Assorted responses

[cbmvax!uunet!munagin.ee.mu.OZ.AU!nsn (Nick Nicholas)]

Languages Spoken: English, Greek, Esperanto, Lojban, C
Date: Tue, 13 Aug 91 18:55:43 +1000
From: nsn


My apologies for flooding the net; Australia was offline for three days last
week. I thought this was caused by a causal relation between a shark and
a cable. A landlubber shark, it turns out, who went ashore and into the OTC
building in Sydney %^1/2

Anyways, nice to know you've been busy in my absence:

>From: dave@PRC.Unisys.COM
>Message-Id: <9108081712.AA03626@gem.PRC.Unisys.COM> 
>Date: Thu, 8 Aug 91 13:11:51 EDT
>Subject: Re: response to Dave Matuszek

>Thanks for your thoughtful reply to my rather acerbic posting.

Acerbic, slightly, but it made a lot of sense.

>The point I want to make is that, while Lojban will not be accepted as
>linguistics, you (or someone) should attempt to found a new subarea of
>linguistics, with some suitably descriptive but catchy title, that is
>devoted to the things one can learn from, or should know in order to
>build, conlangs such as Loglan, Lojban, Gua!spi (did I spell that
>right?), etc.  [I'd like to see it include the problem of constructing
>languages for communication between humans and computers, but that's
>just my personal interest showing.]  "Constructive Linguistics"? --
>surely we can do better than that!

Interlinguistics already exists; however, until now, it's been the domain
of either AL antiEspists with a new AL to push (the lojban for AL, btw, is
{runbau}. Yum.) or Esperantologists. The infrastructure is there, but 
exclusively run by Espists until now (and I doubt it has that good a rep
among linguists; Klaus Schubert's book might have helped, and it is certainly
the first interesting book about Esperanto (incidentally about inter-
linguistics, but that can't be helped) I've seen in English.) It would be
good for the field for Lojbanis to get involved in it; how much this improves
or worsens Lojban's rep amongst linguists is another matter.

>Date: Wed, 7 Aug 91 18:24:00 -0700
>From: fschulz@pyrps5.pyramid.com (F. Schulz)
>Message-Id: <9108080124.AA15115@pyrps5.pyramid.com>
>To: lojban-list@snark.thyrsus.com

>The translations are only needed for very short pieces of lojban
>text, say less than 5 sentences. Longer text is too overwhelming
>and I immediately delete the message. Shorter lojban is saved,
>edited to remove junk, and reviewed periodically.

This issue of how to translate and how much nonEnglish text to go for is a
recurring problem; it showed up, for example, when I was editing the Australian
Esperanto Youth Group's monthly last year. I had a lot of fun in Esperanto,
and the four or five members whose Esperanto was good enough for them to
follow were not complaining. As soon as I faced the rest of the readership,
at the annual Esp summer school, there was hell to pay.

Being the chief cuplrit in voluminous Lojban text (with Mark and John also
guilty), I really don't know what I can say about long text. The word-to-word
translations ultimately do more harm than good; I'd be loath to have anything
I'd written and considered artful subjected to such treatment. Short sentences
are another matter; yet usually here they are used as grammatical or morpho-
logical, and requests for translation often seem to me to miss the point (even
if even Lojbab requests 'em, as has happened.)

My vocab is about 600 gismu, and I welcome interlinear colloquial translations;
I wouldn't have been able to handle the recent long texts without them. I can
admit my weaknesses, and consider this one of them. Word-to-word, though, are
rarely communicative, don't really give any more semantic information than a
colloquial translation, tell you nothing about syntax you shouldn't have already
known directly in Lojban, and to me distort the text inexcusably. They were
a nice crutch in my first months of Lojban, but reading Helsem or whoever in
the original is far more helpful.

So then: I don't deny the usefulness of word-to-word, but I don't write for
paedagogy (but to explore Lojbanic expression), and I won't be in it.

That said, I think some systematically paedagogical content on this list would
be welcome. Any suggestions?

>Date:     Thu, 8 Aug 91 17:51:45 EDT
>From: "Arthur W. Protin Jr." (GC-ACCURATE) <protin@PICA.ARMY.MIL>
>Subject:  Re:  response to Dave Matuszek
>Message-Id:  <9108081751.aa17225@COR4.PICA.ARMY.MIL>

>    As to the issue of lujvo, I am sick and tired of not having
>a reference for the collected ahd anotated lujvo.  PLEASE SEND
>ME ANY WORDS THAT YOU CONSTRUCT.  I will try to organize them.
>If I get any set of non trivial size (>100) I will regularly
>include a comment about how many I have and will answer requests
>for copies.

You asking this of Lojban Central (nice phrase there, John; jbozda, perhaps?)
or of individuals? Best way, I think, is a corpus of grep-able text on the PLS.

>From: cbmvax!snark!cowan@murtoa.cs.mu.oz@uunet.uu.net
>Message-Id: <m0k8dcg-0002VSC@snark.thyrsus.com>
>Subject: Re: Definition of lujvo
>Date: Fri, 9 Aug 91 16:38:31 EDT

>I don't think there's a genuine dispute that regularities exist that make
>the interpretation of lujvo straightforward.  

And so forth. John, as you have claimed, you do speak for me %^) And for
historical accuracy, let's start referring to Rene de Saussure (Ferdinand's
brother, subsequent inventor of Esp Mark II, Esperantido, Return of the Killer
Esperantos, Abbot and Costello meet Esperanto, Esperanto III: the Omen, Son
of Esperanto, The Esperantio Strikes Back, and Esperanto: the Mini-series)
as the formulator of those Esp wordmaking rules (more word-grouping, often,
than semantic), which Kalocsay twenty years later codified.

>I think, therefore, that jimc's "diklujvo" rules would better be interpreted
>as a set of lujvo heuristics, something like the following:

How to choose between these heuristics? gismu calssification, I suspect, which
will have Lojbab fuming. That part can wait.

>1) Some two-part lujvo A-B can be interpreted as "B be FA lo nu A".
>(FA means a member of lexeme FA specifying some place of B.)
>This interpretation is plausible when B has a place which can be filled
>most naturally by an event; the lujvo typically has the places of B, except
>that the event place is replaced by the places of A.  

This is how I treat {ri'a}, to give me the lojbanic equivalent of -igi in Esp:
xy. ctiri'a .y'ybu. zy.
xy. rinka lenu .y'ybu. citka zy.
X causes that Y eats Z
X feeds Y Z

BUT (and this realisation put me off diklujvo):
I want a teddybear
I want that I should have a teddybar
mi djica lenu mi ponse lo cribe se kelci
mi posydji mi lo cribe se kelci
I want me a teddybear/ I want for me a teddybear,

which will be universally ditched, I suspect, for

mi posydji lo cribe se kelci
I want a teddybear

since noone will want to go as pedantic as

mi selposydji lo cribe se kelci
mi djica lenu lo cribe se kelci cu se ponse mi
I want that a teddybear should belong to me
I want a teddybear for me

at least not in the early stages of the language. (As happened in Esp, these
rules do tend to generalise and have their simpler alternatives applied after
a while - forty to fifty years, in Esp's case.)

Thus two paradigms for this heuristic:

Bridi A (args A1 A2 A3 A4 A5) and Bridi B (args B1 B2 B3 B4 B5), with say B2
able to take an abstraction as its value, "diklujvoes" into EITHER

AB: B1 A1 A2 A3 A4 A5 B3 B4 B5, OR
AB: B1 A2 A3 A4 A5 B3 B4 B5.

And what exactly happens to choose between these two is a matter for 
description, not orders from Lojban Central (and I suspect the ukase on {tu'a}
will not meet with wide acceptance either.

>2) Some two-part lujvo A-B can be interpeted as "A je B".

Often, the version {le A poi B} might be clearer.

>3) Some two-part lujvo A-B can be interpreted as "A be FA lo B".
>(FA means the same as before.)  This interpretation is plausible when
>"lo B" could plausibly fill one of the places of A.  The lujvo typically
>has the places of B, except that the place being filled is removed and
>replaced by the places of A other than the first.  Example:

Again, this place is not always removed in the admittedly anglo-coloured
lujvo we are likely to come up with:

To go up a mountain
garkla le cmana

{le cmana} is {le gapru} (the up thing), which has not been elided here. Once
more, a complication.

A faint analogy in Esperanto: 

Iri en cxambron
To go into a room
should transform to
Eniri cxambron
To enter a room
with room the direct object, but the "pleonastic" form
Eniri en cxambron
To enter into a room
is often found.

It is this split in two of the three Carter categories which made me conclude
that the time for diklujvo was not yet upon us. But they are damn useful
heuristics (heuristics, not algorithms), and should be included in the 
dictionary. I'd also like a commitment on whether the places of a lujvo may
only be a subset of the places of the component bridi under normal circum-
stances; this is, as I've said before, a bold step, yet essential in a Jimc-
like perception of lojban place regularity.

Oh, and I think both "tiger talk" and "breathe water" suck as lujvo. Sorry,
Jim.

>From: jimc@math.ucla.edu
>Message-Id: <9108121715.AA11673@njoshua.math.ucla.edu>
>Subject: Re: Some cmavo comments, if y'all please. 
>Date: Mon, 12 Aug 91 10:15:02 -0700

>> >>>fa'e      BAI

Your comment on sumtcita use of {fa'e} was quite helpful, Jim. Not convincing,
but helpful %^)

Thanks for the formalisation of ne'x too. Hope Lojbab was listening %^)