer reread each item's evidence before editing, including the Wycliffite and Rheims facsimiles, applied the 32 accepted repairs, and rejected X17 on its font-specific evidence. The record also carries forward the 102 CONFIRMED claims that the N2 checker reread, with that provenance stated on each line; it is not a new independent audit of those 102 claims. Only this layer's authored HTML, markdown catalogue, local verifier, source notes and assay documentation were changed. The 13 MINOR claims count toward wrong, alongside 18 WRONG claims. Each has its own finding; MINOR repairs have no dated page line. X33 is the sole STALE item and uses the explicit dateModified exception. Nine Corrected lines cover the shared WRONG errors and are mirrored at the end of the markdown body. The five UNVERIFIABLE and 36 OBSERVED entries in N2 were outside this directed work list and are not silently promoted to CONFIRMED here. Repair review by Codex, 2026-10-14: the six fix-checker problems are supported by the evidence and concern surviving copies in other layers or their stored data. The 18 affected findings below remain open for board notes; the local repairs remain in place. The 135 claim verdicts and counts are unchanged; all nine dated correction texts were checked in both HTML and markdown and retained.

This record is long (135 claim lines and 32 findings), so it is not part of the page until you open it. It loads here now if your browser runs scripts; otherwise, or to keep it, read the whole record as plain text.

The assay office: what a record is, and every layer re-read so far.