Proto-Sinaitic c. 1900 BCE
└─ Phoenician ──┬─ Greek ──┬─ Latin
│ ├─ Cyrillic
│ └─ Coptic
├─ Aramaic ──┬─ Hebrew (square script)
│ ├─ Arabic ── (Persian · Urdu · Jawi …)
│ ├─ Syriac
│ └─ Sogdian ─ Old Uyghur ─ Mongolian (rotated 90°, top-to-bottom)
├─ South Arabian ── Ethiopic (vowels added independently)
└─ Libyco-Berber ── Tifinagh
Brahmi? ── descent from Aramaic contested — recorded, not resolved
Simplified. One Bronze Age consonant inventory, three continents, four thousand years. See WI·BIO, Biographies of Letters.
Proto-Sinaitic Wing
10 scripts on displayOne ancestor: Egyptian pictures reassigned to single consonants by Semitic workers in Sinai. Nearly every alphabet and abjad on Earth hangs from this wall.
Reading Room → WI·BIO Biographies of Letters · WI·MCH The Machinery of Text
Brahmi Wing
7 of ~200 descendantsThe Indic radiation: some two hundred scripts across South and Southeast Asia descend from the chancery hand of Ashoka's edicts. Whether Brahmi itself descends from Aramaic remains genuinely contested.
Reading Room → WI·ORT Orthographies (Tibetan, the frozen spelling)
Han Wing
4 scripts on displayThe East Asian sphere: Chinese characters and the systems distilled from them. The deep collection lives with our partner institute; this wing holds the comparative frame.
Partner institute → hanzi.design holds the Han deep collection
Inventors' Room
12 scripts on displayScripts designed, not evolved — each with a named creator, a date, and a brief. Writing as an act of deliberate design, from a fifth-century monk to a 2006 language committee.
Reading Room → WI·INV The Inventors — one chapter per designer
The Undeciphered Vault
3 sealed casesVisible but unread. Decipherment needs a large corpus, a known underlying language, or a bilingual key — the Vault holds the cases where all three are missing.
Reading Room → WI·DEC Decipherment Dossiers — the open files