In a previous entry I promised a series of posts with excerpts from Dixon's Memoirs of a Field Worker, and since three commenters in that thread mentioned the story of the Mbabaram word for 'dog,' I think I'll start with that. Here's the setup:
When Ken Hale had sent the Jabugay tape, he'd urged me to find a speaker of Barbaram, the apparently aberrant language that Lizzie Simmons ["eighty years old, toothless, and cranky"–p. 54] had declined to speak to us. Certainly Dyirbal and Jabugay had very normal Australian grammar and vocabulary, not radically different from the Western Desert language, almost two thousand miles away. But from the few words that Norman Tindale had published of Barbaram, that language looked really different.And here's the payoff, from his visit the following year:People at Mareeba had mentioned Albert Bennett, at Petford, and early one Sunday morning I set out to try to locate him... Albert was an oldish, square-framed man with curly grey hair. He was sitting stolidly on a bench just outside his open front door. I introduced myself, but he really wasn't very interested. He didn't remember any Barbaram language, but who'd want it anyway? What good was it?... Finally he volunteered a word.
"You know what we call 'dog'?" he asked. I waited anxiously. "We call it dog." My heart sank... [pp. 105-107]
Barbaram was still a major priority... I met the third and last living member of the Barbaram tribe, Jimmy Taylor, who had walked down from his barracks near the store... We had a good session, getting another seventy-five words and—even more important—bits of grammar... Most exciting of all, I could see a relationship between Barbaram and the other languages I'd studied. "Stomach" is bamba in Dyirbal but mba in Barbaram; "we two" is ngali in Dyirbal and Wagaman but li in Barbaram... Barbaram had simply dropped off the initial vowel and consonant... So Barbaram did seem to be a language of the Australian family, only it had undergone a quite regular change that had produced odd-looking words. Stress probably shifted from first syllable (as in Dyirbal) to second syllable—bámba to bambá. Then the first syllable was gradually dropped off in pronunciation, yielding modern mbá... [Dixon discovers at this point that the name of the language is actually Mbabaram and not Tindale's "Barbaram."]For a linguist, that kind of insight is as thrilling and beautiful as a really nice proof for a mathematician (which is what I once intended to be, and I still remember my excitement on understanding Gödel's proof when I read this excellent bookFour years later, when I was spending a year at Harvard and first met Ken Hale, he pointed out that the e and o had developed in Mbabaram in the same sort of way as in some languages he had worked on from further up the Cape York Peninsula. An a in the second syllable of a word had become o if the word had originally begun with g. So from guwa "west", Mbabaram had derived wo. We were sitting on a bench near Gloucester, Massachusetts one Sunday in September when Ken suddenly saw the etymology for dog "dog". It came from an original gudaga, which is still the word for dog in Yidin (Dyirbal has shortened it to guda). The initial g would have raised the a in the second syllable to o, the initial gu dropped and so did the final a (another common change in the development of Mbabaram). Ergo, gudaga became dog—a one in a million accidental similarity of form and meaning in two unrelated languages. It was because this was such an interesting coincidence, that Albert Bennett had thought of it as the first word to give me. [pp. 125-129]
). And now I have another instance to set beside my standard Persian bad 'bad' when explaining to people that similar words are not necessarily related.
Posted by languagehat at January 12, 2006 04:23 PM
LH, did you know that Dixon was also a mathematician before switching to linguistics?
Posted by: nomis at January 12, 2006 05:01 PMChance resemblances between Indo-European and Austronesian words are always fun...
"two"
dua (Malay, Sundanese, Balinese, Ilokano, etc.)
due (Italian), duo (Latin), dua (Pashto)
"eat"
mangan (Javanese, Ilokano, Kapampangan)
mangiare (Italian), manger (French)
"same"
sama (Malay)
samma (Swedish), same (English)
"heart"
hati (Malay)
hjärta (Swedish), heart (English)
nomis: No, I didn't, but there seems to be a correlation. I've known other cases.
Posted by: language hat at January 12, 2006 06:17 PMChance resemblances between Indo-European and Austronesian words are always fun...
Hmmm, Ben.
Let us first note that Australian languages are not Austronesian languages. I imagine, but do not know, that some northern Australian languages may have some influence from Austronesian languages, since there was intercourse between our continent and SE Asia for many centuries.
Next, while I not an expert in the area, I put it to you that some of the resemblances you remark upon are not due to chance. At least in the case of sama, there seems to be a direct link with the Sanskrit (and to English, Russian, and so on). Malay and other Austronesian languages have a good deal of vocabulary from Sanskrit (and also Pali, etc.?) because of the ancient dominance of Indian religions and culture in the region.
A good resource:
http://crcl.th.net/indic/indo.htm
Why should any Australian language have a truly ancient "indigenous" and un-Indo-European word for dog, in any case? Indigenous Australians have been here for tens of thousands of years; but dingoes (the "native dogs", which came from SE Asia) for only an estimated five thousand. And then, any extension of vocabulary beyond dingoes to cover dogs in general must come after European settlement, which brought dogs other than dingoes.
It would take some argument to convince me that the Mbabaram word is not derived (simply or complexly) from the English word dog.
Huh? Did you read the second part of the post? The word is derived by perfectly regular sound change from the inherited and well-attested gudaga. To think that they happened to borrow a word which happened to coincide with the word they would have had if they'd inherited one is... well, "unlikely" is so severe an understatement I'm at a loss for words.
Posted by: language hat at January 12, 2006 06:58 PMHuh? Did you read the second part of the post?
Sure I did! Let's focus especially on this:
It came from an original gudaga, which is still the word for dog in Yidin (Dyirbal has shortened it to guda).
If this is correct, whence, in the light of my observations about dingos and dogs generally in Australia, did gudaga come from?
I wanted an argument, recall accurately, that the Mbabaram word is not derived (simply or complexly) from the English word dog. And I would want that argument to show that the current Mbabaram word is not etymologically overdetermined.
Do you really think the inhabitants of Australia spent 5,000 years waiting for someone to come and tell them what to call dingoes?
Posted by: nomis at January 12, 2006 08:00 PMDo you really think the inhabitants of Australia spent 5,000 years waiting for someone to come and tell them what to call dingoes?
No. Did you think I might? We're discussing Aboriginal words for dog; I was providing some background (accurate, I hope: but I'm ready to be corrected concerning it), since many here will not be aware of these facts.
Noetica, I never claimed that Australian languages are Austronesian. I was simply providing LH with more examples of chance resemblances across language families. And also, I'm well aware of the Sanskrit influence in Malay and other regional languages, but none of the examples I gave are Sanskrit-derived.
Posted by: Ben Zimmer at January 12, 2006 08:10 PMI never claimed that Australian languages are Austronesian.
Ben, sorry if my intention was unclear. For the record I do not say that you make that claim, in your quite interesting post. It's just that, in the context, it may not have been clear to all readers that the focus had shifted to Austronesian from Australian. I have seen the two confused before.
As for sama, what is your evidence for it not being derived from Sanskrit? The resource I cite certainly suggests a connexion.
Why should any Australian language have a truly ancient "indigenous" and un-Indo-European word for dog
If dingoes have been in Australia for 5,000 years, the chances are pretty good that Australian languages have non-Indo-European words for 'dingo'. If we assume that gudaga > dog originally referred to 'dingo' and was extended to cover other types of dog, there's no need to postulate any Indo-European influence.
Posted by: nomis at January 12, 2006 08:24 PMYes, Nomis. If! I'd want more information on whether this is a safe assumption.
The relations between the classes dingo and dog in modern Australian English are problematic. Relatedly, so are law and regulation concerning them.
If Lindy Chamberlain had cried out "A dog's got my baby!", very few who heard her would think she meant to include the case covered by "A dingo's got my baby!" Of any who did think that, most would have thought that there was either a misuse of the language or else a misperception.
For all we know, on the evidence given here so far, the relations between dingo and dog are similarly problematic in indigenous Australian languages and cultures, and also in the complex pidgin and creole languages and the milieux associated with them.
Posted by: Noetica at January 12, 2006 08:41 PMHere's a site where you can find many 'Amazing Coincidences in Linguistics' including the Mbabaram word for dog. (I sure hope the html works properly in these comments, otherwise readers will see silly code!)
I'm not sure, but I'm willing to bet there's some connection between at least some of those Malay words (dua, etc.) and Sanskrit. One scholar of the Malay language suggested that only three Malay words are actually of Malay origin: batu (stone), kayu (wood), and babi (pig). Everything else came from Sanskrit (and later Hindi), Arabic, Javanese, Portuguese, Chinese, and of course English. I think there must be more than three words of Malay origin, but I can also see that, given the fact that so much of what you can find in a Malay dictionary came from elsewhere, the large number of coincidences might not be coincidences at all, kan?
Posted by: Jordan at January 12, 2006 08:52 PMOK, much to comment on. Sorry for the list and the lecture-tone.
* Dingos aren't dogs, they're mostly closely related to wolves of SE Asia, hence Noetica's comments.
* Words for both dog and dingo are highly varied and hard to etymologise over much of Australia. Dingo itself is supposed to come from the word for camp dog in (I think) Guugu Yimidhirr - I can't check this now but will do so tomorrow.
* Malay dua has no connection to Indo-European, as far as I know; there are cognates with r- and l- all over Oceania.
* There are indeed many loanwords from Austonesian languages in Northern Australian languages. Nick Evans wrote a nice paper on Iwaidjan loan stratigraphy (they got borrowed over about 500 years and show different participation in sound changes). There are about 800 Austronesian loans in Yolngu, according to David Zorc, such as compass points (dhimurru - east, dhalathaŋ - south, bärra' - west), there's a class of uninflecting verbs that are all Austronesian in origin, pipe terms, etc.
This gets more and more interesting...
What would be good to know is whether Mbabaram (or any related language) has a separate word for dingo. Anyone know what the general pattern is in Australian languages? I just had a look at dictionaries for three languages, of which Kayardild appears to have one word for both, while Ngalooma and Warlpiri have separate words.
Posted by: nomis at January 12, 2006 09:47 PMMark Rosenfelder has a great page on this very topic: "How likely are chance resemblances between languages?". It's very much worth reading even if you have no more than a smidgin of probability theory.
Posted by: John at January 12, 2006 09:50 PMBardi has separate words, as do all Nyulnyulan languages where both are documented. Yolŋu Matha (can't speak for all varieties, but the ones I know) has different words. Arrernte and Kaytetye have separate words, I'm pretty sure.
Posted by: Claire at January 12, 2006 10:07 PMAny idea where the words come from Claire? Particularly those for 'dog', which must be more recent.
Posted by: nomis at January 12, 2006 10:22 PMnope, sorry. They'll all a pain for etymology. but fwiw the Nyulnyulan word for dog, *yiila, has regular correspondences in the languages (Bardi iila, Nyulnyul yil, Nyikina yila, for example) so it's a good candidate for Proto-Nyulnyulan.
Posted by: Claire at January 12, 2006 11:45 PMThe words I gave in Malay, etc. for "two", "eat", and "(figurative) heart" are pure Austronesian. My apologies about sama -- didn't realize that one had a Sanskrit root. Live and learn.
Posted by: Ben Zimmer at January 13, 2006 12:33 AMWikipedia says that "the name dingo comes from the Eora Aboriginal tribe who were the original inhabitants of the Sydney area." Couldn't say whether this is true or not.
Posted by: David Costa at January 13, 2006 02:34 AMThis gets more and more interesting...
It sure does. Sorry I was so dismissive, Noetica; I had no idea the situation was so messy. This is all making me very nostalgic for my days as a proto-linguist...
Posted by: language hat at January 13, 2006 07:07 AM"Dingos aren't dogs, they're mostly closely related to wolves of SE Asia, hence Noetica's comments."
I think that at the level of ethnozoology, this fact is unimportant. Dogs are closely enough related to coyotes, wolves, and foxes to interbreed, and some dogs resemble wolves more than they do other dogs. Furthermore, the aborigines had never seen wolves. To me the statement "dingos are dogs" is true, because dogs are just tame canines.
Whether native Australians agree is unknown to me. If there are distinct words, does that mean that native Australians think that dingos aren't dogs and dogs aren't dingos; or does it mean that dogs are a subset of dingos; or that dingos are a subset of dogs (as we would say)?
not having the vaguest idea of how Australian languages work but having some knowledge about how semantic fields operate and how native peoples react to new, imported animals, my working hypothesis would be:
in the great majority of languages the native word for Dingo (however it arose) would be extended to the dogs that arrived with Europeans (perhaps with some sort of derivational modification, perhaps not). Perhaps in some cases, the word for dingo would come to be applied exclusively to european dogs and the word for Dingo would have some sort of derivational modficiation (this happened with pig and possum in a couple of SE US Indian languages).
I think English borrowings for one or both might also displace the local name(s) for dog/dingoes but in this case, it does look more like regular phonetic changes yielding a chance resemblance. Of course it's quite possible that the last few speakers thought the word was borrowed from English.
Posted by: michael farris at January 13, 2006 08:37 AMIn support of Michael's working hypothesis: I did fieldwork on a coastal language of New Guinea in which the 'native' neologisms for new kinds of passenger vehicles extend the meaning of 'canoe', so that submarine was 'underwater canoe' and airplane was 'treetop canoe'. Of course, most of the time, people just used borrowed words, like Tok Pisin balus 'airplane' (itself extended from the Tolai word for 'pigeon', I believe).
I'm not sure how to render 'submarine' into Tok Pisin. Probably bot bilong ananit long solwara (boat of underneath in/to/at sea [< saltwater]).
Another of my favorites along these lines is the Romanian way to distinguish two vegetables that originated in the New World: tomatoes are 'reds' (rosii) and eggplants are 'purples' (vinete), both subtypes of patlagele (whatever a patlagea used to be).
Posted by: Joel at January 13, 2006 04:08 PMMust have been an eggplant; that's the pan-Mediterranean word. Persian bādingān was borrowed into Arabic as bādinjān (which is now the Persian word as well) and from there it spread in a wondrous variety of forms, such as Catalan albergínia (the source of English aubergine). The Romanian word clearly comes from Turkish patlıcan (also the source of Russian баклажан baklazhan), but I'm not sure of the relation of the Turkish word to the Arabo-Persian. I should really do a post on this.
Posted by: language hat at January 13, 2006 04:26 PMThanks, LH. Yes, Rom. patlagea (final stress) patterns like such Turkish borrowings as cafea 'coffee', saltea 'mattress', pijama 'pajama', basma 'headscarf', mahala 'slum'.
Posted by: Joel at January 13, 2006 04:38 PMyes, please do a post on words for eggplant, they are incredibly cool! don't forget brinjal while we're at it.
Eora is probably right, I couldn't remember if it was a Sydney word or a CYP word.
Relevant here before we all say what semantic extensions we expect everone to make is that there's often a split between words for wild dog and tame dog. E.g. yolŋu rirrpi-rirrpi or wärrany (scary dog) vs wuŋgan or waṯu (tame dog).
Michael, I wish I had your confidence in how semantic fields and extension operate. For example, if I saw a kangaroo without having a name for it, I think "rat" is not the first I'd pick. But that was what Dutch settlers to WA called them. And if I saw a horse, I don't think I'd call it a kangaroo, but that's an attested source of a word for horse in some Aboriginal languages. The more I do historical linguistics the less confidence I have in generalisations like this.
Posted by: Claire at January 13, 2006 04:47 PM
The principal I'm talking about is that new animals often get names of already known animals though the similarities may not be immediately obvious. Both rat-kangaroo and kangaroo-horse seem completely reasonable though sheep as big-rabbits [Koasati] seems a little odd to me.
I'm not saying that happened every time just that it's a reasonable assumption rather than assume that English semantics (dingo =/= dog) apply to aboriginal naming patterns and that it's economical to assume that dogs would most likely be incorporated into whatever category dingoes inhabited previously (real evidence they didn't would trump hypotheticals though).
Posted by: michael farris at January 13, 2006 05:06 PMLH:
Dismissive? No problem. An understandable reaction.
Claire:
For example, if I saw a kangaroo without having a name for it, I think "rat" is not the first I'd pick. But that was what Dutch settlers to WA called them.
I had thought that the Dutch in WA considered the quokkas to be rats, and this is how Rottnest Island got its name (= rat-nest). I was on Rottnest very recently, and handfed some rather tame quokkas. Cute, and not entirely unratlike in size and general conformation.
SOED on quokka:
A small rare short-tailed wallaby, Setonix brachyurus, of coastal scrub in SW Australia.
Are you sure that other macropods were also called rats?
I wonder also about the propriety of calling dingoes wolves (as you do above), in any but a very loose and assimilative sense. Some binomials to conjure with, which may be germane:
Canis familiaris
[common, normally tame and domestic, dog]
Canis lupus
[wolf]
Canis dingo
[dingo]
And while we're at it:
Vulpes
[fox genus; Vulpes vulpes is the common red fox]
I know that Canis dingo and Canis familiaris can interbreed freely, for what that classic criterion is worth. (I was for a year the custodian of a dingo-German Shepherd cross, and a beautiful and smart beast it was, too.) And it is said that Canis familiaris arose from Canis lupus. Can they, and do they, still freely interbreed? I doubt that foxes can, with any of the other types mentioned here.
Just more grist, for those still interested. I'm sure that these classifications are disputed anyway, and are of limited and provisional use only.
Michael:
principal
= principle [*ahem*!] A mere slip, I'm sure.
Posted by: Noetica at January 13, 2006 06:07 PMClaire:
O, correction. You don't say that dingoes are wolves, but they are most closely related to "wolves of SE Asia". I should have more accurately wondered, and now do, whether this is right. Are there wolves properly so-called in SE Asia, and if so, are dingoes most closely related to them, or rather to some other Canis of SE Asia?
re my previous point, David Nash rightly points out that the rat-kangaroo thing isn't right - they were quokkas. BUT, and this kind of does support my point (assuming we can rule out bad eyesight), in 1658 they were assumed to be a type of cat. It was when De Vlamingh landed in 1696 that he thought they were rats (hence "Rottenest" (= rats' nest) Is off the coast of WA). And, incidentally again, while I clearly got that wrong, I'm sure I've read of another case of kangaroos/wallabies being classified as rats ... off to hunt for it.. actually, I think it's in a book that lives in Boston (the perils of a long-distance relationship).
Posted by: Claire at January 13, 2006 06:19 PMAren't they supposed to be more closely related to Canis lupus? I'm going on my memory of a 2004 study of Dingo DNA.
Yeah, they're lovely creatures. Except when they live under your house and get upset at 3am.
Posted by: Claire at January 13, 2006 06:28 PMDingos *and* vinete. This is very, very good.
Posted by: zaelic at January 13, 2006 09:46 PMA useful site concerning dingoes and others of genus Canis:
http://www.lioncrusher.com/animal.asp?animal=168
See especially the taxonomic note at the end.
Noetica, no, actually probably not a slip but a consistent generalized tendency in internet writing for phonetic substitutions (I don't do it in other kinds of writing). I'm not the only one either, some bloggers do so with some frequency. (I don't confuse they're, there, their in other types of writing either but do so continuously in posting, sometimes I catch it, sometimes I don't).
Posted by: michael farris at January 14, 2006 03:23 AMI understand, Michael. It happens to me too.
What, it happens to you too, N(A)? And to you, Michael?
I feel better and better, what with the previous vindication of my "to no end" usage and this phonetical mix-ups (in the ;etter to a friend I just substituted thing for think)
Yes, please, LH, I too would love to see a post on baklajan, my favorite August vegetable. Do you know in Southern Ukraine we still call it синий (синенький) >/i>?
Dingo: since reading this book at the tender age of 9 I'm convinced dingo is indeed "a dog from the wild".
Posted by: Tatyana at January 14, 2006 04:17 PMNomis:
> LH, did you know that Dixon was also a mathematician before switching to linguistics?
Good thing he changed fields then! In his original Dyirbal book (1972), he gives what he says is a mathematical justification for "50% equilibrium level for adjacent languages" hypothesis. His maths is hopeless -- based on the assumptions he makes in his mathematical model, adjacent languages would end up 100% identical, not 50%. Alpha and Nash (Austr J Ling, 1999) point this out. In Dixon's second Cambridge Green Book, "Australian Languages", 2002, Dixon not only repeats his derivation completely unchanged, but makes detailed reference to other aspects of A&N's article in the same section! Apparently, he read their refutation, but didn't realise what it was.
As a mathematician, he's a great linguist.
Posted by: John Atkinson at January 15, 2006 04:30 AMDavid Costa:
> Wikipedia says that "the name dingo comes from the Eora Aboriginal tribe who were the original inhabitants of the Sydney area." Couldn't say whether this is true or not.
True. Both the words used by the white settlers for dingos are from the Sydney language Dharug (of which the "Eora" were a subgroup). "Dingo" is from /din-gu/ or /dayn-gu/, domesticated dingo. "Warrigal" is from /warrigal/, wild dingo. (Ref: Dixon, Australian Aboriginal Words in English")
All the Australian languages I've come across seem to have separate, unrelated, words for domesticated and wild dingos. The actual words vary widely.
Posted by: John Atkinson at January 15, 2006 04:49 AMAnother vote for an eggplant word roundup.
Recall (from, among other places, Hackers Greenblatt and Gosper chapter) that Chinese restaurant menu for tomato is 番茄 (fan1 qie2) 'barbarian eggplant'.
Posted by: MMcM at January 16, 2006 03:41 PMNO, NO, NO - this is the book you want to read to find out about Gödel's proof: Gödel, Escher, Bach by Douglas R. Hofstadter.
http://www.amazon.com/gp/product/0465026567/104-0376951-2066313?v=glance&n=283155
It's probably the single best book I've read over the past 5 years.
Posted by: Guy at January 17, 2006 04:03 AMGuy, no offense, but you haven't a clue what you're talking about. I loved GEB too, but it's no match for Nagel and Newman's mathematical insight. Hofstadter is a Professor of Cognitive Science and Computer Science; Adjunct Professor of History and Philosophy of Science, Philosophy, Comparative Literature, and Psychology. Do you see "Mathematics" in there? No, neither do I. I don't suppose you've actually looked at the book I mentioned? No, I didn't think so. You might want to do that before dismissing it.
Posted by: language hat at January 17, 2006 08:13 AM