Re: conlang maturation

Hi, Tom.

I've got four long grammars.  One is the euroclone that I've just posted.
Two need sample texts and glossaries.   The fourth is even more of a work in
progress.

It's interesting that you should speak of moving in the direction of longer
grammars.   After I finish my big four, I intend to confine my conlanging
activity to short grammatical sketches.   I need to figure out ways to be
more concise!

How do my grammars grow?

I start with an idea either about the theme of the language, or about the
core grammar.   (e.g.  Let's make it look kinda Germanic.    Let's have
clause-initial particles give us the mood and syntactic information.   Let's
have obligatory instrumental complements in basic sentences.   Let's make
the language simple yet exotic to English speakers.)

From there, the language grows section by section:

First, I work on sounds and spellings until I've got a straightforwardly
phonemic spelling system and a simple set of phonotactic rules.

Other conlangers, especially the linguists in this group, often give
phonology a much more thorough treatment than I ever do.   I've seen conlang
grammars with "historical" sound changes and quite a few sound-change rules,
like vowel harmony schemes, for instance.

After I decide on the spellings, sounds, phonotactics, & stress, I sketch
the core grammar, and use that as a basis for chapters on the parts of
speech, (nouns, verbs, etc.).    Making up the core grammar before
specifying the word classes is a good idea.   That way, you don't do things
like write whole chapters on adjectives and prepositions, only to find out
that you'd rather not use those parts of speech in your language.

After I flesh out the parts-of-speech chapters, I describe the core grammar
in as much detail as my limited patience, knowledge, and talent will permit,
and then do the same for coordination and subordination.   (I tried to
create a language with no sentence embedding once, but failed.  I don't know
enough about verb cross-referencing systems and other markers that can
obviate embedding.)

I've always found that writing a sample-text makes it necessary to revise
the grammar at least a little.

Writing sample texts also helps me decide what vocabulary to use.   I
created a sample text called "The Strange Child."   It's not much of a
story, but it's loaded with preschool vocabulary.  AFAIK, there's no such
thing as a definitive list of basic concepts, but you can cook up your own
practical list of elementary words by consulting preschool picture books,
lists of first words, and kindergarten picture dictionaries.   If your
conlang culture is alien, complete with alien techology, flora, and fauna,
you'll REALLY have to cook your list up from scratch, but it still might be
helpful to think about alien toddlers/hatchlings when you make your list of
elementary words.

About lexicons:   others on this list do a much more meticulous job at
creating lexicons than I do.   The first thing I do to develop vocabulary is
write sample texts in English.   Then I translate them into my language
using what I call dummy words:   English words marked with an asterisk.
So, for example, "I jumped." in the Goesk sample text became "Ik jump*eus."
After writing the entire sample text in this fashion, I used the computer's
"replace" feature to convert all instances of "jump*" to "husp."   So we
wind up with "Ik huspeus."

With this method, there's a danger of having the lexicon look too English,
but that danger isn't as great as you might think.   e.g.

If your language makes extensive use of derivational morphology, as many
conlangs do, "hop," "jump," and "prodigiously leap" in English can be
rendered as the dummy words "jump*-diminutive*," "jump*," and
"jump*augmentative*" in your prelimary sample text.

If different senses of a word are rendered as different words in your
language, the dummy words can be marked to reflect this, as in these
examples:   "jump-a*" (referring to a leap on a level surface), "jump-b*"
(referring to a leap from a height to a low point), and "jump-c*" (referring
to a leap from a low point to a high point).

If different syntactic environments change the translation of your word, the
dummy word doesn't change.   The dummy word "see*" won't change in my sample
text for Mno, even though it translates as "see" in one sentence type, "look
at," in another, "come into view" in another, and "look (like) in another.

After I've written sample texts with dummy words, I use Jeff Henning's
Language Maker software to create list of word-forms that conforms to my
language's simple phonology.   Then I assign word forms to dummy words,
substitute reasonably clear glosses for the dummy words, and thus create the
glossary.

Even so, I admit that my method is rough-and-ready.   Developing the lexicon
as you're doing the language and writing the sample text may be more
conducive to the development of truly exotic vocabulary, and more conducive
to the process of learning one's own conlang--something I have yet to do.

To sum up, I write the sample text after I've got a long grammar.   Hope I
haven't rambled too long.  Best of luck with your work;   hope you post it
on the net.

Jim G.




----- Original Message -----
From: "Tom Tadfor Little" <tom@...>
To: <CONLANG@...>
Sent: Monday, July 16, 2001 9:19 PM
Subject: conlang maturation


> I'm curious about the process that leads from a language sketch to a
> full-blown conlang. Most conlangs begin with a phonology, grammatical
> sketch, some orthographic system, perhaps a few vocabulary words or roots,
> and some derivational strategies. This is enough to give one the feeling
of
> trying out any new "bright ideas" that seem fun or inspiring, and enough
to
> generate some plausible names and brief snatches of dialog for fiction,
> role-playing, etc. Most of my own projects have stopped at that stage,
> partly because that is what seems to give me the most fun, and partly
> because of no one to share the further growth of the language with (a
> situation remedied by my recent discovery of this list!).
>
> After that initial sketch, further work on the language is what I'm
calling
> the "maturation" process. The vocabulary increases beyond a short
exemplary
> wordlist, and syntactical structures are strained into greater
> sophistication to express the subtleties of extended sentences with
> interrelated clauses and the like. I'm wondering how others manage this
> process. When I've gone that far, it tends to be a case of "necessity as
> the mother of invention"--I have a need for extended texts in the conlang,
> so I grit my teeth and do it, expanding the language as I go.
>
> Is this the case for others, too? Or do you systematically enlarge
> vocabulary? Do you translate into the conlang for no other reason than to
> stretch it? Do you use the conlang yourself in daily life, to stay
immersed
> in it? Do you solicit friends to help? Do you try to compose original
texts
> in the conlang, rather than writing in your primary language first and
> translating? Or do you have a conculture project that makes steady demands
> on the conlang?
>
> Has anyone kept track of the lexical growth rate for their conlang? Has
> anyone maintained a steady growth for one language while starting a swarm
> of new ones? (this is always my great temptation)
>
> Cheers, Tom
>
>
>
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Tom Tadfor Little               tom@telp.com
> Santa Fe, New Mexico, USA
> Telperion Productions           www.telp.com
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Reply