Hallo conlangers!

On 15.10.2015 11:03, R A Brown wrote:

> On 14/10/2015 22:10, And Rosta wrote:
>> On 14 Oct 2015 15:25, Jörg Rhiemeier wrote:
>>>
>>> The recent discussions of loglangs (as a tangent of the
>>> (p)aaS thread) has raised a question in me.
>
> I was thinking along similar lines as that thread neared its
> end    :)
>
>>> I have for long been of the opinion that loglangs are
>>> beside the point because language and formal logic
>>> serve "different purposes".
>
> While natural language and formal logic may serve different
> purposes, I don't think that in itself invalidates loglang
> experiments.  In so far as they achieve or fail to achieve
> their objectives, or have to be modified to be humanly
> usable, they may give some insight into the way language works.

Sure.  The value of logical languages may not lie in applying formal 
logic to conversation in order to weed out inefficiency and ambiguity, 
nor in enforcing more "logical" thought in a Whorfian way, but in 
testing what kind of "unnatural" systems the human language faculty can 
cope with.

> But I was left wondering both about their exact purpose and,
> indeed, the definition of them.  I said in one of my emails
> in that thread that my definition of loglang was probably
> too narrow.

Fair; we know of Loglan that its purpose is to test the Sapir-Whorf 
hypothesis.  We also know that some transhumanists want to use Lojban to 
enforce more precise and "correct" reasoning in a Whorfian way, but that 
is, like the proposal of using Loglan or Lojban as an IAL, not part of 
the original design goals.

>>> I used to say that the purpose of formal logic was to
>>> prove or disprove assertions, but I have been
>>> criticized for that here as being "too narrow".
>
> To prove or disprove propositions was the purpose of
> Aristotelian logic.  But, as I observed in an earlier email,
> formal logic has moved on over the two and half thousand
> centuries since Aristotle.  Mark gave a brief and rough
> outline of some of these developments.

Yes.  While the original purpose of formal logic indeed was to test 
assertions, it has been put to other uses later.

>>> So, it is upon you: what is (or are) the purpose(s) of
>>>  formal logic, in your eyes?  And how does language
>>> benefit from designing a logical language?  (And, of
>>> course, to which degree do such loglangs as Loglan or
>>> Lojban actually implement formal logic?)
>
> I'll pass on the first question.
>
> As regards the second question, those who still pursue the centuries 
> of goal of finding the "perfect language" will surely find modern 
> logic formalism tools they would wish to use.  And suggested that 
> James Cooke Brown was in fact in pursuit of just that language and 
> that the Sapir-Worf thing was an afterthought. He may well be right.   
> That is possibly still the pursuit of some loglangers.

It is.

> (Umberto Eco's book "The Search for the Perfect Language" gives an 
> interesting history of the pursuit over the centuries)

Yes, I have read that book, and it is not only interesting, but 
something every conlanger ought to read!

>> I think your question implies a perhaps off-beam
>> understanding of what loglangs are. Tho loglangs are
>> described as "based on formal logic", that's really just
>>  a quick and dirty approximation, and is needlessly vague
>>  as a definition.
>
> But not IMO nearly as vague as defining a loglang as a "logical 
> language."  Many an auxlang enthusiast has claimed that her/his chosen 
> auxlang is 'perfectly logical'.  When I was at school long years ago, 
> our headmaster used to bang on about Latin being a logical language; I 
> have seen that claim made for French - and I have not the slightest 
> doubt that many have made similar claims about other natlangs.    :-)

A much-colported case is Aymará, to name just one.

"Logic" is a word with many meanings, and one is easily misled by that.  
For instance, I once entertained the notion that the Vulcans of the Star 
Trek universe, being devoted to "logic", should speak a loglang.  I no 
longer think so, as I realized that Vulcan "logic" has very little to do 
with formal logic, and that speaking a loglang is in no way a 
consequence of such an ideology.

>> A loglang, in a definition that I use explicitly and that
>> most loglangers use implicitly even if not fully aware of
>> it, is a language that unambiguously encodes
>> predicate--argument structure of unlimited complexity.
>
> [snip]
>
>> Natural languages encode PASs of limited complexity and
>> do it ambiguously;
>
> To do unambiguously what natlangs do with ambiguity - this seems to 
> smack of the search for the perfect language. Though would a language 
> in which ambiguity is impossible really be a perfect language for us 
> humans?

A valid question.  Ambiguity is not the devil which many searchers of 
the perfect language consider it to be.  It has also its virtues: our 
languages would be much less expressive without metaphors, which are 
ambiguous almost by definition.  Another case of people trying to fix 
what isn't broken!

>> natlang syntactic structure is in part
>> a PAS (or representation thereof). So the value of a
>> loglang is that it encodes thought with greater fidelity
>> than natlangs do. There are thoughts that a loglang can
>> encode fully and unambiguously that natlangs can encode
>> only partially and ambiguously.
>
> But, in fact, do any existing loglangs encode thoughts fully and 
> unambiguously?  I think not.

I don't know; but you are probably right here.

> Also are human thought themselves ever fully formed and unambiguous?  
> I somehow doubt this.

So do I.

> But, once more, I say that the aim to encode fully and unambiguously 
> what natlangs do imperfectly and ambiguously is IMO the search for the 
> perfect language.

Indeed!

>> I haven't answered your question about the purposes of
>> formal logic, because the answer probably isn't relevant
>>  to loglangs. Another question one might pertinently ask
>>  is what value logic has for linguistics.
>
> I would say that it may give some insight into the workings of human 
> language and that loglang _experiments_ may help in this regard.

Right.

> Um - I must return to my Glossopoeia page and redefine 'loglang', I 
> think.  but it will have to wait as I am actually in the process of 
> updating stuff on Britainese   ;)

Do so.  Why not do both in a single update?

--
... 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