Date: Fri, 14 Nov 1997 13:21:35 -0500 (EST) Message-Id: <199711141821.NAA09567@locke.ccil.org> Reply-To: And Rosta Sender: Lojban list From: And Rosta Organization: University of Central Lancashire Subject: Re: le/lo X-To: LOJBAN@cuvmb.cc.columbia.edu To: John Cowan Status: OR X-Mozilla-Status: 0011 Content-Length: 5520 X-From-Space-Date: Fri Nov 14 13:21:43 1997 X-From-Space-Address: LOJBAN@CUVMB.CC.COLUMBIA.EDU Jorge: > >I think {loi taksi} pretty much entails {su`o pa taksi}. > > Well, I don't know in this particular case, but I can think of > other cases where it certainly doesn't. For example: > > le klaji cu culno loi plejykarce > The street is full of taxis. > > It is not the case that there is at least one taxi such that it > fills the street. It is the case that there are some taxis such > that they (en masse) fill the street. You are of course correct. > >> There's no problem either with: > >> > >> mi denpa tu'a lo plejykarce > >> "I'm waiting for something about a taxi." > >> > >> because the quantification is within the abstraction: > > > >Was that actually established? I don't remember that. > > I think it was established that {tu'a lo plejykarce} meant > {le nu/du'u/ka/... lo plejykarce cu co'e}. If it wasn't then it > should, shouldn't it? It was established that {tu`a ko`a} meant {le su`u ko`a co`e}. But it is not established whether {tu`a lo broda} means {da poi ke`a broda zo`u ... le su`u da co`e} or {le su`u lo broda cu co`e}. > >> mi denpa le nu lo plejykarce ti klama > >> "I wait for the event that there is a taxi that > >> comes here." > > > >That I think is an improvement, though I'm not too happy about > >the {le nu} bit. Which nu are you referring to? > > I don't know, events are tricky beasts. Currently I believe that > not all events happen, so that it is possible to wait for an event > that may or may not end up happening, just as you can want > an event to happen without saying that it will or won't. I've addressed this in another reply to John. The gist of my response is that I don't see why nu is privileged from among other predicates to mean is-(not-(necessarily-(actually))). Happening is essential to eventhood. But not all imaginable happenings happen in this world. So we can define nu as an imaginable happening. But why then is xrula not defined as an imaginable flower? There is then a question of how to distinguish between imaginable and actual events. The objects of our desideration can be imaginable events, but causes, for example, need to be actual events. > >But {lo nu} > >would be no improvement, for it might be that the taxi will > >never come. Better would be "mi XXX zei denpa le du`u lo > >plejykarce....": "I wait for it to become the case that there > >is a taxi that arrives here". > > Yes, perhaps that would be clearer, although I think it would > be too restrictive to say that nu could only refer to events that > actually happen. I don't see why du`u can't serve for these imaginable events. True a du`u is not an imaginable event, but it is easy to define denpa as "x1 waits for x2 [du`u] to become the case", whereas I can't easily think of a good definition of denpa is x2 can be an imaginary event. Maybe "X1 waits for x2 to become actual". > >I say "XXX zei denpa", because "denpa", like virtually all other > >intentional gismu, is defined in a different, and ultimately > >illogical way. The only solution I can see, if the baseline is > >respected, is to abolish the use of these gismu and use > >alternative correctly-defined selbri instead. > > Could you explain why they'd be ultimately illogical? Because logically they involve an embedded proposition, yet syntactically they don't. They mean "x1 brodas it to be the case that...". > Is it still illogical if nu can be used to refer to any event, > whether it happens or not? The crucial thing is that there is one proposition embedded within another. It is then a secondary issue about whether the abstractor is the appropriate one. At the moment, I currently think du`u is apter than nu. > >> mi sisku lo'e plejykarce > >> "I'm looking for a taxi." > >> > >> I don't agree that the conclusion we reached was that > >> the right gadri to use was {loi}, either. I think that the correct one > >> is {lo'e}. > > > >I can't believe you're correct. This is partly because it seems to > >me that the solution must involve a subordinate clause, and > >partly because noone really has a clue what lo`e means. I know > >from experience that when we've discussed it before we basically > >sat around inventing candidate meanings for it. (Same for le`e.) > > Well, this is the way I've been using lo'e for some time now. > The argument filled with lo'e is "curried" out (was that the term?) > and does not intervene in the quantification. {sisku lo'e plejykarce} > is like a new selbri meaning "x1 is taxi-searching". Sort of like a > lujvo, but more explicit. The new gismu list asks for a property in > the x2 place of sisku, though, so I'm using the old definition. > I wouldn't know how I would say "I'm looking for my hat" > with the new definition, if I wore one. I remember you adopting this, but you did so unilaterally; it's not official, though this doesn't inhibit you. (And nor should it; if we emulated your usage, we'd generally end up speaking better Lojban.) But anyway, I would beware of anything that can't be mechanically translated into logical terms. "I am book-reading" would mean "there is a book and I am reading it", while "I am taxi-seeking" would not mean "There is a taxi and I am seeking it". I would be happier if you used lujvo rather than this lo`e method. At least with lujvo we know they can have idiosyncratic, nonmechanical translations into logical terms. --And