The author's own structure, recovered from the linguistic evidence — freely licensed, machine-readable, for the languages that have no commentary at all.
Discourse Flow
The Problem
Billions of people read the Bible in languages that have no commentary, no study Bible, and no linguistic resources explaining how the text is structured. Those resources exist elsewhere, and they are proprietary: a translation team in a minority language cannot legally translate one for their community, distribute it freely, or build software that reads from it.
Â
That gap will not close by asking publishers to give away their catalogs. It closes by building new, freely licensed resources on the same scholarly foundations.
What It Does
The Bible was written to be heard. Its authors marked which events are central and which are background, where a scene begins, how topics shift, and what they most wanted an audience to carry away. Modern readers miss those signals — and chapter and verse numbers, added roughly a thousand years after the text was written, have trained everyone to read in chunks that are not the author's.
Â
Discourse Flow recovers that structure for a whole book, in Greek or Hebrew: where each unit of thought actually begins and ends, what it is doing, how it is built inside, and how the book as a whole is constructed. Boundaries are decided from the linguistic evidence, bottom-up, and each one carries the strength of the evidence behind it.
Â
Standard translation checking largely cannot reach any of this, partly because there has been no freely licensed, machine-readable account of the structure to check against. Output is published under CC BY-SA 4.0, in formats for software, for people, and for the tools translation teams already use.
Where It Sits
Third in the stack. The environment above it runs the analysis; the work below it is built on the structure this produces. Eleven books have been through the pipeline so far, chosen across genre rather than in canonical order.
