[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
John: > I finally read JCB's two articles on gadri posted to the Loglan site. > These date from 1995, long after the split, and clarify very thoroughly > just what JCB meant by all his distinctions. If only he had figured > this stuff out before 1988, Lojban would have been much better off. > > JCB has five gadri (I'm going to use Lojban jargon and Lojban cmavo, to > avoid confusion despite the anachronism): le, le'i, loi, lo'i, lo'e. > He uses quantifier+brivla in the same sense we do, and consequently > dispenses with lo. He has no analogue of lei. But what does he mean > by these things? > > Le as is well known is Distributive when it's plural; when it's singular, > of course it's Singleton. (JCB's term is "multiple", but he means > Distributive.) So far so uncontroversial. > > In the first essay, http://www.loglan.org/Articles/sets-and-multiples.html > , JCB tells us that le'i and lo'i are about Collectives, not mathematical > sets, and that this is what he has *always* meant by the term "set" > in his writings! (JCB doesn't believe that mathematical sets deserve > a gadri; in particular, the "empty set" is not a useful notion for him.) > > In the second essay, http://www.loglan.org/Articles/sets-and-masses.html, > JCB corrects several places where he had used loi to represent Collective > in _Loglan 1_. (The online edition has the corrections made already.) > So loi is *not* Collective, except by error. What is it then? Well, > it is and always has been Kind, he says! And he conflates Kind with > Substance, on the grounds that it is all one whether you myopically > singularize all birds into Mr. Bird, or goo-ify all quantities of water > into Mr. Water. This also explains why he doesn't have lei: it's not > too useful to talk about the Kind of something subjectively defined > (= specific). His lo'e is CLL-lo'e, since he has loi for Kind. > > I suppose it is too much to ask that Lojban return to its roots in > this fashion. But how bad would it be to conflate Set with Collective, > and Kind with Substance? I'm half converted already. Conflating Set with Collective would be a good thing. Conflating Kind with Substance is not foolish, but still not a good thing, for two reasons. The first reason is that it obscures the distinction between Mr Bird and Mr Bird Goo -- Mr Brick Object and Mr Brick Stuff. The second reason is that it obscures the distinction between "some paint, all of which is red" and "all paint that is red". However, conflating Substance with Collective (contra JCB's system) is a good thing, provided that there is a way of marking null cardinality. This was xod's idea -- that the difference between Substance and Collective is one of cardinality (bcs Substance lacks cardinality). This has the virtue of restoring some sense to Lojban Mass, and to Lojban's neutralization of the singular/plural distinction (Collectives of countables have cardinality su'o; singulars have cardinality pa; plurals have cardinality za'u). I agree that JCB-lei (e-Kind) is dispensable. And his le could be replaced by role'i. But the JCB system is certainly greatly superior to Lojban's, especially if you subtract the Kind--Substance conflation. I think our discussions have proved that at minimum, only 3 gadri are indispensable: Group, Kind, Specific. (Plus an extra gadri for Named, if Namedness is to be marked by a gadri, which logically it needn't be.) --And.