Sources

The Greek base text. Every reading starts here; every claim is keyed to a Greek word. Morphology via MorphGNT; lemma, part of speech and the Louw–Nida semantic-domain codes for each Greek word via MACULA Greek (Clear Bible, CC BY 4.0).

The Aramaic spine, consulted for every token: it gives each sentence its order and tenses and supplies the Syriac form shown throughout, as a witness to the Aramaic rather than the original. Electronic text by ETCBC; morphology from SEDRA 3.

The Curetonian and Sinaitic Old Syriac witnesses. Burkitt’s 1904 text is recovered where legible and shown as a provisional witness, never backfilled. The Sinaitic palimpsest edition of Lewis (1910) has been re-read from the page and woven into the interlinear as an Old Syriac band on 731 verses, placed to within about a verse and shown for comparison. Where it agrees with the Peshitta reading it also enters the consensus as an independent witness; it governs no word and moves no tier by itself.

Christian Palestinian Aramaic: the Western Aramaic Gospel witness closest to Yeshua’s own dialect family. Transcription is under way page by page under a two-pass gate (183 of 318 pages as of 2 September 2026, the drive running to completion); the gated lines are shown on the interlinear as a display witness (626 verses at the last rebuild) and do not yet enter the coverage counts in the Galilean essay.

A second Christian Palestinian Aramaic Gospel witness, independent of the lectionary: Lewis’s 1909 edition of the palimpsest. Its printed CPA text is not recoverable from OCR, so a heading index was built first (260 scripture headings; the Gospel pages are mostly Matthew), and 22 of the 24 Gospel pages have passed the same two-pass vision transcription and gate as the lectionary (as of 2 September 2026). A display witness only, 212 Gospel verses at the last rebuild.

CCR II (the Pauline epistles) and CCR XI (Acts and the Catholic epistles) in a modern edition. Not Gospel text: it can attest Christian Palestinian Aramaic word forms beside the lectionary and the lexicons, and its text pages are queued for a vision read. Whether non-Gospel CPA may count as Western attestation is an open policy decision recorded in the method; until it is made, this is a display lane.

Bensly, Harris & Burkitt, The Four Gospels in Syriac (1894): the same Sinaitic palimpsest as Lewis 1910, in its earlier transcription; the 1910 text already stands above. Pusey & Gwilliam, Tetraevangelium Sanctum (1901): the volume in hand is the Latin translation and apparatus, not the vocalized Syriac, so SEDRA 3 remains the vocalization lane. The OpenGNT edition text: licence unverified, held; only Levinsohn’s separately licensed quotation index is used.

The Old Testament quotations render from the Hebrew source. Amalgamated Hebrew OT with morphology, from Tyndale House.

The Greek OT for quotation comparison; the Byzantine text supplies the Greek of verses the critical text omits, plainly flagged.

The Galilean dictionary: the project’s primary witness for Palestinian Aramaic attestation, woven entry by entry across the glossary.

The Christian Palestinian Aramaic dictionary, indexed page by page from the print: 250 pages, 3,604 entries with their Latin glosses and scripture citations. It anchors CPA forms beside the Galilean dictionary and cross-checks the lectionary transcription. A reference lane, not a scoring witness.

The dialect map of the whole language. All 1,845 distinct Aramaic lemmas in this corpus were queried lemma by lemma, particles included, not only the words that earn a glossary card. CAL files a lexicon entry for 1,756 of them; the rest are mostly proper names, Greek loanwords, compounds, and a few words it files under another headword. Where an entry exists and we can say which homograph is ours, it supplies the vocalized form, the part of speech, the root, and the numbered senses with the dialects that attest each one. Its Gal (Galilean) tag is the attestation this project is otherwise short of, and it stands beside Sokoloff’s DJPA as a second reading of largely the same Palestinian texts. Its CPA (Christian Palestinian Aramaic) tag is recorded on the card and not counted: the CPA corpus is itself a Gospel translation, so it attests a Gospel word by translating it. A high-confidence Common Aramaic match counts as Galilean and Western lexical attestation because the family-wide word includes Galilean; the exception screen removes only a demonstrated same-sense case where Western Aramaic prefers another lemma, and no pronunciation is claimed. It works as a label layer, corroborating a word rather than choosing one. What it corroborates, and what it actually adds, is counted in the Galilean essay.

The classical Syriac and Targumic dictionaries, consulted by hand for semantic range. Payne Smith’s scan has no usable text layer and is not indexed. Both volumes of Jastrow are in hand; a targeted index of the entries the Galilean dictionary does not cover (Palestinian-Targum citations) is queued behind the Christian Palestinian Aramaic drives.

Each glossary entry carries its Greek semantic-domain code and label. The copyrighted sense-glosses themselves are not reproduced.

Supporting Greek and Hebrew lexica. The legacy Strong’s gloss layer is being retired wherever a curated gloss serves better.

Retroversion claims and the poetic structure of the sayings.

The classic treatment of the Aramaic behind the sayings tradition, woven as a scholar witness.

The grammar of Jewish Palestinian Aramaic, indexed page by page from the print — 20,301 Hebrew-script forms, 6,734 of them with Dalman’s own vowel pointing. Where a Galilean form’s vowels are shown in the apparatus, a pointed form printed here (or in Sokoloff’s dictionary) is the anchor, and the display says whether the vowels are printed or reconstructed.

Modern reconstruction scholarship for Mark and the double tradition.

The caution layer: documented inner-Syriac translator habits that must not be mistaken for Aramaic substrate. Where Williams warns, the claim is flagged.

The control corpus of dated Palestinian Aramaic (2nd c. BC–2nd c. AD) against which Galilean claims are checked.

The Galilean register itself: which forms, spellings, and constructions are Western rather than Babylonian or Syriac. Read as a dialect check on the reconstruction, not as a source of readings; its 206 claims carry no word into the text and open no tier.

Torrey read the Gospels as translations of Aramaic originals and explained hard Greek as mistranslation. That is a minority position and it is weighed as one. His 425 claims, 381 of them keyed to a verse, are recorded and scored at the lowest scholar weight; they carry no word into the text and open no tier on their own. Where his transliterations are unreadable in the scan, the claim is held for review.

Grammar authority for aspect and tense decisions, now indexed by section so that a ruling can cite him by § (382 sections, 50 of them on the verb’s tenses, participles and periphrasis); and the map of Aramaic’s dialects and history.

Public-domain English baselines used for comparison only.

Locating the scriptures inside the scripture: which Gospel words quote which Hebrew verses. The quotation list is cross-checked against the standard loci citati apparatus (NA28/UBS5); no edition text is reproduced from it.