Comparing languages. Examples [gothic-l]

[keth at ONLINE.NO]

Hi Cory !

I thought some more about these problems, while going to sleep
last night!  I should like to compare it to a tree in the garden.
Since mine have grown too tall, blocking the view, and also the 
neighbor complaining, you can imagine that I want my tree to
be exactly 3 meters tall. I then cut off all branches at exactly
this altitude. Thus, the top of the tree is now entirely horizontal.
(and then I smear some tar on - but that's not the topic ;)

The branches of the tree are of course the various languages
and the distance between the branches tell us how far the
different languages are removed from each other. Altitude
represents time, and the horizontal cut represents a synchronous
comparison.

However, there is no logic inherent in the matter that
dis-allows us to compare different languages at different times.
If a language, for instance, is isolated on some island, and
its development in consequence is retarded, it may actually 
make more sense to compare different languages at different times.
But if that is done, a reason ought to be given. Some explanation
of why one wants to compare different languages at different times
and what kinds of conclusions one hopes to be able to obtain from
such comparison. Thus, it is my opinion that the "default" 
comparison level must be the synchronous level (equal time comparison).

Another approach that may be more general and hence "better" in 
some sense, is the topological approach. Here time has no bearing
on the problem at all. The only thing that counts here are the
branch points, i.e. such historic moments (events) at which it
is possible to say: "yes, now this one language has clearly
split into two languages". On my garden tree, this corresponds
to a point where one branch splits into two.

A family of languages may now be described enirely in terms of
its branch-points, where one totally disregards the length of
the relevant time-intervals, and ONLY describes the tree in
terms of its bifurcations. In that way one obtains a timeless
and purely logical description of how the languages are related.

It is my understanding that the latter method is the one used
by Francisc in his analysis. (but you'd have to ask him, of
course) I regard his method as superior to mine, because he
is able to decribe the differences in a purely logical fashion --
in a timeless fashion.

There are then only two processes that describe the development
of the language tree. One is the innovations, and Two is the
disappearance of archaic features. "Archaic feature' do not
have to be 'old' in the sense of time or duration. But perhaps
one can say that they are such features as survive a couple of
bifurcations.

Of course one soon sees that the analogy with the tree in the garden
cannot be fully maintained. It is nevertheless a picture that may
be helpful to obtain an overwiew of the different possibilities
of development that may happen. The growth of a language is a 
continual process of the arisal of new features and a disappearance
of old ones.

Well, maybe you knew al this  :)
I just thought it might be fun to write down some of the thoughts I had.



>Hi Keth,
>        Looked at from that point of view, your objection is well taken, and I
>agree with your points in the technical context in which you placed them,
>but I wasn't attemping a scientific comparision.
>        Your methodology would seem to limit comparisons with Gothic to a small,
>contemporary portion of the runes, making other information "irrelevant";

That is true. We lack contemporary material from neighboring languages.
Nevertheless, runologists have done exactly that, and compared the grammatical
suffixes of the oldest runic inscriptions with those of Gothic.
Ottar Grønnevik has written about this, and on this basis he divides the
development of Proto-Nordic into 3 levels (I & II & III). He says
Gothic corresponds to the second level. For myself I can "explain" this
by saying that because the Goths were constantly on the move, constantly
meeting new peoples, and thus constantly receiving new impulses, their
language changed faster than the language "back home" (in Scanza or Gothiscanza
as the case may be). Thus level II occurred somewhat later in the North 
than with Gothic. And thus, this is in agreement with what Francisc
said about the language of the oldest runic inscriptions of ca. 200 AD
and some centuries forward in time  (level I), belongs to a more archaic
level of Germanic than Gothic does.


>however, what is irrelevant to one methodology may be quite relevant to
>another.  Your methodology would also seem not to take into account the
>problem of comparing texts of different Germanic languages, where the
>writer in Language A has two equal synonyms available, one with a cognate

Well, when I translated from German to Dutch, I *always made sure to choose
the Dutch word that had the same etymology as the German word. *Only
when this was not possible, did I choose a different word, since I thought
this was the fairest way to compare. (otherwise you might get the result,
in the case of English, that it is a romance language, whereas we know
it is not).  Thus in the case of German/Dutch, when I came to "heute"
(=today), I could not find a Dutch word with the same etymology,
and thus I wrote "vandaag" (=today in Dutch), and counted this word
as one of the differences between the two languages. Of course "heute"
*may also be related to *Tag. But since this isn't obvious to a normal
user of the language, I count "heute" as different from "vandaag".


>in language B, and one without.  If the writer in Language A chooses the
>synonym that does not have a cognate in Language B, the two languages
>will seem further apart than they actually are, which could seriously
>skew the results of a comparison; for example, Gothic has two words for
>"father":  "atta" and "fadar"; Old High German has "fater."  Bishop
>Wulfila wrote:  "Atta unsar"; the Old High German scribes wrote:  "Fater
>unser."  The texts seem far apart.  Yet Bishop Wulfila could have
>written:  "Fadar unsar,"  and the presence in Middle High German of the
>word "Etzel" (from Gothic Attila, meaning "Little Father") may point to
>an Old High German form of "atta", which may have dropped out as a result
>of the scribes choosing "Fater."  In any case, a comparision of the
>actual texts would make the two languages seem further apart than what
>they actually are.

Yes, I agree with that.
Conider it as if you are grading exams, and you always try to give the
examinand the benefit of the doubt. But of course if words become so 
"far fetched" that they are hardly used any more, I think one ought to 
choose the newer and more widespread word. But such choices are always
subjective. That's why I compared it to grading examinations.

Any way, I actually said nothing about the relationship of Gutnic
to Gothic. The only thing I wanted to point out was that Gutnic 1350
is VERY close to Icelandic 1350. Which of these has more in common
with of Gothic I am unable to say. But of course Sophus Bugge's
remarks show that Gotlandic *may have been somewhat closer to Gothic,
but if it was, it was *so little that even the genius of Sophus Bugge
was not able to say so with much certainty, and thus all he could
do was to hand over certain research proposals to those specialist
who might follow later, and be able to devote more time to the question.
Of course Sophus Bugge was one of the great language experts of his day.
His 1867 critical edition of the "Sæmundar Edda" still holds the
position as the basic reference work today, alsmost 150 years later.

With best regards
Keth

 

>Sincerely yours,
>Cory       
>
>On Tue, 24 Jul 2001 14:54:05 +0200 keth at online.no writes:
>> Hi Cory!
>> Thank you for your reply!
>> My objection is methodological:   :)
>> You compare 3 samples from 3 related languges at vastly
>> different points in time. Since modern German is direct descendant
>> of Old High German, but separated by a thousand years, it does not
>> belong in the comparison. Gothic from ca.350 AD  with Old High 
>> German
>> from ca. 950, is also a wide separation.
>> 
>> In my examples I did something entirely different.
>> I compared Gutnish (by an example sentence) from ca. 1350
>> with Icelandic, also from ca. 1350.
>> 
>> Then as a control I also compared modern German with modern Dutch,
>> because those are living languages today and many people who
>> have visited both countries are intrigued by the question how
>> close they are. Well, in my opinion Dutch and German are not
>> *very close. Because a speaker of one of the two languages does not
>> automatically understand the other language. It takes a while to 
>> become
>> accustomed. That is different between f.ex. Norwegian and Danish
>> which are virtually the same language. Or even Swedish.
>> Modern Icelandic is, however, considerably further away - because 
>> the
>> languages have drifted apart over the last 700 years. And it is the
>> Scandinavian languages that have done most of the drifting.
>> But if you have chance to ask an Icelander how close he feels
>> that Old Gutnic is to Icelandic, then that would, I think, be the 
>> best test
>> that one could devise. Because only a very seasoned user of one
>> of the two languages that are being compared, can directly "feel"
>> how far the languages are apart. And that is, in my opinion,
>> that which counts.
>> 
>> Best regards
>> Keth
>> 
>> 
>
>________________________________________________________________
>GET INTERNET ACCESS FROM JUNO!
>Juno offers FREE or PREMIUM Internet access for less!
>Join Juno today!  For your FREE software, visit:
>http://dl.www.juno.com/get/tagj.
>
>
>You are a member of the Gothic-L list.  To unsubscribe, send a blank email to <gothic-l-unsubscribe at egroups.com>. 
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/ 



------------------------ Yahoo! Groups Sponsor ---------------------~-->
Small business owners...
Tell us what you think! http://promo2.yahoo.com/sbin/Yahoo!_BusinessNewsletter/survey.cgi
http://us.click.yahoo.com/vO1FAB/txzCAA/ySSFAA/wWMplB/TM
---------------------------------------------------------------------~->

You are a member of the Gothic-L list.  To unsubscribe, send a blank email to <gothic-l-unsubscribe at egroups.com>. 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/