Re: Fw: Intro to natural language processing
On Tuesday 28 December 2004 12:35, John F. Sowa wrote:
>
> The point is that symbols are the prerequisite.
> You can have a natural language with symbols
> and just rudimentary syntax. But without symbols,
> you can't have anything like a natural language.
If you define NL to be fundamentally symbolic I imagine symbols do become very
important to your model. But I don't think symbols explain syntax.
What are symbols for Peirce? Convention?
Collocation is not just habit (convention?) built on weak syntax. You don't
really think it is, do you?
> I suggest that you read Deacon's book
There are so many books to read. If you could give me a hint that Deacon even
begins to explain syntax (other than to say it is "rudimentary") perhaps I
would read his book.
Could you just give me the broad strokes for Deacon's explanation of syntax,
the stuff that explains not only the rigidity, but also the flexibility of
collocations? (In particular the limits on the flexibility!)
For example, could you sketch how Deacon would explain the following:
The most interesting analysis of language learner "errors" I have come across
was a pre-print I read by one Peter Howarth, at Leeds at the time. What
interested me was his analysis of two types of collocational disfluencies he
characterized as "blends" and "overlaps".
By "overlaps" he meant an awkward construction which was nevertheless directly
motivated by the existence of an overlapping collocation:
e.g.
"Those learners usually _pay_ more _efforts_ in adopting a new language..."
*pay effort
PAY attention/a call
MAKE a call/an effort
So "*pay efforts" might be motivated by analogy with "pay attention" and "make
an effort" (because of the overlapping collocations "pay a call" and "make a
call".)
Similarly:
"..._attempts_ and researches have been _done_ by psychologist to find..."
*do an attempt
DO a study
MAKE an attempt/a study
To continue in Howarth's own words:
"Blends, on the other hand, seem to occur among more restricted collocations,
where the verbs involved are more obviously figurative or delexical in
meaning and the nouns are semantically related, though there are no existing
overlapping collocations.
'*appropriate _policy_ to be _taken_ with regard to inspections'
TAKE steps
ADOPT a policy
...
It is remarkable, firstly, that NNS writers produce many fewer blends than
overlaps and, secondly, that it is the more proficient (by informal
assessment) who produce them."
What I understand Howarth to be saying is that "overlaps" tend to be produced
first, and conceptual "blends" only later. It is the opposite of what we
would expect if language learning started by combining words according to
general concepts. It is more what we would expect language started by
analogizing between examples, and concepts only began to emerge later from
collections of many such analogies between examples of usage.
-Rob Freeman