[YG Conlang Archives] > [engelang group] > messages [Date Index] [Thread Index] >
In a message dated 6/10/2002 7:58:44 PM Central Daylight Time, a-rosta@hidden.email writes:First of all, I deliberating spoke of Lojban, i.e. dating from 1987 Agreed that that was a group more interested in Logic. But there were remarkably few decisions left to be made at that point: the broadened negation notion was largely linguistic (Horn's book), the expanded tense system was as much linguistic as logical. And so on -- the goal at that point (left over still from JCB) to make anything any language could say easily be sayable in Lojban in a non-off-putting way (roughly). Aside from the disputes about existential import and your problems about {makau} and {ce'u}, I can't think of any logical issues that got taken up anew or for the first time in Lojban (and the first of those was bruited about already in Loglan). <Second, I get the impression that even if the connection with logic was originally just a good solution to a problem in designing a Whorfian experiment, it soon came to be an attraction in itself, and if you look at the history of major figures involved with Lojlan/ Lo**an during its history, most have been far more interested in logic than in Whorfianism.> Well, I will admit that Whorfianism was rarely the dominant concern -- even for JCB -- but I don't think that Logic (in any sense narrow enough to be meaningful -- and certainly not the goals you have set) was the alternative. I think computer control or MTIL or archiving or even IAL -- any of which might, I suppose, incidentally involve logic -- were more important that logic per se. But that Lo*an is a logical language essentially only in the syntactic sense has been consistently maintained (right up there with "unambiguous in the syntactic sense"). If someone wants to read more into it than that, we warn them, but let them go on with their enthusiasm. <I see no evidence in the many discussions about Lojban, or the few discussions about loglangs, of debates about what logical or semantic form should be being a stumbling block. And I think it is a matter on which it would be relatively easy to reach consensus. Relative to how easy it would be to reach consensus on pretty much anything else about language design, that is. I realize that if professional logicians themselves were involved, then consensus might be harder to achieve, but not if linguists (or your average engelanger) were involved.> I admire your optimism. But... I've been in the linguistic end of this since 1958 and the logic end for a couple years more. I've seen new structures proposed in both arenas (which have become largely indistinguishable over a large part of the project). I have not seen a proposal definitely refuted and dropped by its adherents. I have seen precious few concurrences either. In logic the problem is roughly 2500 years old, with an upsurge since 1918 or so. In linguistics it is about 50 years old (but not quite, I think -- not much happens of any value before Chomsky). And then there is neuro-physiology/psychology/whatever, that got into the act less than 20 years ago (if you don't count the relatively respectable phrenology of the 19th century). <what I had in mind was how well Lojban (or natlangs, for that matter) do when compared to how well an invented language could do if they actually set out with the goal of being fully baked. It's by this measure that I think Lojban is half-baked> Well, we'll have to see the fully baked language before we can test this one out. I suspect that the extent that it is more baked will be largely directly related to the extent it is less a language and more a formulary. But time will, I hope, tell. Mike: <1. The parse, which determines grammaticality (legal strings of words). 2. The syntactical structure (most of the meaning, but not things like scope), which can be obtained from the parse by removing some of the divisions. 3. The logical structure, which can be obtained from the syntactical structure and the original utterance by various "other means", most notably by inspecting word order.> & has reminded me that separating parse from syntactic structure is not justified, in that the parse does give the syntactic structure of the sentence according to this grammar: an appropriately boxed and labeled string. But I took you to be meaning by "syntactic structure" somethign with a bit more universal appeal that "what this grammar gives," something that showed g-and-d or function-and-argument or some of the categories of your favorite syntactic theory. In that sense, the parse is a long way (though not an unbridgeable --even an automtically unbridgeable --way) from a syntactic structure. Some of the grammar's binary groupings have to be ignored (like "bridi tail" most of the time) with the subsidiaries raised to some higher level. But this can be done pretty mechanically. Some other factors may take more tickling -- but I haven't really tried doing any of this for any syntactic theory. The move to a logical structure should come pretty much from the syntactic structure and may or may not be complicated depending upon what syntactic theory you favor and what you think the logical structure should be like (and how forethoughtful you've been in your syntactic theory toward you logical theory). Many combinations ought to be pretty easy to mechanize without losing connection with the parse. But, again, I nor anyone else that I can think of has really done much with this. <Incidentally, are you familiar with lambda calculus? I am revisiting this concept from my studies of LISP/Scheme. Along with their "lists", as I mentioned to And in another posts, I think that these things might possibly be applied profitably to the parsing of loglangs.> Yes, I learned lambda from Church, but nearly 50 years ago, so I am a little rusty (I don't remember, when I tried just now, how to deal with universal quantifiers. I'll look it up) and I have worked with that abomination LISP. On the good side, everything is explicit and precise. At the cost of being some complex as to be unreadable -- and usually uninterpretable. Just remembering what the head was when you get to the significant tail three pages later is marked worse than academic German at its fullest ("I can't say what he said, he hasn't come to the verbs yet"). There are ways to achieve lambda precision without lambda's logorhea. |