[Corpora-List] Semantic primitives

Tue Jan 18 19:31:10 UTC 2011

Hi Ken,

Recently, Yorick Wilks posted two papers on this forum.  One explained how
semantic primitives were viewed differently by each of the research groups,
none of which settled for the previous group's semantic primitives.  

In the second paper on Semantic Preferences, he states his belief that many
choices of primitive sets are useful, adequate, and helpful in
characterizing semantics.  

So my conclusion is that many semantic primitive sets exist, and each person
chooses (or invents) her own.  

The reason semantic primitives can be expected to exist, but not be exactly
the same from person to person, is well expressed by Steven Pinker in his
book How the Mind Works:  

Tooby and Cosmides point out a fundamental consequence

of sexual reproduction: every generation, each person's blueprint is
scrambled with someone else's. That means we must be qualitatively alike. If
two people's genomes had designs for different kinds of machines, like an
electric motor and a gasoline engine, the new pastiche would not specify a
working machine at all. Natural selection is a homogenizing force within a
species; it eliminates the vast majority of

macroscopic design variants because they are not improvements. Natural
selection does depend on there having been variation in the past, but it
feeds off the variation and uses it up. That is why all normal people have
the same physical organs, and why we all surely have the same

mental organs as well. There are, to be sure, microscopic variations among
people, mostly small differences in the molecule-by-molecule sequence of
many of our proteins. But at the level of functioning organs, physical and
mental, people work in the same ways. Differences among

people, for all their endless fascination to us as we live our lives, are of
minor interest when we ask how the mind works.

The differences are more pronounced than the similarities because we use
them in communication to stamp our purpose on the world.  But the people who
study odd languages have found enormous variation among the 6,000 known so
far, and culture has obvious effects on encouraging some and discouraging
others, so that the primitives may not be fully linguistic in nature.
However, as the movie title went, there will be primitives.  

JMHO,

-Rich

Sincerely,

Rich Cooper

EnglishLogicKernel.com

Rich AT EnglishLogicKernel DOT com

9 4 9 \ 5 2 5 - 5 7 1 2

  _____  

From: corpora-bounces at uib.no [mailto:corpora-bounces at uib.no] On Behalf Of
Ken Litkowski
Sent: Tuesday, January 18, 2011 9:47 AM
To: corpora at hd.uib.no
Subject: [Corpora-List] Semantic primitives

In a recent <http://mailman.uib.no/public/corpora/2011-January/012085.html>
posting, John Sowa criticized so-called primitives as being the results of
analysis by adults writing dissertations and expressed the belief that there
are no primitives that are truly primitive. I would agree with John to the
extent such work may be a priori. However, I am concerned that his
statements may have a discouraging effect on research into primitives,
particularly given his well-deserved reputation. Similar statements by
Veronis and Ide in 1991 had such an effect on research with
"machine-readable dictionaries" (all the rage during the late 1970s and
1980s).

I have spent 40 years with my digraph analysis of dictionary definitions,
most notably in early 2002 helping Oxford identify hypernyms
(superordinates) for a noun hierarchy derived from the Oxford Dictionary of
Engliah. This work has been of some use in Oxford's latest roll-out
<http://oxforddictionaries.com/> . My techniques have proved useful in
analyzing prepositions
<http://www.clres.com/online-papers/PrepositionClasses.pdf>  and FrameNet
frame elements <http://www.clres.com/db/feindex.html>  (which was aided by
the Oxford noun hierarchy). I view this work as being very a posteriori.
Robert Amsler recently expressed his enthusiasm for Google n-grams as being
useful identifying component elements of noun compounds, likening this to
analyzing chemical compounds into their component elements and binding
properties. Thus, I hope we don't take John's words to stifle the hunt for
semantic primitives (look at the billions in the hunt for the Higgs boson).

I have an expanded version of this note at my blog
<http://www.clres.com/blog/?p=172> , where I provide further links to
salient materials.

    Ken

-- 
Ken Litkowski                     TEL.: 301-482-0237
CL Research                       EMAIL: ken at clres.com
9208 Gue Road                     Home Page: http://www.clres.com
Damascus, MD 20872-1025 USA       Blog: http://www.clres.com/blog
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/corpora/attachments/20110118/5709a6f4/attachment.htm>
-------------- next part --------------
_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora