Hallo conlangers!

On Monday 08 April 2013 14:23:37 BPJ wrote:

> On 2013-04-07 19:03, Jyri Lehtinen wrote:
> >>> I don't think it's a trap to want all the forms to be distinct. I think
> >> 
> >> it
> >> 
> >>> would be a good thing.
> >> 
> >> It depends on your goals for a conlang.  If you are trying for
> >> naturalistic, forms colliding are not necessarily a big deal, since, as
> >> Padraic pointed out, natural languages often end up with identical
> >> shapes for different conjugations or suffixes.  I would expect that in
> >> most cases, those identical forms would actually merge into one form.
> > 
> > Identical word shapes are also a great source of word plays which in turn
> > give great pleasure for a vast amount of people all around the world.
> > This is actually a good challenge for us all, trying to write puns in
> > our conlangs.
> 
> Agreed.

Agreed by me, too.

>       I actually have worked homonymy into (Proto-)Sohlob
> over the last year.  When I started the lang 10+ years ago I
> avoided homonyms in the the protolanguage (which is now the
> oldest 'attested' daughter language) since I was afraid that
> too many homonyms in the PL would lead to disastrous results
> in the daughter langs

Protolanguages can get away with high regularity as they are
only *models* of lost ancestors of attested languages and can
therefore be *expected* to miss some details.  And there are
some *very* regular natlangs around (Turkish is an example
mentioned often).  My own Proto-Hesperic is also highly
regular with very little (if any) morphological homonymy.

Letting loose sound changes on a protolanguage is the best way
of introducing irregularities into a naturalistic artlang.
After all, the prevailing opinion on irregularity among
historical linguists is that irregularities are residues of
old regular patterns that have been rent apart by language
change.  For example, the synchronically irregular vowel
alternations in English strong verbs date back to a neatly
regular PIE ablaut pattern *e ~ *o ~ *0.

>       -- as if: any natlang would 'fix' that
> with compounding, semantic drift and derivation, and I do so
> now.  There is even a good deal of form overlap between the
> verb 'to be' and demonstratives/deictics/articles (which
> latter are the same words) but the languages work regardless!

Old Albic (a descendant of Proto-Hesperic after about 5,000
years of imagined linguistic change) does not have much
morphological homonymy because some analogous changes have
happened that remedied the worst problems.  For instance,
in Proto-Hesperic 'I' was *mi and 'we' was *mis in the
agentive case, which would both have come out as _mi_ in
Old Albic - which would hardly have been acceptable.  But
most oblique forms of 'I' in Proto-Hesperic have *ma- as the
stem (the result of sound changes in the prehistory of Proto-
Hesperic), and this was generalized on the way from Proto-
Hesperic to Old Albic, so 'I' is _ma_ and 'we' is _mi_. 

> BTW artlangers and engelangers are *really* talking about
> two different things in this thread!  This is a case where
> what's good for the one group is not good for the other,

Sure.

> so one should really not make this kind of statements without
> also noting one's design goals at least in broad terms:
> arty or engy.  Maybe we should have ARTLANG and ENGELANG tags?

I don't think we really need those tags, though they may
sometimes be useful: indeed, artlangers and engelangers
sometimes talk past each other.  But I did not notice much
"talking past each other" in this thread, only statements
that some people have different design goals in their
conlangs than others.  (Also, not all artlangs are meant
to be naturalistc: there are what could be called "abstract
artlangs" which embody a personal idea of beauty which is
independent from resemblance to natlangs.  An example is
Jim Henry's gjâ-zym-byn.)

Naturalism is simply a *non-issue* to engelangers.  They
care about such things as self-segregating morphology,
depending on the project.  While I am not much into
engelanging myself, I find some of the engelang ideas
discussed here fascinating and entertaining.  At least,
they keep one alert of how little of what we find in
natlangs and apply to our naturalistic artlangs can be
taken as self-evident.

What regards the term "naturalistic", it is the auxlangers
and the artlangers who talk past each other, when they
talk with each other at all - they mean different things
when they use the word.  (The auxlangers' usage predates
the artlangers' usage, but the artlangers' usage is also
well-established by now, so one cannot say the artlangers
use the word "wrongly".
  
--
... brought to you by the Weeping Elf
http://www.joerg-rhiemeier.de/Conlang/index.html
"Bêsel asa Éam, a Éam atha cvanthal a cvanth atha Éamal." - SiM 1:1