Message-Id: <9203082030.AA23426@relay1.UU.NET>
Date:         Fri, 6 Mar 1992 19:10:07 GMT
Reply-To: cbmvax!uunet!cuvmb.cc.columbia.edu!C.J.Fine
Sender: Lojban list <cbmvax!uunet!cuvmb.cc.columbia.edu!LOJBAN>
From: CJ FINE <cbmvax!uunet!cuvmb.cc.columbia.edu!C.J.Fine>
Subject:      Re: Re: Hans Christian Andersen
To: John Cowan <cowan@snark.thyrsus.com>
Status: RO

            >
             >
             > >[Note - I am using "ki'u" and "ni'i" the way suggested
             in a previous
             > >mail on the subject]
             >
             > Allowing an ambiguity in the interpretation of
             .i+BAI+bo is intolerable; and
             > if you can't change the lojban community's mind on this
             one (well, not that
             > most of them have an opinion) by protodictionary time,
             your text will have
             > to be changed back to the traditional inconsistency.

             I have now changed my view on these, as explained in a
             mail I have just posted.  However, I wrote the text
             before my first mail on the subject got answered.

             >
             > >.i ko'a djica lo nolraixli   .i ri mulno
             > >beloka nolraixli be'o jo se zanru beko'a
             >
             > My first problem is that {mulno} is starting for me to
             mean {ba'o}. OK, that's
             > just me. The second is that your use of {zanru} is for
             me vague: I'd rather
             > you state what the princess was approved for. But the
             places of zanru are not
             > unclefted (you can't say x1 approves x2 for x3-ing;
             only x1 approves that x2
             > happen).

             The gi'uste says "x1 approves plan/action x2
             (object/event)" I don't understand quite what is implied
             by saying plan/action, and then putting object/event in
             brackets, but I took the "object" to indicate that the x2
             could be a thing (eg a person), translating the English
             "I approve of him". I now realise that I was confusing
             "approve" and "approve of", but I can't think of how to
             express the latter. I suppose "maptypai" (fit judge)
             might do in this context, and "xaurpai" in some, but I'm
             not happy with them. I certainly don't want any speni-ing
             in the translation at this point.

              Thus, if we suppose marriage (to be approporiately
             reactionary :) ,
             > we need to say {.i ri mulno loka nolraixli .ijo lenu
             lego'i speni ko'a cuse
             > zanru ko'a}. But you lose your {jo}. You'd get it back
             with my experimental
             > cmavo {xe'e}: {.i ri mulno beloka nolraixli be'o joi
             xe'e se zanru be ko'a
             > beifai lenu ri speni ko'a}.

             Please explain "xe'e" - I can't get it from the example.

             > utter princess, he'd say "Jolly Good!", and move off to
             the next."

             I don't think the original does either, to be honest.
             It's really turning on the meaning of "zanru" - I've
             already admitted that it doesn't quite mean what I wanted
             it to mean - further, I was trying to get it to mean
             "accept" too - and I don't know how else to translate
             that either!

             > >gi lorinika nolraixli ku ko'a na se birti .uu
             >
             > OK, having seen the English, I know what you mean by
             {ni ka}, but I still
             > think the phrase {leni ri nolraixli} is sufficient
             (assigning a fuzzy-logic
             > truth value to all predications, like the amount by
             which *the predication*
             > "they are princesses" is true.

             I would be afraid that this would mean "how many
             princesses they were".

             > Yes I know, not all princesses are princesssish; what
             you are really saying
             > is {leni ri mele'eka nolraixli}.
             I don't think you mean the "ka" in there, do you?
             > Still, if you can see nix better, stick to {nika}.
             >
             > >.i ri selkecmlu .uuse'inai ri'a lo carvi .e lo xlali
             viltcima
             >
             > {ri'atu'a}. Actually, the {tu'a} rule is becoming
             irritatingly rote-ish, but
             > what can you do... cause != causer (inanimate !=
             action)

             I haven't seen any discussion on this (was it in JL15,
             that I've still not got hold of?), but I get very
             irritated at all these more-or-less meaningless tu'a's
             everywhere. For the grammar to require a structural cmavo
             to convert a word or construction to a different
             syntactic class is one thing. For it to require an
             operator to convert its semantic class in some ill-
             defined way, is quite another. The rot set in with "la'e"
             (Loglan "lae") - it's got to ridiculous proportions. At
             least the meaning of "la'e" is well-defined. I find
             "tu'a" an occasionally useful tool to assist me in
             achieving a requisite level of vagueness. I find it
             ridiculous to import something as vague as "tu'a" in the
             name of precision!
              I believe it does no injury to the language to allow
             gismu like "rinka" to have their places overloaded to the
             extent of allowing either an event or a causer, and
             accepting that the semantics are slightly different in
             the two cases. If you wish precision, by all means shove
             a "lenu" in, and think about what bridi you want in
             the "lenu" - but "tu'a" is a humpty-dumpty word and,
             while it has a place in the language, it is not the place
             it currently seems to hold.
                            [End of flame - I had no idea I felt so
             strongly!]
             >
             > >.i ni'a flecu lo djacu vi le kerfa .e le taxfu
             >
             > {ni'a} is "below"; {mo'ini'a} is "downwards".

             Thanks. I've never seen a discussion of <tense>, so I'm
             learning it by induction.
             >
             > >.i cusku fara ledu'u ra nolraixli mulno
             >
             > I don't think {ra} works; the referent is *way* back.
             But {ru} could pick up
             > the king. Either anaphorise on appearance (which will
             be hell in unselfconscious
             > prose, I know, all this mess of {goi}), or say {le
             nixli}.
             >
             Yeah, I wondered ....

             > >ni'o co'i le cerna cu preti fofo'a feleli'i fo'a capu
             sipna ge'ekau
             >
             > I feel like {ge'ekau} is a son I sired and no longer
             recognise. Though I
             > force myself to do so comfortably, I aknowldedge that
             this is a perceptive
             > expression.

             I really must get hold of JL15 and read the discussion on
             this. It's another one I've picked up by induction.
             >
             > >.i lu .oi mabla seisa'a selcu'u fo'a
             >
             > A {sei} clause can only have SOV syntax. The above is
             equivalent to {mabla
             > fo'a seisa'a selcu'u}.
             You're right, but your explanation is wrong. The grammar
             says [term ... ] selbri.
             "seisa'a selcu'u be fo'a" would be OK.
             >
             > >.i mi su'eso'uroi .uu ganga'i le kanla ca'o piro le
             nicte
             >
             > *ganga'i*?! Well, maybe, but I don't like it. Lojban
             may never obtain a
             > general-purpose factitive (-ri'a necessitates tu'a [or
             xe'e]), but I don't
             > think -ga'i is it, though I must admit it is plausible.
             I'd just say gangau:
             > {mi gansu lenu lemi kanla cu ganlo} -> {mi kangau lemi
             kanla}. As opposed to

             kanri'a lemi kanla}

             I think that "galfi" and "binxo" are a much underused
             resource, and intend to continue to use them in this way.

             > >ni'o ni'ibo co'i djuno ledu'u fo'a nolraixli mulno
             ki'u lenu fo'a fi le
             > >reno sairdicne jo'u le reno gacykicne cu ganse fele
             dembi
             >
             > {fi le}... strictly speaking, this should be {fi
             tu'ale}; the matresses are
             > not themselves the conditions of sensing.

             See flame above.
             >
             > >.i lo ckaji loka ganse du'ila'edi'u cu nolraixli mulno
             ju'o
             >
             > I think you mean {lo ckaji be loka ganse pedu'i
             la'edi'u}

             You're right about the "be". I'm not convinced about the
             "pe"
             It could be "characterised that much by sensitivity" as
             easily as "characerised by that much sensitivity".
             >
             > >ni'o le nolrainanla goi ko'a galfi fo'a le speni
             >
             > {tu'a le nolrainanla}, to be Lojbab-pedantic; and I
             don't think {galfi} is
             > necessary. {co'a speni} or {spegau} should be enough.

             Fair enough. I was trying to capture the rather archaic
             "took her to wife" ("havde henne til Kone, I think -
             haven't got the Danish here).
             >
             > >.i le dembi pu se punji fi la larkumfa
             >
             > {pu}? I'm never sure about story-telling time, but I
             think you should escape
             > story time into the present (flashforward) before
             saying this {pu} ({pu} for
             > us, {ba} in story time): {.ikibo le dembi pu se punji}

             Good point. I don't think I did this very well - I wanted
             to get the "go'i" into the next sentence, and let myself
             be ruled by that. I think it would be better not to tense
             them, and use something like ".i caca'a selstu ri"

             > Apart from the points mentioned above, you handled the
             grammar and vocab
             > well (I particularly liked the use of {co'i}); the
             style is straightforward,
             > though not as "colourful" as the English translation.
             >


             > In contrast to what I get with Ivan's text, what I get
             with Colin's is a lot
             > of "Huh? Oh, no, that's right." cpana punji, fi'o sefta
             and ge'ekau were
             > examples of that here. This suggests to me (and Colin's
             choppy sentences,
             > almost reminiscent of his earlier "colloquial" piece,
             seem to confirm it)
             > that my stereotype of Colin is more of an "explorer"
             with the language than
             > my stereotype of Ivan is. (To detour: even Ivan's
             counterexample of his
             > being explorative with {fi'a} is revealing. {fi'a} is
             the more natural,
             > less detail bound way of asking ("um, it's just an
             argument"). Most of us,
             > having in mind the answer for the second use of {fi'a},
             would have veered
             > off to {cu'e}.
             >
             > In any case: Colin's tale seems to me to set the tone
             for standard prose;
             > its sentences are choppy and to the point, ellipsis is
             applied but not abused
             > (as I suspect I have done), UI are used moderately
             (neither too few, like me,
             > nor too many, like Lojbab, whose JL articles - I say
             this not having read them
             > for 3 months - read like eyebrow raising exercises :) .
             >
             > It thus annoys me that, despite the fact that the
             above-mentioned points are
             > not critical at all, the tale still comes out somewhat
             dry, certainly so in
             > comparison to the English. I don't think this is
             Colin's fault; I think
             > the problem lies in the language's lack of a culture
             and colloquial usage.
             >
             > I certainly know that I could only do worse than Colin
             has.
             >

             ki'e nik. ckire ko ledi'u velxe'o pinka .inaku do snada
             seme'a mi .i leni leima'a jboselsku cu camvrici kei
             ci'iroi .o'a so'imei

             (Thanks, Nick, for those kind words. You do no worse than
             me. May our (and others') Lojban always be various!)

                            kolin
                                 c.j.fine@bradford.ac.uk