Message-Id: <m0peTJd-0000R2C@xiron.pc.helsinki.fi>
Date:         Wed, 9 Mar 1994 13:49:43 EST
Reply-To:     jorge@PHYAST.PITT.EDU
Sender:       Lojban list <LOJBAN%CUVMB.bitnet@FINHUTC.hut.fi>
From:         Jorge Llambias <jorge@PHYAST.PITT.EDU>
Subject:      Re: The Mad Proposals
To:           Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Content-Length: 4799
Lines: 109

Matthew says:

> I have a number of reservations though:

They are valid, but I'll try to answer them:

> 1.  Any change to the language, even if it is a compatible improvement, has
>     to be considered very carefully before being taken on board.  There is
>     now a significant amount of published material in Lojban and thus a
>     significant amount of effort that has to be expended to update this
>     e.g. the textbook and the paper on logical connectives for a start.

I don't think updating the textbook and paper would be a big problem
(maybe because I'm not the author :), but I agree that the published real text
material should be kept as usable as possible. Since with the exeption of {gu},
no word would change its meaning, I think this is not a big concern. Old text
that used the old forms would still remain perfectly understandable. This
relates to your second point:

> 2.  The proposed changes involve the redefinition of the meaning of two
>     cmavo (ji and gu) which would thus render invalid any existing text with
>     these words in.  This is a big change to make (more than a proposal that
>     just caused the previous words to become "archaic").

That's partly true. However, much of the proposal can be salvaged to take
that into account:

- {ji} does not change meaning, but its meaning is extended, requiring
  the use of {ku} in some places where before could be elided. If wanted,
  we may instead use {je'i} as the general question connective. Proposal 4
  was just the final touch, but not essential. Even if {ji} is extended,
  the old texts would require at most (but not always) a {ku} to make them
  grammatical again, and the sense would be clear even if ungrammatical,
  anyway.

- Instead of {gu}, use another cmavo for this function, perhaps {gu'i}.
  However, I think the number of appearances of {gu} in real texts can
  probably be counted with the fingers of one hand, and you'd have plenty
  of fingers left. And the number of appearances of {gu} in its new
  function, would be of the same order of magnitude. I found a single
  appearance of {gu} in the texts of the ftp archive, and a single
  appearance of a GUhA (strangely enough, not {gu'e} but {gu'o})

> JL: Change {ji} from selmaho A to selamho JA
>
> Doing this means that there is no longer a question word for selma'o A
> (which you haven't completely done away with).  This would force people to
> use the unelided forms of sumti when asking questions, e.g.:
>
>     do visku le nanmu ku ?ji le ninmu

Yes, I left this unsaid on purpose, with the hope that noone would notice. :)
It's not really a problem. As you say, the question can still be asked using
the terminator. I think it's nicer if it's in selmaho JA, but if prefered we
can keep {je'i} for JA and {ji} for A.

> I think that this also means that the answer ".e" (or any other A) would be
> invalid, though since you can answer "joi" (of JOI) to a question "ji" (of
> JA) at the moment, perhaps not.

As you conclude, the answer is still valid.

By the way, {le nanmu .e le ninmu} is really {le nanmu ku .e le ninmu ku}.
The terminators need not be explicited, but they're there anyway. The question
{le nanmu ji le ninmu} is saying the same as {le nanmu ku ji le ninmu ku}.

> On a slightly different tack:
[...]
> If the proposals were adopted, I think I'd prefer gije, gijoi, guje, gujoi
> etc. to match the existing .ije etc. (which can't be reversed in order).
> This would mean having to change the existing joigi to gijoi though.

No, that can't be done. You're mixing the forethought and afterthought forms:

        proposed:               today's:

Afterthought:

        ....   je ....          ....   je ....          for tanru
        ....   je ....          ....   .e ....          for sumti
        .... .ije ....          .... .ije ....          for bridi
        .... gije ....          .... gi'e ....          for bridi-tails

        ....   joi ....         ....   joi ....         for tanru
        ....   joi ....         ....   joi ....         for sumti
        .... .ijoi ....         .... .ijoi ....         for bridi
        .... gijoi ....          not possible           for bridi-tails

Forethought:

         je gi.... gi....           ge .... gi....      general
         je gu.... gu....         gu'e .... gi....      for tanru

        joi gi.... gi....       joi gi .... gi....      general
        joi gu.... gu....         not possible          for tanru


You can see that the right hand column has many more different forms for
the same type of function, and is even missing some. In addition, the
irregularity of the question cmavo doesn't show up in this table. If {ji}
was the only question cmavo, it would be completely regular.

> Terveisin,
>
> Matthew

Saludos,

Jorge