Re: creating words (was Re: "Language Creation" in your conlang)

Quoting JS Bangs <jaspax@...>:

> Andreas Johansson sikyal:
>
> > > I think you can get away with leaving it incomplete.  Come up with a
> > > basic phonoloyg, some gramatical suffixes (or have the protolang be
> > > isolating, depending on the present lang), and vocab.  Syntax, et al.
> > > can be pretty much ignored, except for some basic issues like
> > > adjective-noun order (for determining compound words), since syntax
> > > tends to be pretty changeable over time.
> >
> > I find even that level of completeness hard to achieve when what I really
> want
> > to do is elaborating the "modern" language. Conjuring up an ancestral
> > phonology and phonotaxis with some grammatical markers isn't that hard,
> but
> > making a large number of ancestral roots before I start the (already
> painfully
> > slow) process of creating vocabulary for the modern language(s) is.
> >
> > So what happens is I invent a "modern", usually Tairezazh, word, work out
> a
> > possible ancestral form, and then goes forward again to the sister
> languages.
>
> What's wrong with that? No one said that you had to *complete* the parent
> language before making the daughter languages, you just have to have some
> idea of what it is.
No-one's said it's anything wrong with it.

It's just that the thought of constructing a language family in the "right"
sequence is appealing in theory, I find I cannot do it in practice.

> What you described is, in practice, what I usually do
> with Yivrian. So long as the roots you create conform to the ancient
> phonotactics, you're fine.
Since the ancestral phonotactics where originally arrived at by running
Tairezazh phonology backwards thru various sound-changes, that's not
particularly hard!

                                                 Andreas