From cbmvax!uunet!CUVMA.BITNET!LOJBAN Sun Mar 8 18:43:27 1992 Return-Path: Received: by snark.thyrsus.com (/\==/\ Smail3.1.21.1 #21.19) id ; Sun, 8 Mar 92 18:43 EST Received: by cbmvax.cbm.commodore.com (5.57/UUCP-Project/Commodore 2/8/91) id AA17786; Sun, 8 Mar 92 14:39:46 EST Received: from cunixf.cc.columbia.edu by relay2.UU.NET with SMTP (5.61/UUNET-internet-primary) id AA06372; Sun, 8 Mar 92 11:36:53 -0500 Received: from cuvmb.cc.columbia.edu by cunixf.cc.columbia.edu (5.59/FCB) id AA24073; Sun, 8 Mar 92 11:36:54 EST Message-Id: <9203081636.AA24073@cunixf.cc.columbia.edu> Received: from CUVMB.COLUMBIA.EDU by CUVMB.COLUMBIA.EDU (IBM VM SMTP R1.2.1) with BSMTP id 1066; Sun, 08 Mar 92 11:35:13 EST Received: by CUVMB (Mailer R2.07) id 7393; Sun, 08 Mar 92 11:34:39 EST Date: Sun, 8 Mar 1992 20:10:33 +1000 Reply-To: cbmvax!uunet!MULLIAN.EE.MU.OZ.AU!cuvmb.cc.columbia.edu!nsn Sender: Lojban list From: cbmvax!uunet!MULLIAN.EE.MU.OZ.AU!cuvmb.cc.columbia.edu!nsn Subject: Proceedings of first phone game X-To: lojban@cuvmb.cc.columbia.edu X-Cc: nsn@ee.mu.oz.au To: John Cowan , Eric Raymond , Eric Tiedemann Status: RO As you know, the phone game involves people being given a phrase, and trans- lating it into or out of Lojban, passing the result to the next person, until the circle is completed. Right now, the circle contains just three people (we'd welcome any new participants). Here's what happened: Phone Game: Phrase 1: "Graff" The original phrase sent out was: "Viceroy Graff, when you talked to him about veterans' pensions, the treasury is empty" (General Makriyannis, 1797-1864) Unfortunately for me, this sentence makes much more sense in Greek than English. What it essentially means is something like "When you talked to Viceroy Graff about veterans' pensions, the treasury would magically be empty" or "the treasury would be empty (that's what *he* said, but I know it wasn't)" Colin understood the phrase as I described it: la graf. poi vipturni zo'u faulenu kansa beri casnu da'i lonupu'o jibysti dikle'i filo prusonci keikei le jecydinsro cu kunti .uanaiti'e (For the one called Graf such that s/he is a deputy governor: (In the event that (someone) accompanying him discusses (suppose) some event of having not yet jobstop-regular-paid to some past-soldiers) the state money store is empty (huh? well, that's what I've heard). Close to how I'd do it, with two exceptions. First, {.uanai} seems inapprop- riate; what's really required is "that's what *he* said, but it's not true". I originally had {.ianaiti'e}, but perhaps it's not strong enough: maybe {je'unaiti'e}? The second fault is the insertion of {kansa be ri}. It was superfluous: if you're talking to him, you're with him. (Colin comments: This was the only way I could think of of saying "discussing with him": "casnu" takes a plural set! On reflection I suppose "zo'e ce ri casnu" might have been clearer, but stylistically I avoid "zo'e" whenever I possibly can.) It was also confusing, as Sylvia showed: If Graf, the vice-governer, were to discuss with himself: if unemployment benifits were paid to vetrans, the state treasury would be empty Sylvia does not actively participate on lojban list, so she doesn't neces- sarily subscribe to the idiomatic modes of expression developed there. The {da'i} was understood; the prenex (the zo'u phrase) was dragged back into the main sentence, which is fair enough. {lenu kansa be ri} was interpreted as {lenu ri kansa be ri}: talking to himself. Since {kansa} looks odd in context, this is not an unreasonable presumption. The distinction between retirement and unemployment (which in cmavo terms is mo'u vs. co'u or de'a) has been lost. I think it nicer to convey unemployed as {jibni denpa} - work-pause, or work-wait. But this interpretation of jibysti is not unrea- sonable. The {.uanaiti'e} is lost. The ti'e refers specifically to {kunti}, and not to the whole sentence, so I think it should have been made explicit. Still, the fact is that the unreality was insufficiently conveyed by the first participant. Perhaps {le dinysro cu cu'u kunti}? - "the treasury is (it is said) empty". (Colin comments: At first sight, that is sheer genius. We (ma'a) don't make enough use of "cu BAI selbri". But on reflection, I'm not sure it does it. I think it means something very tightly attached to the selbri: something like "is-empty-as-expressed-by" - meaning the kind of emptiness or way in which emptiness is characterised.) I'm not sure the "if benefits *were* paid" is correct. The {da'i} applies only to the word it is attached to. Furthermore, {fau} is explicitly non-causal, whereas this translation implies causality. The original sentence implies the treasury is empty before any payment, and that such payment is thus impossible. I think it my fault for not pointing the full extent of the Greek idiom out, and will be ritually castigated on Wednesday. On the other hand, Colin *had* used {pu'o}, and I would have interpreted {casnu} as contemporary with {kunti} by default. Mark had: lenu la graf. noi vipsi cmagu'etru cu casnu vo'a lejei loi na'e gunka jdini cu dunda loi pu sonci kei kei cu rinka lenu le cmagu'e dinsro cu sorcu noda (The event that {Graf who is a deputy small-country-governor discusses with him/herself whether or not it is true that the non-working money is given to the past soldiers} causes that the small-country money-store stores nothing.) (Mark, having seen the original, suggests: cu'u la graf. poi vipturni zi'epoi ca ve preti lesi'o [pensions given to verterans] ku'o le jecydinsro cu kunti {or whatever}. See? Using the relative clause to indicate that it was Graf, when he was asked! Not sure about the whole thing, and especially not sure about the {cu'u}, but you get the picture. Nick: I must say this makes a lot of sense. Mark still: When it got to me, in English, the phrase was in exactly the form that Lojban would never leave it: namely, it was two if's with one then, and there's no telling which if is thened ("If Graf, the vice-governor, were to discuss with himself: if unemployment benefits were paid to veterans, the state treasury would be empty"). I'm not sure it's a complete sentence, but one way or another it's ambiguous. Nothing wrong with that, English is ambiguous and the English phases of the game should be colloquial, not stilted. But of course, the Lojban can't reflect that kind of ambiguity. Which entailed the state's treasury's being empty, Graf's self-discussion, or benefit-paying? I picked the first, sort of at random, though the second probably made more sense. Turns out it was closer to the meaning of the original. I *did* like my use of {le dinsro cu sorcu noda}. The English "empty" is so broad... The Treasury building undoubtedly held workers, the safe probably had drawers and cabinets. {kunti} is certainly not wrong, but I think {sorcu noda} is more precise.) That should be {se dunda}. And {lejei} is not enough; at least {cu.ei} (whether it should be true that...). A bit of deontic logic won't go wrong (deontic logic is based on "shoulds" and "oughts"; they're notated by underlining a yes/no predicate symbol.) (Colin comments: I don't agree. *As a translation of #2*, "lejei" is perfectly OK. Besides which, you can't put UI there, unless you're talking about the *speaker's* deontology. I respond by pointing out the debate on the "knower" of {kau}: in {la djan. djuno ledu'u ri klama zo'ekau}, is it that John knows where he's going, or that I know where he's going?) An additional ambiguity in Sylvia's phrasing is resolved upsettingly: Mark has the discussing, not the payment, emptying the treasury. Thus we started with "Viceroy Graff, when you talked to him about veterans' pensions, 'the treasury is empty'" and ended up with "Vice-Governor Graf's discussing with himself whether pensions are giving away veterans emptied the treasury." Ouch. Graff was stingy and a liar; Graf is fiscally irresponsible and a surrealist. The moral: * I should pick less ambiguous sentences. * Extraneous words with ellipsis cause more confusion than clarification. (Colin comments: Not sure about this one. My penchant for ellipting x1 (particularly in a case like this when it's an indefinite anyway) caused some of the problem. But what is interesting is that I used the construction I did because the brivla had a particular place structure - and the next person construed the construction as meaning more than that. It is possible that with more practiced speakers (than any of us!) this sort of misunderstanding would not arise.) > * Tense might not be that optional. (Colin comments: True. Looking at in the light of what followed, I did need a tense on the last bit (or maybe a "ca'a"). As it stands, there is no way to tell whether Graff would have said "But the treasury is empty" or "But the treasury will then be empty".) * Truth value is very tricky, and UI might be insufficient to convey all the subtleties; metalinguistic comments may be required. Any comments are welcome. For example, is this level of discussion too involved? Is the typographic presentation too muddled? ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Nick Nicholas, Melbourne Uni, Australia. nsn@{munagin.ee|mundil.cs}.mu.oz.au "Despite millions of dollars of research, death continues to be this nation's number one killer" - Henry Gibson, Kentucky Fried Movie _______________________________________________________________________________