Entropy of Meaning.
Rollin Thomas
Postulate.
A statement is in the state of Maximum Entropy of Meaning when
translations between languages by a translating device fail to
change the statement.
Consider Babelfish.
English:
The quick brown fox jumped over the lazy dog.
Spanish:
El zorro marron rapido salto concluido el perro perezoso.
English:
The brown fox fast concluded jump the sluggish dog.
Spanish:
El salto concluido rapido del zorro marron el perro tardo.
English:
The concluded jump fast of the brown fox the dog I take.
Spanish:
El salto concluido rapidamente del zorro marron el perro que tomo.
English:
The quickly concluded jump of the brown fox the dog that volume.
Spanish:
El salto rapidamente concluido del zorro marron el perro que volumen.
English:
The jump quickly concluded of the brown fox the dog that volume.
Spanish:
El salto concluyo rapidamente del zorro marron el perro que volumen.
English:
The jump I quickly conclude of the brown fox the dog that volume.
Spanish:
El salto concluyo rapidamente del zorro marron el perro que volumen.
English:
The jump I quickly conclude of the brown fox the dog that volume.
Hence the statement
The quick brown fox jumped over the lazy dog.
Becomes
The jump I quickly conclude of the brown fox the dog that volume.
I wonder what other statements lead to the aforementioned resultant statement.
Funding for a study of meaning entropy degeneracy?
Rollin
Sean T Barrett
Rollin Thomas <tho...@mail.nhn.ou.edu> wrote:
>I wonder what other statements lead to the aforementioned resultant statement.
>Funding for a study of meaning entropy degeneracy?
Are there states of maximum entropy that are
cyclic? A goes to B, and B goes to A? Or perhaps
3 or more levels of circularity? What is the
_furtherst_ starting point from the entropofied
"The quick brown fox jumps over the lazy dog?"
How many maximally entropic (that is self-translating)
sentences of a given length are there?
Are there English sentences that are self-translating
when put through more than one SysTran-supported
language?
Technically the maximum-entropic ones _are_ a cycle
of two, an English "sentence" and a Spanish (or whatever)
"sentence". Study maximum entropy under cycles like
English->Spanish->French->English, English->French->Spanish->English,
alternating both of the above, E->S->F->S->E, etc.
I believe there is room for an entire new academic
discipline here, not just a research grant or three.
Too bad it's merely the study of incredibly naive
symbol shuffling routines, not a study of truths
about those languages.
buzzard
act...@greenheart.com
buz...@world.std.com (Sean T Barrett) wrote:
>Too bad it's merely the study of incredibly naive
>symbol shuffling routines, not a study of truths
>about those languages.
Unfortunately, there are no "truths" to be found
out about any language. Language is itself a
distortion, and even the simple act of naming
creates nothing but ghosts. A cursory decon-
struction of any sentence easily reveals the
meaninglessness of language. I can refer to a
"rock" or and "apple," yet all I do is manipu-
late abstractions. And those are merely names
for objects. Imagine the lack of meaning in-
herent in such words as "justice" or "fairness."
Language contains its own entropy.
act...@greenheart.com
Sean T Barrett
In article <353d4...@209.136.2.9>, <act...@greenheart.com> wrote:
>buz...@world.std.com (Sean T Barrett) wrote:
>>Too bad it's merely the study of incredibly naive
>>symbol shuffling routines, not a study of truths
>>about those languages.
>
>Unfortunately, there are no "truths" to be found
>out about any language.
Careful here, buddy mine. This is seriously debated
amongst 20th century philosophers: "what is truth (in language)",
necessitated, unfortunately, by the "semantic shift"
in philosophy. (The question-begging presumption that
since we can only reason using language, philosophy
should debate language--e.g. "how can language _really_
refer to things in the real world?")
I even have two books on my philosphy shelf: "Realism and Truth"
and "Language and Reality" (I think--the books are at home),
both of which are largely about how language connects to the
real world, has meaning, and can be considered "true", in
the context of "common-sense realism". The arguement is
roughly that those who start by assuming you know things
are led down an anti-realist path; if you instead start from
realism and salvage what you can, you can get fine theories
of reference and truth and knowledge, but they're a bit less
resplendent than you might hope. Seems reasonable, though,
since human knowledge is a frail, fallible thing in reality.
As I recall (I confess I have't read them in at least six years),
a late chapter in "Realism and Truth" tackles a very hard problem--
someone has shown that you can dispense with truth to explain
communication--people act in certain ways (speech acts), which
results in responsive acts, and this can all be described
behaviorally without "truth"--their acts can be in response
to events in the real world without being understood as
"referring" to the real world, and one can dispense with
theories of reference and truth entirely. The author clearly
thought long and hard about this issue, and offered the
half-hearted theory that the _success_ of education via
communication was most effectively understood by talking
about truth (and reference).
I will grant you, with respect to your criticism of my comment,
I was oversimplifying, trying to provide a convenient sound
bite. If I hadn't offered some alternative, one could argue
that all human thought is "naive symbol shuffling" (there are
anti-AI arguments that involve decrying this possibility:
"clearly human thought isn't naive symbol shuffling, but a
human could hand-execute an AI program, and all the human
would be doing was naive symbol shuffling, so how could
the results be naive symbol shuffling?") The loaded word
I'm really relying on here is "naive" but that's a little
subjective for a technical debate.
I'm only belaboring this tediously because certain
people have been publically challenging my knowledge,
putting me into explain-everything mode.
>Language contains its own entropy.
But language is a virus.
buzzard
eggplant or zucchini?
jul <j.z. ALHURIYEH>
On 21 Apr 1998 20:42:53 +0500, act...@greenheart.com opined:
}buz...@world.std.com (Sean T Barrett) wrote:
}
}>Too bad it's merely the study of incredibly naive
}>symbol shuffling routines, not a study of truths
}>about those languages.
}
}Unfortunately, there are no "truths" to be found
}out about any language.
Not so. Cultural paradigms are buried deep in all languages. Analyzing
the structure might not lead to any truths, but being able to compare
one language to another brings those paradigms to light. Being able
to speak a language fluently means developing a second social self.
In fact, however flawless your grammar and pronunciation are, you are
NOT fully fluent until you have master the protocol too, even the body
language, the "filler" words, the baby talk, etc. etc.
jul
== "Stealing kisses from
the lepers' faces ==
act...@greenheart.com
buz...@world.std.com (Sean T Barrett) wrote:
>I'm only belaboring this tediously because certain
>people have been publically challenging my knowledge,
>putting me into explain-everything mode.
I understand. I often find myself in the same
position, and it unfortunately usually exposes
my profound ignorance. Once I begin to dig
into what it is I'm actually saying, I find
that I'm not saying anything.
>>Language contains its own entropy.
>
>But language is a virus.
It reminds me more of a subatomic particle that
paradoxically exists as both an "object" and a
"concept," but only if you consider each separately.
But then, what do I know?
Brian Cash
jul <j.z. ALHURIYEH> (jul...@themermaid.net) wrote:
: On 21 Apr 1998 20:42:53 +0500, act...@greenheart.com opined:
: }buz...@world.std.com (Sean T Barrett) wrote:
: }
: }>Too bad it's merely the study of incredibly naive
: }>symbol shuffling routines, not a study of truths
: }>about those languages.
: }
: }Unfortunately, there are no "truths" to be found
: }out about any language.
: Not so. Cultural paradigms are buried deep in all languages. Analyzing
: the structure might not lead to any truths, but being able to compare
: one language to another brings those paradigms to light. Being able
: to speak a language fluently means developing a second social self.
: In fact, however flawless your grammar and pronunciation are, you are
: NOT fully fluent until you have master the protocol too, even the body
: language, the "filler" words, the baby talk, etc. etc.
IHNJH,IJLS "zug-zug"
Brian /-|-\
--
You might not agree with what I have to say,
but I will defend, with your life, my right to say it.
Rollin Thomas
bc...@bnr.ca (Brian Cash) writes:
>jul <j.z. ALHURIYEH> (jul...@themermaid.net) wrote:
>: On 21 Apr 1998 20:42:53 +0500, act...@greenheart.com opined:
>: }buz...@world.std.com (Sean T Barrett) wrote:
>: }
>: }>Too bad it's merely the study of incredibly naive
>: }
>
Punnet Poem.
Noam Chomsky. Noam Chomsky.
He called my house today. He talked to me today.
My cat answered the phone. My cat was making breakfast.
"Hello. This is Rollin's," "Hello. Is this Rollin?"
He said. He asked.
"Virgil, is Rollin there?" "This is Rollin,"
asked Chomsky. I assured Chomsky.
"Sure. Hang on," "You're a hooligan,"
Virgil said. Chomsky told me.
Noam Chomsky. Noam Chomsky.
He called Art Bell today. He talked to Art Bell today.
His cat answered the phone. His wife answered the phone.
"West of the Rockies, "He not here now,"
you're on the air." Ramona said.
"Comet, is Art there?" "That's okay. All we need
asked Chomsky. is ten minutes."
"Of course. Wait," "Ten minutes?"
Comet said. Ramona asked.
Rollin
The Reverse Psychology Major
act...@greenheart.com wrote in message <353e3...@209.136.2.9>...
>buz...@world.std.com (Sean T Barrett) wrote:
>
>>I'm only belaboring this tediously because certain
>>people have been publically challenging my knowledge,
>>putting me into explain-everything mode.
>
>I understand. I often find myself in the same
>position, and it unfortunately usually exposes
>my profound ignorance. Once I begin to dig
>into what it is I'm actually saying, I find
>that I'm not saying anything.
>
>>>Language contains its own entropy.
>>
>>But language is a virus.
>
>It reminds me more of a subatomic particle that
>paradoxically exists as both an "object" and a
>"concept," but only if you consider each separately.
>
... oh wow! look at all them big words!!!
beelzibub
ps;
*really*impressed