Theiling Online    Sitemap    Conlang Mailing List HQ   

Keep On Conlanging

From:Jim Grossmann <steven@...>
Date:Wednesday, September 12, 2001, 2:50
Hi, everybody,

It may seem odd for us to be conlanging in the shadow of today's
incalculable tragedy.   But for Americans, going about one's usual business
can be a way of affirming that we will not be demoralized.

If part of that usual business is conlanging, let's keep doing it, with
thanks to our friends and fellow conlangers in other countries.

To business...

1st)    A lot of people on the list have been doing a lot of interesting
things with ergativity.  If I haven't done so already, may I recommend
Dixon's book, "Ergativity," number 69 in the Cambridge Studies in
Linguistics?

In addition to ergative/absolutive systems, the book discusses split
ergative languages, along with ...

a)    split S languages.   In such languages, we have two kinds of subjects,
one kind that stands for a patient, and another that stands for an agent.
A split S conlang might have clauses like these:

subject-A    verb

e.g.    he jumped

subject-P    verb

e.g.     he drowned

subject-A   verb  object

e.g.    he ate muffins

subject-P   verb    agent

e.g.   he was eaten by crocodiles

Also interesting is direct marking, in which arguments are marked according
to the semantic relationship that they have with the verb in a given
sentence, rather than according to the prototypical meaning of the verb, as
in nom/acc;   erg/abs systems.

Quoth Dixon:    "Turning now to languages of the second type, we find that
in any instance of use of a verb, its arguments are marked not by a
syntactic rule relating to any prototypical scheme, but so as to directly
describe its meaning in that instance.   Thus, in 'John hit Bill', 'John'
would be marked as (volitional) agent, and 'Bill' would be marked as
(affected) patient.   In 'John hit Bill accidentally," and 'The falling
branch hit Bill,' 'Bill' would again be marked as patient, but 'John' and
'the falling branch' would not be marked as agent, since they do not exert
volitional control over the activity.   In 'John hit at Bill' (but the
implement didn't come into contact with him),
then 'John' would be marked as agent but 'Bill' would not be marked as
patient, since he was not affected by the action."

..."Languages with 'direct marking' may also have more fluid grammatical
requirements (compared with English, for instance, where a subject is
obligatory in non-imperative clauses).   Corresponding to English 'John was
hit' we could, in a language with of the second type, find 'John' marked as
patient, but no statement of any agent (and nothing like a passive
derivation)."

(from Dixon's 'Ergativity,' page 24)

2nd:   Welcome back, Matt.

3rd:    I have a book on Tolkien's languages that I never use verb much.
It's called "An Introduction to Elvish" and has more information about
Tolkien grammars than I've ever found in the commercial books.   Is anyone
interested in having this book?

4th:   All on the list are once again invited to check out my
language-on-the-web, Goesk, at...

www.eccentrix.com/misc/rainchild/goesk1.html

Its synchronic phonology is baby simple;   its meager vocabulary is an
etymological disaster area;   and this supposedly Germanic language has a
future tense suffix and no participles.

BUT the reference grammar is pretty complete, and the reflexive pronoun
phrases, prepositions and prepositional modifiers, and relative pronouns are
worth a grin if I do say so myself.

Keep on conlanging.

Jim G.