Print

Print


Muke Tever wrote:
> What _is_ the 'canonical accent' of Esperanto?

Presumably, the way that Zamenhoff (sp?) described it.  For instance,
using a trill for "r", using combinations like /kn/ (rather than [k@n])
and the like.

> Is that on the web anywhere? (other than the list archives of course)

Yes, on Irina's page.  But, I don't have the URL handy.

> That ain't necessarily so.  Weren't there some culture somewhere that
> allowed (or required) leaving part of the farm produce for people like that?

Certainly, more than one I'm sure.  But the one that comes to mind is
the Jews.  Biblical law required that part of the grain be left after
harvest for the poor.  It's a major part of the storyline of the Book of
Ruth, as well as being mentioned in the gospels, e.g. Matthew 12:1, "At
that time Jesus went though the grainfields on the sabbath; his
disciples were hungry, and they began to pluck heads of grain and to
eat", the Pharisees, of course, attack them for doing that on the
Sabbath.

> I found a (nother) tool to do this just today, at zompist.com --a different
> system than langmaker's, it lets you define sets to apply changes to...

An even better, IMHO, is Phono.  You have three files, one with all the
sounds, each defined as a large number of binary variables (for
instance, +cons -syllabic +voice +labial -dental ...), one file that has
rules (for instance, voiceless stops (+stop -voice) become voiced
(+voice) between vowels, and finally a third file that puts those rules
in order, so that it's very easy to rearrange rules or remove rules from
acting, without losing the rule itself (e.g., if you want to test the
consequences of taking out a specific rule).  The only major problem is
that you can't tell it to, say, ignore the first 20 rules, which would
come in handy for borrowings that occured before the final stage but
after the initial stage.  You can simply create a backup of the order
file, and then delete the first 20 rules, but it's a bit of a hassle.

There's also a function wherein you can compare predicted forms [derived
from ancestreal forms] with actual forms.  E.g., say in Old Speak you
had "ma'aste", and by the sound changes you've programed it should come
out with "Ma:te" in New Speak, but the New Speak form is "Mate", it
would indicate a bad pair.  You can have long lists of such ancestreal
forms and descendant forms, useful for figuring out sound changes in
natlangs, by indicating how many errors you have, if any.  It comes with
files for deriving Spanish from Latin, as well as a file for Pig Latin
(Just for fun, I guess)

Anyways, Phono is at http://www.siu.edu/~nmc/phono.html

There are also a whole bunch of computing tools for linguistics at
http://www.sil.org/linguistics/computing.html

> That is, in Langmaker (so far as I know) every change has to be done
> explicitly:

Langmaker's good for creating words, but I don't think that the
transformation utility is all that useful.

> Hmm... a kind of wizard setup to create a kind of default grammar and
> something to start with... [sound of grinding gears turning inside a skull]

To me, that would take all the fun out of conlanging.

--
"If the stars should appear one night in a thousand years, how would men
believe and adore, and preserve for many generations the remembrance of
the city of God!" - Ralph Waldo Emerson
ICQ: 18656696
AIM Screen-Name: NikTailor