The difficulty in displaying lexical relations is that they vary widely in
type and closeness. By closeness I mean how salient they are in the mind of
a speaker. For instance I looked at the Visual Thesaurus entry for 'cook'.
There was a link to 'Captain James Cook' and to 'skilled worker' but no link
to many words that are closely linked in my mind to 'cook', such as 'bake'
and 'fry'.
The basic problem is that words are not merely linked to other words in a
spider web of relations such as is displayed in Visual Thesaurus (VT). The
network is much more highly organized and systematic. If you think about the
domain of 'cook' and the words linked to it, you will begin to see some of
this complexity. First we have to eliminate secondary meanings, such as the
link in VT from 'cook' to the set 'wrangle, misrepresent, manipulate, fudge,
falsify, fake'. (The basic problem here is that a thesaurus lists 'synonyms'
of a word, including all its senses, rather than listing forms that express
a particular meaning, or listing forms that are related to a topic.) So if
we limit our links to those that belong to a single sense, 'cook v. to
prepare food for eating', we find that the words linked to this sense fall
into sets: (1) methods of cooking (bake), (2) people who cook (chef), (3)
things used to cook (oven), (4) foods that are cooked (bread), (5)
procedures (recipe), (6) places (kitchen).
This particular word is an example of only one of several different types of
words. It is a 'scenario' word and links to other words in that scenario.
Not all words are of this type. For instance the word 'dog' belongs to a
very different kind of set. Scalars such as 'freezing, cold, cool, ambient,
warm, hot, burning', are a very different type. So the way in which a set of
lexical relations is displayed will vary according to the characteristics of
the set.
Please do not misunderstand me. I am not opposed to developing a more
sophisticated way of displaying lexical relations. I believe it would be a
huge advantage to see 'cook' and its "relatives" displayed in a format that
suits the pattern of relations it has. But we would need to carefully study
patterns of lexical relations. A common misconception is that lexical
relations are pairs of words. They are not. Even antonyms are not
necessarily pairs (good:bad/evil). Lexical relations hold between sets of
words. Not every lexical relation applies to every word. There is no antonym
of 'one', because one belongs to an ordered set. I suspect that there are a
limited number of types of lexical sets. Some of the primary types that I am
aware of are Scenarios, Hierarchies, Scalars, Ordered sets, Parts of a
whole.
Another complicating factor is that words belong to more than one set. For
instance 'knife' belongs to 'cook' as a 'thing used to cook', to 'tool' as a
specific kind of tool, to 'cut' within the scenario of cutting things, to
'weapon' within the scenario of fighting, and so on.
I have discussed with the programmers working on FieldWorks the issue of how
to display lexical relations and semantic domains in some kind of visual
format. Apparently the programming issues are not as difficult as designing
an underlying model that would capture the theoretical issues well. In
lexicography we are moving from a paper-bound orientation to an electronic
orientation with all the power that computers offer. It surprises me to
realize how paper-bound my perspective sometimes is. For instance last week
I was working on how FieldWorks handles reversal indexes and I was surprised
to find that I was arguing for a paper-bound format. I had to rethink the
different requirements of a paper reversal index versus an electronic
reversal. They are actually quite different and I had to request two
reversal strategies, one for paper and one for electronic dictionaries. We
are in the middle of a paradigm shift in lexicography. Much work is needed.
Ron Moe
-----Original Message-----
From: lexicographylist at yahoogroups.com
[mailto:lexicographylist at yahoogroups.com]On Behalf Of Mike Sangrey
Sent: Tuesday, March 21, 2006 9:46 AM
To: lexicographylist at yahoogroups.com
Subject: Re: [Lexicog] a question about profession
I would also be VERY interested in any responses to this.
In fact, I would be interested in any discussion about how this might be
done. Think outside the box! What would you like to see if you had a
bucket of pixie dust and two Aladdin's lamps.
On Sat, 2006-03-18 at 15:58 -0800, Wayne Leman wrote:
> I've been fascinated with words all my life. Both my mother and father
> were/are, as well. I have spent the last 30 years studying the
> Cheyenne language spoken in Montana and Oklahoma. It's part of the
> large Algonquian language family of North America. I thoroughly enjoy
> rooting around in the Cheyenne lexicon. It's a high point of any day
> for me if I discover some new Cheyenne word. I'm glad that we have
> computers now and decent programs for working with lexical data. The
> program I use most of the time for lexical work are Shoebox and
> Lexique Pro. Actually, I seldom use Shoebox anymore, but, instead, do
> everything directly from Lexique Pro. It's one of the most
> user-friendly computer programs I have used. I would like to see
> better ways of displaying semantic networks. Semantic domain lists are
> a good start but there is a lot more to semantic networking than
> linear lists. Of course there are plenty of semantic links available
> within Shoebox, Toolbox, and other similar lexical programs, but I
> have yet to use a program that *displays* the variety of linkages very
> well. I have been fascinated using a lexical program for English
> called Visual Thesaurus. It would be nice if we could display semantic
> linkages for other languages as VT does for English.
>
> Wayne
> -----
> Wayne Leman
> Cheyenne website: http://www.geocities.com/cheyenne_language
>
>
> SPONSORED LINKS
> Science kits
> Science education
> Science kit for kid
> Cognitive science
> Science education
> supply
> My first science kit
>
>
> ______________________________________________________________________
> YAHOO! GROUPS LINKS
>
> 1. Visit your group "lexicographylist" on the web.
>
> 2. To unsubscribe from this group, send an email to:
> lexicographylist-unsubscribe at yahoogroups.com
>
> 3. Your use of Yahoo! Groups is subject to the Yahoo! Terms of
> Service.
>
>
> ______________________________________________________________________
>
--
Mike Sangrey (msangrey AT BlueFeltHat.org)
Exegetitor.blogspot.com
Landisburg, Pa.
"The first one last wins."
"A net of highly cohesive details reveals the truth."
Yahoo! Groups Links
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/lexicographylist/
<*> To unsubscribe from this group, send an email to:
lexicographylist-unsubscribe at yahoogroups.com
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/