Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!nntp.sei.cmu.edu!cis.ohio-state.edu!math.ohio-state.edu!usc!elroy.jpl.nasa.gov!ncar!uchinews!ellis!deb5
From: deb5@ellis.uchicago.edu (Daniel von Brighoff)
Subject: Re: Hard Languages
X-Nntp-Posting-Host: midway.uchicago.edu
Message-ID: <DGFFCA.6Kw@midway.uchicago.edu>
Sender: news@midway.uchicago.edu (News Administrator)
Reply-To: deb5@midway.uchicago.edu
Organization: The University of Chicago
References: <17427131B6.C1737016@UMSLVMA.UMSL.EDU> <rharmsen.843.000B2970@knoware.nl> <44omlv$9er@oravannahka.Helsinki.FI> <455l51$nsh@gordon.enea.se>
Date: Sat, 14 Oct 1995 06:57:46 GMT
Lines: 53

In article <455l51$nsh@gordon.enea.se>,
Erland Sommarskog <sommar@enea.se> wrote:
>simon@cc.Helsinki.FI (Julia A M Simon) wrote:
>>On the other hand, I had a lot less trouble with the different genders
>>and inflection classes when learning Russian and Latin. But that might
>>be because I was already used to three nominal classes and different
>>nominal inflection classes from my native language (German)...
>
>Of course genders are not too much of a problem in Russian where 
>you can deduce the gender of most words by a quick glance. And I
>assume this is true for Latin as well. Now German with is three
>seemingly random genders is another matter.

I don't particularly want to get involved in an endless exhibitions of
language-difficulty anecdotes, but, as someone else pointed out, gender
is largely predictable in German.  For example, every word ending in -ung,
-heit, and -igkeit (which are extremely common nominalising suffixes) is
feminine, as are almost all words ending in -schaft and most words ending
in -e (except a small class of deverbalised collectives, like Gestoehne
and Gefaelle, and the so-called "weak masculine nouns").

Furthermore, no word has more than five distinct forms (nominative singular,
genitive singular [obsolescent], dative/oblique singular [obsolete], dative
plural [same for all declensions], non-dative plural) and most have fewer.
The feminines listed above, for instance, have only two.

Compare this to Russian, where most nouns have ten forms.  It's small com-
fort knowing that all nouns ending in -o are neuter when you find that, in
cases other than the nominative singular, this ending changes to -a, -akh, 
-am, -ami, -e, -om, -u, or -0.  Cross-linguistic studies have shown that a
single consistenly-applied morpheme (e.g., -n for German dative plurals) is 
much easier to learn than a variety of allomorphs (e.g. Russian -am, -enam, 
-jam, -jatam).

>Generally I would say that I find all features that require learning
>forms word by word are the most difficult ones when learning langauges.
>Unpredictable genders is only one of them, there are also declensions -
>another killing German feature - or unpredictable stress - which is
>what makes Russian a nightmare - or unpredictable or irregular verb 
>conjugations - the aspects in Slavic languages can make you quite lost
>here.

Yup!  The fewer lexically-specific rules, the better.  I believe it's also
been shown that form-dependent rules (e.g. nouns ending in -a pluralise by
dropping it and adding -e) are easier to remember than those that are seman-
tically linked (e.g. verbs denoting an action take -(nu)nta instead of -ta
in the declarative plain form), but I don't have a reference handy for this.


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