On Thu, Aug 7, 2008 at 3:26 PM, Jörg Rhiemeier <[log in to unmask]> wrote:

> A friend of a friend of mine once said, "Art is when someone says
> 'Now'."  There is a point in most works of art when the artist
> declares it finished and refrains from further additions.  Art
> always means to choose what to do and what NOT to do.

Without disagreeing with you and Ray that Andrew should probably
refrain from major changes in Brithenig, I wonder whether the saying
you quoted really applies to artlangs in general, and whether, if it
doesn't, that has anything to say about how conlanging differs
fundamentally from other artforms.

I suppose many of us who work one one conlang long-term, whether
engelang or artlang or even auxlang,will never consider it really
finished.  One of our ideal goals, which we can never reach in a
single lifetime, is to make it as expressive as the typical natlang.
Unless we start out with a severe degree of simplicity in the grammar
as one of the design goals, we may never be finished with the grammar,
and unless we start out with oligosynthesis or oligoisolation as a
design principle, we'll never be finished with the vocabulary.  The
open-endedness of our goal of maximum expressivity doesn't necessarily
mean we keep tinkering with existing parts of the language; but it
does mean we can never say it's finished in the way a novelist or
painter says he's finished with a novel or painting.  There's always
another species of beetle you haven't got a name for yet.

Is there any equivalent in other art forms to the state Brithenig is

>> Grammatically I consider Brithenig a closed canon.  Lexically I'm still
>> researching words and phrases for translation exercises.

Novelists, for instance, occasionally do second editions of their
novels (James Branch Cabell, J.R.R. Tolkien); but I've never heard of
one saying something equivalent to this, like "I may keep making
slight additions to the worldbuilding detail until I die, but the plot
and characters aren't going to change".  That doesn't mean that
conlanging is a better or worse kind of artform than literature (let's
not open that can of worms again) but it may tell us something about
how they differ.

To return to another point,

> Art always means to choose what to do and what NOT to do.

I agree that applies to artlanging (and engelanging insofar as it can
be an art or craft) as much as to any other kind of art, even though I
disagree that (baring certain unusual goals or design principles) an
artlang should at some point be considered finished and left in an
unchanging state thereafter.  There can be art in the ways you
deliberately choose to leave certain grammatical capabilities out of a
language or in what concepts you deliberately choose not to
lexicalize.  The term "kitchen sink conlang" was coined for a good

Another point: it seems that many conlangers think of their conlangs
primarily in terms of *langue* rather than *parole*.  I may be in a
minority in focusing on *parole* -- more and more in recent years, I
think.  gzb is stabilizing, I'm becoming semi-fluent in it, and its
corpus is growing, so there's more scope to observe and describe my
usage and less scope to creatively make arbitrary, unconstrained
design decisions about how it's going to work.  But my last few
sketchlangs have also been *parole*-oriented, almost all my notes on
them consisting of sentences in the language with general (not
interlinear) glosses, having little or no lexicon or analytical
description of grammar.  Even with my engelang project säb zjeda,
there is more grammar implied by the corpus than what is explicitly
described in the grammar documentation.  It may be that this
orientation is part of why I see artlanging as an inherently
open-ended process; the *langue* may at some point seem to be as complete
and expressive as necessary, but as long as *parole* is going on, it's
going to change the *langue* in various subtle ways.

Every natlang is a collaborative conlang in a sense, each speaker's
*parole* acts of speech and writing constantly modifying the *langue*
structures in their own and other speakers' brains, which influence
the form future *parole* will take, which determine the *langue*
structures that will eventually form in the brains of speakers yet
unborn.  And it seems to me that something similar may take place on a
smaller scale in the interaction between a lone artlanger and his
conlang, if he works on it intensely enough for long enough, even if
not enough to become truly fluent in it.  Once you get that kind of
positive feedback loop going, why would you ever want to stop?

1. -- I Googled the term and got only 5 hits; I'd thought it was more
commonly used than that.  The earliest datable uses were from 2006.
I thought I'd seen it used before then.

Jim Henry
Conlang fluency survey -- there's still time to participate before
I analyze the results and write the article