Re: building from primitives (was Re: Langauge Constets)
| From: | Jörg Rhiemeier <joerg_rhiemeier@...> |
|---|
| Date: | Thursday, November 29, 2007, 15:51 |
|---|
Hallo!
On Wed, 28 Nov 2007 15:43:00 -0800, Gary Shannon wrote:
> --- Jörg Rhiemeier <joerg_rhiemeier@...> wrote:
>
> > Hallo!
> >
> > On Wed, 28 Nov 2007 18:45:33 +0200, Benct Philip Jonsson wrote:
> >
> > > While I agree with Jörg about the essential dangers of making people
> > > think like computers I can sympathize with the desire for a limited
> > > vocabulary.
> >
> > The problem I see with such schemes is that the world is way too
> > complex. How do you say 'spaghetti' or 'kimono', or 'quantum
> > chromodynamics' or 'morphosyntactic aligment', in an oligosynthetic
> > conlang?
>
> When I think of "building from primitives", and what my original intention
> was
> when I first brought up the subject, was to have a basic vocabulary of 1000
> or
> so words that would be used to DEFINE new words. It was not my intention
> that
> new words be derived from, or built up from the basic words, only that they
> be
> definable in terms of the original core vocabulary so that once a new
> learner
> learned the core vocabulary he or she could use the dictionary to learn
> learn
> new words without recourse to his or her native language.
>
> The new words themselves, beyond the core vocabulary, could be loans from
> other
> languages or complete neologisms created by pulling Scrabble tiles out of a
> hat.
I see. That is of course something else than restricting oneself to
a small number of lexical roots that cannot be expanded. Using a
well-chosen limited set of lexemes for definition purposes may indeed
be quite useful.
On Thu, 29 Nov 2007 13:06:24 +0100, Henrik Theiling wrote:
> Hi!
>
> Jörg Rhiemeier writes:
> > The problem I see with such schemes is that the world is way too
> > complex. How do you say 'spaghetti' or 'kimono', or 'quantum
> > chromodynamics' or 'morphosyntactic aligment', in an oligosynthetic
> > conlang?
>
> Easily done: you lexicalise a compound or derivation (with a
> specilised or even slightly adjusted meaning). Start with translating
> those loans -- this makes them easy to grasp in English for a start.
> Then try to find a mapping to not too many lexemes in your conlang and
> lexicalise the result. Done. If you allow portemanteaus, you can
> further shorted long words.
Sure, but this tends to produce clumsy circumlocutions - unless your
morphemes are *very* short, you are likely to end up with very long
words. And with compounds, you easily run into the "black bird vs.
blackbird" sort of problems.
> One mis-conception of Toki Pona that is often spread and repeated for
> some reason is that it has only 118 words. It only has 118 roots plus
> a lexicon of compounds. Look at Sonjas page -- both are given there,
> the lexicon being quite small yet, but the idea is quite clear and I
> hear the community has more words they mostly agree on.
Yes. TP has 118 *roots*, not 118 *words*.
> If there wasn't another thing about Toki Pona, namely that the creator
> does not *want* it to be used for, say, scientific conversation, then
> you could very probably make it an all-purpose language. Well, ok,
> you might want to think about whether the selected 118 roots are a
> good choice to start with for an all purpose lang, but the mere number
> is not a problem. Some natural polysynthetic languages do not have
> that many more roots (e.g. Kalaallisut, Náhuatl or Blackfoot IIRC),
> although for a direct comparison I suppose they do have too many.
Indeed, some natlangs get along quite well with a few hundred roots;
I don't know how much they borrow from other languages these days
in order to cope with the multifarious new concepts of the modern
world. I remember reading somewhere that in some Native American
languages, human/animal body part terms are used to refer to the
parts of a car, as in "car's leg" for 'wheel'. Yet, things such
as 'desoxyribonucleic acid' are difficult to circumscribe this way,
I think, and native-material compound words probably end up being
very, very long.
Sure. Auxlang advocacy is a can of wyrms better left unopened here,
and it wasn't my intention to open it. I just wanted to point at
something I have said before on the taxonomic/oligosynthetic
language matter, although in a different context. My point was
that borrowing is often much more convenient than verbose
circumscriptions, which in turn speaks against an oligosynthetic
design, no matter whether the whole thing is to be used as an
auxlang or not.
... brought to you by the Weeping Elf
Reply