Job Discussion Forums

[eslstudies]

[Brian Caulfield]

11.59 if we are trying to figure out if the person is a 70 or a 75 we are not listening to what a person is saying. I tell my students the same thing about this language. Unlike French , Russian or Korean to name a few languages where people judge you by how you use a language , English is about what you say not how you say it. The queen speaks beautiful English but really is a quite boring person to lilsten to. The ex prime minister of Canada addressed the leaders of the Pacific region a few years back and made over 150 mistakes in English in a thlirty minute speach. He was well recieved and listen to. Jean Chretian is disliked by the Quebecoise because he uses bad French but liked by the rest of the world because he speaks from the heart and calls things straight.

[11:59]

Brian, I take your point, but, with all due respect, I think my point is radically different to yours and furthermore that it has gone totally over your head. Perhaps I am guilty of not explaining myself adequately.

You stated, interestingly I think, that 'English is an art', and therefore cannot ever be gauged or measured by any 'test'. First, I don't think you really mean that English alone is an art, so I assume you mean that 'language' is an art (if I am mistaken here please correct me, but I will assume this is what you meant). Wittgenstein, too, was of the opinion that language (at least 'public language' - as opposed to 'private language') was indeed an art, but as he stressed, it is a precise art. It certainly is not mathematics, but it is also the case that it is not painting with watercolours either.

This is important as I think you err when you say that competence in English cannot be truly measured or gauged through tests. In fact there are numerous ways competence in (or knowledge of) a language such as English can be determined. As way of example, consider the following.

In the following sentences 'she' can refer to 'Mary', that is, they can 'co-refer':

1) Mary left town after she won a lot of money.

2) After Mary won a lot of money she left town.

3) After she won a lot of money Mary left town.

Now, it is also the case that 'she' in the above sentences may refer to someone else, say, 'Sue', who perhaps was mentioned in the preceding sentence. Thus (1) would be something akin to 'Sue played the lottery. Mary left town after she (i.e., Sue) won a lot of money'. That is, in the above sentences 'she' can receive a 'deitic' interpretation, depending on the context of the discourse. What is important to us here though is to establish that native speakers (and those with native speaker competence) agree that in (1) through (3) 'she' can refer to 'Mary'.

Now consider the following:

4) She left town after Mary won a lot of money.

Can 'she' now refer to (co-refer with) 'Mary'? Native speakers respond with an overwhelming 'no!'.

Now, a (subtle and complex) algorithm has been developed in computational linguistics that can actually reflect and 'mimic' (i.e., replicate) native speakers' grammatical intuitions on these points (if you wish I will refer you to the relevant literature, though some is a tad technical for those with no prior training in the field of computational linguistics).

So, my question for you is, how is that art? Is the computer (or the algorithm) 'doing art' when it determines whether or not 'she' can refer to 'Mary'? I think not. In short then, language competence can indeed be measured through tests, and to a very accurate degree, too.

[malu]

I hardly think receptive skills in language are too artistic to assess in an exam.