Re: Introducing myself, and several questions

Hallo!

On Mon, 14 Feb 2005 00:01:43 -0500,
Damian Yerrick <tepples@...> wrote:

> My name is Damian, and I'm a conlanger.
Welcome!

>       I've dabbled for
> years, never "finishing" anything to the point that arbitrary
> conversation is possible.
Well, only few people ever get to that point in conlanging.
Most conlangs remain incomplete.

>       I come to this list to ask for help
> in getting past roadblocks.  I've read some of the archives,
> but not all 6 1/2 years of them.  Here's where I want help:
>
>
> LEXICAL ICONICITY
>
> When creating the a priori lexicon for Qenya (early drafts of
> Quenya), Tolkien chose sound patterns that he felt "fit" a given
> meaning.
> http://www.uib.no/People/hnohf/vice.htm
>
> However, I seem to have a dulled sense of aesthetics, possibly
> caused by my Asperger syndrome that causes me to distrust vague
> hunches.  Much of the time, I can't seem to do better than creating
> phonotactic rules and then randomly assigning Swadesh-list glosses
> to sound patterns, possibly with the aid of a computer program.
> Are there some general procedures that govern lexical innovation
> in natlangs and naturalistic conlangs?  Has anybody successfully
> implemented ding-dong or ta-ta in their conlangs?
The relationship between sound and meaning in languages is generally
arbitrary, with only few words approaching iconicity.  And aesthetics
is a very subjective issue.  Beauty is in the eye of the beholder.
I find naturalistic conlangs (i.e., conlangs that look like natlangs,
with a sense of historicity) beautiful and conlangs that give away
their artificiality at first sight ugly, but there are people around
here who have a different taste.

There have been attempts to create conlangs that avoid the
arbitrariness inherent in natlangs.  They are called "philsophical
languages", and they were failures.  There is no such thing as
an "iconic language".

> DEFAULT SETTINGS OF GRAMMAR
>
> I don't want to make euroclones all the time, but I don't want to
> make an unspeakable language that violates fifty-two universals
> either.  What structures are "easier" for the developing hominid
> brain to parse?  For example, do learners intrinsically prefer
> object-verb order or verb-object order?  What about adjective-noun
> or noun-adjective?  Is there any appeal to iconicity for this?
The variation among human languages is enormous, and it is hard
to say whether left-branching (object-verb, adj-noun, postpositions)
or right-branching (verb-object, noun-adj, prepositions) structures
are easier to parse.

> CULTURAL-PHONETIC CORRELATION
>
> Does tendency for open or closed syllables, for softer or harder
> sounds, or for tones or no tones, depend on culture?  I've heard
> of the Inuit and the Arabs, whose languages have fewer distinct
> vowel heights and more back consonants because their harsh
> environments make it painful to open the mouth to the elements
> in order to produce low vowels.
Few linguists would subscribe to that.  There seems not to be
any correlation between culture and phonology.

>        In addition, Tolkien's chaotic
> orcs speak a phonaesthetically "harsher" language than his
> lawful elves.  Is such correlation the rule or the exception?
Again, it's subjective.  Tolkien decided that the good guys in
his story would speak languages he'd consider beautiful, and the
bad guys languages he'd consider ugly.  The next author will have
different ideas about what is beautiful, and build his languages
accordingly.

> CULTURAL-GRAMMATIC CORRELATION
>
> Likewise, are any grammatical qualities correlated to aspects
> of the culture?  Does an environmental or cultural constraint
> correlate with an OV or VO preference, with obligate marking
> of various properties of a noun or verb, or anything similar?
> I can see how a more paranoid culture might lead to evidentiary
> markers becoming grammaticalized; are there other examples?
There is perhaps more of a correlation between culture and grammar
than between culture and phonology.  A rigidly stratified culture
is perhaps more likely top develop an elaborate system of honorifics
than an egalitarian one, for example.

> SIMPLIFICATION
>
> I understand that the lexicon can be reduced to sizes that
> may initially appear absurd while retaining expressiveness.
> Evidence: A conlang called Toki Pona manages to convey every
> meaning one can think of in 120 basic words.
Closed vocabularies don't really work.  You will find that the
words that aren't there must be replaced by circumlocutions
which must be learned individually just like words.  No natlang
has a closed vocabulary.

> Is this true of grammar as well?  For instance, in computing, the
> problem called 2-SAT is not NP-complete, but 3-SAT is NP-complete.
> Does this result have an analog in human language?  Is it possible
> to make a fully expressive language that uses two-word clauses?
I don't know, but I doubt it.

> Specifically, is the narrator's description of the language of
> the Eloi in chapter 5 of HG Wells's _The_Time_Machine_ unnatural?
>
> "Either I missed some subtle point or their language was
> excessively simple - almost exclusively composed of concrete
> substantives and verbs. There seemed to be few, if any, abstract
> terms, or little use of figurative language. Their sentences
> were usually simple and of two words, and I failed to convey
> or understand any but the simplest propositions."
Sounds like a pidgin.

> (N.B.: The description doesn't match movie Eloi by John Logan.
> But then little in the book matched the movie.)
This is a recurrent problem with movies based on books ;-)

> FURTHER READING
>
> When I search for some of these topics, Google often gives me
> results that look promising but say "Download this article for $30".
> Once Google fails me for gratis web resources, and my local public
> library's search engine fails me for print resources, what are some
> good resources for learning about these subjects without spending
> $500 on buying books and buying individual PDF article downloads?
> Or is conlanging a rich man's hobby?
It is not.  There is plenty of good free stuff on the Net.
For a good start, read the Language Construction Kit, which tells
you how to build a conlang and addresses most of your questions:

http://www.zompist.com/kit.html

Greetings,

Jörg.

Reply