[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Konrad on Glosa
From: hmiller@ea.com (Herman Miller)
>Subject: Re: Konrad on Glosa
>
>In article Don Harlow <donh@netcom.com> writes:
>>Well, let us substitute "ability" for "power" here (assuming that this is
>>what Konrad meant), and I get the following set:
>
>> lia pova helpo / his able help / an pote auxi
>
>> li povas helpi / he can help / an pote auxi
>
>> lia povo helpas / his ability helps / an pote auxi
>
>> li pove helpas / he helps ablely / an pote auxi
>
>>We can safely throw out the first expression -- since it is not a
>>sentence, it is not likely to appear out of some other context (except
>>maybe in English). Which means that you have three remaining complete
>>sentences, meaning three different things, which are expressed identically
>>in Glosa.
>
>...
>Lojban, from what I've seen browsing the Lojban WWW page, has gone to
>the other extreme: having so many ways to mark the function of a word
>in a sentence that it is excessively confusing to a beginner (unless
>he/she is a computer programmer or mathematician).
Lojban has many ways to use a word in a sentence because there is a lot
of hidden semantics in the ways words are used which Lojban makes
explicit. Thus, to translate from a natlang to Lojban, you have to
really think about what the words mean not merely what their syntax is.
Let me tackle the above examples.
Assuming that "he" in this contact is the Lojban pronoun-letter ny.
(N.: in this case standing for "nakni" = "male")))
> lia pova helpo / his able help / an pote auxi
1. le ny kakne nunsidju
le ny kakne nunsidju
the N.'s able help-event
An incomplete sentence: an argument/sumti which could serve in the role
of subject or object. "le ny" indicates a possessive. "nunsidju"
reflects that "help" as a noun is semantically different from help as a
verb - in this case, I have translated it as an abstraction: "the event
of helping". An alternative:
1a. lenu ny kakne sidju
le nu ny kakne sidju
the event of N. ably(?) helping
is akin to the 4th example below, being a specific event of the referent
of that sentence occurring.
> li povas helpi / he can help / an pote auxi
2. ny kakne lenu sidju
ny kakne le nu sidju
N. is capable of the event of helping
This one is probably the most straightforward, but it still has
helping at a different abstraction level than ability.
Lojban also has potentiality expressible as a modal:
2a. ny ka'e sidju
ny ka'e sidju
N. potentially helps
> lia povo helpas / his ability helps / an pote auxi
3. le ny ka kakne cu sidju
le ny ka kakne cu sidju
The N.'s quality of ability helps
(ableness)
or
3a. le ka ny kakne cu sidju
le ka ny kakne cu sidju
the quality of N. being able helps
Ability is here shown to be a different type of abstraction, a property
abstract. These tend to be associated with concepts that are adjectives
or adverbs in English, while event abstracts tend to be more often found
with noun and verb concepts. But there is no restriction:
nu kakne = event/state of being able
(possibly but not necessarily suggesting that N.'s ability is limited in
time in some way.)
ka sidju = property of helping
= helpfulness
> li pove helpas / he helps ablely / an pote auxi
4. ny kakne(?) sidju
ny kakne sidju
N. ably(?) helps
N. is an able helper
I am wary about this translation, because my sense of the English
"ablely/ably" is that it usually implies "with skill/prowess" rather
than merely "with capability", which is the correct interpretation of
the above Lojban. Lojban would use "certu" (akin to "expert") instead
of "kakne" for my "skillfully" interpretation:
4a. ny certu(?) sidju
ny certu sidju
N. skillfully helps
or N. is a skillful helper.
I will add a fifth sentence, that shows one other natlang semantic that
is typically hidden, but revealable in Lojban. Compare with 2. above:
5. tu'a ny. kakne le nu sidju
tu'a ny. kakne le nu sidju
"N." is capable of the event of helping
"N" can help.
where the quoted "N." refers not to N, but to something unspecified that
N does or some abstract property that N. has. For example, his height
may be of assistance is reaching an object in a high location. Or his
running could help in getting a message to a critical person. This
is an explicit marking of "raising" - a phenomena of taking an object
as representing an abstract event or activity. A better example is to
compare:
6a. The evening meal is done.
tu'a le vanci sanmi cu mulno
6b. N. is done
tu'a ny [cu] mulno
The Lojban explicitly marks that each sentence is raised from an
abstraction like:
6c. N. is done eating the evening meal.
The event of N.'s eating of the evening meal is done.
lenu N. citka le vanci sanmi cu mulno
There are many contexts in the natural languages where we will use a
short form like 6a, or 6b with raising, to (briefly)) stand for 6c. We
ignore the semantic confusion - after all, N, a person, is not "done"
until he is dead. And unless you are having a formal dinner where
everyone rises from the table in unison after synchronizing their last
bite, each person will finish their meal at a somewhat different time,
and the meal is over - for them. (I am ignoring the social aspects of
eating, which are a culturally specific property of meals. If you
dislike this example, the same can be shown for "The race is done" and
"He is done." as short, raised, forms of "He is done running the race."
which would be pseudo-Lojbanized as "His running the race is done." Or
"The sock is done." and "He is done." as raised forms of "He is done
knitting the sock"/"His knitting the sock is done."
>Esperanto does seem to be a good middle ground, in that it allows any
>root to be used as a noun, verb, adjective, or adverb, and it easily
>resolves the ambiguities in these test sentences. But the disadvantage
>is that every word (except for prepositions, numerals, conjunctions, and
>a few others) must be marked, which adds an extra syllable. (The -o may
>be dropped in poetic usage, but I haven't seen that in ordinary prose.)
Lojban allows any root to be used as a noun, verb, adjective, or adverb,
but the above examples don't really show this since they hinge on the
fact that English "help" and "able" when used in those roles various
mutates semantically. Glosa loses that fact, while Esperanto seems to
make the syntax clear while retaining the muddy semantics. Lojban
resolves the ambiguities by making it clearer just what semantics are
actually involved. How about a different example set:
ny bajra le cutci
ny bajra le cutci
N. runs to the shoes. bajra is a verb or
ny bajra le cutci
N. is-a-runner to the shoes. bajra is a noun
ny bajra cutci
ny bajra cutci
N. runningly shoes bajra is an adverb and cutci is a (stative) verb
(i.e. N. is serving as a shoe/shoes, and doing so in a running manner))
ny bajra cutci
N. is a running shoe bajra is an adjective and cutci is a noun
are running shoes
ny cutci bajra
ny cutci bajra
N. shoelingly runs
N. is a shoed runner
ny cutci le bajra
ny cutci le bajra
N. is-shoes-of the runner
ny nelci le bajra cutci
ny nelci le bajra cutci
N. is fond of the running shoe(s)
ny nelci le bajra be le cutci
ny nelci le bajra be le cutci
N. is fond of the runner to the shoes.
ny nelci tu'a le bajra
ny nelci tu'a le bajra
N. is fond of "the runner".
(i.e. something the runner is doing - probably running. Or perhaps the
third sentence following.)
ny nelci lenu bajra le cutci
ny nelci le nu bajra le cutci
N. is fond of the event of running to the shoes.
ny nelci leka bajra le cutci
ny nelci le ka bajra le cutci
N. is fond of the qualities (involved in) running to the shoes.
ny nelci leka le bajra cu bajra le cutci
ny nelci le ka le bajra cu bajra le cutci
N. is fond of the qualities of the runner's running to the shoes.
Dare I ask for Glosa and Esperanto forms for all of these. I'm sure
they are expressible. I'm curious as to how much semantics is lost.
----
lojbab = Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA 703-385-0273 lojbab@access.digex.net
Ask me about the artificial language Loglan/Lojban, or see the Lojban WWW Server
<a href="http://xiron.pc.helsinki.fi/lojban/";>The Lojban Archive</a>
We also have material available via ftp (ftp.cs.yale.edu, directory pub/lojban).
email mailing list (listserv@ and lojban@cuvmb.cc.columbia.edu). The
LLG is funded solely by contributions, and are needed in order to
support electronic and paper distribution.