Re: Odp: joatsimeo-Loan Words

----- Original Message -----

Sent: Tuesday, December 14, 1999 3:54 AM

Subject: [cybalist] Re: joatsimeo-Loan Words

joatsimeo-wrote: (and Piotr, see PS)


-- you can generally tell which are cognates, yes. Cow and Tocharian
'ku' 
and Latin or Greek 'bos' and so forth, to continue the examples.

You can't account for every word that way, however. 

For example, the Germanic languages, besides having some characteristic 
sound-shifts from PIE, also have a common group of lexical items --
mostly 
having to do with the sea and sea products -- which aren't reduceable
to 
Indo-European roots. 


-- sound changes are pretty well universal and they're extremely
regular; 
they always happen one and only one way in any given context.

------------------------------------------------------------------------
M.S.

I think you answered my question, in the affirmative – that linguists
do make the assumption that all, or nearly all words that are shared in
common, and not shown to be loan words are assumed to be cognates.

I don't understand how your comment about the directional nature of
sound change, and its predictably can insure that all shared words are
derivative words. 

I apologize for not making myself clear, I don't know the jargon used
by linguists to express the concepts clearly. Perhaps if I tell you a
couple of reasons why I am asking this question, and maybe that will
make it clearer, what I am wondering about.

In a communication to a  member concerning the origin of the word salt,
I directed them to the Web page where it state the word came from the
name of a town in the Jordan called Es Salt, and that the site was
older than Jericho!!!
I thought that if this was the case, and since salt was an essential
component for survival, that salt must have very deep origins, probably
before PIE. But later I realized that salt was  highly traded, so the
word may have come from the main suppliers.  It may have had an origin
that was very deep and transmitted down the branches – or it may have
been a product that moved freely across cultures, and took the names
with it, with each branch making those adjustment necessary – and I
couldn't see how one could tell the difference.

A more recent  examples is on the Web site Piotr just posted about
Hurrians/Mittanni – and Proto-indic language.  At that site the author
states that he does not believe that Dravidic and the Austro-Asiatic or
Munda languages should be excluded.   In other words, one very large
family of languages.  He further states that there is less difference
between between the groups he was proposing than between Celtic and
Sanskrit.

He might be right, as far as I would know.  But what I did pick up on,
was that many of the common/shared words were with items that either
had a trade value, or were likely to have been developed once, or
discovered once – and assigned a name at that time. If so, they could
have been imported into a culture, and not have been carried down an
ancestral line.  He states:

quote... the identity (was) most striking in the names of instruments,
places, and acts connected with a simple life
Also: The dominance of economic activities in the lives of ancient
Indians will be apparent from the semantic clusters compiled in this
lexicon. Semantic clusters include words expressing cognate
'thoughts'17.
The ancient economic court was dominated by plant products such as
fragrances, incenses and exudations which were highly valued and in
great demand. unquote.

Fragrances, incenses etc are high demand items, with limited sources,
hence the name assigned by the producers would likely be the one used
to identify it in all cultures.  Likewise, instruments would tend to
keep the names assigned by the inventor – and place names can be very
resilient.  The words related to a simple life do suggest a common
origin – but the others were suspect in my mind.

The words you referred to in German are another case in point, Sea and
Sea products also suggests trade – but here they make themselves
obvious by being words that do not fit within IE structure.  Assume
that these same words found themselves across many of the IE cultures,
but from the same external source.  Wouldn't the assumption be that
they were involved in a derivative chain, rather than a coincidental
introduction across the various branches?

I have run across other instances, in my short time looking into this
that makes me wonder how to sort it out.

Upon reflection on how a word ends up in a language I can think a whole
lot of sources from outside of the culture:

a) Words from Interaction of cultures:
Borrowed, trade goods or technology originating elsewhere that is
brought to the second culture and brings its name with it.
Also – words used often in trade, so coinage, barter and exchange,
contracts of exchange etc.
Diplomatic and Inter-regional political concepts, exchanges or presence.
b) Physical convergence of two groups::  
Conquest of one culture by another, 
Mutual association
Wars,  enslavment of a population (by raids, indenture or purchase)
Population exchanges through marriage (of large amounts of population,
	or influential members of the population)
c) "Sought" enculturation because of the belief of its superiority
(Hence 			Greek by Macedonians and Romans).
Possibly cultural ideas: art from exchanges items.  Stories and
mythologies.  Possibly religion (later religions were certainly
inter-national)
d) Adjacency of populations: Prolonged or intense (non-hostile)
interaction of two adjacent populations.  Reasonably free flow of ideas
and communications between to adjacent groups.  Especially if border
peoples feel the need to communicate with adjacent populations.

So I wonder, how are you sure it derivative and not introduced – if
don't have the obvious give away, you pointed out concerning loan words
with discordant sound evolutions.

For instance, in looking at the discussion on the Armenian lineage, I
had to wonder how much of the perceived associations there was from a
derivative relationship – and how much was from contemporary
interaction between adjacent non-related populations.  And wondering
how in the world you go about sorting it all out.

I would think that establishing some sort of hierarchical evaluation of
the type of words that would be acceptable for tracing relationships
would be essential.
For instance, single words related to trade might be very low in the
hierarchy and suspect. Words that were not related to trade, technology
new to the period, placenames, or involve possible cultural
contaminations would be more creditable.  Traceable changes in syntax
and grammar would be the most creditable.  I assumed there was some
such criteria, hence my previous questions.

Brent

PS I checked the board before posting and saw Piotr's posting.  He
suggest one instance of the same nature, Karl etc.

Brent:

The "hierarchical evaluation of the type of words ..." which you regard as essential indeed exists. It is good comparative practice to start with the "core vocabulary" -- terms like TWO, THREE, FOUR, FIVE; FATHER, MOTHER, SON, DAUGHTER ...; HAND, FOOT ...; STAND, SIT, LIE ... -- which are known to be far more resistant to replacement than words of technology, commerce and culture.

In practice, if there is systematic agreement involving a number of "core vocabulary" terms, genetic relationship becomes a plausible working hypothesis; lexical evidence of this type serves as a starting-point for a more detailed analysis, which concentrates on establishing the nature and relative chronology of the sound changes involved and reconstructing the phonology and morphology of the protolanguage. A successful reconstruction of this kind gives us more confidence in our initial hypothesis and suggests where to look for more evidence to verify it.

Comparative reconstruction is a dynamic process in which the output is used to test and re-test the initial assumptions. The more feedback, the more credible our reconstruction becomes. In fact, reconstructed words are often quoted not for their own sake but merely in order to illustrate the operation of phonological and morphological processes; even an early loanword will do the job. Distant relationship hypotheses tend to rely solely on etymological equations, ignoring morphological analysis -- an error which cannot be excused, no matter what kind of justification is offered.

I share your distrust of culture-laden words; this is why I'm suspicious of "linguistic palaeontology" -- that is, the reconstruction of the protoculture on the basis of the protolanguage. If a technical word displays, say, extremely archaic morphological structure betraying an old derivational paradigm, non-productive already in PIE -- I will admit that it isn't likely to be a Wanderwort, but trivial derivatives of the kind I mentioned in my last posting are of little value.

Piotr