Germanic — The Germanicist

The Germanicist — Gothic and Old English to the Norse sagas and the runestones, the word-hoard of all three branches.

Gothic on the proiel/ud treebanks beside the West and North Germanic branches: Old English verse (ASPR) and prose (ISWOC) with Bosworth-Toller, Old Icelandic (IcePaHC via ud), Old Norwegian and the Poetic Edda (Menotec), the Medieval Nordic manuscript archive (Menota), the Corpus of Middle English (CME), early-modern English at scale (EEBO-TCP), Beowulf again in Klaeber’s edition with its aligned English (perseus-anglit), the Old Saxon Heliand (HeliPaD), Middle High German manuscripts (ReM), Middle Low German and Low Rhenish (ReN), the German print era whole (DTA, 1473–1969), and the runic inscriptions (Rundata, dual-tagged epigraphy).

The desk spans all three Germanic branches. East is Gothic (Wulfila, gold-lemmatized in PROIEL). West runs Old English (ASPR verse, ISWOC prose), the Old Saxon Heliand (HeliPaD), and Middle High German (the ReM manuscripts). North arrives in force: Old Icelandic through IcePaHC — a rule-based UD conversion of the Icelandic Parsed Historical Corpus, filed under is — Old Norwegian and the Poetic Edda of Codex Regius through Menotec (non), and, dual-tagged with the epigraphy desk, the runic inscriptions of Rundata. The wave went live 2026-07-22 with the owner-verified first syncs — all five shelves hold their corpora below. Newest on the desk: ReN, the Middle Low German / Low Rhenish reference corpus (1200–1650, gml — the Hanseatic rung between ReM and the Norse shelves): 235 texts, 1.49 M gold-annotated tokens, live below.

New here? The Quickstart sets up the library in minutes.

The shelves

A source wears every desk it serves — these nineteen answer this desk. Holdings are read live from the catalog and dated; a shelf with nothing synced yet says so.

Languages on this desk (live doc-or-entry counts as of 19 September 2026): ang 63,170 · en 59,433 · non 20,449 · eng 6,814 · de 5,668 · swe 3,397 · dum 2,226 · lat 532 · gmh 406 · enm 297 … and 42 more (nabu axis germanic lists all).

The desk’s instruments

• Gold-lemma languages: got (Gothic, on the PROIEL and UD treebanks), ang (Old English — ASPR verse and ISWOC prose, the West-Saxon Mark), and — live since the 2026-07-22 wave — gmh (ReM’s 2.10M rows, instantly the corpus’s third-largest gold pool), non (Menotec’s PROIEL-scheme Old Norwegian and the Edda, 258K) and osx (HeliPaD’s gold form-lemma Heliand) and is (IcePaHC via the ud sync — 812K rows, straight in at #4 of all the gold pools: the deepest North Germanic lemma lane).
• Dictionary: Bosworth-Toller (nabu define aethele --lang ang folds æ/þ/ð to find æþele).
• Alignment work: nt — Gothic (Wulfila) and Old English (the ISWOC West-Saxon Gospel of Mark). The Gothic × OCS cognates join is strongest from this desk.
• The runic five-lane design (Rundata): each inscription is one document keyed on its signum (urn:nabu:rundata:u-344); the bare urn is the scholarly Latin transliteration (the notation legend is content, never stripped), and up to four sibling lanes — -fvn/-rsv normalisations, -eng/-swe translations — ride beside it, reached with show --parallel.
• The timeline desk: Rundata is the germanic desk’s first dated lane — the RundataDates extractor puts each inscription’s SRDB year envelope and find-spot on the calendar, so --from/--to, --century and --place scope the runestones. (gmq-pro tags the urnordisk-dated inscriptions — Proto-Norse, for which no ISO code exists.)

Working the germanic desk

The generic axis surfaces — every desk answers to these, in working order (enable once, sync, then query):

nabu enable germanic               # first time: put this desk's shelves in this box's profile
nabu sync germanic                 # fetch/refresh the desk's enabled members
nabu list --axis germanic          # the shelf census, this desk only
nabu axis germanic                 # the desk card: members, holdings, gold coverage
nabu search WORD --axis germanic   # a query scoped to this desk's shelves

This desk’s own surfaces:

nabu show urn:nabu:aspr:A4.1:1        # Beowulf, line 1
nabu show urn:nabu:rundata:u-344 --parallel  # the Yttergärde runestone beside its normalisation and translation lanes
nabu show urn:nabu:menotec:non-edda-regius-dep  # the Poetic Edda, Codex Regius (Menotec)
nabu search --word cyning --lang ang  # whole-word match on an Old English form (ASPR/ISWOC)
nabu search liudi --lang osx --axis germanic  # the Old Saxon Heliand text (HeliPaD)
nabu search --lemma konungr --lang non --axis germanic  # the Menotec gold lemma lane — kings across the Edda and the sagas
nabu define aethele --lang ang        # Bosworth-Toller, with the æ/þ/ð fold
nabu formulas urn:nabu:aspr:A4.1      # the Old English poetic formulas of Beowulf

Ask your model

With the MCP server connected, this desk answers conversational research questions. Each example ran live against this library:

• “Where does Gothic guþ come from, and who are its cousins?” → nabu_etym guþ (lang: got) — Proto-Germanic *gudą with 64 cognates (nine attested in this library, counts shown: Gothic 𐌲𐌿𐌸 ×914, Old Norse guð ×1,337, Middle High German got ×6,995 …) and the PIE ancestors *ǵʰutós “libated” / *gʷʰew- “to venerate” above it.

Terminal setup

• Gothic (got): nabu does nothing; install font-noto-sans-gothic.
• Old English (ang): Latin script with æ/þ/ð — any extended-Latin font (Noto Sans Mono covers it); no install needed.
• Old Icelandic / Old Norwegian (is/non): IcePaHC and Menotec store the text in Latin script (þ/ð/æ/ǫ) — the terminal default suffices, no install.
• Old Saxon (osx): the HeliPaD Heliand is Latin-script Penn bracketing (uu for w); nothing to install.
• Middle High German (gmh): ReM’s diplomatic transcription keeps the long ſ (U+017F) and combining editorial marks — any Unicode font renders them (Noto Sans Mono); nabu’s search fold maps ſ→s, so you need not type it.
• Runic (Rundata): the transliteration IS the text — a scholarly Latin transliteration with zero runic codepoints stored, so no runic font is needed; the notation legend (§A/§B sides, | boundaries, a leading " marking a name) is content and never stripped, and the normalisation and translation lanes are sibling documents reached with show --parallel (docs/display.md §4).

The full guidance, per script, is on the display page.

One of the twenty-five research desks; the flat shelf map is The Library and the reasoning is docs/axes.md.