Re: [lojban] observatives & a construal of lo'e & le'e

[And Rosta <arosta@uclan.ac.uk>]

>>> <pycyn@aol.com> 10/29/01 09:19pm >>>
arosta@uclan.ac.uk writes:
#> in Lojban a sequence
#> of one or more sumti can count as a sentence.
#
#Technically not; every sentence must contain a bridi tail. But such a string 
#could be a fragment.

I meant "count as a macrosyntagm", a licit maximal syntactic unit. (But 
actually that wouldn't work, because lojban syntactic units go up to 
text level.)

#<I'm talking about current Loglan, or at least my desultory reading of it.>
#
#Even then the sumti-observative seems to ahve been rejected as too unclear 
#(i.e., how to tell it from a fragment or the beginning of a longer -- though 
#slowly emitted -- sentence ("...derry.")

I don't have my loglan files here, so can't check.

#<So anyway, it seems the rule for Lojban is that a zo'e x1 cannot be
#elided.>
#
#No, observatives are possible (actual, even).  It just is that the role of 
#such elided {zo'e} is restricted to objects at the center of one's 
#observation.

Right. A normal-zo'e x1 in main bridi cannot be elided. Where normal-zo'e
= zo'e with its normal meaning.

#<What le/lo have to do with observatives is that the sentence
#{lo gerku} could mean "Lo! A dog", while {le gerku} could mean
#"Lo! The dog".>
#
#Well, it is not a sentence, and I suspect a bit of subconscious punning to 
#get to this reading (it was surely present when the idea was first presented 
#-- so far as I know -- in '76 or so).  

I don't understand the punning.

#Oddly, the object of observation is always {le} by definition, so the pun fails.

"lo! the dog" = "le gerku is here"
"lo! a dog" = "here, da gerku"

For the latter, the object of observation is not a dog but a soa of there being
a dog.

#<No. Jorge's usage was intended to be canonical/defining, replacing the
#meaning that has to do with typicality.>
#
#Yes, that mistake, too; even xorxes can't set up a new canon for a fixed term.
#
#<, the whole 
#point of {tu'a} is to prevent the quantifier on -- or one derived from -- the 
#sumti from rising to the upper level, so of course the quantifier goes on the 
#{le nu ... co'e} that {tu'a} indicates.  Where has this been doubted?  
#
#I don't remember this ever having been the whole or even the partial
#point of tu'a. Tu'a marks so-called sumti-raising, and I don't recall the 
#quantification either having been settled or much discussed. My memory
#may be unreliable, though.>
#
#Well, I claim that MY memory is non-veridical, so I won't dispute on what has 
#or has not been said.  Somethings are so obvious as not to need mentioning 
#even here (until the mention is required, of course).  The only logically 
#interesting feature of subject raising (and, indeed, almost the only 
#linguistically interesting one, aside from cooccurrence restrictions) is that 
#it gives a surface structure (in English and most languages -- but NOT in 
#Lojban) that appears to allow quantifications that are in fact illegitimate 
#and the point of reconstructing the invisible clause is to explain why the 
#quantification is illegitimate.  

None of this is so obvious as to not need mentioning. Based on my time in
Lojbanistan I'd say that the main point of sumti-raising is to be briefer or
vaguer than would be the case if the full bridi were used in stead of the
raised sumti. 

This argument is a bit of waste of time, though. I think tu'a is more useful
if it blocks the usual quantification rules, so the only thing we disagree about
is whether this makes it exceptional.

#<It's a good exception, but you can't escape it being an exception, given
#the rule that sumti quantified in situ are interpreted as being quantified
#in the prenex of the localmost syntactic bridi.>
#
#I can and do.  First, as you know, I do not believe there is a rule
# of the superficial and erratic kind that you seem to be appealing to here, 
#from which a raised subject is an exception.  Secondly, in the real rule of 
#which yours is a parody (and, but for the erratic element, in yours too), the 
#sumti in {tu'a sumti} does not occur in the bridi in which {tu'a sumti} #occurs.  Thirdly, that sumti is bound in the bridi in which it does occur.

I don't know if this is discussed in the Refgram. If there's no documentation
anywhere, then it's hard to settle this thing. I say what I say based on 
a decade of relatively attentive reading of this list, but even if in any
verifiable sense I am correct, the consensus I report is destroyed by
your dissent, and the new situation is that there is no consensus.

So I'll just note that this is an unresolved issue that can be left until the
Elephant is running.

#<Come on, you chide others for saying "Lojban can't say X". Just make the
#appropriate lujvo.>
#I didn't say it couldn't be done; I just said that we have lost a natural way 
#of doing it somewhere along the line.  As for making a lujvo, people dislike 
#mine as much as my translations and a lujvo does not seem the appropriate way 
#to go here anyhow.
#
#<#Well, it will fail -- at least to be useful -- if it cannot be given some 
##meaningful content.  Historically, it has been used as a magic wand to cover 
##cases that could not be made to fit otherwise  
#
#you're talking about {lo'e}, I take it. Or prototype theory?>
#
#Prototype theory.(this version of {lo'e} a fortiori).
#
#<The main answer to all of this is that lo'e and its implementation of 
#prototype
#theory's categorial individual (= 'prototype') makes available an alternative 
#ontology. Users who find that ontology useful can use it; those who don't
#needn't. All pretty Lojbanic.
#
#To answer your "Is it a blueprint or a member of the class or a way of 
#talking 
#about the class fuzzily", those are indeed answers given by weak forms of
#prototype theory, but the strong form is best understood as saying that
#either there are no categories, only individuals, or that everything is a
#category. Either way, the traditional member-of relation is replaced by the
#version-of relation.>
#
#Still lacking is any explanation of what this might conceivably mean that is 
#different from either bad statistics or worse Platonism.  I suppose bad 
#Platonism is a possible ontoology, but it doesn't help here, because, as 
#noted, the prototype has very few properties in common with the versions and 
#so fails to do its job.  Alternate ontologies are welcome, but tey ought not 
#be brought in to solve a problem they in fact make worse.

I have done my best to explain. 

So you think prototype theory is bad statistics or worse Platonism: so be it,
but there are plenty of people who don't agree with you, or who nonetheless
find it valuable; they should not be denied their gadri.

#<I think it isn't what I would think of as Platonism. For example I can see
#touch and smell the pc prototype, so prototypes aren't inherently abstract
#(-- I understand Platonic categories to be inherently abstract).>
#
#Whoa!  You can (or could in certain situations) see, touch and smell ME, but 
#I am not a me prototype in any interesting sense.  

Yes, you are. 

#I suppose, we are back now 
#to st-worms and  temporal (or maybe even some kinds of spatial) cross 
#sections.  

among other sorts of version, yes

#That works for an individual, but not for a natural kind (let 
#alone a {le} group). This is not coming across as an ontology now, so much as 
#a verbal formula that covers several ontologies that inherently have nothing 
#to do with one another.  That judgment may turn out to be wrong (see 
#Lesniewski's mereology), but it sure needs a lot of work to make it plausible.

I think you're setting unreasonably high philosophical standards here. 

Is there a linguistic case for prototype-theoretic gadri? Yes. Are their logical
properties well-understood or well-defined? No. Is their conceptual essence
adequately understood? Yes.

#<#-- in which case the question of connection -- which may not be 
##important in the present sense -- remains unsolved, 
#
#I don't know what that question is.>
#
#What is the relation between the prototype and the version by virtue of which 
#the version is a version of the prototype, rather than something else.

Ah. Resemblance, is the usual answer.

#<#and of course means that 
##the individual has properties that none of its versions has and 
#conversely).  
#
#I think this is held (by prototype theorists) to not be the case. I would 
#favour
#going along with that view in the interpretation of {lo'e broda}, and taking
#{lo'e/tu'o du'u ce'u broda} to be the Platonic essence.>
#
#Yes, but HOW would they do it?  

sorry -- do what?

#I take {lo'e [or whatever] du'u ce'u broda} 
#to be a property, and thus not the right category to be an Ideal or a 
#prototype or...

Not the right category to be a prototype. I'm not sure enough what an Ideal
is, so can't suggest how to refer to it.

#<I don't want to debate the ontology itself. I just want to argue that it is 
#important
#enough that Lojban should have a way of expressing it, and furthermore that
#it's useful if it is expressed by a gadri..>
#
#I am sure we can come of with an expression meaning "is a/the prototype of" 
#to express this ontology, I'm trying to find out just how this helps explain 
#a gadri.  Byron on Coleridge: "explaining metaphysics to the nation, how I 
#wish he'd explain his explanation." I have a pretty clear idea what {lo'e 
#cinfo} means, I have rather less of what a prototypical category of lion is.

As I say, I've done my best to explain. There's also a vast literature which you
may or may not know. Either way, let's assume you know the literature but
still think the theory's crap -- well, then, lots of people don't, and they have
language rights too!

I am unsure whether "we can come of with an expression meaning "is a/the 
prototype of" to express this ontology". How do we avoid quantification and
treat a category as an individual? That's where, as I see it, the gadri is needed.
Someone prototype-theory-minded doesn't want to be faffing about choosing
among different quantifiers etc. when they want to refer to Lion.

#<Apparently KOhA gets quantified, not just LE, so (without my knowing the
#formal grammar) it seems plausible that the grammar allow "li" and "me'o"
#to be quantified.>
#Oh, it probably does, but I am not sure that a default was specified 
#(presumably {pa}, but I wouldn't be surprised at anything).
#
#<However, my ulterior question is: What should one do when the grammar
#requires a gadri but, as with "li" and "me'o", the sumti attaches always to
#unique objects? Answers so far are "tu'o" and "lo'e".>
#
#I agree, but, on the other hand, since it won't in the end make a difference, 
#take whatever you like (well, {lo'i} is clearly a mistake as is any number 
#but 1).  I like {tu'o} for its in-your-face nose-thumbing and dislike {lo'e} 
#because it is obscure and possibbly wrong.
#
#<Those examples, on certain construals, take us into an imaginary realm
#where there is only one customer and only one postman, and situations
#in other realms where there are many customers and many postmen are
#versions of the imaginary one. To me this is not an unnatural mode of
#thought, and it is desirable that Lojban be able to express it relatively
#effortlessly.>
#
#Well, I don't like it to be too easy to move into imaginary realms -- I 
#certainly want a least one major flag up when I move.  It does seem to me 
#that Lojban is more than adequately equpped -- though not yet very well 
#trained -- for making such moves easily and visibly.  I am less sure about 
#versions of an imaginary postman, however.

This is fair enough: you wish for indicators of shifting realms and ontologies.
But the lojbanic way is to make such things optional.

--And.