Hallo!

On Fri, 23 Jul 2010 09:23:35 +0100, R A Brown wrote:

>  Lars Finsen wrote:
>  [...]
>  >  I do not believe in isolates, simply because there is no
>  >  way any language, except in the very first stages of the
>  >  origin of the phenomenon of language itself, could have
>  >  arisen anywhere with no predecessors or relatives of any
>  >  kind.
>
>  ?? But I've never understood _isolate_ to mean that.  Of
>  course Basque, for example, will have had a history and
>  other related languages will have been spoken.  But that was
>  long ago - way, way before any written records existed - and
>  we have [at the moment] no way of recovering this.  Now
>  Basque remains an _island_ (that, after all, is what an
>  isolate means) surrounded by a 'sea' of Indo-European
>  languages.  Although many theories (some extremely whacky)
>  have been proposed as to possible connexions with Basque, so
>  far have none have found general acceptance.
>
>  The position is, as far as one can tell, similar with
>  Etruscan: a language of non-IE origin 'isolated' within a
>  sea of neighboring IE languages.

Indeed.  A language isolate is *not* a language that never had
any relatives; it is a language for which no relatives are
*known*.  It may always turn out that some other language later
turns out to be related to it, in which case it loses its isolate
status.

>  [snip]
>  >
>  >   From what I have seen of Etruscan I would guess a
>  >  probability of around 15-30% for any close relation to
>  >  IE, though I'm no expert at all.
>
>  I put 'isolated' in quotes above since, although some
>  communities do consciously cut themselves off from outside
>  contact, this was not so of the Etruscans. As their non-IE
>  language developed they are very likely to have picked up
>  lexical items and, maybe, other features from the
>  IE-speaking neighbors.

Certainly.  AFAIK, the loanwords from Greek and various Italic
languages found in Etruscan number in the dozens, if not in the
hundreds.  The language never was isolated from its neigbours
but interacted with them.  It borrowed words from other languages,
and also lent words to others; there are a handful of Etruscan
loans (some ultimately coming from Greek) in Latin, for instance;
some have even become internationalisms (such as _person_, which
is from Etr._phersu_ 'mask; role (in theatre)').

>  Such things are not exactly unknown. An obvious example are
>  the Nguni languages of southern Africa.  No one doubts that
>  they are Bantu languages; yet by long contact with Khoisian
>  languages they have not only borrowed lexical items but have
>  also developed a whole system of click consonants.

Contact-induced change is commonplace.  Some scholars even think
that most language change was contact-induced, but that overstates
the matter.  Languages with little contact to other languages also
undergo change, even if they indeed tend to be more conservative
than languages in close contact with others.

>  Jörg Rhiemeier wrote:
>  [snip]
>   >
>   >  The Indo-European languages of Anatolia are the least
>   >  well understood members of the family, but we understand
>   >  them well enough to tell that Etruscan is not one of
>   >  them, if that is what you are hinting at.
>
>  Yep - in view of the ancient tradition that the Etruscans
>  came from Lydia in Asia Minor, that was the first place that
>  scholars looked for evidence of linguistic relationships. No
>  such connexions have been identified that win any general
>  acceptance.

Right.  We don't really know whether the Etruscans came from Asia Minor
or not.  The ancints said so, but they aren't always reliable.  The
Lemnos stele is better evidence, but even that could have been an
interloper.  I don't know what to say about the "Raetic" inscriptions
in northern Italy, but AFAIK they are so fragmentary that nobody can
really say whether they have anything to do with Etruscan or not.
I have seen other interpretations as well.

>   >  When I said
>   >  that I discarded the idea of a relationship between
>   >  Indo-European and Etruscan, I didn't mean that the
>   >  languages are in my opinion utterly unrelated; I just
>   >  meant that *as for now*, the evidence is *not sufficient*
>   >  to posit a *particularly close relationship* between IE
>   >  and Etruscan.  I consider it very likely that the closest
>   >  living kin of Indo-European is Uralic, and Etruscan
>   >  doesn't seem to be closer than that.
>
>  Yep - we're probably going back into 'nostratic' territory
>  and that's a whole different can of worms   ;)
>
>  I think these theories are valuable in that they get people
>  thinking and looking more closely at evidence, as long as we
>  remember these are theories. Since ....

Indeed.  Few historical linguists would say that something like
Nostratic, Eurasiatic or Indo-Uralic were impossible the way a
perpetual motion machine was, but most are not yet convinced by
the evidence adduced so far.

>   >>  Anyhow, science build models and make progress dealing
>   >>  with realities based on probabilities and informed
>   >>  guesses as well, not just hard evidence.
>   >
>   >  Sure.  Science will progress, and new linguistic
>   >  relationships are being established.
>
>  Of course - no one, as far as I know, has denied this.  Work
>  goes on patiently with Etruscan and our knowledge gradually
>  increases.  The same with other 'isolates' (in the sense
>  that I understand the word). But the point I was making is
>  that in the case of Ugaritic the cognates languages are well
>  known and well-attested.  The computer program had a whole
>  body of sound data to make comparisons with.  In the case of
>  Etruscan we simply do not know what languages it is more or
>  less related to.

Exactly.  It is well-known that Ugaritic is a West Semitic
language closely related to Phoenician, Hebrew and Aramaic.
Etruscan, on the other hand is what is commonly termed an
"isolate", i.e., a language whose external relationships are
not yet known.

>  [...]
>
>  I think the case of sign languages is not exactly
>  comparable. It is all too easy to see how deaf communities
>  could become isolated from other similar groups and have to
>  begin again.

I know virtually nothing about the history of sign languages, but
couldn't it be possible that there is much undocumented and
unexplored history involved, and some sign languages are older
than we think?  And isn't it conceivable that some sign languages
evolved from finger-spelled spoken languages?  (However, I *am*
aware that the major sign languages of Europe are structurally
very different from the major spoken languages of Europe.)

>        But I find it more difficult to imagine a
>  situation in which a group of hearing people could find
>  themselves together without any previous speech.  I suppose
>  it is possible that on occasion feral children with no human
>  speech did meet up and then evolve some spoken communication
>  from scratch, so to speak. But will this have happened often?

Probably not.  We can't be sure that language wasn't developed
more than once, but many kinds of human-usable languages are
conceivable and yet not found in "the wild".  If language was
"invented" more than once, one would expect that some languages
use whistling, or melodies or whatever, rather than consonants
and vowels - but apart from some secondary encodings of ordinary
spoken language (such as El Silbo de Gomera), none have been
discovered so far.

>   >
>   >  A similar thing goes on with the evolution of pidgins to
>   >  creoles (for example, in Hawaii).
>
>  Ah, now that's a whole different matter.  I think it very
>  likely that as communities drifted apart and their language
>  changed and evolved beyond recognition, there would be many
>  occasions of groups with languages which were no longer
>  mutually intelligible forming pidgins and creolization
>  taking place. It must, I think, have happened time and time
>  again ever since the first speaking hominids set out from
>  Africa - how many millennia ago?

Sure.  In my opinion, the exceptionality of pidgins and creoles
is overrated anyway.  Isn't it just an extreme case of imperfect
learning combined with interference between differently-structured
languages?  The phenomenon may be related to substratum-induced
restructuring of languages (as some scholars assume to have happened
with Insular Celtic).

--
... brought to you by the Weeping Elf
http://www.joerg-rhiemeier.de/Conlang/index.html