Yes they did Latro, the dingoes are descended from tame dogs that went feral thousands of years ago.
Yes, there were dingoes in Australia for thousands of years, they perhaps came over with the first humans.
No, I’m not in an Ivory Tower (at least one concerning philology in general, my education is limited to classical languages). That’s why I posted what I did. I would never dare to make a conclusive statement on a theory of a language superfamily, because I know my opinion is worth nothing without legitimate training. I was just asking why other posters don’t feel the same way. Why debate something if real experts aren’t listening?
That kind of crap has spurred countless generations of cranks.
Anyway, I’ll hijack no more. Someone can start a GD or Pit thread if they disagree with my points, I’d post further there (if I get a chance, I will be on vacation for 10 days starting tomorrow).
UnuMondo
I’ve never seen lack of formal training stop a Great Debate before – I’d venture to guess that less than 1% of those who post in political debates are formally trained political scientists, and AFAIK there are only a couple of people on here with four years of training in theology, yet religion threads proliferate and opinions get voiced as if fact. Are we to adopt a rule that from now on nobody argues with Diogenes on reqligious questions, with Unu Mondo on linguistic questions, and with Collounsbury on MENA affairs? And David B. can get a well-deserved rest?
I started this thread because I am interested in the limits of glottochronological research and whether it can be taken back before Proto-Indo-European (which someone scholarly might address – “when come back, talk about PIE,” please! ;)) – and because I had hopes that people with more knowledge than myself might address whether there are any sound grounds for accepting the Nostratic hypothesis. If you want to take that back further to a hypothetical proto-language, that’s your choice.
But, unu, do not sit back and snidely inform me that as an interested layman I have no right to an opinion. I have a right to an informed opinion, and I started this thread in the hopes of forming one.
I’m not clear on the genetic data, but I’d give the Nostratic idea more credence if there was some good DNA link between the various populations that showed them to be more closely related than they are to other living populations. I haven’t gotten around to reading Genes, Peoples, and Languages by Cavalli-Sforza, but it might be a good place to start.
As for the mono vs poly language origin debate, I think a good case can be made for either one. Even if one assumes that we went thru a “genetic bottleneck” about 75k yrs ago, and there were approx 10k people at the time, it’s unclear if those people existed pre or post the appearance of fully articulated language. If post, there could be much older language roots than we might expect.
And as for the Neanderthal question, the data is pretty mixed. The Neanderthal hyoid bone that was found a few years back is, in all key features, identical to that found in Sapiens. Given the key role this plays in the physical structures support spoken language, it is very reasonable to assume that Neanderthals were capable of spoken language as well.
One way of thinking about this is that the mitochondrial DNA evidence that shows that we’re all descended from one woman who lived however many thousands of years ago (don’t remember how far back, but I’m sure someone can come up with this) also shows rather graphically how many lines of descent just up and disappear over the course of time.
Ditto for languages and cultures. As they disappear, so too do the lines of descent for those languages and cultures. This has been going on throughout human history, so the chance that all modern languages might be descended from a common ancestor is probably higher than we might otherwise think, even if polygenesis is correct. But the evidence is staggeringly thin. (I did take historical linguistics in college, and what matt_mcl said is true, from what I remember. There’s just about no evidence of a connection between the different language families that holds up to any real scrutiny. OTOH, I always found it curious that, for instance, most Amerindian languages are polysynthetic.)
A few points for your consideration:
[ul][li]Check Language Miniature for the issue of “primitive language.” The same site has a number of other rather interesting miniatures on the subject of language.[/li][li]Way back in Intro to Linguistics, the instructor also taught that there’s no such thing as a primitive language. All languages are complex.[/li][li]DNA “trails” of a particular group do not necessarily correlate to the linguistic history of that group owing to the fact that some groups have abandoned, either willingly or by force, their language in favor of another.[/li][li]Those Historical Linguistics who claim reconstruction in the neighborhood of 12,000 years appear to have played fast and loose with the tools of that particular science.[/li][li]Linguists (also known as Linguisticians) use a slightly different meaning of the word cognate than its meaning for someone learning a foreign language. For us, cognates are words (reflexes) in the languages we’re comparing. To be a cognate, those reflexes must have descended from the same word in the proto-language. The reflexes do not necessarily resemble each other (i.e., sound alike).[/li][li]Proto-language does not mean “that stuff folks used before they had language.” It means the language from which the current language developed. Both the daughter language and the proto-language are full-fledged languages.[/li][li]Linguistics is an actual science. I personally didn’t believe that until I started my training for this field.[/li]Lastly, Linguistics is FUN! It’s never too late to change your career, Poly![/ul]
As always, when making a semi-long post, something gets left off. Drat.
Add to the list above, preferably at the end of the 3rd item:
[ul]Sometimes languages from different phyla absorb aspects from each other when they come into contact. Just because a language has certain features (even many features) common with another language, doesn’t mean that those languages stem from a common proto-language.[/ul]
I don’t know about linguistics to make much of an argument myself, but Mark Rosenfelder has some good essays online on the subject of reconsructing pre-PIE protolanguages (he’s a skeptic):
Deriving Proto-World with tools you probably have at home
How likely are chance resemblances between languages?
Proto-World and the Language Instinct
There’s also some relevant info in the sci.lang FAQ, also maintained by Rosenfelder: How do linguists decide that languages are related? and What about Nostratic and Proto-World?
It’s certainly true that DNA evidence would not be proof of linguistic connections, but it would be one more piece of evidene to be weighed. It’s astonishing how much inof there is to be had by intelligent DNA mining. In addtion to the book I mentioned above, there are a number of good ones that have been published in the last few years (Seven Daughters of Eve; Journey of Man; for example) that I’d highly recommend to anyone interested in the subject of decerning human migration patterns from genetic data.
John, I recognize that it’s a piece of evidence; however, I don’t recognize that it’s the evidence to rely upon. I’d say that’s it’s more an indicator than a proof.
Monty: Agreed. But do we have anything other than “indicators” to work with in the case of something like Nostratic?
Is my question invisible, or just interesting?
:mad:
Uhh…uninteresting, dammit.
Add me to the list of linguists on the SDMB. Anyway, it’s a plausible theory. It’s just unprovable.
I’ve seen the big lists of cognates across language families, and it’s just not that impressive. To begin with, they’re superficial similarities. As matt_mcl said, a lot of cognates don’t often superficially look like cognates unless you understand the underlying sound changes the languages have gone through over time. The more laterally you are comparing across languages and language families and the further back in time you are required to go to establish a relationship between such languages, the more “disguised” these cognates are going to be.
Even just considering the superficial similarities when building these lists of cognates, how do you decide what sounds alike enough to qualify as a hit? Different sounds have varying significance in different languages, and the “closeness” of sounds is quite relative. There’s also a lot of semantic leeway in deciding definitions are “close enough” to be a hit. Ideally you would want to be very, very strict in defining what you will allow.
On top of that, the number of cognates found through these questionable methods isn’t any greater than what you would expect to find through sheer chance. I’d want more evidence, since languages are far more than just lists of words. Where is the syntactic and morphological evidence to corroborate the lexical matches?
-fh
What are the tests for differentiating “false cognates” from legitimate lexical kin?
I recall a phenomenon that seemed spread across a wide assortment of non-Indo-European languages where the subject of a transitive verb was in a given case, while the object of the transitive verb and the subject of an intransitive verb were in another – and that this phenomenon (for which I’ve forgotten the term) was intriguing in its distribution. As I recall, Dravidian, Kartvelian, and Chadic nouns were among those which exhibited it.
AHunter, I suspect it is less an unwillingness to address your question than that it’s one of the extensions of the entire phenomenon being discussed, of whether cross-phyletic connections are “real” or coincidental, which requires to be resolved before any intelligent answers to your question can be given.
(FWIW, native American languages, almost all of them extremely rare if not extinct now, are considred to fall into as many phyla as the rest of the world put together. A five-county area in California may be more complex in its linguistic history than anywhere outside the Caucusus or Northern Territory. And, save for some hypothetical connections between the Innuit languages and Siberian forms, there are no known interrelationships between any of them and any of the Old World groups. AHunter’s question relates to some hypothecated connections of that sort.)
On the same subject, I recall seeing a highly controversial theory that all native American languages except the Athabascan and Innuit groups were actually related in a single phylum. I’m sure that’s even more outré than Nostratic.
You mean ergative-absolutive case systems? (It’s example E on the linked page, which gives a rundown of the different approaches languages can take to this problem.) I don’t know anything about the distribution of ergative-absolutive languages, unless you count the fact that I think Basque is ergative-absolutive and that various crackpot linguists like to relate Basque to pretty much every other language in the world. 
I have studied Dravidian languages; I know of no ergativity in them (I could be mistaken, but I can’t look it up right now because I’m moving and all my books are packed up).
There is, however, ergativity in Indo-Aryan; specifically, Hindi and Pashto.
Time does not necessarily destroy the resemblance of related words; careful. Don’t assume that related words will inevitably lose all resemblance over time. Here’s a good counterexample: The Proto-Indo-European word for ‘nephew’, 5000 years old, is *nepot-. The modern Romanian word for nephew is nepot. Five thousand years and still the same. This is an unusual example, but it goes to show that resemblance does stand a chance of persisting over time.
Some minor comments: Indo-European is not a phylum. Neither is Uralic, Altaic, Sino-Tibetan, Afro-Asiatic, etc. The standard word is family. The proto-language for those families was probably spoken closer to 6,000 years ago than 4,000. Some people want to call a grouping on a larger scale (like Nostratic) a phylum, while others call it a super-family.
I can’t add anything useful to the debate on the existence of Nostratic.
Hate to break this to you, WW; however, there are a number of people in the Linguistics field who use phylum for that particular designation, and designate the various language groups within that phylum into families. Example would be: Germanic Family is part of the Indo-European Phylum.
Merriam-Webster’s online dictionary provides this as definition 2:
That being said, there are also those in the Linguistics field who make the division the way you describe.