Y/HA

[Danko Sipka]

I would first like to thank to Oscar Swann, for his
information about Cienki's book, and to Frank Joseph Miller, who has
informed me about Dorothy Soudakoff's dissertation related to
Slavic prepositions at Indiana University 10-15 years ago, as well
as about textbooks by M.V. Vsevolodova (Moscow), where she contrasts
Russian prepositional usage with that in the other Slavic languages.

Wayles Browne writes:

>I wonder about one point. As I understand it, na selu (like the Polish
>na wsi) means 'out in the countryside' (the opposite of 'in the city'),
>whereas u selu (Polish we wsi) means 'in a/the village'. So the branch
>with u gradu doesn't seem like an exception.

Yes, you may interpret these examples differently, but it will not
eliminate this branch. You will still have examples like 'u brdima'
'u vukojebini', etc., which stay even after a new branch (inhabited:
yes/no) is added. A part of the problem might be eliminated by
giving a specific definition to 'being enclosed', the definition which
produces the shortest list of exceptions. In general, the aim would be
to have the number of branches as low as possible (people will have to
learn less, and machines would have to go through less loops = will be
faster).

>There should be a branch
>labelled 'inhabited places' including grad, selo, kolonija,
>zemlja 'a country', drz^ava 'a state', as well as such concrete items as
>Vars^ava, Sarajevo, S^estine, Slovenija, Vorarlberg, Arizona...
>and then a branch for exceptions, such as Rijeka (takes na or u),
>Florida (takes na), etc.

Yes, this is one solution. The other would be to add a more general
rule for onyms, saying that they are governed by the same preposition
as their appellative, with certain exceptions. This would take care not
only about toponyms, but also about onyms such as names of restaurants,
buildings, galleries... This rule might be implemented at the very top
of the algorithm - onym: yes/no, and if yes, jump to appelative if the
onym in question is regular one.

------------------------------------------------------------------
Dr. Danko Sipka
Slavic Department, Adam Mickiewicz University, Poznan
Institute of Philosophy and Sociology, Polish Academy of Sciences
Translation Experts Poland
------------------------------------------------------------------
e-mail: sipkadan at hum.amu.edu.pl (or sipkadan at plpuam11.amu.edu.pl)
Web: http://www.amu.edu.pl/~sipkdan/ja.htm
phone: ++48-61-535-143
mail: ul. Strzelecka 50 m. 6, 61-846 Poznan, Poland
------------------------------------------------------------------
I think where I am not, therefore I am where I do not think

                                                    Jacques Lacan
------------------------------------------------------------------