From cowan@MAGPIE.LL.PBS.ORG Ukn Aug 2 18:57:11 1993 Received: from YALEVM.YCC.YALE.EDU by MINERVA.CIS.YALE.EDU via SMTP; Mon, 2 Aug 1993 18:57:10 -0400 Received: from CUVMB.CC.COLUMBIA.EDU by YaleVM.YCC.Yale.Edu (IBM VM SMTP V2R2) with BSMTP id 3883; Mon, 02 Aug 93 18:56:03 EDT Received: from CUVMB.COLUMBIA.EDU by CUVMB.CC.COLUMBIA.EDU (Mailer R2.07) with BSMTP id 7748; Mon, 02 Aug 93 18:56:49 EDT Date: Mon, 2 Aug 1993 18:44:11 -0400 Reply-To: John Cowan Sender: Lojban list From: John Cowan Subject: Re: [long] Re: On the tense system X-To: lojban@cuvmb.cc.columbia.edu To: Erik Rauch In-Reply-To: <9308022108.AA25842@nycenet.nycenet.edu> from "Jorge LLambias" at Aug 2, 93 05:03:30 pm Status: RO X-Status: Message-ID: <_4WUcQNe6fF.A.7dH.oy0kLB@chain.digitalkingdom.org> la xorxes. cusku di'e > You say "pu'o is the time before the event" > > I think this means "a point in time before the event", where the point > can be taken as a co'i type of point, but it doesn't mean > "all of time, from time immemorial, to the begining of the event" > The same goes for the other two. On the contrary. It does in principle mean "all of time to the beginning of the event", because pu'o/ca'o/ba'o represent spans of time, not points of time. > (If the definitions of pu'o and ba'o were reversed, as I think would > be natural, this would be: mi ba'o damba. = I'm on the verge of fighting > which shows clearly that the fighting is in the future) I agree that such a reversal would be more perspicuous, but it's too late now. This is of a piece with the PU vs. BAI confusions. > In your picture, you write the word at the place of the reference point, > and use the same graph to show the three tenses. I prefer to separate > them because some parts of the graph are irrelevant to some tenses, eg > the end of the event is irrelevant to pu'o, the beginning is irrelevant > to ba'o, and both boundaries are irrelevant to ca'o. Both boundaries are relevant, not irrelevant, to "ca'o"; they are precisely the boundaries of that span. > For co'u the existence of a natural ending may or may not be relevant. > For za'o I'm not certain whether I got the 0 in the right place, but > from the example in the paper it seems right: > > ____le xirma ca za'o jivna bajra___ > The horse [present][superfective] compete-type-of-runs > The horse keeps on running a race too long. Again, "za'o" is a span, the span whose boundaries are the natural ending point (co'u) and the actual ending point (mo'u). > 2) the semantics of the ZAhO as sumti tcita is unnecessarily different > from that of the rest of the tenses. This I think is important, and this > is the cause of the switched ba'o and pu'o. You will perhaps notice the version number on the published tense paper, namely 3.11. One reason for this value is that I tried many times to make ZAhO consistent with the rest of the paper. I could not; usage had already settled several points, consistently or inconsistently. I hold consistency to be a virtue, but not the only virtue; sometimes usage wins. (In the end, usage always wins, and we are very close to the end now.)