Hallo!

On Tue, 11 Nov 2008 13:03:55 -0800, Gary Shannon wrote:

> My pet theory is that natlangs were created by people who knew nothing 
> whatsoever about linguistics, and linguistics is the educated attempt to 
> explain the verbal behavior of the uneducated. Pidgins and Creoles are also 
> the creation of the linguistically naive.    
> 
> Consequently, I have always preferred the notion that conlangs are best 
> created by people who know nothing about theory, or if created by 
> knowledgeable people, they should be created without paying too much 
> attention to theory.   

Indeed, I don't find linguistic theory very useful in working
on my conlangs.  When designing a naturalistic conlang, it is
useful to have a feeling of what occurs in natlangs and what
doesn't, to know some basic grammatical terminology in
order to conveniently describe your language (but keep to the
basics, or people won't understand your grammar), and, if you
are working diachronically, how languages change, but that's
about it. That's "general knowledge" rather than "theory";
the emphasis is on *practice* and *experience*.  Actually,
I know only rather little about linguistic theory, and am not
aligned with any of the current (or deceased) competing
theoretical traditions.

What enabled Tolkien, for instance, to create such masterful
conlangs as Quenya or Sindarin, was not so much knowledge of
linguistic *theory*, but a good knowledge of and feeling for
the diversity of actual *languages*. He drew his inspirations
from languages such as Welsh or Finnish, not from some kind
of theoretical edifice.

> Somewhere in the dim recesses of pre-history, people who had some single 
> utterance for some basic experience like "fire", decided, for some 
> unfathomable reason, to begin inflecting that word with case endings (or 
> maybe case endings started out as post positioned particles?). I'm sure they 
> did so for some simple utilitarian reason, without any deep philosophical or 
> theoretical ponderings.      
> 
> In other words, I tend toward the pragmatic belief that before there was 
> language, there could not have been any theory of language, because theory 
> only describes what the uneducated masses have sort of accidentally gotten 
> in the habit of doing. So naturalistic conlangs should still feel somewhat 
> "accidental".    

That's self-evident, I'd say. Theory without practice is
meaningless. It's an idle mindgame.

On Tue, 11 Nov 2008 14:19:23 -0800, David J. Peterson wrote:

> I have to echo what Gary said: Naturalistic languages should
> look like natural languages, and thus should involve natural
> processes.  With or without theory, one should probably take
> a look at as many natural languages as possible, so as not to
> copy one's own, or to produce something too artificial.  Insofar
> as theory helps one to achieve these ends, it is useful.

Concurred. Studying grammar sketches of dozens of languages
from all over the world is vastly more useful to the
naturalistic conlanger than studying dozens of volumes of
advanced linguistic theory. The former gives you great
insight in the way human languages work, and lots of ideas
to try out in your own languages; the latter may help you
understand academic linguistic treatises - which, however,
usually turn out to be less than useful for conlanging.

> [...]
> 
> This, of course, all has to do with languages that attempt to
> look natural.  When it comes engineered languages, visual
> languages, logical languages, artistic non-natural languages,
> philosophical languages, etc., it seems to me that one has to
> create one's own theory.

Or, at least, a kind of manifesto stating what the author
wishes to achieve with the language.

>        Linguistics is the study of *natural* 
> languages, so when it comes to what one can or cannot create,
> linguistics doesn't have much to say.

Certainly.  Linguistics is about describing and understanding
natural languages; conlanging is about inventing hypothetical
languages.  That's something different.

>       Certainly, one can take  
> claims that are made and test them, but unless they're particularly
> grandiose claims or foundational assumptions, testing them
> won't be that interesting, it seems, since linguistics attempts to
> explain what is, and not what can be imagined.

Yes.  Surely, there are people who make up languages in order
to "test" the Sapir-Whorf hypothesis or whatever, but most
linguists don't take such "experiments" seriously, for good
reasons.  When it comes to describing and explaining human
natural languages, such "experiments" are meaningless.

> Frameworks, though, can be interesting constructs to tool
> around with.  One of the best examples I've seen of this type
> of language is Fith: a language built use LIFO grammar,
> which, in real time, I think is impossible for a human to use.
> There, the framework becomes a self-imposed constraint,
> as opposed to a rule one has to follow just because.

Fith is a nice example of a conlang designed to do something
purportedly different from human languages.  It takes a basic
idea and explores it in full.  But does it take much heed of
linguistic theory?  I don't think so, or at least, its grammar
(fortunately) doesn't delve deeply into theory, and makes the
endeavour perfectly understandable without it.

>       If someone 
> claimed that languages could *only* work if the followed
> LIFO grammar, we'd all think it was a joke (and would
> probably want to know much more about the person who
> claimed it).  The same applies for other frameworks and
> theories.  Why look at any of them as mandates?

Indeed.

... brought to you by the Weeping Elf