From pycyn@aol.com Fri Aug 31 18:06:51 2001
Return-Path: <Pycyn@aol.com>
X-Sender: Pycyn@aol.com
X-Apparently-To: lojban@yahoogroups.com
Received: (EGP: mail-7_3_2); 1 Sep 2001 01:06:50 -0000
Received: (qmail 77760 invoked from network); 1 Sep 2001 01:06:48 -0000
Received: from unknown (10.1.10.142)
  by l7.egroups.com with QMQP; 1 Sep 2001 01:06:48 -0000
Received: from unknown (HELO imo-r03.mx.aol.com) (152.163.225.99)
  by mta3 with SMTP; 1 Sep 2001 01:06:47 -0000
Received: from Pycyn@aol.com
	by imo-r03.mx.aol.com (mail_out_v31_r1.4.) id r.5b.1afe8e06 (4013)
	 for <lojban@yahoogroups.com>; Fri, 31 Aug 2001 21:06:38 -0400 (EDT)
Message-ID: <5b.1afe8e06.28c18ea2@aol.com>
Date: Fri, 31 Aug 2001 21:06:42 EDT
Subject: Re: [lojban] Symbolic Logic and Lojban
To: lojban@yahoogroups.com
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_5b.1afe8e06.28c18ea2_boundary"
X-Mailer: AOL 6.0 for Windows US sub 10535
From: pycyn@aol.com
X-Yahoo-Message-Num: 10363

           
--part1_5b.1afe8e06.28c18ea2_boundary 
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 8/31/2001 6:03:17 PM Central Daylight Time, 
samuelriv@yahoo.com writes:



> Bertrand Russel was an excellent mathematician,
> philosopher, and logician, but his attempt at linking
> the worlds of symbolic logic to natural linguistics
> failed miserably.
> 





Well, it depends on how you look at it.  From Russell's point of view (most 
of the time), natural language was a disaster area, the chief source of 
metaphysical and just ordinary nonsense.  His aim (among many others, to be 
sure) was to, purify language so that the form in which something was said 
would accurately reflect the structure of the fact being described 
(beautifully spelled out in young Ludwig's early madness and set to music by 
Wuorinen).  Russell's own attempts along this line -- and several others 
later -- did not really succeed in getting rid of metaphysical howlers, 
though they did help to make more clear just how language did help perpetrate 
and perpetuate them.  And Russel's basic stuff, after a century of tinkering, 
provides the underlying framework for virtually all linguistic studies -- 
more or  less remotely. Not too bad for a miserable failure.

<I've been studying symbolic and semantic logic, and
have been trying to incorporate their principles into
an IAL (international auxiliary language) which I am
creating.>

Welcome to the throng.  We have a half-century jump on you and rest on a 
tradition of another 5 centuries.  If you lean on us, do give credit (we tend 
to forget our ancestors rather too often).
What is semantic logic?  (or is the right question, What is non-semantic 
logic)?

<Obviously Bertrand Russel is GOD of logic>
Oh would he hate that expression! (not that his attitudes would not count as 
evidence in its favor).

<Essentially he designated all sentences in the form
F(x,y,...) where the F function is the verb and the
arguments act as each segment acting as a noun,
adjective, or adverb. For example, the sentence "I
went to school at three o'clock today" would be
symbolized as 'went(I,school,3:00,today) where the
place value system would denote how each argument
would act in the sentence.>

Common nouns, adjectives and adverbs -- also prepositions and various other 
what-have-yous from other language traditions -- are all along with verbs in 
the function category.  The arguments are all proper names or variables for 
such.  Your sentence might do as a first pass, but would have to be unpacked 
about four-fold before it would pass a Russell inspection.

<Now
let's see what happens when we want to CHANGE a
sentence while preserving the function. For example,
instead of "I went to school today" you want to say "I
went to school in Chicago today". These would be
represented by:
'went(I,school,today)' and
'went(I,school,Chicago,today)'

Now the addition of Chicago creates a fourth place
value denoting location instead of time, and the only
way this can happen is if the FUNCTION changed. So
essentially if you want to use a Russelian
logic-language, you would have to create your verbs
(or functions) to account for all possible arguments,
otherwise there must be an influx of modified verb
forms to adjust for necessary arguments, or, more
likely, language use will be constricted and never
creative, and usage will collapse.>
Some of these factors might well merely mean replacements in other parts of 
the fully analysed sentence (it does matter, by the way, what year you are 
talking about for deciding what the fully analysed sentence should look like; 
Russell shifted and others after him but in his tradition moved even more).  
At one point or another most logicians who are serious about oardinary 
language do work out ways to introduce those modifications in systematic 
ways.  They may amount to new predicates (though they may just be different 
sentence types)  but one related in rule-governed ways to older ones.  Which 
is, of course, what languages do -- except for the rule part.

<lojban, to get around this problem, used complicated
functions have up to five place values, some making
little sense. For example, "lanci" is lojban for flag.
If you want to say "that is a flag" you say "za
lanci". Now, there are four place values, so you can
say anything (the arguments are capitalized) on the
lines of "THAT is a flag symbolizing AMERICA with
pattern of STARS AND STRIPES on material of COTTON".
However, if you want to say "That is a flag hanging on
a flagpole" or "next to my house", then you have
serious problems, and lojban attempts to resolve them
with essentially new functions and new arguments, and
meanwhile all the arguments you DON'T want to fill in
must be replaced by "zo'e", so essentially "za lanci
zo'e zo'e zo'e mopni" which means "That flag is made
of cotton">

{ta lanci}, {za} is the adverb.  And, since {zo'e} can be dropped freely as 
long as we keep track of where we are, {ta fanci fo lo mopni}  ({fo} for 
fourth place, {lo} because an argument, a name, is needed here, not a bare 
predicate :"a piece of cotton") That sentence literally means "that is a flag 
made of cotton." As for resolving the problem of adding places, we do it 
pretty much as we do in English: attach phrases or clauses as need be.  
Lojban is particularly fond of compound functions: {ta lanci co dandu lo 
grana}  "That is a flag hanging from a stick"
This may seem a deviation from strict logic but it is not all that great a 
one, and it does marvels for the language.

<This is a 
completely linear interpretation of linguistics. It allows for 
complete transcription of any sentence at any moment into root 
functions, but it fails to realize the amount of language that is 
creative. Language is about building and demolishing and shaping; by 
mathematical survey chances are that over half the sentences we speak 
each day have NEVER been spoken by ANYONE in human history.

IMHO, lojban hits high on every note except its key, its foundation, 
if you will, and this I feel is fatal.>

Lojban, being a human language, is not linear any more than any other human 
language is (it comes out a drib at a time over time).  I would argue that 
the same is true of the language of logic, too, but won't work on it here.  
In any case, Lojban has the same horrendous (and largely unknown) structure 
as any other language and we are constantly finding new, often dismaying, 
aspects of that structure.  Further, in Lojban we have a lot of hard evidence 
that most of what we say has never been said before, because we have records 
of so much of it from close unto day one. Sounds reasonably creative -- and 
if you see what some of its practitioners do, the problem of creativity (as 
opposed sometimes to intelligibility) would be the last of your worries.

Would you mind passing these replies on to whatever group you wrote the 
original to?  a balance is always useful. 





           
--part1_5b.1afe8e06.28c18ea2_boundary 
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><BODY BGCOLOR="#ffffff"><FONT  SIZE=2>In a message dated 8/31/2001 6:03:17 PM Central Daylight Time, 
<BR>samuelriv@yahoo.com writes:
<BR>
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">Bertrand Russel was an excellent mathematician,
<BR>philosopher, and logician, but his attempt at linking
<BR>the worlds of symbolic logic to natural linguistics
<BR>failed miserably.
<BR></BLOCKQUOTE>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>Well, it depends on how you look at it. &nbsp;From Russell's point of view (most 
<BR>of the time), natural language was a disaster area, the chief source of 
<BR>metaphysical and just ordinary nonsense. &nbsp;His aim (among many others, to be 
<BR>sure) was to, purify language so that the form in which something was said 
<BR>would accurately reflect the structure of the fact being described 
<BR>(beautifully spelled out in young Ludwig's early madness and set to music by 
<BR>Wuorinen). &nbsp;Russell's own attempts along this line -- and several others 
<BR>later -- did not really succeed in getting rid of metaphysical howlers, 
<BR>though they did help to make more clear just how language did help perpetrate 
<BR>and perpetuate them. &nbsp;And Russel's basic stuff, after a century of tinkering, 
<BR>provides the underlying framework for virtually all linguistic studies -- 
<BR>more or &nbsp;less remotely. Not too bad for a miserable failure.
<BR>
<BR>&lt;I've been studying symbolic and semantic logic, and
<BR>have been trying to incorporate their principles into
<BR>an IAL (international auxiliary language) which I am
<BR>creating.&gt;
<BR>
<BR>Welcome to the throng. &nbsp;We have a half-century jump on you and rest on a 
<BR>tradition of another 5 centuries. &nbsp;If you lean on us, do give credit (we tend 
<BR>to forget our ancestors rather too often).
<BR>What is semantic logic? &nbsp;(or is the right question, What is non-semantic 
<BR>logic)?
<BR>
<BR>&lt;Obviously Bertrand Russel is GOD of logic&gt;
<BR>Oh would he hate that expression! (not that his attitudes would not count as 
<BR>evidence in its favor).
<BR>
<BR>&lt;Essentially he designated all sentences in the form
<BR>F(x,y,...) where the F function is the verb and the
<BR>arguments act as each segment acting as a noun,
<BR>adjective, or adverb. For example, the sentence "I
<BR>went to school at three o'clock today" would be
<BR>symbolized as 'went(I,school,3:00,today) where the
<BR>place value system would denote how each argument
<BR>would act in the sentence.&gt;
<BR>
<BR>Common nouns, adjectives and adverbs -- also prepositions and various other 
<BR>what-have-yous from other language traditions -- are all along with verbs in 
<BR>the function category. &nbsp;The arguments are all proper names or variables for 
<BR>such. &nbsp;Your sentence might do as a first pass, but would have to be unpacked 
<BR>about four-fold before it would pass a Russell inspection.
<BR>
<BR>&lt;Now
<BR>let's see what happens when we want to CHANGE a
<BR>sentence while preserving the function. For example,
<BR>instead of "I went to school today" you want to say "I
<BR>went to school in Chicago today". These would be
<BR>represented by:
<BR>'went(I,school,today)' and
<BR>'went(I,school,Chicago,today)'
<BR>
<BR>Now the addition of Chicago creates a fourth place
<BR>value denoting location instead of time, and the only
<BR>way this can happen is if the FUNCTION changed. So
<BR>essentially if you want to use a Russelian
<BR>logic-language, you would have to create your verbs
<BR>(or functions) to account for all possible arguments,
<BR>otherwise there must be an influx of modified verb
<BR>forms to adjust for necessary arguments, or, more
<BR>likely, language use will be constricted and never
<BR>creative, and usage will collapse.&gt;
<BR>Some of these factors might well merely mean replacements in other parts of 
<BR>the fully analysed sentence (it does matter, by the way, what year you are 
<BR>talking about for deciding what the fully analysed sentence should look like; 
<BR>Russell shifted and others after him but in his tradition moved even more). &nbsp;
<BR>At one point or another most logicians who are serious about oardinary 
<BR>language do work out ways to introduce those modifications in systematic 
<BR>ways. &nbsp;They may amount to new predicates (though they may just be different 
<BR>sentence types) &nbsp;but one related in rule-governed ways to older ones. &nbsp;Which 
<BR>is, of course, what languages do -- except for the rule part.
<BR>
<BR>&lt;lojban, to get around this problem, used complicated
<BR>functions have up to five place values, some making
<BR>little sense. For example, "lanci" is lojban for flag.
<BR>If you want to say "that is a flag" you say "za
<BR>lanci". Now, there are four place values, so you can
<BR>say anything (the arguments are capitalized) on the
<BR>lines of "THAT is a flag symbolizing AMERICA with
<BR>pattern of STARS AND STRIPES on material of COTTON".
<BR>However, if you want to say "That is a flag hanging on
<BR>a flagpole" or "next to my house", then you have
<BR>serious problems, and lojban attempts to resolve them
<BR>with essentially new functions and new arguments, and
<BR>meanwhile all the arguments you DON'T want to fill in
<BR>must be replaced by "zo'e", so essentially "za lanci
<BR>zo'e zo'e zo'e mopni" which means "That flag is made
<BR>of cotton"&gt;
<BR>
<BR>{ta lanci}, {za} is the adverb. &nbsp;And, since {zo'e} can be dropped freely as 
<BR>long as we keep track of where we are, {ta fanci fo lo mopni} &nbsp;({fo} for 
<BR>fourth place, {lo} because an argument, a name, is needed here, not a bare 
<BR>predicate :"a piece of cotton") That sentence literally means "that is a flag 
<BR>made of cotton." As for resolving the problem of adding places, we do it 
<BR>pretty much as we do in English: attach phrases or clauses as need be. &nbsp;
<BR>Lojban is particularly fond of compound functions: {ta lanci co dandu lo 
<BR>grana} &nbsp;"That is a flag hanging from a stick"
<BR>This may seem a deviation from strict logic but it is not all that great a 
<BR>one, and it does marvels for the language.
<BR>
<BR>&lt;This is a 
<BR>completely linear interpretation of linguistics. It allows for 
<BR>complete transcription of any sentence at any moment into root 
<BR>functions, but it fails to realize the amount of language that is 
<BR>creative. Language is about building and demolishing and shaping; by 
<BR>mathematical survey chances are that over half the sentences we speak 
<BR>each day have NEVER been spoken by ANYONE in human history.
<BR>
<BR>IMHO, lojban hits high on every note except its key, its foundation, 
<BR>if you will, and this I feel is fatal.&gt;
<BR>
<BR>Lojban, being a human language, is not linear any more than any other human 
<BR>language is (it comes out a drib at a time over time). &nbsp;I would argue that 
<BR>the same is true of the language of logic, too, but won't work on it here. &nbsp;
<BR>In any case, Lojban has the same horrendous (and largely unknown) structure 
<BR>as any other language and we are constantly finding new, often dismaying, 
<BR>aspects of that structure. &nbsp;Further, in Lojban we have a lot of hard evidence 
<BR>that most of what we say has never been said before, because we have records 
<BR>of so much of it from close unto day one. Sounds reasonably creative -- and 
<BR>if you see what some of its practitioners do, the problem of creativity (as 
<BR>opposed sometimes to intelligibility) would be the last of your worries.
<BR>
<BR>Would you mind passing these replies on to whatever group you wrote the 
<BR>original to? &nbsp;a balance is always useful. 
<BR>
<BR>
<BR>
<BR>
<BR></FONT></HTML>
           
--part1_5b.1afe8e06.28c18ea2_boundary--