Re: Lislakh

From: johnvertical@...
Message: 66927
Date: 2010-12-04

> Here below some 20 criticism points I wrote about IE theory
> About the some 500 supposed (constructed or more accurately guessed) hypothetical proto indo-european roots (you can find all the proto indo-european roots in wiktionnary)

I'm pretty sure Wiktionary doesn't have a complete list. Sergei Nikolayev's database, for example, has 3178 entries:
http://newstar.rinet.ru/cgi-bin/query.cgi?basename=\data\ie\piet

> 3/Many semantic shifts are very very broad to such extent that with such lax semantics many (constructed or attested)[proto or not]words&roots of different languages in the world can fit as proto Indo-European!
> 4/Many sound shifts look very unlikely and are against the sound laws!

These points would need elucidation to address. But some irregular developments always occur, that is inevitable.

> 6/Many supposed proto ie roots are most likely Semitic loans(star,three,sun,six,seven,eight,home,tree,field,pilaku[axe],barley,field,snow,door,corn,dher,goat,buck..... ) as they could not be explained from indo-european intrinsic phonetico-semantic pradigmas

Most languages have some loanwords that don't quite respect the same structure as inherited words. In English, for example, you normally only find final unstress'd syllables ending in consonants or -i ("lucky", "doggie"), or -o ("hobo", "window"). However, that doesn't stop words like "lava", "voodoo", "cafe" existing, and having existed for centuries.

So a Semitic origin does not necessarily imply a non-common origin. On the contrary, even, it is simpler to explain how a given word ended in numerous IE languages if it was only loaned once: into PIE.


> 7/Many other roots could be loans from Kartvelian,NW Caucasian,NE Caucasian,Altaic,Uralic&pre Indo-European languages of Europe(Vasconic,Pictish,Tartessian,Pelasgian,Iberi an,Aquitanian,Ligurian,Raetian,Etruscan,Wiik's Saami substratum...)

Some of these could well have donated loanwords to PIE or some late dialects of it... however, it is geographically rather improbable to have words from Caucasian languages directly end up in Celtic, or Italic, or Germanic, etc. Substrates like Pictish, Basque or Pre-Samic are even worse here.


> 8/Many supposed proto ie roots are shared eurasiatic and nostratic roots and thus could be loans.

The theory of Nostratic starts from the assumption they (or most of them) aren't loans. If you take a stance that loaning is always more likely than inheritance, you will be unable to set up any Nostratic roots to begin with.


> 9/Many supoosed proto ie roots are supported by examples of very few Indo-European branches and sometimes by only 1 ie branche,or from only 1-2 branch

> 15/The number of the common Indo-European roots shared by at least 3 Indo-European branches is very limited (106 roots) when compared for example with the number of roots shared by languages such as Semitic (more than a dozen of thousands of common shared roots wich do have own meanings) Malay languages etc...

These again sound you're relying on Wiktionary or some other not quite accurate list of IE roots and their descendants.


> 10/The sound change du=>er in the Armenian erku=2,if included will make many languages Indo-European,for example proto Semitic thnay(2)is by far closer to proto Indo-European *two(2)than the Indo-European Armenian erku(2) is.

Sounds like you don't quite understand sound change. The relevant criterion isn't the degree of apparent similarity, it's whether the sound laws are regular (occur repeatedly). For example, there is a law in Armenian that *w > g in certain positions, including this word. Later there is another law which makes all instances of *g (of whatever origin) turn to /k/.

To link the IE and Semitic numerals you'd need to demonstrate the correpondence th ~ d (it's *duwo, not *two) to be general; likewise n ~ w. Some random similarities exist between any languages imaginable. (Acehnese for example has _duwa_ "2", like a whole lot of other Malayo-Polynesian languages.)


> 11/A great number of proto indo-european roots are somehow impossibly proununciable as
> dngh2wleis=tongue

PIE reconstructions can look daunting because they're phonemic, not phonetic. For starters, any laryngeals (*h1 *h2 *h3) when occuring between consonants, should be pronounced as weak vowels, similar to e, a, o.

Words like "tongue" with syllabic consonants aren't really any worse than, say, English - we could write "buttn" rather than "button", "shrt" rather than "shirt", etc. Sanskrit (or should I say Sãskrt?) for example also has several syllabic consonants.


> 12/Very often semitic(as well as other languages)roots are closer to some indo-european reflexes than proto ie do for example proto semitic "lis" (tongue)is closer to Armenian "lezu" and Baltic "lesvis" than proto ie "dngh2weis" do.

You'd need to enumerate on that "very often". One example doesn't cut it.


> 16/Since the guessed proto indo-european roots are not found in any indo-european language(ie of course there is no for example an english word nebhos=cloud)

Old English actually has _neowol, nifol_ "deep", corresponding to German _Nebel_ "fog", Old Norse _nifl-heim_ "underworld", and from there on we can work out way down to this PIE root. But anyway…


> the reflex words in the various indo-european languages are cutted from their roots and their semantico-phonetic derivations and inner pradigmas are rather arbitrary ones lacking a clear system,

Generally true in the modern day. All this means is that the IE family has spread far and wide over a long time, and because of that diversified far more than Semitic, or Turkic, etc.

However what prompted the IE theory in the first place was the greit degree of similarity between the classical languages of Latin, Greek and Sanskrit - not only in lexicon, but also inflection, conjugation, derivation, syntax and so on.


> for example in Semitic every speaker who knows the meaning of a root automatically will know the meaning of its derivatives(example from the root *ktb, the derivative kVtVb(V stands for vowel) is always connected with the active form and nkVtVb with the passive form and so on... ie with clear and well defined paterns a system that is lacking in the indo-european daughter languages in respect with the constructed proto indo-european roots)

This kind of long-lasting transparency is a peculiarity of Semitic, not an universal property of language families. And I suspect even that transparency starts dimming up once we get into Semitic roots whose consonantal skeleton has undergone some actual changes.


> 17/
> if we would reanalyze proto indo-european roots with a semitic model perspective(ie purely consonantal roots with the ablaut of vowels serving as well established models to derivate words and various grammatical forms as well as conjugation paters)perhaps it would open a new horizon for "clogged" indo-european studies

Well, perhaps? You're free to try.

>(albeit in the same time it would create internal problems such as the below example:

That's precisely why, probably.

> ie why the initial 3 "bher" have so different semantics when they have 100% identical phonetics and why the 4 th and 5 th "expanded root" "bheres"&"bhergh" have no semantics similar with at least one of the 3 initial "bher"!?

Because they are different roots, how about?

> 20/We know from other language families(Turkic,Semitic...)that some succession of particular consonants/vowels or consonants&vowels or particular morphemes or words starting with a particular vowel and consonant are not possible but since proto indo-european is a hypothetical language we could not know much or verify the phonemic clusters and other phonetic paterns that the intrinsic proto indo-european language pradigmas DO NO ALLOW!

If PIE did not allow some particular phonotactic structure, this gap will be inherited by its descendants, and can be identified that way. For example, *tl did not exist in the beginning of a word, and this continues to be the case in most IE languages.

It is true that we cannot verify the correctness of a reconstruction, but this has to be accepted with all reconstructions, not PIE in particular.

John Vertical