Return-Path: Received: from SEGATE.SUNET.SE by xiron.pc.helsinki.fi with smtp (Linux Smail3.1.28.1 #1) id m0tBqby-0000ZSC; Sat, 4 Nov 95 23:58 EET Message-Id: Received: from listmail.sunet.se by SEGATE.SUNET.SE (LSMTP for OpenVMS v1.0a) with SMTP id 91C2A0AD ; Sat, 4 Nov 1995 22:58:50 +0100 Date: Sat, 4 Nov 1995 13:47:36 -0800 Reply-To: "John E. Clifford" Sender: Lojban list From: "John E. Clifford" Subject: bal}Nirely, three dogs, on the web X-To: lojban list To: Veijo Vilva Content-Length: 5255 Lines: 122 Another thread reminded me that there is a rafsi for _da'a_, "almost all, all except...," _daz_. Although the metaphor is not very good, it suggested _dazypencu_ for "almost touch" and _daznalpencu_ for "barely touch." I think the two are related this way, one just shy of the positive side, the other just shy of the negative, thoubh I admit that the predicates seem wrong way round. This numerical may not be the right one either, but perhaps there is a a better one, "approximately," say. But then, a numerical does not seem the best way to do it, nor does a modified brivla as opposed to a modification elsewhere in the hierarchy of the sentence. The numerical could not perform this outer modifying function effectively, if at all. So that suggests something in the tense/modal area, which seems right for other reasons as well: the connection with the infamous inchoative aspect, for one thing ("I barely made it" is clearly related factually to "I was on the verge of not making it but pulled it off" and correspondingly for "almost"). And that, of course sends us to predicates that take event sumti and the resulting mare's nests (about references and raising and what have you). But that does seem right (and the clear exposition from and and xorxes of the ways that an "almost" might expand makes the rightness even clearer). I do not know whether _jibni_ can take an event second term now, but I do not think there is anything against the permission to do so, unless there is another gismu or lujvo occurs to someone to take on the ask. I am sure that _jibni_ does not provide a good guide to an unused cmavo to take on the reduce form but the cmavos' relation to brivla has always been pretty tenuous. Note though that I think only one brivala and one cmavo are needed, the "almost" one as I now am thinking, with the "barely" handled by negation -- in the embedded event description or _nai_ attached to the cmavo. This does seem backwards to me, the event that does not occur is affirmatively expressed, the one that does negatively. Perhaps a better image would be that of the glancing blow, the spam (spasm? spaz? tempus: tense:: spatium:x) marker for tangential motion, perhaps. But then the parallelism is lost, for "almost" is not a tangential blow on not doing something but more like the missed swing, a fly-by as contrasted with a touch-and-go. And I do not know whether there is a gismu brivla for either of these nor a cmavo for the second. On the whole, I like the _jibni_ (and related cmavo) solution best. Back to the three dogs. We are pretty much agreed that _ci (da poi/lo/-) nanmu cu pencu ci (etc.) gerku_ involves three men and from three to nine dogs (whatever else the different forms may involve). Further, we agree that, however we get the dog quantifier in front of the man one, the result will only be to involve only three dogs but now three to nine men. We also agree that the problem that there is no way to get both of the numbers fixed absolutely, rather than one being relative to the other, is unique to the two-or-more-number cases. As long as one quantifier is a unversal or particular, we can free either quantifier from the influence of the other by getting it to the front at least (universals are always free from the influence of the other, whatever and wherever that other be). Indeed, this freedom is the main reason for fronting things, matters of elegance aside. I wonder, then, whether the afterthought fronting marker proposed a while back (does anyone remember what it was? xa'a?) might not be better thought of as an independence marker, so that it might be applied also in the case of two or more number quantifiers to achieve the effect that mere fronting will not manage. Since universals never need this effect, the marker could be kept as a literal fronting for them, for afterthought fronting for them is fairly common, though less so than independence moves for other quantifiers. So, for three men and three dogs we might have _ci nanmu cu pencu xa'a ci gerku_ (or variants). I have never seen the Lojban Web page (I just got Navigator installed and never learned the raw commands for some of this stuff), but if it does -- or ever did, for that matter -- tell newbies to send stuff to Lojbab, it should stop it immediately. Lojbab has other functions to fill, ones that only he can fill (in the real world of time and money, anyhow), while checking exercises is a job that any of several dozen people can do pretty well (and the community of them all can do excellently -- asa witness the jobs that it does do). It might be a good idea to get up a set of answers for the minilessons, since none exist, but these would still have to be handled sensitively, since many things other than exactly what is on the answer sheet -- even if all the active proctors in the community got an answer for each question on it -- would be good Lojban answers. And surely we can be as warm as the Elgin dyke brigade (whom I admit to finding neither warm nor dykey and not very clear on Elgin either), though the pseudoevangelical chumminess of some Esperantist prostelytizers may be beyond us seething rationalist types. pc>|83