Re: Language comparison

> >* little effort to use
>
> If there were human languages that would require more effort to talk, then
> they would have died out in the many thousands of years of language evolution.
Hardly. The only thing evolution produces is sufficiency, not
efficiency. I think it's a mistake to belive that (any) evolution
results in "the best", just probably the best amongst the local
competitors.

And in any case, it might well be possible to intentionally design an
even easier language. I don't know, but I'm at least willing to give
the idea an opportunity.

> Language is not a "translation" of thought, but merely a form of thought used for a
> specific purpose.
Okay. Perhaps, by your framing, it would be better phrased as being
more similar to non-linguistic forms of thought? Or would you say that
it's impossible (or inapplicable?) to move between them?

> "Über die Verschiedenheit des menschlichen Sprachbaus und seinen Einfluss auf die
> geistige Entwicklung des Menschengeschlechts" ('about the heterogeneity of
> language construction and its influence [of language construction] on the
> intellectual development of mankind'), which is said to be one of the most
> interesting books on languages (and one of the first). Maybe I should try
> again to read it! :)
Sounds interesting; any chance of an English translation? (Ich spreche
keine Deutsch...) ([sic], probably)

> However, there's no such directionality in the evolution of language.
Does that mean you believe the current (~1 bit per character IIRC)
amount of density is the maximum feasible?

> >* as clear/ambiguous as desired
> >* possible to carry multiple meanings (if desired)
>
> Unlike in computer "languages" (which is a misleading metaphora in this
> discussion) or Logics, meaning in natural languages is always flexible.
Yes, but it's not always possible to have it flexible to the exact
degree you want. I cannot talk about "demons", for example, without
invoking a bunch of extra meanings that I don't necessarily intend. Or
for another, in English, it's difficult to talk about groups of people
of a specified gender-ratio; I could add that through (cumbersome)
explanation, but the grammar doesn't support it natively. Some do.

I think that's not a very clear (ha) explanation of what I meant, but
perhaps it's better.

> All writing systems can be read from any direction/inversion. We're just
> more used to read in a specific way. If you would have learnt reading in a
> class with a single book so that all had to sit around it in a circle, then
> you'd be able to read equally fluent from any direction.
But English, at least, is certainly "meant" to be read in a particular
direction/inversion. I would say that some - Wine Script on the other
thread, for example - don't necessarily have that, and therefore don't
have the added difficulty when trying to read it in a nonstandard way.
Is it *equally easy* in English? No.

> >* writing system that fully uses its medium (e.g., non-serial use of 2d
> >space)
>
> The aim of writing is not to fully use the paper, but to represent speech,
> which is essencially bidimensional.
I strongly disagree on that. The aim of writing is to convey meaning.
If you use it merely as a code for speech, then fine, but then you
loose most of what you might have been able to do through the
different medium. It's like only using TVs for slideshows; the medium
is capable of more.

> >* multimodal - maximally using whatever means of information transfer
> >available
>
> Face-to-face communication does this always. It involves rhythm and melody
> and any movements we make.
Yes, but not in ways that are taken advantage of linguistically.

And in this, I can even give a real-life example - hearing people I
know who also know ASL start to use gestures in a *linguistic* manner
- when talking (not signing per se) with other hearing folk - and thus
their gestures much more meaningful. Having a full-on sign language
going side by side (not parroting as in TC, but additive content) with
speech would be a much better use than simply relegating it to "any
movements we [happen to] make".

The latter is like saying that someone who can sign, but only make
incoherent (although equally meaningful to uninformed "body language"
/ gesture) speech sounds is using their speech maximally to complement
there signing. I don't think this is true.

> All speech organs have a more original function that is different from
> producing speech. When that original function is used, you can't produce
> speech. All natural languages are either sound languages (respiratory
> system, mouth) or sign languages (arms, hands, face, body). There are no
> languages that use either.
Fallacy of argument, there. I never asked whether there *are* any; I
asked whether there *could be* any.

Are you willing to really support an actual rebuttal, like that it is
for some reason *incompatible* for someone to use a language that is
(when possible) simultaneously signed and spoken, in an additive way?
(I.e., I don't count that used by interpreters, where their signing
and speech are translations of each other.)

> >* advanced version: carry multiple meanings, some of which are only
> >understandable if you have a different mode of access (e.g., a more
> >advanced version of shaking your head when you're talking, such that
> >hearing-only listeners don't get the "this is false" message)
>
> Why should any language keep its speakers from showing what they think about
> the facticity of what they're uttering? I can't imagine a language where it
> wouldn't be possible to show through intonation (or some other device) that
> you are ironical about what you're saying or very convinced or very sceptic etc.
Yes, but that wouldn't be differential. I can conceive of situations
where this would be useful (e.g. where the message to one part of the
audience needs to be filtered). Nevertheless, the point here is only
that current languages are not capable of doing this (except in an ad
hoc manner I described above, which you'll get in any movie with an
"we've been bugged" scene)... and that it is probably possible to do
it better.

> I don't understand these two (they sound pretty much like computerese to me).
> >* corollary: not overspecified (e.g., ubiquitous gender when wasteful
> >of space or otherwise undesired)
E.g., Romance languages have ubiquitous gender. You can't *avoid*
having to spend (some) semantic space on it, on an ongoing basis, even
when it's already quite obvious and adds nothing to the conversation.
I say you should only specify as much information as needed; that
semantic "space" could be used for something more valuable.

> >* runtime encryption (if wanted)
Ability to say something so that only those who also know your "key"
would be able to understand the content; people who don't, even if
they're fluent and know your method of encryption, would not.

It would take some work to conceive of a good way to do this that can
be done "runtime" - i.e., while you're talking - by humans, and isn't
as simplistic as a variant of Pig Latin (which doesn't meet the second
criterion above). I think it's possible.

 - Sai

Reply