Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!oitnews.harvard.edu!yale!gumby!newsxfer.itd.umich.edu!ncar!uchinews!ellis!deb5
From: deb5@ellis.uchicago.edu (Daniel von Brighoff)
Subject: Re: The whole language tree thing.
X-Nntp-Posting-Host: midway.uchicago.edu
Message-ID: <DF28LE.Av4@midway.uchicago.edu>
Sender: news@midway.uchicago.edu (News Administrator)
Reply-To: deb5@midway.uchicago.edu
Organization: The University of Chicago
References: <DEIxrB.8J0@crash.cts.com> <AC7F8164966813F9B@yarn.demon.co.uk> <DEz84H.17A@midway.uchicago.edu> <AC80D0F196686EEE@yarn.demon.co.uk>
Date: Sun, 17 Sep 1995 17:29:38 GMT
Lines: 116

In article <AC80D0F196686EEE@yarn.demon.co.uk>,
Paul Talacko <taka@yarn.demon.co.uk> wrote:
>In article <DEz84H.17A@midway.uchicago.edu>,
>deb5@ellis.uchicago.edu (Daniel von Brighoff) wrote:
>
>>>Try an experiment by translating from French and German into English. 
[snip]
>>Qu'est-ce que vous avez dit?  	What is this that you have said?
>>Was haben Sie gesagt?      	What have you said
>
>Ich kann den Brief Peter schreiben.
>Je peux ecrire la lettre a Pierre.
>I can write the letter to Peter

(A quibble:  A more accurate translation of the German you have written
would be "I can write the letter to *Peter*."  The unmarked order is
"...Peter den Brief...".)

Keep in mind what a straw man you've constructed here.  Modern Neuhoch-
deutsch is not the ideal candidate for the canonical Germanic language,
having preserved inflections lost in all the others except Icelandic and
having been subject to more syntactic influence from Latin than most
others.  Why not compare English to its close relatives:  Dutch and 
(better still) Frisian?  

>>And it always bugged me that Chinese was segregated off in some completely
>>different family.  Analytic languages unite!
>
>Now that is just silly.

Why?  You've gone on record saying language trees are outdated, misleading
bunk.  So if you are going to toss them out and reclassify the world's 
languages based on structural criteria, why ignore the strong resemblances
between English and Chinese vis-a-vis inflected languages like German?

>As you know there is very little chance of Chinese influencing Western
>European langagues.  There is a very great likelihood of French and English
>influencing each other.  I was taught that languages may borrow words from
>each other, but not structure.  This appears to be quite wrong.

Yes, it is wrong.  This is why, as many have pointed out, typological
evidence is just about the weakest kind for proving genetic relation.
So, tell me, how do you quantify this "influence" you keep bringing up?  
How many structures do language X and Y have to have in common with each 
other but not Z before you consider language Y closer to X than to Z?  What 
weight do you give to the relative frequency of variant structures?  For 
example, the two genitives in English and German vs. one in French.

la lettre de Peter  =  the letter of Peter  =  der Brief vom Peter
			Peter's letter	    =    Peters Brief

"Peter's letter" is immensely more frequent than "the letter of Peter."
Does this make the English genitive more like that of German than French?
How much more?  Does it outweigh the (optional) use of the article with
person names in German?

In short, what proof can you give that your replacement for strict genetic
classification is anything more than subjective?  English and French may be 
a clear-cut case to you, but what about Chinese and Vietnamese?  Malayalam 
and Marathi? Manchu and Mongolian?  Zulu and !Xhosa?

>>You will never hear a German speak like that either.  I think you are
>>putting style (very formal written style, heavily influenced by the
>>ideas of post-Enlightenment grammarians) before basic syntax and morpho-
>>logy.  Certainly, I don't think Juristendeutsch is a firm basis for 
>>linguistic toponomy any more than I think topiary is relevant to Linnaean 
>>classification.
>
>Yes, alright I admit it, it was a bit silly to suggest that, but I will not
>let the basic idea go.  That is the obsession with classifying languages
>according to 'family', according to 'genetics' is misguided and past its
>use-by date.

When exactly, pray tell, did it expire?  Do you at least know the year?
>
>It is something that nationalist obsessed 19th century scientists tried to
>do, in the same way some went around collecting fairy tales as a true
>expression of nationalist consciousness.  The whole PIE project is similar
>- based on nationalism, and, in any case, is doomed to failure.  Language
>is so fluid and so infinitly varied that any reconstucted langauge must be
>wrong - there are just too many variables involved.

Ah yes!  I remember, for example, how test cases (such as historical
reconstruction of Vulgar Latin based on the modern Romance languages)
were completely off the mark and revealed the inherently bogus nature of 
historical reconstruction.

>Socio-linguists (am I allowed to move this group away from historical
>linguistics for just one second?) know that everyone has his/her own
>mini-dialect and own speech patterns  that are influenced by that person's
>own micro-society.  Sometimes these idiosyncracies are dispersed into the
>wider community and become dialectical or, even, part of the standard
>langauge.
>
>In this sense it is obvious that French and English must be closely
>related.  More French onto English that the other way around.  Similarly,
>because Latin was the educated language for so long it influenced all the
>major langauages of Europe.

I'm not arguing this.  Models based on influence have their place--I've
seen some particularly nice modifications of the tree used to model
lexical borrowing, though I've yet to see a satisfying diagram of 
syntactic influences--alongside genetic trees.  How you display the data 
all depends on what you're trying to emphasise.  Your interest is socio-
linguistics.  Fine!  Go and make some lovely wave-pattern, latticed,
web-like, 3-D, etc. models that capture your notions of mutual influence 
and synchronic resemblance and let me, an inveterate historical linguist,
have my trees.
  
Daniel "Save the Trees!" von Brighoff


-- 
	 Daniel "Da" von Brighoff    /\          Dilettanten
	(deb5@midway.uchicago.edu)  /__\         erhebt Euch
				   /____\      gegen die Kunst!
