The word for fire in some modern Australian languages:

Tiwi            yikwani
Djinang         junggi
Maung           yungku
Walmajarri      yakun

This is strikingly similar to that in PIE:

PIE     h₁n̥gʷnís

So, what was the form in proto-Australian?

  • 3
    Pease add some info to the Australian languages (e.g., which Australian language family they belong to). While Proto-Pama-Nyungan is reconstructed, I am not aware of any Proto-Australian as a whole. Commented Jan 25, 2021 at 9:57
  • 1
    slightly more strong than jk's comment, I don't believe there is any serious proposal for all Australian aboriginal languages belonging to a single family. Instead the use of the term "Australian" is generally taken to solely be a term of convenience as they appear to constitute a language area with several typological similarities. In this respect, they are similar to the Khoi-San languages, which comprise several families with no known connections
    – Tristan
    Commented Jan 25, 2021 at 10:07
  • @Tristan well, rthere is much similarity in numerals as well: preview.redd.it/…
    – Anixx
    Commented Jan 25, 2021 at 10:10
  • 2
    I see at least two distinct sets there. Regardless, of the Australian languages, most belong to the Pama-Nyungan family mentioned by jk, so there is no surprise in these languages having similar numerals. There are a few other small families, but also several isolates. It's also worth noting that some Australian aboriginal languages appear to have developed their numeral systems relatively recently (presumably from a system more like that of Pirahã), since contact with Europeans, and so similarities between their numerals, and their neighbours', even if unrelated, is to be expected
    – Tristan
    Commented Jan 25, 2021 at 10:15
  • @Tristan these numerals do not seem similar to any modern European language though.
    – Anixx
    Commented Jan 25, 2021 at 12:01

2 Answers 2


Unfortunately, there's no real consensus on any sort of "Proto-Australian" reconstruction. Some linguists like Dixon proposed "Australian" as an actual language family, with a common ancestor, but my impression is that this isn't popular any more: most linguists now consider Australian to be a Sprachbund containing multiple not-necessarily-related families.

So even if there was, at some point, a common ancestor of the Australian languages, there's never been a widely-accepted reconstruction of it.


Shapper et al in "Tree, firewood and fire in the languages of Sahul" calls the word "dyung(u)". It is unclear if this is scientific reconstruction or not, but it seems to be the closest we have to one. The word apparently initially meant "tree" or "firewood".

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.