The granted doors — the sagas and the Gawain-poet arrive

26 August 2026 · Nabu news

Both of this month’s correspondence grants are now shelves. The library stands at 988,944 documents — 75.2 million passages — in 139 languages.

The Nordic manuscripts. The Menota archive enters at manuscript grain: 91 medieval codices and fragments, each word transcribed at up to three levels — what the scribe drew, what the scribe wrote, what the philologist reads. Nabu stores the diplomatic reading, line by manuscript line: Laxdæla saga, the Codex Wormianus (whose runic entities the file declares itself — the parser learned to listen), Konungs skuggsjá, the Old Norwegian homily book, the Landslǫg law manuscripts. Eight manuscripts that stumbled on their first pass were recovered by reading the files more honestly, not by guessing: every character mapping comes from upstream’s own bytes. And the language registry grew its Norwegian and Danish ladders to receive them — sixty-two manuscripts now sit on their language’s own timeline, staged by nothing more than their catalogue dates meeting the registry’s period bands.

The Middle English corpus. From Michigan, the corpus behind the Middle English Dictionary: 297 texts — the Canterbury Tales, Sir Gawain and the Green Knight, the York and Towneley mystery cycles, Piers Plowman in its versions, and the wills, gild records and medical treatises that carry the working language. It rides a new parser family built for the Text Creation Partnership’s schema — machinery deliberately shaped so that the sixty-thousand-text EEBO-TCP world, whose door the same grant opened, can walk through it one day.

Thanks are owed twice: to the Menota network, whose archive is openly licensed and who blessed our polite crawl while their bulk endpoint is built, and to the Middle English Compendium’s editors, who pointed us at files better than the ones we asked about.