Hallo conlangers!

On 02.10.2015 12:22, R A Brown wrote:

> On 02/10/2015 08:47, And Rosta wrote:
>> On 30 Sep 2015 14:08, "Patrik Austin" wrote:
> [snip]
>>>
>>> If there weren't any problems to solve, the holy grail
>>>  would have been found long since.
>
> ... assuming that the Holy Grail actually exists     ;)
>
> IMO the search for the perfect loglang is only another
> aspect of the centuries old search for the perfect language
> which Umberto Eco discusses in his book "The Search for the
> Perfect Language" (ISBN 978-0-00-686378-6).

It is, in my opinion too.  The loglangs are perhaps the modern 
equivalents of the 17th-century "philosophical" languages, though the 
means by which their authors attempt to construct a language where 
argumentation can be formally tested have changed a lot.

>> Maybe. But nobody has really tackled the loglang problem
>>  before Loglan, and neither Loglan nor Lojban worked hard
>>  on ergonomic aspects of their designs.
>
> True. Indeed, it was James Cooke Brown's Loglan that began
> it all. His main aim was to test the Sapir-Worf hypothesis.

Yep.

>  Ergonomics was indeed not high in his thinking. Lojban was
> an attempt to produce a 'workable' version of Loglan but, in
> the process, AIUI took on board other aspects and tried to
> be wider in its appeal.

AFAIK, the initial motivation of the Logical Language Group to design 
their own loglang was to evade copyright claims from James Cooke Brown 
and his Loglan Institute!  But as they were designing their own loglang, 
they "improved" it in other respects, too, such as better computer 
tractability.

> Several other loglangs have followed
> but none AFAIK got a following even as great as the Loglan
> Institute's Loglan, let alone Lojban.

Indeed.  Lojban is by far the most popular loglang now, thanks to the 
activities of the LLG on the Internet.  Second ranking is the original 
Loglan; all others are below significance.

> [snip]
>
>> I'm not sure I get what point you're making. I suggest
>> that natlang syntax is optimal in being roughly the best
>>  solution within the limits of mental processing power.
>
> I agree.

I would say so, too.  It is the result of 2 million years of evolution, 
one would guess that there is not much to be improved after all that 
time.  At least, the notion that a single conlanger could do better than 
the thousands of generations who lived before him, seems pompous and 
hubristic to me.

As you know, I have no intention to go "beyond" natlangs in my 
conlanging.  Only to explore the possibilities of the human language 
faculty and the diversity of language structures that evolved over time.

>> And I suggest that altho we can work on optimizing
>> loglang designs, no possible optimizations suffice to get
>> the loglang across the threshold of holygrailhood.
>
> I agree also.  It JCB had a specific aim: to test the SW
> hypothesis. It seems to in developing a loglang, one should
> either do it just for a fun mental-exercise (like doing a
> cryptic crossword) or have some _clear, specified_ aims in
> mind.  The latter is certainly required IMO if for an engelang.

Concurred.  An engelang needs clearly specified design goals against the 
language can be tested.

> [snip]
>>
>> I don't see that a loglang needs to serve as a good
>> programming language.
>
> Nope - they are different beasts.

They are!  Loglangs are an attempt at applying formal logic to the 
communication *between humans*; programming languages serve to give 
orders to computers.  Loglangs may be closer to programming languages 
than natlangs are in some respects, such as the formal strictness of 
their specification; but they aren't programming languages.

>> We don't habitually need to use programming languages.
>
> Exactly. Programmers will use whatever programming is
> convenient and/or most suitable for whatever task they are
> doing.  I know I have been retired for 11 yeas now, so I am
> not as up-to-date as I'd like to be with current
> developments.  But when I was actively working as a computer
> scientist, it was not uncommon for people to devise their
> own programming language for some particular task[s] -
> often, I seem to recall, as part of doctoral research    :-)

I am also out of the business, but I think little has changed.  If 
anything, there are *more* programming languages designed now than in 
earlier times, as the diversity of computer applications (and the number 
of computer scientists who could design new programming languages ;) ) 
has increased.

> I have no reason to think this does not still happen.

It still happens.

>> And as soon as machines are able to understand human
>> language at all,
>
> Or even understand anything at all in the human sense of
> understanding.

Touché.

> [snip]
>>
>> The virtue of the loglang is simply that it doesn't force
>> you to discard PAS-structural information from the
>> sentential encoding just because the language has no way
>>  to encode it. That is, it allows the encoding to be more
>>  faithful to the encoded thought.
>
> That's interesting. I am neither agreeing nor disagreeing
> with this.  But JCB's idea with Loglan AIUI was to devise a
> a language so different from natural languages that people
> learning it would _think in a different way_.  You appear to
> be saying that you consider a well designed loglang allows
> more faithful encoding of human thought.
>
> Interesting.

I understand it like you, Ray.  I may be wrong about this, but I have 
the impression that many loglangers (though perhaps neither JCB nor the 
LLG) view loglangs as a means of enforcing clearer and more logical 
thinking in a Whorfian way.  For instance, loglangs have been embraced 
by the hyper-rationalist Transhumanist movement for long.  If you look here

http://www.aleph.se/Trans/Cultural/Linguistics/index.html

for instance, one of the big topics is Lojban (the link is dead, though; 
the whole page hasn't been maintained for 15 years!); the other big 
topic is "General Semantics", another attempt at enforcing "clear 
thinking" by means of language engineering.

--
... brought to you by the Weeping Elf
http://www.joerg-rhiemeier.de/Conlang/index.html
"Bêsel asa Éam, a Éam atha cvanthal a cvanth atha Éamal." - SiM 1:1