[Corpora-List] Stubbs' analogy?

Dominic Widdows widdows at maya.com
Wed Dec 14 18:00:01 CET 2005


Dear Ramesh,

I would be wary of an analogy that tried to view lexical information as
discrete and particle-like, and grammatical information as continuous
and wave-like. I think your instinct for "token" rather than "type" is
more appropriate, and of course the type / token distinction happens in
both lexicon and grammar.

There are definitely models of the lexicon in which word senses have
distributions over regions of "semantic space", which is rather like
the idea of the position of a particle being represented as a
wave-function. A rough analogy then arises between observing a word in
context and inferring its appropriate sense, and observing a particle
in an experiment and inferring its actual position.

Colin Cherry's 1956 book, "On Human Communication", contains a lot of
insight on the logical necessity of "quantization" in human language.
Obvious examples include concepts such as "size" which may be a
continuous variable, but we have a discrete number of words ("small",
"large", etc.) to describe sizes. Arguably, we can generate phrases to
describe any sizes we want to ("not too small", "quite large but not
very large", etc.), but we can't get very fine-grained without
resorting to number words. The process of transcribing sounds into
symbolic representations (e.g. strings of alphabetic characters) in the
first place is a form of quantization.

More recent implementations of these ideas include Schutze's Word Sense
Discrimination (http://citeseer.ist.psu.edu/schutze98automatic.html),
which clusters word senses and "quantizes" new context-vectors my
mapping them to their nearest cluster. The close similarity between the
vector model used in this work and the vector model used in quantum
mechanics is investigated in my own book, "Geometry and Meaning" (see
http://infomap.stanford.edu/book/chapters/chapter7.html).

This work relates mainly to the lexicon, and does not compare lexicon
with grammar at all. My only suggestion here is that there are enough
quantum analogies in studying just the lexicon, and I therefore doubt
whether it is possible to consign the lexicon to playing just one of
the many confusing roles that abound in quantum theory.

Best wishes,
Dominic


> Hi Jörg,

>

>

> Not a lot, unfortunately, although your question has prompted me to

> find out more....*

>

> (I think it was) John Sinclair (who) once described lexis and grammar

> as looking at

> language through opposite ends of the same telescope...

>

> Somewhere or other, I picked up the idea that if lexis and grammar

> were looking at the same

> phenomenon (language) from different points of view, the dichotomy

> might be similar to one that

> has confronted physicists: looking at light as particle and wave at

> the same time.

>

> I'm sure this is an ultra-naive understanding on my part, but if you

> can help, I'd be grateful.

>

> *e.g. Queen Mary College London ( http://www.qmw.ac.uk/~zgap118/1/

> etc)

> has some information that might help:

>

> Energy and matter we have learnt from Einstein's theories are

> analagous, matter can be simply described in terms of energy. So far

> we have only discovered two ways in which energy can be transfered.

> These are particles and waves....

> Particles are discrete, their energy is concentrated into what

> appears to be a finite space, which has definite boundaries and its

> contents we consider to be homogenous (the same at any point within

> the particle)... [lexical item?] Particles exist at a specific

> location. If they are shown on 3D graph, they have x, y, and z

> coordinates. They can never exist in more than one place at once...

> [so "token" rather than "type"?]

> Waves unlike particles cannot be considered a finite entity. Their

> energy cannot be considered to exist in a single place since a wave by

> definition varies in both displacement and in time.... In an area of

> space, unlike a particle, a wave can propagate until it exists in all

> locations and at all times... [grammar?]

>

> Best

> Ramesh

>

> At 11:00 14/12/2005, you wrote:

>

>> I also use the 'particle/wave' analogy for the 'lexico-grammar'

>> continuum

>>

>>

>> Could you tell us more about this? I have never heard of it.

>>

>> Jörg

>>

> Ramesh Krishnamurthy

> Lecturer in English Studies

> School of Languages and Social Sciences

> Aston University, Birmingham B4 7ET, UK

> Tel: +44 (0)121-204-3812

> Fax: +44 (0)121-204-3766

> http://www.aston.ac.uk/lss/english/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/enriched
Size: 4521 bytes
Desc: not available
Url : https://mailman.uib.no/public/corpora-archive/attachments/20051214/0091bf1e/attachment.bin


More information about the Corpora-archive mailing list