Date: Sun, 19 Apr 1998 05:25:44 -0400 (EDT) Message-Id: <199804190925.FAA26160@locke.ccil.org> Reply-To: Logical Language Group Sender: Lojban list From: Logical Language Group Subject: Re: Lojban ML: Syllogism and sophism X-To: jorge@INTERMEDIA.COM.AR X-cc: lojban@cuvmb.cc.columbia.edu To: John Cowan X-UIDL: 857ed5eb5aa20c2d656c0d056ec06c25 X-Mozilla-Status: 8011 Status: RO Content-Length: 7403 Lines: 164 First of all, re your unit suggestions in the first message, I find your ideas acceptable (this may be the first time in a long time). I still prefer the gradu approach, but if you want to make explicit unit lujvo, your approach is a suitable convention. %^) >The same happens with {klani}. I might be interested to talk >of a quantification of John or one of Paris. The scales would be different, >even if the value is the same. The scale would be different because >the ce'u would go in a different place, just like the properties above >are different, even though the underlying event is the same. > >>(e.g., if John had travelled to Paris, your sentence would >>become >>la paris. klani li pasexa le ka se litru la djan lo ki'otre > >Not exactly, because {litru} doesn't have a destination place, >--More-- >you have {lo ki'otre} in the place of the vehicle. I see. I intended that paris be part of the route, so I needed ... leka lo ki'otre joi ce'u se litru la djan, or something like that. >But the principle >is correct: > > le vi karce cu klani li pasexa le ka la djan litru lo ki'otre [be >ce'uxire] [ce'uxipa] > This car is quantified as 176 in the scale of kilometers travelled by >John > driving it. I haven't the vaguest idea how you determine the links of those two ce'u terms. I mean I can see it using careful word-analysis, but it seems unwieldy to resolve. Might I suggest that pe vo'a and pe vo'e would be clearer than the subscripts? But I still say that it is the travelling that is measured, and not the car or the traveller. The car is no more a quantity than John is or any of the other places, but it takes all of them in order to make the quantity meaningful. You are putting the focus on the objects, whereas I think the essence of Lojban is to focus on the relationships as gestalts, insofar as it is possible. > la djan cenba le ka djuno makau > John changes in what he knows. > > la djim cenba le ka te djuno makau > Jim changes in what is known about him. Well you are using makau here instead of ce'u, I note, so there is some difference. I am never entirely sure why you do one instead of the other. But what if the change is in what John knows about Jim. Both of the above convey part of the picture, but not all of it. I want to say that leka la djan djuno fi la djim cu cenba JOhn's knowingness of Jim changes. But then the problem is what to put in the x2 of cenba. Sloppy thinking Lojbab wants to be able to focus on the x2 of djuno as what is changing, so my first guess would be leka la djan djuno fi la djim cu cenba leka djuno ce'u But I suspect you will object for some reason I will not understand %^). >>One solution >>then would be to just use the leka abstraction in x1, but this leaves the >>x3 place of klani unfilled and possibly superfluous/redundant. > >In my opinion that would not be right. You could put the lenu abstraction >--More-- >in x1, but not leka. leka is a scale with open slots. That scale does not >get quantified. I think then that you have identified the nature of leni. leni talks about properties without necessarily there being open slots. It is working however on the properties of the relationship, unlike nu which is a gestalt on the relationship happening as one or more instances (possibly with a particular time structure as would be shown by the subtypes of nu). Whenever you use nu and are not talking about the relationship as a real or potential event, but rather as a shorthand for the properties/aspects of the relationship, I think you are using the wrong abstractor. You seem to have settled on ka for solely that relationships with open slots, and it isn't even clear you really care about the property nature of the abstraction, merely the openslottedness. Then you use nu for any abstraction that has no open slots. At the very least this seems to short cirtcuit the power of abstractions to manipulate relationships as entities. (While I do not pretend to understand the requirements of 2nd or higher order levels of predicate logic, if Lojban has a way to handle them, it will likely involve some form of abstraction) >Just like it is not the property of knowing that changes, >but John that changes in that property, or Jim that changes in the >property of being known about, or the event (nu) of John knowing >about Jim that changes in what is known. Ah I see you dealt just with my question. But the event of John knowing about Jim does not change. That is nu clauses are not openslotted. "lenu la djan djuno fi la djim." means "lenu la djan djuno zo'e la djim" for some specific value(s) of zo'e. Just becuase those values are unspecified does not make zo'e any more a variable than la djan. or la djim. It is almost as if you could say that lenu la djan djuno fi la djim cu cenba leka makau djuno John's knowing about JIm changes in who knows (about Jim). But if such a change takes place, then John no longer necessarily knows about Jim. Similarly, if zo'e changes value, then the nu relationship has become a different relationship and is no longer necessarily true for the old value of zo'e. This is like fingernails on my mental Lojban blackboard it seems so wrong. (I wonder how to translate that into Lojban without necessarily preserving the idiom., but preserving the picturesqueness %^) >Just like it is not the property of knowing that changes, >but John that changes in that property, or Jim that changes in the >property of being known about, or the event (nu) of John knowing >about Jim that changes in what is known. But it is not the event that changes because then it is not the same event. It is indeed the values (but not the nature) of the abstract relationships between the various places that is changing when a value changes. At least this is what I envisioned "property abstract" to mean. Put in colloquial terms, a change in nu is a change in the event THAT he knows, whereas a change in ka is a change in HOW he knows (characterization of the knowing, not means of knowing),a nd a change in ni is a change in HOW MUCH he knows (degree of relationship, not a quantification of x2 of djuno). As my parentheticals show, djuno again is an abysmal example because of the perversions that can come about in English translation. If the event of John being blue changes then he is no longer blue, or at least there is a change in the time signature (maybe he is intermittently blue?). If the ka of John being blue changes, then something about the relationship changes which could include a variation in one or more of the values. If the ni of John being blue changes then I would expect that the degree to which he is blue is changing (how much of him is blue, or perhaps how close to the archetype of blue that he achieves). And if the si'o of John being bliue changes, then I suspect that means that something about the nature of what it means to have John be blue that is changing. How you say all of these with cenba, I do not expect to resolve to Jorge's satisfaction %^) >If you put the property in the x1, >then the change has to be in a property of properties, I don't think that should have to be the case. >and the >scale has to be a scale that associates a number to an >open property. I think that all scales associate numbers and properties. lojbab