Some references on lexicons and ontologies
I'd like to mention some resources that may be of interest to
people working on lexicons and ontologies:
1. The Unitex project in France has developed dictionaries in
both French and English and offers some open-source tools
under the GPL license:
http://www-igm.univ-mlv.fr/~unitex/
Excerpt: "With this tool, you can handle electronic
resources such as electronic dictionaries and grammars and
apply them. You can work at the levels of morphology, the
lexicon and syntax."
2. Some publications by two linguists who have spent their
careers on analyzing multiple languages in the search for
universal semantic types and primitives:
Talmy, Leonard (2000) Toward a Cognitive Semantics,
Volume I: Concept Structuring Systems,
Volume II: Typology and Process in Concept Structure,
MIT Press, Cambridge, MA.
Wierzbicka, Anna (1992) Semantics, Culture, and Cognition:
Universal Human Concepts in Culture-Specific Configurations,
Oxford University Press, Oxford.
Wierzbicka, Anna (1996) Semantics: Primes and Universals,
Oxford University Press, Oxford.
3. And on a related note, Science magazine published an article
about research on the evolution of Nicaraguan Sign Language,
which suggests that the semantic primitives for signed and
spoken languages are very closely related. Some excerpts
from a BBC summary below.
John Sowa
________________________________________________________________
Source: http://news.bbc.co.uk/1/low/sci/tech/3662928.stm
A new sign language created over the last 30 years by deaf children in Nicaragua
has given experts a unique insight into how languages evolve. The language
follows many basic rules common to all tongues, even though the children were
not taught them. It indicates some language traits are not passed on by culture,
but instead arise due to the innate way human beings process language, experts
claim.
One key trait that the children adopted is called "discreteness". This refers
to the process of breaking down information into small manageable packages.
Expressions of motion are particularly useful for studying discreteness in spoken
and sign languages. In developed languages, we break up the idea of continuous
motion into separate words. So, in the expression "rolling down the hill",
one word (rolling) conveys the movement, while another (down) conveys the direction.
But if a hearing person were asked to convey this idea in gestures alone, they
would almost certainly do it with a single continuous movement.
Dr Senghas and her colleagues showed the deaf people from each of the age groups
a cartoon, in which a cat swallows a ball and then wobbles down a steep road.
Then they asked the participants to tell the story.
The oldest group, who invented the initial "crude" form of NSL, told it with
one continuous gesture as a hearing person might. But the younger groups did
something different. They separated the movement and direction into separate
signs as is done in spoken language.
"If they were just clever at learning they would have learned to do it the way
they had seen it being done," said Dr Senghas. "But that isn't what they did
- they ended up acquiring something different. They ended up breaking down the
gestures into something they could build a language out of."
This is compelling evidence that humans are predisposed to develop language
in this way, say the researchers. In other words, children instinctively break
information down into small chunks so they can have the flexibility to string
them back together, to form sentences with a range of meanings.
Dr Senghas does not claim her findings support the extreme "nature" camp, but
that they do suggest there is an instinctive component to the way we learn language.
"It doesn't prove that language is hard wired to the degree some people say
it is, but it does prove the fundamentals of language are part of the innate
endowment," she said. "So you don't have language or grammar in your head when
you are born, but you do have certain learning abilities."
Professor Pinker said the results of the study showed something that had always
been suspected by some psychologists. But, he said: "Since children's language
ordinarily ends up the same as their parents' language, one couldn't easily
pinpoint what their minds added. "It takes a case in which the language children
end up with is more complex than the language they hear to identify the creative
contribution of the child."