Spot the mistakes

The ABC science department seldom thinks much about Aboriginal languages, and I guess we should be grateful that they did. However, there are four major factual errors in this article, aside from the fact that Clendon’s own proposal (“innovative new theory to challenge the establishment“) cannot possibly be true. Factual error corrections are below. I was really irritated that this article was presented as a challenge to current doctrines, mostly because there isn’t really an orthodoxy – there are several possibilities, and it’d be really great if instead of coming up with possible scenarios that fit the current distribution there was a bit more work on looking for evidence. Like, you know, actual research.

(The article on which this abc report is based on can be found here. Confession: I was one of responders to the original article.)

Here are the errors I sent to the ABC.

1) Mark Clendon’s article concerns only the current distribution ofAboriginal languages – he makes no claim about how long Aboriginal languages have been spoken in Australia.
2) The claim only relates to the 150 or so “Pama-Nyungan” languages. It does not apply to all Aboriginal languages, including the languages of the Kimberley region and the Northern Territory between Daly River and Southern Arnhem Land.
3) the term is Pama-Nyungan, not Pama Nyugen. It is a language family (like Indo-European), not a single language. Saying it’s a single languageis like saying French, German, Swedish, Russian, and Latin are all the same language.
4) There are over 100 languages in the Arafura region, not one, as is claimed.


3 responses to “Spot the mistakes

  1. David Marjanović


    I can’t leave a comment to the original article, which I’ve just read, so I’ll do it here. First of all I’d like to nitpick on the term “cladistics”. As an evolutionary biologist, I’m a member of the culture that owns that word ;-) , and so I do proclaim that cladistics finds the most parsimonious (and thus most scientific) explanation for the distribution of shared derived features. Unfortunately cladistic analyses have been very rarely employed in linguistic phylogenetics so far (I’m aware of 2 or 3 papers, published in Nature and Cladistics, not in linguistics journals). Mass lexical comparison and lexicostatistics are phenetics; they are mere measures of similarity in %. Phenetics does not give you a phylogeny, unless there happens to be little enough homoplasy* in the data, which in turn is among what you are trying to find out in the first place.

    In his reply Clendon claims that reconstructed PIE is not pronounceable. If you can produce voiced aspirated plosives, it is completely pronounceable, except that you’ll probably need to smuggle in a little schwa or two like when e.g. a palatalized velar stop is supposed to be followed by a syllabic nasal ( = zero-grade of Ablaut).

    Oh, er, and… Proto-World and Amerindian are products of mass lexical comparison and thus of phenetics. Nostratic, on the other hand, has been brought to you by the Good Old Comparative Method®, regular sound changes and all. It may still be wrong, but it must be taken seriously.

    * Convergence + reversals ( = convergence on an ancestor) + borrowing/horizontal gene transfer.


    Another question: What evidence is there for the commonly cited age of 5000 years for Pama-Nyungan?

  2. Reversal isn’t usually a problem in historical reconstruction, but undetected parallel convergence definitely is.

    The main evidence for 5-6000 years for Pama-Nyungan is mostly from a combination of these factors:
    a) the languages aren’t hugely different, so a very long time period is implausible (judging from elsewhere in the world)
    b) there is evidence for early-mid Holocene intensification in quite a few parts of Australia, although the timing varies (e.g. aquifers in Victoria, more intensive marine exploitation in Queensland, and greater permanent settlement in the Central deserts). It’s all a bit sketchy but the impression I get is of very small transient hunter-gatherer bands in most of the Pama-Nyungan area until about 5,000-6,000 years ago, when we start to see an inferred increase in population, greater exploitation of resources, and so on. There’s no direct evidence to link that expansion to Pama-Nyungan, but it’s plausible, since intensification and population rise would provide a trigger for linguistic expansion (that’s my idea, anyway – haven’t seen it articulated like that).

  3. David Marjanović

    I see, thanks.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s