On Thu, 19 Nov 2009 20:21:58 +0100, Benct Philip Jonsson wrote:

> This really veered out on a tangent...
> On 2009-11-17 Daniel Bowman wrote:
>  > To all you artlangers out there:  how do you use ambiguity?
>  > What does it say about your language?  Have you used it to
>  > bring out certain aspects of the language's artistic
>  > expression?  Is it a stumbling block, or an opportunity?
> Ambiguity, homonymity, synonymity and the like are
> not only tolerable but desirable in my artlangs.
> In fact I only wish I were better at creating
> them! This is because all natlangs contain a
> degree of ambiguity.

Indeed.  I sometimes feel that there simply is not enough ambiguity
and irregularity in Old Albic, and I am going to retouch some points
in order to give the language more depth.  Currently, I am researching
the prehistory of Proto-Indo-European in order to reconstruct a common
ancestor of Albic and Indo-European which I shall use to give Albic a
deep linguistic history running beyond Proto-Albic itself.  (And I will
be able to branch off even more languages later!)

> [...]
> Of course mileage may vary as to what weight
> artlanger gives to natlang precedent. My artlangs
> are meant to be 'naturalistically plausible', i.e.
> they should have a structure which could plausibly
> exist in a Terran, human natlang.

Same to me!  As you probably know, Albic is a family of "lostlangs",
i.e. languages spoken in a version of Earth differing from real Earth
only in having a few extra languages which are assumed to have been
lost in the real world.  The native speakers of the Albic languages
are humans just like you and me, so their languages ought to have
structures which could plausibly exist in human natlangs here on

>       (The possible 
> exception would be my Sohlob conlang family whose
> speakers are possibly not human -- but as long as
> I have not decided or not 'found out' what they
> are like biologically I play on the assumption
> that Sohlob should also be naturalistically
> plausible).

Some conlangers exploit the non-humanness of their conlang's speakers
as a cheap excuse for linguistic implausibility.  An example is
Klingon, where the implausibility of its phoneme inventory is often
brushed off with the words "The Klingons aren't human!"  If they are
as much like humans as they are portrayed in the _Star Trek_ TV series,
one would expect their languages to follow similar constraints as human 
languages do.

Of course, non-human speakers open up new avenues for conlanging.
Examples include Dennis M. Moskowitz's Rikchik, who use a visual
language based on tentacle gestures, or Jeffrey Henning's Fith,
whose grammar is grounded in the concept of the LIFO stack and
allows for syntactic transformations far beyond anything found
in human languages.  That is alienlanging as it ought to be.

> I think that in order to be meaningful any game or
> art (and artlanging is both) must have rules --
> that is constraints -- on what you can and cannot
> do, since otherwise you cannot evaluate it. 

Absolutely.  Art always involves a decision what to do and what
not to do.  This is the reason why many artists write manifestos
in which they lay down by which principles they create their art.
Without a set of rules and constraints, art degenerates into chaos.
Of course, there is no one true way of conlanging, or merely of
artlanging; but each of us has his ideas on what a good conlang is,
and attempts to build languages that live up to them.

Most of my artlangs are lostlangs.  This first of all means that
these languages must be perfectly plausible as human natural
languages, and what is more, they must fit into the locations where
I imagine them to be spoken.  This means diachronic and typological
explorations: what kind of language could plausibly be spoken in
that location?

Consider Old Albic.  Assuming that there was a great civilization
in the British Isles around 600 BC.  What kind of language could
these people have spoken?  Probably not an isolating, monosyllabic,
tonal language: that's for South-East Asia.  Probably not a radically 
head-marking polysynthetic language with switch reference and fourty-
something consonants: that's for the Pacific Northwest.  Probably not 
something with clicks: that's for southern Africa.  And so on. 
So here are some "inputs" for the Albic project:

1. The language may have been the substratum responsible for the
   typological aberrancy of Insular Celtic among the Indo-European
   languages.  Thus I get VSO word order and a tendency towards
   running syntactically closely associated words together in speech,
   thus later causing initial mutations to crop up, etc.

2. The British Isles take part in the "Old European hydronymy", a
   network of recurring river names spanning much of Europe, which
   look like coming from a language related to Indo-European, but
   branching off before the PIE ablaut alternations developed.
   So there is the idea of having Albic be a sister of IE.

3. And while we are at a sister of PIE: there is some evidence that
   PIE evolved from an active/stative language.  I always wanted to
   do just that - bingo!  Old Albic would be active/stative, with
   the system of which we find traces in IE still intact.

4. Such a language would have left words in Celtic and probably also
   in Germanic; hence, I went to the library, checked out Celtic and
   Germanic etymological dictionaries, and picked all the words that
   have no good IE etymologies.  And there are a few hundred words
   for Old Albic.

5. What were pre-IE languages of Europe like typologically?  Ask those
   that survived: Basque and the Caucasian languages.  That is what my
   language ought to look like then.

With Germanech, my second project, the rules are much simpler: take
Vulgar Latin and apply the sound changes of German to it, as well as
possible.  (Applying the sound changes of one language to another
language always requires some creative adaptations, as no two
languages have the same phonology!)  Make sure that the result is a
plausible Romance language of western Germany.

> [...] My artlanging is mostly concoction, but the
> gratification lies in exploration rather than in
> contraption[^5].

Same with me.  With Albic, I am *exploring* something that feels as if
it had always been there and only waited for me to discover it.  I could
have it easier, e.g. by using a random vocabulary generator or just
drawing morphemes out of thin air.  No, that is not for me.  I want
something that feels *right*.  It is more like *reconstructing* a lost
reality than making something new.  It is different, however, with such
projects as Quetech, the speedtalk engelang.  That's just something I
construct anew, not something which feels as if it has always been there
waiting for me to explore it.  (And naturalism is not an issue.  I know
that the beast is utterly unnatural.)

... brought to you by the Weeping Elf