# Nostratic?: a Debate **URL:** \ **Category:** Great Debates\ **Created:** [July 24, 2003, 3:43am UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541 "2003-07-24T03:43:41Z")\ **Posts on this page:** 12\ **Page:** 3
**Author:** ![Monty](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/monty/32/47_2.png) [@Monty](https://boards.straightdope.com/u/Monty)\ **Post date:** [July 25, 2003, 8:21am UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541/41 "2003-07-25T08:21:24Z")
I suppose a link would’ve been helpful. From [http://ling.lll.hawaii.edu/faculty/stampe/Linguistics/lgsworld.html](http://ling.lll.hawaii.edu/faculty/stampe/Linguistics/lgsworld.html) > [@](#): > > A few points of terminology should be explained before further discussion of the world’s languages is afforded. Language family is the label often used for a conservative genetic classification, one that can be attested only when an abundance of cognates (related words) is available. Phylum is the label for a liberal genetic classification that is attested with fewer cognates; it encompasses language families. Although a given phylum will have greater extension than any of the families included in it, only fragments of phonology will be reconstructible in the protolanguage. In actual linguistic usage, however, the term family is often employed to refer to a group that is technically a phylum–e.g., the Afro-Asiatic (Hamito-Semitic) family, the Sino-Tibetan family. ---
**Author:** ![matt\_mcl](https://avatars.discourse-cdn.com/v4/letter/m/7ab992/32.png) [@matt\_mcl](https://boards.straightdope.com/u/matt_mcl)\ **Post date:** [July 26, 2003, 1:45am UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541/42 "2003-07-26T01:45:36Z")
> [@](#): > > What are the tests for differentiating “false cognates” from legitimate lexical kin? It has to do a lot with reconstruction. Historical lxists painstakingly trace back words through to their ancestors by studying words from different modern-day languages, applying what we already know about sound changes, and slowly trying to reconstruct the parent languages. False cognates (in this context) are ones where there is no reason to believe that they are related. For example, there’s no reason at all to believe that “dog” in English and the Aboriginal word I mentioned are related, just based on what we know of the ascent of both languages, the sound changes that produced the English word, and so forth. The same goes for other cases of paired cognates. Another example: I mentioned the true derivation of the word “wheel”. If it somehow came out that the word for “wheel” in Salish or something was also /wi:l/, it’s very improbable that those two words came from a common source, because several dozen thousand years ago, the latest time a common ancestor could have existed, the cognate hadn’t yet even become “chakra”, let alone “wheel”, and it would have had to follow exactly the same path of sound changes to turn into /wi:l/ in Salish, despite the fact that no other such word turned into anything resembling English in that language. Furthermore, it would be extremely unusual for two languages to share a common cognate for, say, “eggplant,” and not one for “mother”, “brother,” “one,” “two”, or so forth. (Actually, this isn’t true. It would be extremely usual - but it would be because they had _borrowed_ the word from the same source, or one from the other, not because that one root happened to have a common ancestor in both. If it’s impossible that they had borrowed the word, it would be more likely that it was a coincidence.) Before I get too woolly, here’s a great discussion on chance relationships between languages: [http://www.zompist.com/chance.htm](http://www.zompist.com/chance.htm) . ---
**Author:** ![matt\_mcl](https://avatars.discourse-cdn.com/v4/letter/m/7ab992/32.png) [@matt\_mcl](https://boards.straightdope.com/u/matt_mcl)\ **Post date:** [July 26, 2003, 1:47am UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541/43 "2003-07-26T01:47:12Z")
(that should be chance _resemblances_, of course.) ---
**Author:** ![pantom](https://avatars.discourse-cdn.com/v4/letter/p/34f0e0/32.png) [@pantom](https://boards.straightdope.com/u/pantom)\ **Post date:** [July 26, 2003, 4:13am UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541/44 "2003-07-26T04:13:09Z")
Has anyone tried making connections through grammar? I was thinking a little more about the polysynthetic North American languages, and about the fact that the general tendency in Indo-European is to become less synthetic over time. Thus the modern IE languages are synthetic only to a small degree. IIRC, the opposite tendency is/was present among native North American language families. Given that, and the fact that evolution within families tend to follow the same paths, does anyone at all consider broad grammatical similiarities like the above when doing historical linguistics? Does anyone think it proves anything? A more specific question: it appears from the reading I’m doing on the Web that Macro-Siouan is generally accepted. Back in my college days a connection between Sioux and Iroquois was suggested but not really accepted. Has this now become less controversial, to the extent that a majority think it a reasonable family? ---
**Author:** ![Monty](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/monty/32/47_2.png) [@Monty](https://boards.straightdope.com/u/Monty)\ **Post date:** [July 26, 2003, 4:21am UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541/45 "2003-07-26T04:21:54Z")
pantom: Languages also tend to have a shift in their grammar over time; therefore, it’s not always a good indicator. Even languages in the same phylum, or those in the same family within that phylum, could have different grammatical shifting. BTW, IMHO, “evolution” implies going from a subordinate to a superior version. Languages aren’t exactly evolving. They’re changing. ---
**Author:** ![pantom](https://avatars.discourse-cdn.com/v4/letter/p/34f0e0/32.png) [@pantom](https://boards.straightdope.com/u/pantom)\ **Post date:** [July 26, 2003, 4:31am UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541/46 "2003-07-26T04:31:32Z")
Well, Stephen Jay Gould might disagree with you about that evolution thing, but noted anyway. ---
**Author:** ![Monty](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/monty/32/47_2.png) [@Monty](https://boards.straightdope.com/u/Monty)\ **Post date:** [July 26, 2003, 4:10pm UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541/47 "2003-07-26T16:10:54Z")
I’ll agree with Gould on languages “evolving” instead of merely changing as soon as someone discovers a primitive natural language. To date, hasn’t happened. 🙂 ---
**Author:** ![Wendell\_Wagner](https://avatars.discourse-cdn.com/v4/letter/w/8491ac/32.png) [@Wendell\_Wagner](https://boards.straightdope.com/u/Wendell_Wagner)\ **Post date:** [July 28, 2003, 11:21pm UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541/48 "2003-07-28T23:21:19Z")
For what it’s worth, a Google search on “Indo-European family” gets 6,800 hits, while a similar seach on “Indo-European phylum” gets 11 hits. I’ve never heard anyone before use the term “phylum” for Indo-European (but I admit I’m not super-current on this field). One reason that someone might use the term “phylum” for Afro-Asiatic is that its proto-language was spoken longer ago than Indo-European (10,000 years as opposed to 6,000, I think), so it’s slightly more conjectural than Indo-European, so perhaps it thus deserves the name “phylum” more than Indo-European. ---
**Author:** ![Monty](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/monty/32/47_2.png) [@Monty](https://boards.straightdope.com/u/Monty)\ **Post date:** [July 29, 2003, 12:36am UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541/49 "2003-07-29T00:36:00Z")
And the more likely reason that one would use Phylum would be that it’s more correct, in a scientific way, than calling a “family of language families” a family. ---
**Author:** ![eburacum45](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/eburacum45/32/8690_2.png) [@eburacum45](https://boards.straightdope.com/u/eburacum45)\ **Post date:** [July 29, 2003, 2:18am UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541/50 "2003-07-29T02:18:18Z")
I think what **pantom** was saying was that evolution does not really imply progression, but simply change; which is more or less what you seem to be saying as well, **Monty** ; Jay Gould always stressed that there is no direction to evolution from simple to complex, except by random variation; this might clear up the debate about linguistic evolution I have seen on this board (but it might not). ---
**Author:** ![Johanna](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/johanna/32/8318_2.png) [@Johanna](https://boards.straightdope.com/u/Johanna)\ **Post date:** [July 30, 2003, 1:18am UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541/51 "2003-07-30T01:18:39Z")
Vocabulary is more persistent than grammar. For example, the systems of case endings in modern Uralic languages, like Finnish, Hungarian, and Khanty, cannot be reconstructed back to Proto-Uralic. However, there has been plenty of Proto-Uralic vocabulary reconstructed. The grammatical forms are a later development, which each branch of Uralic mostly developed independently of the others. No one has proposed any Nostratic grammatica reconstruction, only vocabulary lexemes. ---
**Author:** ![Polycarp](https://avatars.discourse-cdn.com/v4/letter/p/82dd89/32.png) [@Polycarp](https://boards.straightdope.com/u/Polycarp)\ **Post date:** [July 30, 2003, 3:59am UTC](https://boards.straightdope.com/t/nostratic-a-debate/190541/52 "2003-07-30T03:59:26Z")
> [@](#): > > \*Originally posted by Jomo Mojo \* > \*\*Vocabulary is more persistent than grammar. For example, the systems of case endings in modern Uralic languages, like Finnish, Hungarian, and Khanty, cannot be reconstructed back to Proto-Uralic. However, there has been plenty of Proto-Uralic vocabulary reconstructed. The grammatical forms are a later development, which each branch of Uralic mostly developed independently of the others. No one has proposed any Nostratic grammatica reconstruction, only vocabulary lexemes. \*\* Go into this a bit more, **Jomo** , if you will. I was under the impression that some of the basics underlying grammar were among the most conservative elements of a language, though forms may become obsolescent and other forms become significantly more common. E.g., the English past indefinite in “I used to…” originally had the very specific meaning of “My use (i.e., wont, _modus operandi_) was to…”; periphrastic constructions may replace simple tenses; it being very rare to see the absolute construction in English outside a few idiomatic usages; verb forms become archaic [only in one hymn and one line in Shelley does the 2nd person singular past subjunctive have a separate English form – “…which wert and art an evermore shalt be” (a subordinate clause in direct address, as it happens) and “Hail to thee, blythe spirit; bird thou never wert.”] Nonetheless these forms survive, sometimes in “fossilized” form as if idioms (“If this be true…”) But English will not suddenly develop an aorist tense, verb aspects, an ergative (about which I’m still most curious), etc. A sixth grader in Athens will be able to wade through Homer (probably with a good dictionary at his side. I have a feeling I’m obfuscating rather than making a clear point. Do you see my basic question from this? [Previous page](https://boards.straightdope.com/t/nostratic-a-debate/190541.md?page=2)