[lojban] Re: Loglish

[John E Clifford <lojban-out@lojban.org>]

--- Ben Goertzel <ben@goertzel.org> wrote:

> 
> Hi,
> 
> I don't mean to re-open the Loglish discussion
> on this list; it was very
> useful to me but has probably run its natural
> course.
> 
> However, for those who are interested, I have
> posted a revised version of my
> initial Loglish overview document, which
> benefited greatly from the feedback
> of those on this list.
> 
>
http://www.goertzel.org/new_research/Loglish.htm
> 
> Naturally, me being who I am, there are
> probably still some errors here and
> there ;-)
> 
> The next step in the development of the Loglish
> concept would be the actual
> implementation of a Loglish parser.  One of my
> colleagues plans to do this
> over the next couple months, but we'll see if
> it actually comes about and at
> what rate!  When/if a Loglish parser is
> completed I'll put it online and
> announce it here so y'all can appropriately and
> good-naturedly rip it to
> shreds ;-)
Minor comments:
You introduce {fi'o} as being like "qui" but have
not said anything about "qui." I assume you mean
{zei}, although {fi'o} does not - in your usage
-- look much like {zei}.

You use "la ben proceed lo to fi?o store" and the
like, where Lojban with English vocab would have
"la ben (cu) proceed fi'o to lo store' I suspect
the {cu} is needed, since the limits of cmene is
not defined in Loglish and the same would apply
(at least in speech) to "la mei_li think vo'a"
and subsequent examples. The point here however
is about the relative order of "prepositions" and
"articles."

Example 2: la rena go le suburb be la Melbourne
probably should be "la rena cu eng go lo suburb
be la Melbourne;" {cu} as noted above, {lo} for
indefinite reference, "eng go" per your earlier
discussion.

mi go le restaurant be loi Pakistan
{mi} probably doesn't need {cu}. "eng go" as
before, "loi Pakistan" is obscure to me for
several reasons: 1) "restaurant" does not have a
place for anything but the restaurant, so some
other device is needed here -- depending on what
the relation is (this is not too clear) 2)
"Pakistan" is a proper noun not obviously a
predicate, so one would expect {la} not {loi} 3)
even if "Pakistan" is a predicate -- something
like "pertain to the land, people or culture of
Pakistan," say --the {loi}, for a collective,
seems wrong.  It does not work for "Pakistan"
="Pakistani" for it is unclear what collective of
Pakistanis might be intended (the nearest thing
to a P restaurant here is run by New Yorkers of
Jewish extraction and staffed by Mexicans, so it
is not the staff and mangement that is
necessarily meant).

mi djuno ledu?u zo?e kau cilre la loglish
Well, {loglic} to begin with (or {loglec}, if the
short vowel seems more natural).  I can't figure
out {zo'e kau} in Lojban: {kau} attaches to
interrogative words to signal indirect questions,
so maybe {ma kau} is meant: "I know who is
learning Loglish." This misses the first half of
the English (except by implication -- if I know
who it is there must be someone).  On the other
hand, without the {kau} in the original, we get
the not very illuminating "I know that (you know
who/someone it doesn't matter who) is learning
Loglish," which misses the second part.  the
problems carry over to the Loglish (I think
{zo'e} is a bad choice here in any case, but that
seems to be idiosyncratic).

la Ben cu murder lo chicken lo weapon fi?o pliers
Realtive order of {lo} and {fi'o} again (and
maybe of {fi'o} and "weapon," but I have to see
how this plays out.
You later claim that this cannot be a chicken
with pliers, but the resolution appears to be
semantic, not grammatical.  Lojban would insist
on a grammatical disambiguation.

Later references to "qui" rather than {zei}.


To unsubscribe from this list, send mail to lojban-list-request@lojban.org
with the subject unsubscribe, or go to http://www.lojban.org/lsg2/, or if
you're really stuck, send mail to secretary@lojban.org for help.