Are there Linguists that have or are currently working to reconstruct proto-languages other than PIE? Or to map the historical relationships between various African, Asian, Native American, or other languages like they have for Indo-European ones? If yes, is there any kind of visualized diagram that displays this mapping, like this Indo-European family tree?


2 Answers 2


There are a great many language families studied by linguists, and part of the demonstration of their relatedness typically involves the classification of the members of that family into subgroups.

Indo-European is one of the best-studied, and one of the families where there is the least controversy surrounding which languages belong to it, and its subclassification (at least as far as living languages are concerned, extinct languages sparsely attested in the historical record are more controversial).

Glottolog lists the following as the largest 14 families (by number of languages):

  • Atlantic-Congo
    • Sometimes seen as a subgroup of Greenberg's Niger-Congo, although the unity of this superfamily is widely considered to be poorly demonstrated. This family contains the Bantu languages of Central and Southern Africa
  • Austronesian
    • Consists of the Malayo-Polynesian languages of the Philipines, Indonesia, Southeast Asia, and the Pacific Ocean, and the indigenous Formosan languages of Taiwan
  • Indo-European
    • Fairly well understood, containing most languages of Europe, the Northern Indian Subcontinent, and much of Southwest Asia
  • Sino-Tibetan
    • Contains the Sinitic languages of China, the Bodish languages of Tibet & Bhutan, most of the languages of Burma, as well as minority languages in India, Bangladesh, Nepal, and Thailand. Subgrouping is uncertain
  • Afro-Asiatic
    • Contains the Amazigh (aka Berber), Chadic, Cushitic, Egyptian, & Semitic branches but there is no consensus on the subgrouping of these branches. Omotic is sometimes included as well, in which case it is typically viewed as splitting off first
  • Nuclear Trans-New Guinea
    • The larger Trans-New Guinea family is widely considered to be poorly demonstrated
  • Pama-Nyungan
    • Contains most of the languages of Australia
  • Oto-Manguean
    • Contains several languages of Mexico, and prior to colonisation, Central America more broadly
  • Austroasiatic
    • Contains the languages of Vietnam, Laos, and Cambodia, as well as minority languages throughout Southeast Asia and India
  • Tai-Kadai
    • Contains Thai and minority languages throughout Southeast Asia
  • Dravidian
    • Contains the languages of Southern India
  • Arawakan
    • Contains many languages of South America
  • Mande
    • Sometimes seen as a subgroup of Greenberg's Niger-Congo, although the unity of this superfamily is widely considered to be poorly demonstrated
  • Tupian
    • Contains many languages of South America

A very non-exhaustive list of some other well-studied and commonly-discussed language families (including some families no longer considered valid) is:

In general, wikipedia has an extensive list of language families where you can find more. Most major families have a "classification" section on their entry where the subgrouping of languages within the family is discussed and/or a "subdivisions" section of the info-box at the head of the article listing subdivisions.

The model usually presented is the tree model where once varieties become distinct they have little mutual effect on each other. In other instances (usually where a dialect continuum has remained in situ long enough that many members have almost no mutual intelligibility, but innovations spread over overlapping regions of the continuum, giving the appearance of links on a chain, which is why these groups of languages are often called linkages), the wave model is seen as more appropriate.

The canonical examples of languages ideally fitting a wave model and tree model both come from the Austronesian family, with the Polynesian languages fitting a tree model near perfectly (as the distant islands provided relative isolation), and the Western Oceanic languages of western Papua New Guinea and the Solomon Islands being the group where the concept of linkages was first developed.

  • Uralic is actually understood the best after Indo-European.
    – Anixx
    Commented Nov 9, 2022 at 17:18
  • @Anixx families are not listed here according to how well-studied or well-understood they are. The first list is ordered by number of languages (per glottolog), and the second alphabetical
    – Tristan
    Commented Nov 10, 2022 at 9:43

Yes, there are very many researchers working to reconstruct proto-languages other than Indo-European. There are a number of languages that are not clearly related to any other language (e.g. Burushaski), and yet there are active attempts to join that language with some other group of languages and include it in the reconstruction. I don't think there are many attempts to put languages in an artsy tree (with birds and mammals as decorations), but there are certainly plenty of simpler tree-like displays available. The volume of research and current activity on reconstruction and classification within language families varies wildly.

These trees assume a strict-splitting view of language relatedness. There are other views of language-relatedness where a language can be "relatively related", using the Neighbor-Net method, for example this.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.