Answers are grounded in this paper and cite their sources.
I map every language-level entry in Glottolog 5.2.1 and publish the result as an interactive atlas with a browsable family tree, endangerment, speaker numbers and the proposed homelands of the great language families. Glottolog's 7,674 spoken first languages belong to 238 families and 183 isolates: 421 lineages that cannot, on present evidence, be shown to be related, of which 314 survive. Five families hold 54% of all languages, while 135 families have five or fewer. Diversity is concentrated in a tropical belt: Papua New Guinea alone has 850 living languages from 77 lineages, and the densest 1% of the world's land, measured in equal-area squares of 200 km, holds 28% of living languages. By Glottolog's Agglomerated Endangerment Status 52.1% of spoken languages are endangered and 12.8% extinct; only 35.0% are rated not endangered. Wikidata gives speaker counts for 1,627 languages (21.2%); the counts mix years and definitions, but they show the 20 largest languages accounting for 62.6% of all speakers counted. Finally, I set 16 published homeland hypotheses for 11 large families beside two reference points computed from Glottolog, each family's present centre and the hotspot where its deepest branches crowd most closely. The hotspot lies within 500 km of a proposed homeland for 5 of 11 families, including Austronesian, Pama-Nyungan and Bantu, but far from all three Indo-European proposals. Interactive version: https://maps.tristanhotham.com/language-origins/. Data and code: https://github.com/Tristan-Hotham/language-origins. Includes data from Glottolog 5.2.1 (Hammarström, Forkel, Haspelmath and Bank, CC BY 4.0) and Wikidata (CC0).
No takes yet. Share an insight, caveat, or question.
Tristan Hotham (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: