Talk:Proto-Australian language

Did you know nomination

The following is an archived discussion of the DYK nomination of the article below. Please do not modify this page. Subsequent comments should be made on the appropriate discussion page (such as this nomination's talk page, the article's talk page or Wikipedia talk:Did you know), unless there is consensus to re-open the discussion at this page. You can locate your hook here. No further edits should be made to this page.

The result was: promoted by TarnishedPath (talk) 21:05, 12 February 2026 (UTC)

(

• Comment or view
• Article history

)

The Top End is the likely home of Proto-Australian.

• ... that all the Indigenous languages of the Australian continent descend from a common ancestor, spoken about 6,000 years ago (likely home pictured)?

• Source: https://theconversation.com/new-evidence-confirms-our-indigenous-languages-have-a-common-source-but-how-they-spread-remains-a-mystery-242576

• • Reviewed:
  • Comment: The reconstruction of Proto-Australian is a major advancement in the linguistic history of Australia, one that some thought would be impossible. The first book-length reconstruction was published in 2024. The article describes the reconstruction in some detail, and is very current.

Created by Ikuzaf (talk). Number of QPQs required: 0. Nominator has fewer than 5 past nominations.

Ikuzaf (talk) 08:22, 9 January 2026 (UTC).

General: Article is new enough and long enough
New enough: Long enough:

Policy: Article is sourced, neutral, and free of copyright problems
Adequate sourcing: Neutral: Free of copyright violations, plagiarism, and close paraphrasing:

Hook: Hook has been verified by provided inline citation
Cited: Interesting:

Image: Image is freely licensed, used in the article, and clear at 100px.
Freely licensed: Used in article: Clear at 100px:

QPQ: None required.

Overall: I really appreciate the submission of this well-written article. However, several sections of the prose are unaccompanied by an inline citation, and would need to be sourced in order for the page to be DYK eligible. Additionally (and subjectively) I don't think the image works for a hook since its unclear what region it describes. If it were some sort of graph documenting the possible range with clearly defined borders, it may work, but as it stands its too non-descript. If the unsourced sections of prose are fixed, particularly the bullet points in "The status of the..." section, then this will be DYK ready. (New enough, long enough, and Earwig suggests copyvio free as it stands, though a source spotcheck may be in order). --^The Robot Parade 06:33, 10 January 2026 (UTC)

OK I've added some more citations. I will do a bit more. Still trying to work on a better image. It's somewhat of a challenge, since any geographical discussion of homelands is necessarily vague. Thanks for your feedback. Ikuzaf (talk) 00:29, 13 January 2026 (UTC)

I'm still working on a free use version of map image to show language families. I can probably also do another one to show approximate PA homeland location. But as an alternative I could also submit

Ikuzaf (talk) 10:22, 16 January 2026 (UTC)

@The Robot Parade and Ikuzaf:, is there anything else to do here? TarnishedPath^talk 07:05, 25 January 2026 (UTC)

@TarnishedPath: IMHO, unhappily, potentially, there is another issue. The main source for the proposal of a proto-Australian language in the article is a book (in an academic series) by two linguists (of good academic standing). The hook is supported by a report (in a fairly acceptable source, The Conversation) written by the two authors of the book. Nevertheless, I would want either a weaker formulation of the hook (perhaps adding a "perhaps"), or the addition of sources presenting (reasonably) independent confirmation of the academic reception of their proposal. I'll bring this up on the article talk page. Note: I've never contributed to the DYK proposal pages, as far as I remember; and I'm not sure that this is the right way to intervene; but I want to avoid any mistakes which might be regretted afterwards. JoergenB (talk) 19:06, 25 January 2026 (UTC)

I will respond on article talk. Ikuzaf (talk) 02:31, 26 January 2026 (UTC)

From my POV, I think the article is OK for DYK. The one question is over the image. My personal preference would be for the NASA satellite photo, or failing that the "hand" diagram I submitted above. But if others really feel a map image will work better, I can create one. I've tried but I haven't been happy with the results so far. So, subject to what others think, I believe it's a good idea to go with what we have. I will defer to Courtesy ping: Robot Parade Ikuzaf (talk) 02:31, 26 January 2026 (UTC)

@Ikuzaf: Apologies, I've been bogged down with school. I've just re-reviewed the article and I agree that the sourcing is now adequate. There are still a few spots here and there but I believe they are covered within the sources already provided. I like the included diagram here but I don't believe this would be easily read at 100px. I really think an image may not be necessary since the hook isn't tied down geographically. For now, I approve the hook in its current state, within reason. I think JoergenB brings up valid, genuine concerns, but I don't believe they are issues preventing a DYK approval. --^The Robot Parade 17:59, 29 January 2026 (UTC)

@Ikuzaf:, In my review just now I counted approx. 10 paragraphs which don't have citations at the end of them. Can you please correct this. Once you do so I will promote a variation of ALT0:

• ALT0a ... that all the Indigenous languages of the Australian continent may have descended from a common ancestor spoken around 6,000 years ago?

TarnishedPath^talk 09:05, 30 January 2026 (UTC)

@TarnishedPath:Thank you, I've added some more citations. Can you please check and let me know if any more need to be added? Thank you. Ikuzaf (talk) 04:31, 7 February 2026 (UTC)

@Ikuzaf, I'll have a look when I get home later on. TarnishedPath^talk 04:37, 7 February 2026 (UTC)

@Ikuzaf, at the risk of seeming to be knit-picking, the sentence "Harvey and Mailhammer also propose two new family-level classifications: Yang and Waykkan" (which is also at the end of a paragraph) does not have a citation. Can you please fix it. If we don't fix it now, it will likely hold up the nomination later down the line. TarnishedPath^talk 13:14, 8 February 2026 (UTC)

@User:TarnishedPath OK, that's done now. Ikuzaf (talk) 20:20, 12 February 2026 (UTC)

Note separation

I'll start with what I think is an easy and non-controversial improvement idea.

@Ikuzaf: I believe that this is one of the articles which would be easier to read, if the explanatory notes were separated from the citations. (Right now, both are collected in the reference list. The change essentially would be to employ {{efn}} and {{footnotes}} – or other templates with similar function – in order to get what is now (e. g.) notes 37, 58, and 77 collected under a separate heading and placed before the citation notes. I'm asking you, not since I consider you as "the owner" of the article; but you started it a few weeks ago, and very definitely is its main contributor.)

If you agree that this would be a good idea, then I could do the separation, or you could do it yourself. Regards, JoergenB (talk) 18:00, 24 January 2026 (UTC)

Thanks, that sounds like a great idea to me. Please go ahead and do it if you have the time, otherwise I will get to it over the next few days. Ikuzaf (talk) 20:01, 24 January 2026 (UTC)

OK, I made a separation. Please check that I caught (all and nothing but) the correct notes; and consider what heading of the section of efn's you would prefer! JoergenB (talk) 15:26, 25 January 2026 (UTC)

Reception

I continue with a (potentially) more severe matter. First, I'd like to say, that I find this one of the most interesting new articles concerning linguistics, and find it both informative and of high importance. It describes its main subject, the proposed language family and protolanguage reconstruction by Harvey and Mailhammer (below: HM) in some detail, and also gives due weight to the work leading up to this proposal. IMHO, the article is a very good addition to enwp, even as it stands now.

However, the HM proposal is presented as if it represent the present consensus among the experts in the field (which should amount to linguists with sufficient knowledge about the languages ijn question). On the other hand, only HM sources are given for the proposal in itself (while the general background and also academic work leading up to this result seems fairly well-sourced, in my opinion). Now, I notice from e. g. the article Macro-Pama–Nyungan languages that just a few years ago there even were dissenting opinions on the existence of that language grouping, not to mention the proposed 'phylogenetic tree' structure of relations within it. Of course, it is possible that (virtually) all dissent disappeared with the publication of the HM book; but in that case there should be some quotable evidence for this fact. Another possibility is that there (also) was some more conventional reception, of the type "This is an important contribution to this field, but more research is needed before a general acceptance of the proposal". A third possibility is that some general or more specific criticism occurred, e. g. of the type "The authors are employing a not statistically significant part of the vocabulary as a basis for their conclusions", or "The proposed phonetic correspondences do not follow the pattern of explicable and exceptionless sound shifts, but use ad hoc rules whenever it suits them". (Yes, such criticism of tries to establish so far unrecognised relationships are not that uncommon; see e. g. the discussions around the Elamo-Dravidian languages proposal.)

Thus, I think that this article would benefit substantially from a section about its reception, if good sources about the reactions from other Aboriginal Australian linguistics researchers are available. E. g., has there been overview articles or book recentions by any (preferrably senior) researchers (such as Evans), which may be sourced and summed up? Has there been research articles refeering to the book or (or the HM forrunner articles), either accepting or criticising the Proto-Australian proposal?

If there are criticism (even of the very common form "more research is needed"), then this ought to be reflected by a few words in the article lead. Also positive reception could be briefly mentioned there. More controversially, perhaps: If there has been tangible groups not accepting the proposal as proven, then there should be some reservation in the very start of the article, e. g., changing the first sentence to

"Proto-Australian (PA) is a proposed reconstructed common ancestor of nearly all Australian Aboriginal languages."

Now, such formulations are not uncommon in the wikipedia. However, we also try to avoid WP:UNDUE weight to non-scientist criticism, especially in the lead. This sometimes may create a dilemma. I do not know if this is the case for you with respect to this article; but I do note that you 20 years ago explained that you were an australian student in linguistics (from which it is not unnatural to assume that you by now have expert knowledge in this field); and you have made a professional work in presenting the HM book. Moreover, you have chosen to represent the HM proposal as established facts. I thus find it rather likely that you have made a professional evaluation of these matters, and have indeed concluded that the proposal holds. I respect that. At the same time, as you know, from the WP point of view, you should present the 'consensus standpoint' (if there is one) or the more important different standpoints (if there are such) among "the experts in the field", even when your own expertise tells you that such dissending opinions are based on ignorance or misunderstandings. (This is a recurrent dilemma for experts editing wikipedia. My professional expertise concerns (at least some subfields of) mathematics, not linguistics; but I've met this problem; briefly put, the contrast between "truth" and "verifiability".)

So, I hope that there are a few research articles or surveys or book recensions by other professionals than Harvey and Mailhammer themselves, which may be used to confirm that indeed their proposal has been essentially accepted as the new consensus in the field. If this is the case, the article need little change, but should be improved by a few addition; and I then also would recommend updating of some other articles, like Macro-Pama–Nyungan languages, which now presents a rather different (obsolete?) view. However, if it is not the case, I hope that you'll be able th downgrade the proposal from "the consensus accepted truth" to "a seriously supported proposal", even though you personally may feel that such a less absolute presentation is against the facts you know to be true. Regards, JoergenB (talk) 21:03, 25 January 2026 (UTC)

For the most part, I agree with what you are saying. I very much would like to include more reactions from credentialed scholars to the book published by Harvey and Mailhammer. I haven't been able to find any. That might be surprising to you, but I think it makes sense, given the nature of the field. The majority of the languages that are germane to the proposal are "dead" (or "sleeping" depending on your preferred terminology), and the documentation on them varies. Specialists are continuing to work on them, but Australianists are a relatively small field, and have to wait for time and funding to do in-depth work (The HM book was the result of an Australian Research Council grant lasting several years, for example). There are other reasons too, which I will explain more below.

What I would say is that the article should exist, independently of Harvey and Mailhammer's reconstruction. They certainly aren't the first people to propose reconstructions for something called "Proto-Australian"; R.M.W. Dixon did that in 1980, even if, as the article explains, the forms of his 1980 reconstructions are taken to be "Proto-Pama-Nyungan" by other specialists, the idea in his mind in 1980 was a "Proto-Australian" that was ancestral to the languages of the entire continent. It's complicated because he rejects PN as a family and he later reversed himself on whether comparative method linguistics could ever work in Australia. His argument in 2002 was that, among other reasons, that he had tried it and found it too hard. But that perspective was rejected by most other specialists in the field; the consensus position is that the comparative method can be applied in Australia. The article explains that as well. There are certainly reasons why it is difficult - the biggest problem is the availability of evidence, because while some Australian Aboriginal languages have been surveyed extensively by qualified linguists, the most evidence we have for some is scraps of word lists, usually written ad hoc by people with no linguistic training. Another major problem is that Australian languages are quite unusual and distinctive in terms of their phonological patterning, and certain sound changes - such as initial dropping - have been well documented to take place in Australia while being very rare worldwide. Cognates are generally quite hard to distinguish from borrowings, as Dixon pointed out, which makes establishing family relationships difficult. Not impossible, I think, but difficult.

Aside from Dixon, for the entire time Australian languages have been under linguistic study, specialists, such as Capell, O'Grady, Wurm, Vogelin, Vogelin, and Hale, have generally supported the idea that the languages of Australia form a genetic family. Somewhat like William Jones and PIE; Jones didn't reconstruct PIE, but he argued that it must exist. So the article should document and reflect that. I've attempted to do that so far but of course more contributions are welcome and encouraged.

It's possible and indeed likely that other Australianists will take issue with at least some aspects of HM's reconstructions. In particular, I note that it doesn't take Gunwinyguan to be a valid clade, which is at odds with a lot of other research that has taken place over the past decade or two. But I haven't as yet found any detailed responses from Australianists, and those could be months or years in coming. Any evidence to the contrary is more than welcome!

Luisa Miceli is one person whose reaction I am quite interested in - she is on record as saying that she found the evidence for Proto-Pama-Nyungan 'weak', although my understanding is that she does view what has been published in terms of reconstructions as plausible enough. And of course other doyen(ne)s in the field such as Claire Bowern and Nicholas Evans. But I haven't found published responses by them and my access to linguists is much less than it used to be when I was doing my degree many years ago. I can only imagine that Harvey and Mailhammer have been in communication with other linguists in the (small) field while they have been working on the proposal, and that they have responded to reviews as part of the publication process. I think it's fair, as of this moment, to say that their book represents the current best scholarship on PA that is available. That can, and hopefully will, change in the future. But just as with PIE and other reconstructed languages, we can only document the state of the research as it is at the moment, not where it might likely be. And "where we are" at the moment is, in my opinion anyway, fairly reflected by the article. Ikuzaf (talk) 02:56, 26 January 2026 (UTC)

(I'm not commenting the raison d'etre of the article, since I've already clarified that in my opinion this is a rather good contribution to enwp. The precise relation to other articles could be discussed later.)

Well, a wise old man at our department (Jan-Erik Björk), commenting on the trend of making bibliometric measurements of resent citations significant for the amount of money from the central university to the departments, noted that really epoch-making progress (within mathemathics) was recognised by the academic community, but could result in few or no references. In his opinion, the reason was that it took years for the rest of the community to catch up with groundbraking new methods. When others did advance to the point of beginning to apply the work, and therefore also providing references, the work was old enough for citations to it not being counted in the statistics.

I do not know if there are similar phenomena in linguistics. However, if there are no references to the HM book found, then I think this information should be added, briefly. Moreover, if I understood you correctly, you might wish to stress that the existence of a proto-Australian language is not depending on the existence of the HM reconstruction. Taken together, this might motivate changing the first section of the lead to something like

Proto-Australian (PA) is the proposed common ancestor of nearly all Australian Aboriginal languages.[the scope efn] It is hypothesised to have been spoken approximately 6,000 years ago. A comprehensive reconstruction was published 2024.[the The Conversation reference]

Likewise, we might add something like

(As of January 2026, there has been little published response to this book.)

@Ikuzaf, what do you think? JoergenB (talk) 17:25, 26 January 2026 (UTC)

I quite like the first part (the ending in 2024 part). I am somewhat less of a fan of the second sentence because I think it might imply that my inability to find a response is the same thing as not being able to find a response, but it's probably decent enough. Thank you for your work on this. Ikuzaf (talk) 07:43, 2 February 2026 (UTC)

@Ikuzaf: OK; I implemented the first change part; the second one I leave to you. (Do you have a better suggestion? Would an inserted "seemingly" improve the addition sufficiently? I might remark that I also made some limited tries to find comments on the book without succeeding; so if you missed some obvious reference, you should not be alone in taking the blame... .) JoergenB (talk) 18:10, 3 February 2026 (UTC)

OK thanks good point haha. I think the second point is fine with me. Ikuzaf (talk) 03:07, 7 February 2026 (UTC)

Proto European

Many years ago I accidentally did much of a applied linguistics degree at Melbourne University. The idea of proto australasian came up, and was quickly questioned by senior staff. We were told that linguists had expected some pattern to the spread of language over the continent based on the traceable spread of words across Europe, but there was none. There are an extremely wide variety of features and they appear sporadically. Walapiri is heavily case marked (Simpson dessert) and so is another language, in completely different environment, a very long way away (South Eastern Victoria). The article is well done, but it seems to be about the theory rather that what we know. Perhaps someone could contact the ANU for comment? Also 6000 years? That is an odd number as first human settlement on the continent is usually considered today to be 40 to 100 thousand years ago. ~2026-10433-24 (talk) 07:14, 16 February 2026 (UTC)

A lot to unpack here. Firstly, I'm pretty sure this is just a typo but just in case, "proto-Australian" and "proto-Australasian" are different ideas. I've only ever heard of the first being discussed. Secondly, I'm not quite sure what to make of the argument about the spread of features (I assume grammatical features). Whether languages are genetically related is determined by the comparative method. Proto-Australian, according to Harvey and Mailhammer's reconstruction, is theorised to use case marking, as do many of the languages that descend from it, which is more or less what you would expect from a pattern of genetic descent. There's actually a discussion of the interaction between typology and genetics in the article, and some of the references in there would be worth looking into. I also don't quite know how to respond to the idea that the article is "about the theory rather than what we know". I've used the published book, which contains detail about the proposed Proto-Australian grammar, as a source. If there's a specific area that you think needs clarifying or expanding, please feel free to mention it. The question of time depth is also addressed in the article, and in particular, I'd direct you to this article from The Conversation where it is quite reasonably pointed out that there's no particular reason why language families in Australia would stay the same for tens of thousands of years . There would have been earlier language families in Australia (such as Tasmanian), which were replaced by the current "Australian" superfamily over time (in most cases by Pama-Nyungan). We don't know how or why that happened, but that's a different question as to whether it happened. Finally, it's a little difficult to respond to something that you remember hearing a long time ago in a lecture. If you have access to published, current sources that address the topic, please do add them to the article. Ikuzaf (talk) 08:46, 16 February 2026 (UTC)