| In a message dated 4/16/2001 9:23:45 AM Central Daylight Time,
biomass@hobbiton.org writes: Without using sets, how can "There are many rats" be said? (The book Literally, {so'i da ratcu} <As I understand lujvo, any lujvo may be defined *W/O* tanru, using be and poi/voi (or je) [assuming that <lo broda je brode> is the same as <lo broda poi brode>, and <le broda je brode> is the same as <le broda voi brode>. This looks inherently implausible, since many lujvo (from many tanru, indeed) are not genus-species (class-subclass) types. A goodly number are pulling up later arguments (and, indeed, subject raised later arguments), for example. Therein lies at least one source of the semantic ambiguity of class. Even within your limits, however, there is the difference between {poi} and {voi} ({je} won't enter much because of the inaccuracy it introduces -- a bigmouse is not both a mouse and big, after all). A shelled insect might be {cakcinki} in the {cinki poi se calku} sense,but a beetle is {cinki VOI se calku} -- the definer's choice and to Hell with anything other than is convenience. <Why does the dictionary have an English gloss (which as I see it is meaningless many times [for example brabloti = ship (?!)]), but not the 'long' version of the lujvo, using poi and be? This would be a 'real' definition, which can include the entire place structure.> A fair question and a good suggestion, if you'd like to carry it out. But it would not always clarify matters completely and the English gloss oftendoes (ship is a big boat -- big enough to carry other boats). Making the constructor think the long form through might make for better lujvo, too. Since tanru are (very) semantically ambiguous, how can we allow ourselves to define language concepts using tanru (e.g. <sumti tcita>, <se steci srana>, etc? Those would mean extremely 'wide' concepts!> Language concepts are extremely wide too. And the narrow definition is in the grammar. <Why the hell does <brivla> mean what it means? How do the twoterms connect, and why would it mean only one word? What's the real difference between a brivla and a selbri, then? I mean, <nu prenu kei> is lo valsi, isn't it?> Skipping a long story of sloppy communication and mutual misunderstandings and misapplied technical terms (in short, the history of Loglan/Lojban technical terminology), this is a hard one to answer -- and the tale istoo long for even a face-to-face. And it is probably too late to try to do it all over again right, but a brief attempt may be ok. {bridi} means "predication," a predicate-centered construction that can be asserted, questioned, applied as a description and so on. At the center of a predication is, obviously, a predicate, which may be expressed in one or several words and with a variety of auxiliary devices. A one-word predicate is a brivla (actually, as the list notes, a selbrivla) and it keeps that status even when it is combined into a more complex predicate in a predication. The predicate of a predication is (automatically) a selbri, whatever it is made up of -- one word, several in an unmarked structure, or in a structure of whatever complexity. And no, {nu prenu kei} is not a valsi but a valsi porsi (porsi lo valsi). It is a potential selbri, thoughnot a functional one as given, but it is not a brivla, since not a valsi. |