Opacity = date confidence · Colours = Okabe-Ito colorblind palette

Each tile = one psalm · Dark lapis = few refs · Gold = many refs

Gold = halal · yadah · shavach · zamar  ·  Click any cell

The Elohistic Psalter (Books II & III) dramatically prefers Elohim over YHWH.

Click any word to see its Hebrew lemma, transliteration, Strong's number, and representative verses from across the Psalter.

Opens in a new tab · Full Hebrew root + verse interactivity

This dashboard is a single, self-contained web page — no server, no login, nothing to install. Open index.html in any browser and everything you see — the charts, the psalm details, the slide deck — lives inside that one file. Behind the scenes it's assembled by a small set of Python scripts: one reads the source material, another lays out the page, and a build step stitches them together into the finished HTML. The interactive graphics — the composition timeline, the New Testament treemap, the voice and divine-name charts — are drawn with D3, a charting library that runs right in the browser. The lapis-and-gold, illuminated-manuscript look comes from the Cinzel and Lora typefaces.

Almost everything you see traces back to a single hand-built research spreadsheet — one row for each of the 150 psalms. Each row records the author, which of the five "books" the psalm belongs to, a representative quotation, talking points, historical context, category, the "voice" (whether the psalm speaks as "I," "we," or impersonally), any instruments or melodies named in the heading, counts of the divine names (God, YHWH, Adonai), an estimated date, and the New Testament passages that quote or echo it. That spreadsheet is the curated, interpretive layer beneath all the numbers.

Two further sources supply the words themselves. The English text in each psalm's detail view comes from public-domain translations — the King James Version and the Modern KJV. The Vocabulary Cloud and the word-by-word literal readings draw on a Hebrew dataset that breaks every verse down into its underlying words, complete with transliteration and Strong's numbers, so word frequencies and roots are counted directly from the original language rather than estimated.

The "Film Reel" runs one illustration per psalm, in canonical order. Each frame was produced with a generative-image model, prompted from a short written brief for that psalm — a description composed alongside the research and stored right next to everything else in the spreadsheet. Working from a consistent brief keeps all the frames in a single visual key, so they read as one series rather than 150 unrelated pictures. A handful of other images in the project — maps, manuscript photographs, and the Bible Project overview — come from their original sources and are included for reference.

This page is a living draft, and the tools and sources may evolve as the project grows.