Hallo!

On Mon, 29 Oct 2007 07:47:41 +0000, R A Brown wrote:

> Henrik Theiling wrote:
> > Hi!
> > 
> > Some things got mixed up here . 
> 
> It would seem so to me also.

Yep.

> > The design goals may be clear to me,
> > but I seem to present them in a confusing way.  Sorry for my fuzzy
> > explanations.
> > 
> > Let's see.
> > 
> > Jörg Rhiemeier writes:
> > 
> >>>Henrik Theiling wrote:
> [snip]
> >>>>Yeah.  My goal of a plausibly Romance diachronical fauxlang that I
> >>>>like is a bit hard to explain. :-)
> >>
> >>The problem I see with Terkunan is that it tries to be two things
> >>at once: a diachronic Romance language, *and* a fauxlang with an
> >>isolating grammar.  You cannot have both at once, I think, and end
> >>up with something which is neither.
> >  
> > But I am not trying to do two things at once.  

> Auxlangs like Volapük, Esperanto, Latino sine Flexione, Novial, Glosa 
> etc, etc (or even a fauxlang like TAKE) do _not_ have diachronic 
> development over the centuries. They are the creation of individuals or, 
> occasionally, a committee or group of individuals. It seems to me a 
> _diachronic_ fauxlang is a bit of a contradiction.

Precisely that is the problem.

> I presented the design
> > goals and said that the historical explanation is secondary at the
> > moment.  
> 
> My own feeling is that if Terkunan is a fauxlang, then the historical 
> explanation of diachronic development should be ditched.

Yes - and if Terkunan is a diachronic language, the root selection
methods you present in your orange boxes should be ditched and
*the Latin words themselves* subjected to the GMP, not the
artificially extracted roots - the latter is simply *not the way
natural languages change*.  Sound changes affect *whole words*,
often even larger units (as in the initial mutations of Insular
Celtic, which resulted from sound changes operating across word
boundaries).

> [snip]
> >>>>The diachronical development is retro-fitted and secondary.
> >>>
> >>>That would seem to me to be making life difficult   :)
> >>
> >>AMEN.  What I see what Henrik is trying to do with Terkunan is
> >>to make a language which has evolved from Vulgar Latin by
> >>naturalistic sound changes, while at the same time appealing
> >>to his taste for perfectly regular, simple engelangs.  And that's
> >>the problem.
> >  
> > I don't see the contradiction, actually.  The GMP is a helpful means
> > to prevent chaos.  It does not contradict the fauxlangish goals of
> > simplicity at all.  It helps me prevent making sloppy mistakes and it
> > constructs the sounds I want.  It is very hard (for me) to do this
> > without machine aid.  If fact, I always hated lexicon construction and
> > wasn't very good at it (and slow) and a GMP in an aposteriori conlang
> > is a great help.
> 
> I agree that the GMP is a helpful means to prevent chaos in the 
> diachronic development of a naturalistic conlang. But I do not see why 
> _simplicity_ per_se has to be fauxlangish. Surely pidgins and creoles 
> have similar simplicity.

The problem, Henrik, is that you appliy the GMP to *artificially
extracted roots* - units that would *never* exist for themselves
in the natural evolution of the language.

> > Also, Terkunan is *not* meant to be an engelang and I am *not* using
> > engelang design goals at all.  The simplicity comes from a fauxlang
> > point of view.  Engelang would mean that I'd question and reconstruct
> > the whole tense/aspect/mood/case/whatevercategory system altogether.
> > Instead, I want an isolating language with typical categories from
> > Romance and in the best case, with analytical structures typical for
> > Romance.
> 
> No European Romancelang has particularly analytic verb structures. 
> _Spoken_ French does best in this respect, tho the written language 
> retains quite a bit of synthetic apparatus.

And spoken French arrived there by sound changes which eroded the
word endings, not by "selecting" roots that didn't exist as
free-standing words *at any point of the language's history*.

>       But Romance based creoles do  
> surely show the simplicity that you are aiming for, and those structures 
> have derived from Romance itself.

Yes.

> [snip]
> > 
> > The construction of a conlang is different from that of a natlang,
> > yes.  It always is: for a conlang, there is a person with design
> > goals, even in diachronic conlanging.  But even if my grammar
> > structures seem implausible, it does not mean they are impossible
> 
> Nothing is impossible in a conlang  :)

Sure, but: your page states that Terkunan was a language which evolved
from Vulgar Latin naturally, i. e. by regular sound changes.  That,
however, is not really the way the language is constructed - the
sound changes are not applied to actual Latin words, but to artificial
units.  Hence, Terkunan falls short of what it is meant to be.

> [snip]
> > 
> > If this conlang really bothers you (and others?), I could have a poll
> > about whether it's so disturbing that I should better remove it from
> > the Internet. 
> 
> No, that is not, I think, a good idea. I have to admit that I share much 
> the same misgivings as Jörg does and we are, I hope, trying to be helpful.

Yes.  That is what Ray and I are trying to do - help you create a more
plausible diachronic conlang.

> > The primary primary meta goal for me is to have fun, of
> > course.  
> 
> Absolutely.
> 
> > And even if I don't seem to succeed in explaining how the
> > design goals fit together, they make perfect sense to me.
> 
> My own personally feeling is that it would be better to drop the 
> fauxlang idea and to seek the simplicity and isolating structure you 
> desire by designing Terkunan as a Romance-based creole *there*.

Yes.  A creole is one solution; sound changes that erode the inflections
are another.  But the current way of constructing Terkunan words is
utterly unnatural.

On Mon, 29 Oct 2007 11:33:30 +0100, Henrik Theiling wrote:

> Hi!
> 
> R A Brown <[log in to unmask]> writes:
> >...
> >> Some things got mixed up here .
> >
> > It would seem so to me also.
> 
> :-)
> 
> >...
> > I agree that the GMP is a helpful means to prevent chaos in the
> > diachronic development of a naturalistic conlang. But I do not see why
> > _simplicity_ per_se has to be fauxlangish. Surely pidgins and creoles
> > have similar simplicity.
> 
> Ok, this seems to be the core of the confusion: that I call it a
> fauxlang giving rise to false expectations.  I did not notice this
> problem before, so thanks for the fine analysis!

Yes.  Either it is a fictional natlang - then the GMP has to be
applied to full word forms and not artificially extracted roots -
or it is a fictional auxlang - then one would not expect diachronic
changes.  It can't be both.

> >...
> >> And even if I don't seem to succeed in explaining how the
> >> design goals fit together, they make perfect sense to me.
> >
> > My own personally feeling is that it would be better to drop the
> > fauxlang idea and to seek the simplicity and isolating structure you
> > desire by designing Terkunan as a Romance-based creole *there*.
> 
> Ok.  I won't call it a fauxlang anymore.
> 
> But 'creole' would also give rise to expectations I currently don't
> have, e.g., maybe more loans (if you want to call them loans in a
> creole when the words where there from the start).  I will call it a
> personal language until I find something more specific that suits the
> goals.

Yes, you can of course whatever odd rules you want in a personal
language.  But as it now is, Terkunan is *not a plausible diachronic
language*, because the sound changes are applied in an unnatural way.

... brought to you by the Weeping Elf