Re: CHAT: Introduction

J. Barefoot wrote:
> I have noticed the gender "bias" on this list, but it seems that women
> are generally discouraged from creative and intellectual pursuits
> anyway, if not by society, then by themselves. I enjoy being an
> exception.
Quite true.  I was on a scholar bowl team in high school.  Altho the top
10 of our class was actually mostly female (in fact, both the
valedictorian and the saludictorian were female) there were very few
female members on our team, and they tended to be rather shyer about
buzzing in than the males.

> First, I don't think it's fair to say one language is more beautiful
> than another.
Saying language X is more beautiful *overall* than Language Y, I
certainly agree with you.  But I do think that it is fairer to say that
Language X is more beautiful phoneticallly, for example, than Language
Y.

> Secondly, I am interested in your methods. What do you do when you get
> an idea for a new language? Do you outline it all at once? Do you do
> each section in detail before moving on to a new one? Do you ruminate on
> it for a year before setting done the phonology? How do you keep track
> of vocabulary, etc.?
Here's my methods.  First, I think a while about the general nature of
the language.  Will it be agglutinative?  What distinctions will it
make?  Will it be ergative?  Will there be adjective concord?  That kind
if thing.  Then I work out a phonology (I don't see how you can go any
further without doing that).  What phonemes will I use?  What syllable
types will I allow?  Don't be afraid to change it later, tho.  For
instance, perhaps you might decide that /stS/ cannot be allowed, and
that those are to be replaced with /S/, thus you'd have to change
/astSin/ to /aSin/.

Once you do that, I like to work out inflectional morphology, if any.
This is where I work out exactly what I want to distinguish.  Most of my
languages tend to have lots of inflectional morphology, so I work out
aspects, tenses, personal endings, cases, numbers (they also tend to be
agglutinative).  You may wish to sketch out an earlier stage and changes
that led to this one.  Especially if you want a fusional language.
Fusional langauges are descended from agglutinative languages.  Again,
don't be afraid to change these several times.  W. has gone from 6 cases
to 10 cases to 12 cases to 27 cases to 19 cases.  Verbal inflections
have changed several times as well, going from very agglutinative to
partially fusional, and back to very agglutinative, tho with several
morphophonemic alternations.

Once you've worked out a provisional inflectional morphology, I like to
work on some basic vocabulary, words like "see", "hit", "person", things
like that, and experiment with basic sentences like "the man hit his
brother" (I don't like to use verbs like "see" or "have" since those
kinds of verbs often have different case-roles than most verbs).
Gradually expand your vocabulary, using pencil and paper.  I also use a
computer.  Programs that will alphabetize are wonderful.  I use
Shoebox.  Not only will it alphabetize, it allows you to parse sentences
and derived/inflected words and seperate roots and derivational
morphemes.  From time to time, I'll print out a new list.  I'll use the
most current list to write new words on.  When I have a chance to add in
the words into the program, I do so.  Then, when I have a lot of added
words, I'll print out a new copy, and throw away the old.  And so on,
and so on.

Most importantly, remember that languages *never* have perfect
one-to-one correspondences.  If you intend to have a naturalistic lang,
work out a conculture, it doesn't have to be detailed, just enough to
know what kinds of differences would be considered important enough to
lexicalize, as well as what kinds of words wouldn't have any
translations, like computer for a hunter-gatherer society, or sword for
a pacifistic society.  Also, DON'T be afraid to change.