mainoff.gif
lastdyoff.gif
lastwkoff.gif
treeoff.gif
searchoff.gif
helpoff.gif
contactoff.gif
creditsoff.gif
homeoff.gif


The Daltaí Boards » Archive: 2005- » 2005 (May-June) » Archive through May 20, 2005 » About linguistics (for those who would be interested) « Previous Next »

Author Message
Top of pagePrevious messageNext messageBottom of page Link to this message

max
Unregistered guest
Posted From: 82.226.74.188
Posted on Wednesday, May 04, 2005 - 09:48 pm:   Edit Post Print Post

I feel that the fact that I'am doing research on Irish which is a language I don't speak is puzzling for a lot of people.

I also feel that the fact that I contradict certain things in which the fluent speakers believe may be very annoying to them.

Let me explain (at least try)


The position in which linguists are is not very easy: first they quite often contradict traditional grammarians who have been here before them (and for a long time), second they deal with language which is something that everybody on earth uses every day and therefore about which people always have some idea (metalanguage, the fact of talking about language itself, is very common), and third people generally don't perceive the difference between modern linguistics and traditional grammar.


Still,

a century ago, linguistics became a science:

Its goal is to describe and explain language, not to prescribe.
It is admitted that the whole human knowledge exists in terms of theories. (A theory is a possible explaination of the data, and holds until new data come to contradict it, meaning there is no such thing as "truth".)
In order to process the data, the concepts are to be strictly defined (never like in a dictionary because it works the other way round).

There are several linguistic theories, but there are certain things that they all have in common (like the concepts of "phoneme" and "morpheme" (or "moneme"), or the fact that there are different levels in a language)


There are levels that require fluency in the language to be analyzed (just as semantics), and other that don't (just as phonology or syntax).


Personally, I deal with syntax, and the points I make are formulated within the frame of a theory called "functionalism".
So far, I haven't invented anything, and what I have written elsewhere is pretty common in the linguistic world.

All I'm saying is that it is possible to view things from other points of view.

Top of pagePrevious messageNext messageBottom of page Link to this message

Aonghus
Member
Username: Aonghus

Post Number: 1345
Registered: 08-2004


Posted on Thursday, May 05, 2005 - 04:18 am:   Edit Post Print Post

The question I have is how you can be sure of the accuracy of your syntax data, without being fluent.

How do you know you are not analysing untypical, incorrect Irish? Where do you get your data?

Top of pagePrevious messageNext messageBottom of page Link to this message

max
Unregistered guest
Posted From: 82.226.74.188
Posted on Thursday, May 05, 2005 - 06:42 am:   Edit Post Print Post

first of all, the word "incorrect" is a very bad word (although i also use it from time to time) because it's very prescriptive.

to say that something is incorrect is an easy way to get rid of a problem that you can't explain.
everything is explainable.
but what is true is that the data have to be sorted out. for instance:

if someone who is german tells you "la soleil", you could think it's incorrect because no french would say it, (we say "le soleil"). but if you know that "soleil" is feminine in german ("die Sonne"), and if you know what "gender" is linguistically, then you'll be able to explain this fact.

now if a french tells me "après qu'il soit venu", you could think it's incorrect because from the prescriptive point of view, you should say "après qu'il est venu". but since practically everybody says "après qu'il soit venu", you must find why.

it's not possible to do everything at the same time.
either you deal with french who only speak french (in which case you won't take into account the 1st example), or you deal with french as spoken by german people and see how their french is influenced by german (in which case you won't take into account the 2st example).


the problem with irish is that everybody speaks english, but speak irish more or less.
in order not to get lost, i draw all my data from manuels and grammar books, and i don't rely on what i could find with "google".


for the record, you wrote:
"the man who is a priest" An fear, gurbh sagart é.
"the man who is the priest" An fear, gurbh é an sagart é.
but then admitted it was incorrect after all.

i believe you are a fluent irish speaker. maybe you would never make this "mistake" in a conversation or while writing. this is a particular situation because i ask translations, which is a difficult exercice. but anyhow, what you wrote means that at this point, the structure is maleable enough to allow a fluent speaker to say it this way (if it were not, you would never have even thought of writing it this way), although i can't explain why and how at the moment.

Top of pagePrevious messageNext messageBottom of page Link to this message

Aonghus
Member
Username: Aonghus

Post Number: 1347
Registered: 08-2004


Posted on Thursday, May 05, 2005 - 06:55 am:   Edit Post Print Post

That is my point exactly. By looking for translations here, you are open to getting incorrect data.

As Lúcas pointed out, the versions I gave would be correct in reported speech.

The problem is that you are looking for translations of fragments. I translated the fragment but in the wrong context.

Top of pagePrevious messageNext messageBottom of page Link to this message

Aonghus
Member
Username: Aonghus

Post Number: 1348
Registered: 08-2004


Posted on Thursday, May 05, 2005 - 07:01 am:   Edit Post Print Post

Peut etre ca serait bon pour ton these: http://www.ite.ie/corpus/

Top of pagePrevious messageNext messageBottom of page Link to this message

Jax
Unregistered guest
Posted From: 159.134.221.222
Posted on Thursday, May 05, 2005 - 09:28 am:   Edit Post Print Post

http://borel.slu.edu/corpas/index.html

this is an aligned corpus of English and Irish texts

Top of pagePrevious messageNext messageBottom of page Link to this message

max
Unregistered guest
Posted From: 82.226.74.188
Posted on Thursday, May 05, 2005 - 09:45 am:   Edit Post Print Post

thanx a lot Jax and Aonghus.


the problem is i still need to obtain some extremely specific data, and unless i spend my life looking for it in data bases hoping to find haphazardly what i'm looking for, i have to have it translated...

Top of pagePrevious messageNext messageBottom of page Link to this message

Jax
Unregistered guest
Posted From: 159.134.221.9
Posted on Thursday, May 05, 2005 - 11:57 am:   Edit Post Print Post

Max,
i get you. That is why I did not string more URLs in, in case they be nere too good.

If you state as a list the EXACT things you want, then I will look-see for any useful thing I know of.

I take it, if any program was useful, it would be a particular type of grammar assitant/analyser?

Top of pagePrevious messageNext messageBottom of page Link to this message

max
Unregistered guest
Posted From: 82.226.74.188
Posted on Thursday, May 05, 2005 - 12:32 pm:   Edit Post Print Post

jax,

your proposal is very kind.


i'm not sure i understand your last sentence. by grammar assistant, do you mean what i have in my computer when i open "word"? if it's the case, that wouldn't work, because these programs are still unable to analyze accurately the context, which, as Aonghus says and I agree with him, is indispensable.

Top of pagePrevious messageNext messageBottom of page Link to this message

Jax
Unregistered guest
Posted From: 159.134.220.103
Posted on Thursday, May 05, 2005 - 04:53 pm:   Edit Post Print Post

Max,
I knwo you are at pains to decline what might appear a very naive question, without 'hurting my feelings' I know you are not looking for a toy program, and i agree contextual sensitivity is required. I just asked in case you were searching for some form of aquisition tool, not an analyser per say.

It turns out you dont. Fair dinkum.

Top of pagePrevious messageNext messageBottom of page Link to this message

max
Unregistered guest
Posted From: 82.226.74.188
Posted on Thursday, May 05, 2005 - 08:44 pm:   Edit Post Print Post

Not at all Jax,

my interrogation was sincerely plain...

I still don't understand what you mean by "aquisition tool".

(computers are to me what language may be to others: I know it works but I wouldn't try to understand it)

Top of pagePrevious messageNext messageBottom of page Link to this message

Jax
Unregistered guest
Posted From: 159.134.221.164
Posted on Friday, May 06, 2005 - 07:07 am:   Edit Post Print Post

By aquisition tool i meant a corpus or database of irish. I assuemd one could have, tho it is not very specific, used the aligned corpus by inputting english to find sentances with examples of clauses. as you say, not very specific.

Top of pagePrevious messageNext messageBottom of page Link to this message

Max
Member
Username: Max

Post Number: 18
Registered: 05-2005
Posted on Monday, May 09, 2005 - 07:25 am:   Edit Post Print Post

well, i can't tell in advance whether i would be able to find specific things or not...

i don't know how it would work, so i trust you in this regard.


maybe it would be too specific: i'm looking for syntactic constructions, whatever the words used in the sentences...if you see what i mean...

Top of pagePrevious messageNext messageBottom of page Link to this message

Jax
Unregistered guest
Posted From: 159.134.221.104
Posted on Tuesday, May 10, 2005 - 09:07 am:   Edit Post Print Post

Max,
I understand, any query of the corpus would pull up examples that contain elements of any particular syntacular set, but they will not be formally expressed in a way that would be useful to way. In other words, the link above provides a sort of parsing of the corpus, but not an analysis.

I think you are saying that what you are looking for is a program that would give out the syntax and cathegories of syntax that work in the langue to structure a) particular examples so that you coudl learn about 'the man who is a uncle of the priest' etc and b) list them in some form of global context (thus placing in them in one set rather than another).



©Daltaí na Gaeilge