Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!europa.eng.gtefsd.com!gatech!psuvax1!news.ecn.bgu.edu!siemens!princeton!bessel!rkim
From: rkim@bessel.Princeton.EDU (Ronald Kim)
Subject: Re: Plurals
Message-ID: <1994Dec21.192649.6223@Princeton.EDU>
Originator: news@hedgehog.Princeton.EDU
Sender: rkim@bessel (Ronald Kim)
Nntp-Posting-Host: bessel.princeton.edu
Organization: Princeton University
References: <19DEC94.11813889.0031@music.mus.polymtl.ca> <D12qsH.FqL@cogsci.ed.ac.uk> <19DEC94.18886466.0035@music.mus.polymtl.ca> <D143Jr.6A0@cogsci.ed.ac.uk> <D14wG5.C4H@inter.NL.net>
Date: Wed, 21 Dec 1994 19:26:49 GMT
Lines: 123

In article <D14wG5.C4H@inter.NL.net>, mcv@inter.NL.net (Miguel Carrasquer) writes:
|> In article <D143Jr.6A0@cogsci.ed.ac.uk>,
|> Ivan A Derzhanski <iad@cogsci.ed.ac.uk> wrote:

[excess previous postings deleted]

|> >Let's take an example.  The Latin word _catena_ `chain' was inherited
|> >by French, and, having undergone a series of regular sound changes, it
|> >assumed the form _chai^ne_.  Then at some point someone felt he needed
|> >a term for `information chain'.  He found the word _catena_ in a Latin
|> >dictionary, gallicised it slightly, and _cate`ne_ made an appearance.
|> >How is that different from the case of a Latin word making its way
|> >into a language less closely related or unrelated to Latin?
|> >
|> >The same for _cha~o_ vs _plano_ `flat, smooth' in Portuguese,
|> ><sonA> vs <svar.na> `gold' in Hindi, ...
|> > [...]
|> >I didn't intend to sound confrontational.  Let's make that `if one can
|> >explain ...' or `if we can explain ...'.  I was simply wondering if
|> >your proposal didn't raise more questions than it answered.  Now,
|> >however, I see that things can indeed work as you say, although I
|> >continue to think that they can also work as I say.  :-)
|> >
|> 
|> Interesting discussion.  When I saw the phrase "nontransformed words"
|> (I think it was), my first reaction was one of rejection.  But then
|> it reminded me of the phenomenon of "cultismos" and "semi-cultismos",
|> as it's called by Spanish historical grammarians.  I have been trying
|> to think of a good example, but I can only come up with the triplet:
|> 
|> trecho / trato / (ex)tracto
|> 
|> all "from" Latin `tractus'.  The first word shows the normal development
|> [something like: tractu > traito > tretyu > trecho].  The second word
|> would be "semi-cultured", the third one "cultured", i.e. a loanword
|> from Latin.

	Good example, Miguel.  When learning Spanish along with Latin, I
quickly realized that any word with the consonant cluster 'ct' must be a
'learned' or 'educated' borrowing.  Although those are elitist-sounding
words, I guess the idea is that the words did not pass through all the
sound changes that characterized the development of Vulgar Latin to Spanish
and hence were presumably not part of the everyday people's vocabulary.
Of course, there are many cases in which such a word is inherited as part
of the common lexicon but is later complimented by a form that has been
taken straight from Latin and barely Hispanicized (e.g. -us > -o.)  Thus
'hecho' and 'dicho' are plainly 'popular' words that exhibit all the normal
sound changes (factu(m) > faito > fetyo > fecho > hecho and sim. for dicho),
but 'correcto' and 'obiecto' are not.

|>              Another example of `semi-culto' would be the word "Dios",
|> with its nominative -s.  It's not Latin, but it's not the normal
|> development either (which would have been *di'o).  

	Yup.  I always thought 'Dios' was natural at first, until I rea-
lized that (almost) no other nominative -s had been preserved in Spanish.
Portuguese shows the even more suspicious 'Deus', though here I believe
there is some kind of sound change of 'eo, io > eu', as in caelu(m) > celo
> ceo > ce'u.  French and Italian both exhibit the popular evolution to
'Dieu' and 'Dio'.

|> There are two ways of looking at this:  the "semi-cultured" words are
|> simply early loans from Latin, versus the "cultured", more recent loans.
|> This is Meyer-Lu"bke's analysis ("...words that don't have their
|> origin in an uninterrupted oral tradition [*]").  

	I believe that perhaps the most striking examples of this pheno-
menon in Europe come from (modern) Greek.  One finds not only doublets that
have their origins in exactly the same word (with one 'learned', one 'popu-
lar'), but doublets in which one is the classical word which has evolved
and undergone a semantic shift and the other is a new replacement for it.
The first situation is exemplified by "eleftheri'a" and "lefteri'a" -- the
second variant has dropped the unstressed initial vowel and changed the
cluster of two continuants to continuant + obstruent.  (I think "eleftheri'a"
is more common, though.)  The second situation is rampant and all-too notice-
able from walking around the streets of Greek cities, where the "artopo:lei~o"
sells "pso:mi'" and one goes to an "ikhthyopo:lei~o" for "psa'ri".  Why???
"a'rtos" and "ikhthy's" are the classical words for "bread" and "fish", but
they have long been replaced by "pso:mi'" and "psa'ri".  ("A'rtos" survives
in religious usage.)  I don't know if this is the same kind of phenomenon,
but it certainly seems to be related.  In any case, even the first situation
gives students of Greek major headaches; I have a hard time dealing with
books in which the word for eight switches back and forth between "okhto:'"
and "okto:".

|> Another way of looking at it is that words change at "different speeds".
|> This doesn't seem to me to be the case for a word like "extracto"
|> above, which I can only interpret as a loan from Latin.  But I'm not so 
|> sure about "trato" or "Dios".

	Perhaps, but then why did "tractus" lead to both "trecho" and "trato"?
Were they already distinguished in Vulgar Latin, i.e. as two words with the
same pronunciation but different meanings?  At least in Greek, the situation
seems to be quite clear:  one says "dhe'ftere:" as the fem. nom. sg. form of
the word for "second" but "e: dhefte'ra" for Monday (literally "the second
(day)"), but the latter must be a classical borrowing since few if any other
adjectives which end in -ros with similar accentuation have that kind of form.

|> I still don't like the term "untransformed words", but Meyer-Lu"bke's
|> phrase quoted above is equally unfortunate.

	I would go along with "archaic borrowings" or something like that.
Certainly languages such as German, which do not have a hallowed and illus-
trious ancestor to turn to for vocabulary (German uses its own enormous
word-formation potential instead), do not exhibit this phenomenon, or do
so to a minute degree.  Greek, Romance languages, Orthodox Slavic languages
(do West Slavs, Slovenes, and Croats exhibit significant Church Slavonicisms?),
and certainly Hindi and other modern Indic languages would seem to be the
places to look in the Indo-European world.  I'm not sure if East Asian lan-
guages also show "archaic borrowings", but I'd bet that Hebrew and Arabic
must.

|> [*] Gramm, I, #2, quoted in an article ("Problemas del cultismo"),
|> by R. Beni'tez Claros.
|> 
|> -- 
|> Miguel Carrasquer         ____________________  ~~~
|> Amsterdam                [                  ||]~  
|> mcv@inter.NL.net         ce .sig n'est pas une .cig 

Take care,
Ronald Kim '96
Princeton University
