Hallo conlangers!

On 07.06.2016 18:59, And Rosta wrote:

> On 6 June 2016 at 15:37, Jörg Rhiemeier <[log in to unmask]> wrote:
>
>> I haven't yet seen an engelang that resembles a natlang. All of those I
>> have seen had "unnatural" features; none had any notion of diachronic
>> development. A "naturalistic engelang" seems a contradiction in terms to me.
>>
> As we know, conlangs, tho they come in many kinds, tend to cluster into
> those that closely emulate natlangs (even if not any specific natlangs) and
> those that more plainly follow some transparent structural design, and
> loosely we label the former artlangs and the latter engelangs.

This is not really how the terms are usually defined.

Artlangs are conlangs that serve an an artistic purpose. Most of these 
are fictional languages, but not all. Some are personal exercises in 
linguistic beauty, or just made for the enjoyment of their creator. And 
not all artlangs emulate natlangs.

Engelangs are conlangs that are designed to meet an objectively testable 
design goal. The main subgroups (not exhaustive) are logical languages, 
"philosophical" languages and experimental languages.

The two categories "artlang" and "engelang", however, are not mutually 
exclusive. A conlang may have aspects of both. Neither of the two is 
furthermore mutually exclusive with the category "auxlang". What we get 
here is the Gnoli Triangle.

> But in my view artness and engeness inhere not in the conlang itself but
> rather in the way we judge and assess it as a creation. When we judge and
> assess something, such as a conlang, artwise, we largely accept it as it is
> and form an aesthetic response;

Yep.

> a rational theory of aesthetic judgements
> may or may not be practicable, but in practice among conlangers it boils
> down to De Gustibus.

Also right. Aesthetic judgments are subjective and not amenable to 
objective tests.

> But when we judge and assess something engewise, we
> form a relatively objective judgement about how well it achieves certain
> goals, whether those goals were declared by the author or merely imputed by
> the observer-consumer.

Fine. The objectively testable goals are what distinguish the engelang 
and set it apart from the artlang.

> Naturalism can be judged engewise.

How? Do you have an algorithm for that? Of course, one could compare a 
conlang against a large number of natlangs or against a typological 
survey such as the World Atlas of Language Structures. There are also 
features of a conlang on which intersubjective consensus could be 
achieved that they are unnatural (such as most self-segregation 
strategies), but the line between "naturalistic" and "obviously 
artificial" is IMHO blurry. I think naturalism is more a way of feeling 
"right" as a hypothetic natlang that is difficult to measure.

My point in saying that I haven't seen a naturalistic engelang is that 
most kinds of language engineering involves the introduction of features 
to conlangs that lie far outside the range of features of natlangs. Is 
there a natlang that implements predicate calculus? Not to my knowledge. 
Is there a natlang with a taxonomic vocabulary? Not to my knowledge. Is 
there an oligosynthetic natlang? Examples have been claimed, but 
disproven. Is there a natlang that is organized around a stack, like 
some programming languages? Almost certainly not, because humans could 
not use it in real time.

> Now, many
> conlangers produce conlangs that emulate natlangs, without really caring
> how naturalistic they are, but many do aim to emulate natlangs. A very
> extreme case is Britainese which is maximally naturalistic, not in the
> least intended as an artlang, and entirely intended as an engelang.

I woudn't absolutize it as that. I still think Bretainese is an artlang. 
How do we objectively test whether it is actually what a Romance 
language of Britain would look like? There is no way knowing. Sure, the 
notion "an attempt at what a British Romance language would look like if 
Vulgar Latin had replaced Celtic in Britain" is perhaps not an aesthetic 
one by a narrow definition, but I don't think it is objectively 
testable. It is a rendition of an imagined reality, and therefore it is 
*art*. Britainese is certainly a fictional language and therefore an 
artlang.

> And if
> you look in particular at what Alex has to say about conlangs, he is almost
> always judging them engewise by the criterion of 'statistical' naturalism.;
> Alex's predilective approach to conlangs is strongly to view them as
> naturalistic engelangs.

Alex is one conlanger, and one whose views not everybody agrees with; 
how is his opinion an objective test?

> The reason this art--enge distinction is little articulated in our culture
> -- us conlangers (or at least, me) have had to articulate it for ourselves
> -- is, I think, that no other domain of creative endeavour lends itself so
> strongly and so equally to the two approaches.

In most creative endeavours, art and engineering are more clearly 
separated than in conlangs, this is true. An artistic drawing is 
something else than an engineer's drawing; there are plenty of 
draughtsmen who excel at one and suck at the other, or vice versa.

--
... 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