From pycyn@aol.com Tue Mar 12 08:04:22 2002 Return-Path: X-Sender: Pycyn@aol.com X-Apparently-To: lojban@yahoogroups.com Received: (EGP: unknown); 12 Mar 2002 16:04:22 -0000 Received: (qmail 34110 invoked from network); 12 Mar 2002 15:54:18 -0000 Received: from unknown (216.115.97.167) by m3.grp.snv.yahoo.com with QMQP; 12 Mar 2002 15:54:18 -0000 Received: from unknown (HELO imo-d10.mx.aol.com) (205.188.157.42) by mta1.grp.snv.yahoo.com with SMTP; 12 Mar 2002 15:54:18 -0000 Received: from Pycyn@aol.com by imo-d10.mx.aol.com (mail_out_v32.5.) id r.f3.17bd2f24 (4540) for ; Tue, 12 Mar 2002 10:54:16 -0500 (EST) Message-ID: Date: Tue, 12 Mar 2002 10:54:15 EST Subject: Re: [lojban] More about quantifiers To: lojban@yahoogroups.com MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="part1_f3.17bd2f24.29bf7ea7_boundary" X-Mailer: AOL 7.0 for Windows US sub 118 From: pycyn@aol.com X-Yahoo-Group-Post: member; u=2455001 X-Yahoo-Profile: kaliputra --part1_f3.17bd2f24.29bf7ea7_boundary Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit In a message dated 3/11/2002 5:15:15 PM Central Standard Time, jjllambias@hotmail.com writes: > >Have you really known about {me'iro} for all these years? > > I used {da'asu'o} before, which is equivalent. But the particular > form of the quantifier doesn't change the system. > I told you about {da'asu'o} years ago, but you never > listen to me... :) > Why, so you did (I had it filed under {da'asu'o} rather than "I"). I seemed to ahve replied that it was too ugly to use; yet I like {me'iro}. Tastes change? Maybe not; I still think {da'asu'o} is too ugly to use. I always listen to xorxes, either he has solved a problem or he has said something so outrageous that I have a nice long discussion in the offing. Rather than going through xorxes comments case by case, let me lay out what I take to be the situation, knowing full well that xorxes will catch any errors Ultimate forms: A+: ge de broda (1) gi ro da zo'u ganai da broda (2) gi da brode(3) A- : [ganai 1 gi] roda zo'u ganai 2 gi 3 E+ : ge 1 gi roda zo'u ganai 2 ginai 3 E- : [ganai 1 gi] roda zo'u ganai 2 ginai 3 I+ : [ge 1 gi] su'o da zo'u ge 2 gi 3 I- : ganai 1 gi su'o da zo'u ge 2 gi 3 O+ : [ge 1 gi] su'o da ge 2 ginai 3 O- : ganai 1 gi su'o da ge 2 ginai 3 In general A+ = E+~ = ~O- = ~I-~ A- = E-~ = ~O+ = ~I+~ Within these two groups, each member can be defined in terms of any other member by apprpriate adjustment quantifiers. No member of one groups can be defined in terms of the other using only ~. xorxes' systems Since these are inadequate for defining the remaining cases, we will use his fuller system, adding I can't at the moment find corresponding {da poi} forms, so we will stick to the {lo... broda} ones only. my system: A+ ro broda cu brode A- ro da poi broda cu brode E+ no broda cu brode E- no da poi broda cu brode I+ [suo /lo /su'o lo] broda cu brode I- (su'o) da poi broda cu brode O+ me'iro broda cu brode O- me'iro da poi broda cu brode Somewhere xorxes ascribes a basic set to me. I tend to think of all of these as basic since they all come out of the same description: {Q da poi broda cu broda} for free, {Q broda cu broda} for importing. But I'll accept the Aristotle set, A+, E-, I+ O-, as basic (or any other with a member from each definition set). Rules to return to ultimate forms xorxes: prefix {ro/no lo su'o} immediately with {ge 1 gi} replace Q...broda cu brode with Q = {ro} {ro da zo'u ganai 2 gi 3} {no} {ro da zo'u ganai 2 ginai 3} {su'o} {su'o da zo'u ge 2 gi 3} {me'i ro} {su'o da zo'u ge 2 ginai 3} work remaining negations through mine: prefix all {Q broda} with {ge 1 gi} prefix all {Q da poi broda} with {ganai 1 gi} replace as above. (To be honest, if xorxes had given other versions of I- and O-, his rules would be one shorter, omitting the last one). Moving negations around. The general rule for moving negations across quantifiers is: change outer sign (before the whole thing), change inner sign (before the {brode}), change quantifier to its diagonal opposite, i.e., different in quantity, quality, and import -- so + and - exchange and A <> O and E<>I. moving ~ in. xorxes' version: If there is an initial {naku}, delete it and stop Else drop initial negation exchange external {su'o} <> {no}, {ro} <> {me'iro} place {naku} for {cu} and reduce {naku} string (Again, the unreduced I- and O- add a rule) mine: Drop initial {naku} exchange {Q da poi broda} and {Q broda} exchange Q as above internal {naku} as above The other negation movings are the same, mutatis mutandis. (I should note that, given his usage, xorxes may have some trouble coming up with reasonable I- and O- forms, since, whatever you may think about {le ro broda}, {lo su'o broda} pretty clearly cannot be empty.) Admittedly, then, in an ideal xorxes system my rules would be a whole step more complicated. I think that extra effort is worth it to be able to tell at a glance that a setence has existential import. --part1_f3.17bd2f24.29bf7ea7_boundary Content-Type: text/html; charset="US-ASCII" Content-Transfer-Encoding: 7bit In a message dated 3/11/2002 5:15:15 PM Central Standard Time, jjllambias@hotmail.com writes:


>Have you really known about {me'iro} for all these years?

I used {da'asu'o} before, which is equivalent. But the particular
form of the quantifier doesn't change the system.
I told you about {da'asu'o} years ago, but you never
listen to me... :)


Why, so you did (I had it filed under {da'asu'o} rather than "I").  I seemed to ahve replied that it was too ugly to use; yet I like {me'iro}.  Tastes change? Maybe not; I still think {da'asu'o} is too ugly to use.
I always listen to xorxes, either he has solved a problem or he has said something so outrageous that I have a nice long discussion in the offing.

Rather than going through xorxes comments case by case, let me lay out what I take to be the situation, knowing full well that xorxes will catch any errors
Ultimate forms:
A+: ge de broda (1) gi ro da zo'u ganai da broda (2) gi da brode(3)
A- : [ganai 1 gi] roda zo'u ganai 2 gi  3
E+ : ge 1 gi roda zo'u ganai 2 ginai 3
E- : [ganai 1 gi] roda zo'u ganai 2 ginai 3
I+ :  [ge 1 gi] su'o da zo'u ge 2 gi 3
I- : ganai 1 gi su'o da zo'u ge 2 gi 3
O+ : [ge 1 gi] su'o da ge 2 ginai 3
O- : ganai 1 gi su'o da ge 2 ginai 3

In general
A+ = E+~ = ~O- = ~I-~
A- =  E-~ = ~O+ = ~I+~

Within these two groups, each member can be defined in terms of any other member by apprpriate adjustment quantifiers.  No member of one groups can be defined in terms of the other using only ~.

xorxes' systems
<A-: ro broda cu brode = ro da poi broda cu brode
E-: no broda cu brode = no da poi broda cu brode
I+: su'o broda cu brode = su'o da poi broda cu brode
O+: me'iro broda cu brode = me'iro da poi broda cu brode>

Since these are inadequate for defining the remaining cases, we will use his fuller system, adding

<A+ ro lo su'o broda cu brode
E+ no lo su'o broda cu brode
I- naku no lo su'o broda cu brode
O- naku ro lo su'o broda cu brode>

I can't at the moment find corresponding {da poi} forms, so we will stick to the {lo... broda} ones only.

my system:

A+ ro broda cu brode
A-  ro da poi broda cu brode
E+ no broda cu brode
E-  no da poi broda cu brode
I+  [suo /lo /su'o lo] broda cu brode
I-    (su'o) da poi broda cu brode
O+ me'iro broda cu brode
O-  me'iro da poi broda cu brode

Somewhere xorxes ascribes a basic set to me.  I tend to think of all of these as basic since they all come out of the same description: {Q da poi broda cu broda} for free, {Q broda cu broda} for importing.  But I'll accept the Aristotle set, A+, E-, I+ O-, as basic (or any other with a member from each definition set).

Rules to return to ultimate forms
xorxes: prefix {ro/no lo su'o} immediately with {ge 1 gi}
            replace Q...broda cu brode with
             Q = {ro}  {ro da zo'u ganai 2 gi 3}
                   {no}  {ro da zo'u ganai 2 ginai 3}
                   {su'o} {su'o da zo'u ge 2 gi 3}
                   {me'i ro} {su'o da zo'u ge 2 ginai 3}
             work remaining negations through
mine:  prefix all {Q broda} with {ge 1 gi}
           prefix all {Q da poi broda} with {ganai 1 gi}
           replace as above.
(To be honest, if xorxes had given other versions of  I- and O-, his rules would be one shorter, omitting the last one).

Moving negations around.  The general rule for moving negations across quantifiers is: change outer sign (before the whole thing), change inner sign (before the {brode}), change quantifier to its diagonal opposite, i.e., different in quantity, quality, and import -- so + and - exchange and A <> O and E<>I.
moving ~ in. 
xorxes' version: If there is an initial {naku}, delete it and stop
                       Else drop initial negation
                         exchange external {su'o} <> {no}, {ro} <> {me'iro}
                         place {naku} for {cu} and reduce {naku} string
(Again, the unreduced I- and O- add a rule)

mine:     Drop initial {naku}
             exchange {Q da poi broda} and {Q broda}
             exchange Q as above
             internal {naku} as above
The other negation movings are the same, mutatis mutandis.  (I should note that, given his usage, xorxes may have some trouble coming up with reasonable I- and O- forms, since, whatever you may think about {le ro broda}, {lo su'o broda} pretty clearly cannot be empty.)

Admittedly, then, in an ideal xorxes system my rules would be a whole step more complicated.  I think that extra effort is worth it to be able to tell at a glance that a setence has existential import.
--part1_f3.17bd2f24.29bf7ea7_boundary--