[lojban] Re: [Llg-members] nu ningau so'u se jbovlaste / updating a few jbovlaste entries

[guskant <gusni.kantu@gmail.com>]

(doi la.piier: la.and has moved the thread to lojban-list because the
topic is technical and not suitable for private list.)

2015-01-22 12:40 GMT+09:00 Pierre Abbat <phma@phma.optus.nu>:
> On Tuesday, January 20, 2015 01:52:26 guskant wrote:
>> Besides, it may be too much advanced thought, but I think all official
>> gismu and cmavo should be defined in Lojban.
>>
>> A language consists of only sequences of symbols regulated by a
>> grammar, but the universe expressed by a language depends on
>> definitions of words. As long as the words of a language are defined
>> by another language, the universe is restricted to that can be
>> expressed by the language used for the definitions. I think the
>> universe expressed by Lojban should be liberated from the other
>> languages.
>
> Defining all words of a language in that language necessarily produces a
> circular sequence of definitions, which makes it impossible for someone who
> doesn't already know the language to figure out what they mean. I'm not against
> defining Lojban words in Lojban, but they should also be defined in all six
> source languages.
>

I welcome any circular definitions:

2015-01-20 8:58 GMT+09:00 guskant <gusni.kantu@gmail.com>:
>
> Your idea will be helpful in our actual group work on Lojban
> definitions, though my suggestion was rather more simple, permitting
> any circular definitions between words in Lojban. My suggestion was
> only to construct a network of Lojban-only definitions between Lojban
> words, so that the network could refine possible models of the
> language Lojban.

The point of my suggestion is to liberate Lojban from any other
language in the following sense:

2015-01-20 8:58 GMT+09:00 guskant <gusni.kantu@gmail.com>:
> In my
> definition, A language is a subset of sequences of symbols. If you
> make any modification to the subset, it forms another language. Two
> different subsets (languages) may have different or isomorphic models.
> A "universe" I meant is a set on which a model is constructed. If each
> of languages P and Q has a model, and if the words of P are defined by
> Q, then the definitions fix a mapping from a universe of a model of Q
> to a universe of a model of P. If words of P are not defined by Q,
> such a mapping is not fixed. That is what I meant by the word
> "liberated".
>

Furthermore, this liberation is not much related to SWH, but rather to
Quine's theory of indeterminacy of translation. The translation
between two independent languages is indeterminate. As long as Lojban
is defined by another language, indeterminacy is meaningless, because
any translation between Lojban and the defining language is
determinable by the definitions. In such conditions, there are no room
to bring up signifiés peculiar to Lojban.

In this aspect, Lojban dictionary is different from any other
dictionaries between natural languages. Each natural language has
already its own signifiés, and compilers of dictionaries between two
languages should somehow find out plausible translations consistent
with many texts, though they are essentially indeterminate. On the
other hand, Lojban dictionary is indeed a list of definitions, because
Lojban has naturally no signifiés before being defined by the
dictionary. My suggestion is to let Lojban have its own signifiés by
defining the words in Lojban itself, including many circular
definitions. If it is implemented, "definitions" of Lojban words in
other languages will become only plausible translations, just as same
as dictionaries between natural languages.

For what? It's only for independence of signifiés of Lojban from
signifiés of other languages, but I also imagine giving AI a set of
Lojban PEG, Martin's tersmu and a network of definitions in Lojban,
and observing how it will develop sentences (at this point, it might
or might not be related to SWH as an experiment of it.) I don't have
skill of it, but I wish someone would do.

mu'o

-- 
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.
To post to this group, send email to lojban@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.