Replying to LO29485 --
Dear Organlearners,
John Zavacki <systhinc@msn.com> writes:
>But as well as we have exploited the electron to convert
>language to electron and back again, we still have no
>excellent algorithms for the translation of one human
>language to another. Why do we think this is?? Can it be
>the cultural differences in thinking about training vs education
>vs learning vs knowledge?? Or is it simply that the
>combination of words in phrases and sentences, even
>outside of the social context (but is there really anything
>outside of the social ! ! ! ! context?) is infinite?
Greetings dear John,
The way in which you contrasted "text to digital and back" conversion
with "languageA to languageB and back" conversion made me realise
how important it is to answer your question:-
Why do "we still have no excellent algorithms for the translation
of one human language to another"?
I will let my mind have a go at it and see where it will take me.
I can think of one reason. In the "text to digital and back" conversion
letter for letter, one after another, is taken and converted. But in a
sentence the sequence of words used is usually different between two
langauges. For example, the following sentence in Afrikaans "ek kan aan
een rede dink" has the word order "i can of one reason think" rather than
the conventional "i can think of one reason".
["ek"=i, "kan"=can, "aan"=at=by=....., "een"=one, "rede"=reason,
"dink"=think]
Another reason is that each category of words is used differently. For
example, the prepositions "aan"=at and "van"=of, but we will never say in
Afrikaans "ek kan van een rede dink". To say in English "i can think at
one reason" is equally alien.
But i think it goes deeper than reasons such as the two mentioned
above. When i create internal meaning from both the sentences
(external information)
"i can think of one reason"
"ek kan aan een rede dink"
i read the words from left to right. But in my mind i change their
order by constructing them into the pattern "being"-"becoming"-"being".
Thus the meaning constructed from them is for me
"i"-"can think of"-"one reason"
"ek"-"kan dink aan"-"een rede"
I do not know how it is for fellow learners, but that is how it is for me.
The English sentence stay the same so that it does not seem that any
construction has been taking place. But in the Afrikaans sentence the
verb "dink"=think is moved from the end to the middle, clearly showing
that a construction was involved. In Afrikaans verbs are put as far as
possible to the end of a sentence to stress their importance in relation
to the other words in the sentence.
Let us contemplate this "act of constructing meaning in the mind". If this
act is reversible, then it might just be possible to construct this
"excellent algorithm" which you write about. But should this act be
irreversible, then such an "excellent algorithm" becomes impossible. How
will we know that this act is reversible or irreversible?
Let us see what happens when i formulate the internal meaning
"ek"-"kan dink aan"-"een rede"
back into the sentence (external information)
"ek kan aan een rede dink"
It seems as if i have reversed the "act of constructing meaning in the
mind" into the "act of constructing information outside the mind". I
think it is indeed the case.
But here comes the snag. Although it is convention in Afrikaans to
put verbs as close to the end of the sentence, this convention can
be broken like many others to really stress something. For example,
i can write
"ek kan d'i'n'k' aan een rede"
where i try to indicate the stress on "dink" with the apostrophes '. To
do the same thing in English would be to allow for the unconventional
"i can of one reason think".
It looks somewhat like Shakespear's English. It reminds me of how
poets can reform skilfully grammer and lexicography to suite what
forms they have in mind.
I think that the conventions of grammer are intended to enforce some
reversibility on the "meaning<=>information" conversion. But the breaking
of such conventions shows an underlying irreversibility. So where does
this irreversibility originate from? It is here where Michael Polanyi's
concept of tacit knowing comes into the picture. He defined it as "we know
more that we can tell". When we tell for the first time something which we
know and have not told before, it is irreversible. In other words, the act
of formulating into information what we know tacitly is irreversible. It
is like a plant setting a flower. The act of flowering cannot be reversed.
It has to do with the irreversibility of any emergence.
In the second part Emergence of his book "The Tacit Dimension" Polanyi
discusses emergence in order to tell what may become of our tacit knowing.
I have combed through it to see whether he has anything to tell on the
irreversibility of emergences. Strange for a physical chemist like him, he
writes nothing directly, except for the following sentence on the concept
of an emergence which he has in mind:
"It resembles Bergson's elan vital en rejects
Kohler's dynamic equilibration".
Now, only one thing can move any system out of an equilbrium
state and that is irreversibility="entropy production".
I think that language had emerged originally to deal with irreversibility
of creativity and how it affects human relationships. It has not emerged
to deposit knowledge as written information. We may suppress this
irreversibility in language by conventions, but we overcome it in our
talking, gradually shaping our language into new forms.
>It's a better day in Ohio without the funnels.
I do not know what these funnels are. Could you please explain them to me.
If they are tornadoes, then i understand you.
With care and best wishes
--At de Lange <amdelange@postino.up.ac.za> Snailmail: A M de Lange Gold Fields Computer Centre Faculty of Science - University of Pretoria Pretoria 0001 - Rep of South Africa
Learning-org -- Hosted by Rick Karash <Richard@Karash.com> Public Dialog on Learning Organizations -- <http://www.learning-org.com>
"Learning-org" and the format of our message identifiers (LO1234, etc.) are trademarks of Richard Karash.