LINGUIST List 6.1267

Mon Sep 18 1995

Qs: Pronouns, Lg data, Root-epistemic contrast, Chinese syntax

Editor for this issue: Ann Dizdar <dizdartam2000.tamu.edu>

Directory

1. "Gisbert Fanselow", 1st person: sg. vs. pl.
2. Torbj|rn Lager, Language data
3. , query root-epistemic references
4. "Robin Sackmann", Qs: Syntactic categories in Chinese

German does not distinguish reflexive and personal pronouns for 1st
and 2nd person.
1st ps. singular pronouns cannot be bound, as can be seen from the
following data:

(1) Nur ich liebe mich (only I like me)
just means
only for x = speaker: x likes speaker

(1) CANNOT mean:
only for x=speaker: x likes x
(in contrast to: Nur Hans liebt sich, only Hans loves himself)

Similarly:
(2) Nur ich denke, dass ich intelligent bin
(only I think that I intelligent am)
means:
only for x=speaker: x thinks that the speaker is intellingt
(2) cannot mean
only for x=speaker: x thinks that x is intelligent
(in contrast to: nur Hans denkt, dass er intelligent ist, only Hans
believes that he is intelligent)

(3) Nur ich liebe meine Frau (ony I love my wife)
O.K.: only for x= speaker: x loves speaker's wife
NOT: only for x=speakr: x loves x's wife

The facts are somewhat obscured in structures of the kind (1): as
soon as an "inherent reflexive" interpretation is available, binding
is strongly preferred:

(4) nur ich wasche mich (only I wash myself)

1st person plural pronouns can be bound:

(5) nur wir lieben uns 'only we love us'
can be interpreted as:
only for x= WE: x loves us
and as
only for x=WE: x loves x

(6) is also ambiguous:

(6) nur wir denken, dass wir intelligent sind
only we think that we intelligent are

(6a): only for x=WE: x thinks that we are intelligent
(6b) only for x=WE: x thinks that x is intelligent

(7) Nur wir lieben unsere Frauen (only we love our wives)
only for x=WE: x loves WE's wives
only for x=WE. x loves x's wife

2nd person pronouns behave excatly the same way: no binding in
the singular, but ambiguities in the plural.

My questions are:
(a) Do other languages share these properties?
(b) Why are the data the way they are?

fanselowrz.uni-potsdam.de
(Dept. of Lx, University of Potsdam)

I am tinkering with my own implementation of the Viterbi algorithm
applied to POS tagging. Now I would like to test my program on some
real text so I need statistical data of the following kind:

 Lexical probabilities: Word-Tag-Probability triples (or something
 similar)

 Collocational probabilities: Tag1-Tag2-Probability triples

Preferably for English (Brown corpus stuff would be great) or Swedish
(SUC?). I know lots of such data has been produced. Is there any available?

Thanks in advance.

Best regards,
Torbjoern Lager

- -------------------------------**-------------------------------------*------

Torbjoern Lager E-mail: lagerling.gu.se
Department of Linguistics			 Phone: +46 31 7731175
University of Gothenburg Fax: +46 31 7734853
Renstroemsparken
412 98 Gothenburg
Sweden

**-*-----*-*------------------*------------------------------------------------

I am continuing work from my 1993 dissertation on the root/epistemic
contrast, and would appreciate help compiling a complete bibliography
on the topic. I will post a list of references received. Thank you.

Ginny Brennan
Vanderbilt University

Dear LINGUISTs,

I am currently working on a paper about syntactic categories / word classes
in modern standard Chinese (putonghua resp. guoyu).
Therefore, I would be very grateful to users of the LINGUIST list for any
references to books and papers on this matter.

A bibliographical summary will be posted.

A had posted this same query already in July this year and promised a
summary. However, I got only very few replies, and there is virtually nothing
to summarize up to now. I think this might have been due to general summer
holidays, and so I decided to try it once again before posting the promised
summary.

(This request has also been posted in the CHINESE list, sorry if you get it
twice.)

Many thanks in advance,

Robin Sackmann
FREIE UNIVERSITAET BERLIN
Germany