[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
| As theology shows, hypothesis without exegesis is the short route to heresy (the long route is *with* exegesis). So, I have been speculating about the meaning of {ka ko'a broda} without thoroughly examining the cases in CLL. This has been, in part, because there are so few cases in CLL, that they were unlikely to reveal much (and are usually taken to be accidents or mistakes), and, in greater part, because I am eitehr lazy or conceited to the point of preferring armschair speculation to research. Anyhow, I have now looked at the two examples that I can find (the indexing being what it is) of {ka ko'a broda} in CLL: 11.4.4 (259) {le ka do xunre cu cnino mi} glossed as "Your redness is new to me" but equated with {do cnino mi le ka xunre}, "You are new to me in redness," either to be said to someone who comes back from the beach with a sunburn (he did not have when he left). and 11.12(called 13).1,2 (269). {le ka la frank ciska cu xlali} "The quality of Frank's writing is bad" (More examples would be appreciated, if you come across them These examples pose problems. 1. they occur only in {le} forms, making it unclear what {ka ko'a broda} means alone, separated from the selection and specification process of {le} (we'll assume the non-veridicality does not apply). 2. the glosses given are ambiguous at least, even for what seems to be the intended reading. None of which is helped much by the official lines-of-chat: "x1 is quality/property exhibited by [bridi] " and "x1 is a property of [bridi]" even assuming the uncertainty about what "[bridi]" means is resolved -- and the related issues of what "property of" and "quality/proeprty exhibited by" and, yes, what the meaning of "is" is. Nor by analogies with other abstractors, most of which are in as bad shape or worse, nor with {ka ce'u broda}, which has become suspect (more than usual) in this process. The whole "is quality of/property exhibited by [bridi]" seems the most fundamental issue, without a definitive answer to which everything else is unsupported. Actually, the old "'is' of identity" - "'is' of predication" conflict does not play too great a role here compared to the "quality of" problem, which more or less absorbs it. Is a quality of [bridi] a quality mentioned/highlighted in [bridi] (and [bridi] thus something something linguistic or closely related to that) or is it a quality which [bridi] has (and [bridi] could be anything. If "quality of" takes the second meaning, then it is clear that [bridi] cannot be a sentence, since things like "short" or "grammatical" are pretty clearly not the right sort of things to be ka ceu/ko'a broda. Even propositions do not make much sense here, for, again, "empirical," "true" and the like do not look right. So, for the "[bridi] has as quality" reading ({se ckaji le su'u [bridi]} or pretty close), only events ({nu} for {su'u}) will do, the referents of sentences. And then, of course, {ka} functions as a real predicate, covering usually a large set of properties. And {le ka [bridi]} selects the ones of immediate interest. This was what I was working with trying to make some sense of {ka ko'a broda} without checking the actual examples. The actual examples, aside from being rather opaque themselves (the glosses look more like {nu} than {ka} or even, in the second case, {le se ciska be la djan}), might just bear this interpretation: some or all the properties of a event of you being red might be new to me and some or all of the properties of events of Frank writing might be bad. But which properties would this make the most sense for? "The quality of Frank's writing is bad" looks frankly like a case of the "is" of identity rather than predication (which it clearly is in Lojban). That is, even if we ignore that "the quality of Frank's writing" seems naturally to refer to the quality of what Frank writes (and "writes" in the {finti} sense rather than the mechanical one, at that -- as well as "quality" in a normative rather than a descriptive sense), the residual reading of the gloss is still that "bad" describes Frank writing, i.e., is a quality of that event, not that it describes a quality of that event (for example, Frank writes very slow and extreme slowness is a bad thing/quality for a person's writing to have). In the first example, what is new to me about you being red is mainly just that (this is really hard not to read as {nu} rather than {ka}): that it involves you being red (as opposed to your usual, familiar color -- and, perhaps, as opposed to Molly -- who always falls asleep in the sun -- being red). While none of these problems raise insurmountable obstacles to the "qualities had by the event described " reading of {ka ko'a broda}, they set the ladder pretty steep and tall. We cannot dismiss this reading entirely, since it is shown in 12.2 to be parallel to {ni [bridi]} where [bridi] apparently has to refer to an event (unless 12.2 is one of those bad jokes, like "Washington was the father and capital of this country"). In any case, we can hope that the "quality-of as quality-mentioned" reading -- and [bridi] as a sentence or proposition -- will give a more useful and coherent theory. Right away with this second interprettion, one problem is solved: the property concerned is that referred to by the selbri -- the brivla or tanru at the center of the sentence. But then, isn't le ka do xunre just the same as le ka ce'u xunre? Apparently not. We are told that {ka dunda} can refer to a number of properties, depending on which places are filled with regular sumti (including {zo'e}) and which left open for {ce'u}. While the issue of whether {do} makes for a different property than {ce'u} is never explicitly addressed, the fact that le ka xe'u dunda zo'e is said to be different from le ka zo'e dunda zo'e and both from le ka ce'u dunda ce'u seems to generalize that way. [The problems in that section are handled by the convention on {ka} that separated it from {du'u} when {ce'u} was involved.] But what, then, is le ka do xunre different from le ka ce'u xunre? What seems to fit in is that le ka do xunre is a particular red, the color you are when you are red -- which may (or may not) be different from the red I am when I am red (le ka mi xunre). The "may be" is needed to keep leka do xunre different from le ka ce'u xunre; the "may not be" is needed to keep {la djan frica la bab le ka xunre} from being a trivial claim, since it comes down to {le ka la djan xunre cu frica le ka la bab xunre}; it is only important to claim x is different from y if x could be identical to y. Notice, nothing has been said here about the respects in which various objects' reds might differ -- whether as colors (hue, intensity, saturation) or some other feature (flashing on and off, say). And, when we move away from colors, the possibilities multiply. But then {ka ce'u} is a rather different thing from {du'u ce'u}. {le du'u ce'u broda} names a function which returns a proposition for each replacement of {ce'u} by a sumti -- a propositional function in the Logicians' sense, a property in extension, as it were. {le ka ce'u broda} is a function that gives not propositions but other properties -- apparently in intension, since they are not function themselves. The {ka} form is in many respects more like a traditonal property, at least more abstract and more directly related to individuals. These differences need some thinking on. Clearly, I think this latest interpretation is the way to go. It does have an occasional odd feature, however, when applied to some of the analogous abstractions. For example, if [bridi] is a propsoition or a sentence, one or the other of the obliques places of {du'u} become redundant, barring some messy further complications. Comments? |