From cbmvax!uunet!math.ucla.edu!jimc Sat Aug 24 19:47:21 1991 Return-Path: From: cbmvax!uunet!math.ucla.edu!jimc Return-Path: Message-Id: <9108242219.AA06670@luna.math.ucla.edu> To: lojban-list@snark.thyrsus.com, A_DENCH@fennel.cc.uwa.oz.au Subject: Translation of The Bladder Slugs of Yik Date: Sat, 24 Aug 91 15:19:01 -0700 Status: RO The Bladder Slugs of Yik Posted 91.8.14 by Alan Dench, Dept. of Anthro, U. Western Australia on conlang@buphy.bu.edu (constructed languages list), reposted by lojbab on lojban-list@snark.thyrsus.com. Challenge is to translate the tense metaphors into Lojban. ------ Inferred physical situation: the Bladder Slugs orbit Arcturus IV, and their orbit decays gradually but inevitably; the rate is under voluntary control to some extent. Offspring ascend to the top of the exosphere by an unknown mechanism, then begin their inward spiral. Metaphors of impact and/or implosion from rising outside pressure are used for the final event -- though physically I would expect more meteoric results. Generalization on tense: insofar as there is a 1-1 relation between time and altitude, the directional properties ga'u-gapru-above and ni'a-cnita-below can be substituted directly where humans would use pu-purci-past and ba-balvi-future. Where needed I have also used vi-"at" or ne'a-lamji-next.to as a substitute for ca-cabna-same.time. in Lojban, both temporal and spatial tenses (directional properties) fit the same syntactic sites and both can be elaborated in the same ways, and hence there is no problem at all making a 1-1 replacement. Comments below on expression problems, therefore, are equally applicable to conventional time tenses. Since this is going to at least one non-Lojbanist, I have put in extra translation data. poi is the structure word linking a restrictive subordinate clause; I mark it RSC in the translations. SSC marks a supplementary or "by the way" clause. EMK is one of various kinds of endmarks. Arguments (analogous to English noun phrases) typically begin with what linguists call a "determiner", but to simplify I have mostly omitted their translations. "le" and "lo" are in this category. The one exception is loi-the* which specifies a "mass individual", one example of which is "the* lion lives in Africa". A. I perceived a large glabbage upperday. mi ganse lo jduboi poi barda ku'o ga'u pa djedi I sense(stimulus) a glabbage RSC big EMK above one day ganse - sense stimulus. I failed to locate a good word for "perceives the external reality" as opposed to "senses a stimulus", which I use here. jduboi - jduli bolci - jelly sphere - glabbage. Parallel diklujvo. ga'u pa djedi: See sentence E for explanation. B. The time for implosion is just below us. le nu mi'o po'afa'e cu ve'a cnita mi'o event we implode "EMK" small-interval below us(you+me) po'afa'e - spoja fatne - explode reverse.order - implode. Abstract diklujvo, somewhat questionable with existing place structure of fatne. Though the English word "time" is used, in Lojban the x1 argument has nothing to do with time. Instead a Lojban event abstraction is used. This pattern occurs several times in this text. C. The pressure increases; the light is dimming; I'm plummeting old. le ni mi se danre cu zenba quantity I x1<->x2 pressed-on increases .i le ni mi se gusni cu manze'a quantity I x1<->x2 illuminated dims [dark increase] .i mi bazo'i no'eselsaifa'u I future-approaching plummet manze'a - manku zenba - dark increase - dimming. Abstract diklujvo except that the places of zenba don't clearly allow this interpretation. zo'i: approaching ( tense). Because the selbri is spatial I have combined zo'i with a time tense "ba" for contrast, rather than with ni'a (below). no'eselsaifa'u - no'e se sarji farlu - scalar-neutral x1<->x2 support fall - plummet (fall unsupported). XXXfa'u is a parallel diklujvo. Lojban allows "le ni se danre" etc., but I prefer a style in which for each phrase as many places as reasonable are occupied explicitly. D. How deep until we fall on dense times. (Original appears to be not a question but a discursive calling the listener's attention to the size of the interval.) condi fa le nu mi'o felda lo denmi deep x1 event we(you+me) fall dense (something that is dense) Again, time is irrelevant; this is translated by an event abstraction. E. Three days above I consumed a large splodj. ga'u lo djedi be li ci ku mi citka lo danlysplodji poi barda Above a day times 3 EMK I eat a (animal)splodj RSC big danlysplodji: I don't have the exact rules for this kind of le'avla but it's at least close to legal. Discussion of the time modal phrase: djedi is defined as "x1 is x2 days long" (ignoring x3 as irrelevant here). At first I was confused trying to make this work in an old Loglan style, but then I recognized that djedi was intended to be used the same way as a -gua!spi unit of measure. (Yes, this guy knows when he wins!) The only problem is that the sentence is placed at the top of an interval 3 days long, but where does the interval start? The place structure lacks an origin. One can reasonably assume, though, that when a unit of measure is used ordinally, as here, the start point is the current tense reference; ra'i-kraji-origin has the right semantics for an explicit origin. It may be useful to list rejected modal phrases. ga'u ci le djedi Above three of the (different in extension) days. ga'u le ci djedi Above the three (different) days. ne'a le ga'u djedi be li ci At the aboveward 3-day interval -- not above the interval. F. In the rarefied days of my youth, I set my life on a helical path. ne'a le za'i mi citno fi'o da'ecma ku At state I young RSCb dense-small EMK mi punji le pu'u mi jmive kei lo pluta poi sarlu I set process I live EMK route RSC spiral da'ecma - denmi cmalu - dense small - rarefied: abstract diklujvo. Again, "days" is irrelevant in Lojban; the modal phrase is translated with an abstraction, in this case for a "state". The bridi restrictive subordinate clause (RSCb) says that "I am young" was low density, which for the Bladder Slugs is clearly a tense metaphor; this is the only past tense in the sentence. (Technically, fi'o is a tense in Lojban.) Lojban has no gismu (defined primitive word) for "helix"; however it does have one for "spiral", which is physically more likely as the Bladder Slug's path. G. The foolish Yik lives like a falling space rock. (2 translations) loi selnatmryika poi bebna cu tarti tai loi ru'u rokci poi farlu The* Yik RSC fool behaves in-form the* orbiting rock RSC fall .i simsa fi le tarmi lo pu'u tarti ku loi selnatmryika poi bebna Similar x3 form process behave EMK the* Yik RSC fool loi ru'u rokci poi farlu the* orbiting rock RSC fall selnatmryika: From natmi = "x1 is a nation/ethnic group of peoples x2" with the Yiks in x2. This is slightly cheating on the meaning of x2. I hope the le'avla is morphologically correct. tarti: x1 behaves/conducts self as x2 circumstance x3. "Alive" really gets applied to the rock, which would be bogus semantically. ru'u rokci: "ru'u-orbiting" is literally a tense in Lojban. I very much dislike the metaphor pattern "space X"; I looked for and found this alternative. simsa: x1 is similar to x2 in property x3. The first translation is quite similar to the English sentence form. But the main point of the sentence is to say that X and Y are similar, not to say that the* Yik behaves, and the second translation, more Lojban style, makes "similar" the jufra selbri. A defect of this sentence is, what is behaving? You have to guess that it's the Yik and the rock. -gua!spi replication rules also can't get past "le tarmi". I wonder if a better x3 might be "fi lo pu'u ta'i tarti". This lets -gua!spi style replication stick x1 and x2 into x1 of tarti, but does this phrase adequately express that the similar property is the form of behavior, not the fact of behaving? H. At darkest bottom, we all meet at the centre. ba le dizrai noi manrai At(time-future) lowest SSC darkest ro ma'a penmi lo fi'e vi le midju all we(you+I+others) meet x1-copy at(space) middle dizrai - dizlo traji - low superlative - lowest. Abstract diklujvo. manrai - manku traji - dark superlative - darkest. Abstract diklujvo. I don't much like any of the choices for "darkest bottom": le dizrai noi manrai the bottom, which by the way is darkest le dizrai poi manrai of bottoms, the one which is darkest le dizrai po'u le manrai the bottom, that is, the darkest place Note with penmi-meet: the Lojban phrase says "we meet each other". I. The aliens, who live for eternity high above the days of our youth, believe the universe is expanding. But according to the great physicist Alblort Einslug, it is merely moving up into its own "past". loi fange noi piro ze'u jmive ga'u ve'u The* alien SSC all long-time-interval lives above long-space-interval le za'i ma'a citno cu krici state we(you+I+others) young EMK believe le pu'u le munje ca'o calze'a process cosmos continuous volume-increase .i ku'i du'o la .alblort einslug. noi banli But known-by (name) Alblort Einslug SSC great le ka skefa'i ku'o le munje nitkla le lofi'e purci property science-discover EMK cosmos ascend x1's past The required meanings are handled uneventfully by Lojban with one exception: "merely". I spent much time (maybe a month on and off) struggling to translate "only", "merely" and "such" in Old Loglan and later in -gua!spi, without success; finally I had to create a gismu. I refuse to repeat that task in Lojban. Somebody else please translate "merely". Diklujvo: calze'a - canlu zenba - volume increase - expand. Abstract diklujvo. skefa'i - saske facki - science discover - scientist. If "science" is defined as "x1 is a scientific fact (about x2)", this works as a transitive diklujvo. nitkla - cnita klama - below go - ascend. Parallel diklujvo, motion word subtype. The reason "below" is used rather than "above" is that x1 is related to x2, and during the motion the x1 (moving object) is below the destination. The following usages are plausible but not backed by known official patterns: piro ze'u: use of the fractional "all" quantifier to suggest eternity. ze'u, ca'o: use of a (space interval) or (aspect) in a (elaborated tense) slot without an actual tense. lofi'e: which bridi does it extract an argument of? Strictly speaking, it grabs the output placeholder of the sumti, producing the meaning "the past of itself" or "something which is its own past" (if that has any meaning). vo'a (or ledei?) hit the main bridi, which in this sentence but not always is the one you want. I'm not sure of what anaphor will always work. le lofi'e purci: The official form would be le purci be lofi'e, but when the s-selbri is the kind accepting transitive arguments, one frequently interprets le X as le be X -- even though there is no official rule allowing this transformation. J. All lives converge. At impact we will share our common destiny. de se klama fo lo pu'u ro da poi jmive cu jmive There exists Y 1 common property destiny Y This is a tour de force of quantification. The Bladder Slugs' situation determines the semantics, but it has little to do with time metaphor. Lojban lacks a proper "share" gismu, but also the semantics of "share" in this sentence are bogus in that X has destiny Y from the beginning, not just at the moment of impact. I have, however, represented the bogus semantics fairly faithfully. But I had a great deal of trouble to get the quantification right in the correct semantics. Can anyone do this right? klama-go: x1 goes to x2 from x3 via x4 (on legs x5). Due to the needs of quantification, the selbri's place order is radically slewed around. x2 = de, x4 = "life". litru-travel: The semantics are much better than farna-orientation or klama-go. If it had a standard motion word place structure it could have been used here. ro da...: "the life process of everything that is alive". It sounds stupid to repeat jmive, but "da" has to be restricted. janli-collide: x1 crashes into x2. It is logical, but heavy-handed, to mention Y (de) as x2 of janli. cmibi'o - cmima binxo - member become: abstract diklujvo. x1-x2 feeds into cmima x1-x2. kampu-common: (abstract) x1 is common to (the members of set) x2. Hence X becomes a member of a set which has a common property Z (see below). The set members of x2 in extension should feed into x1 of an abstraction which is in x1 of kampu. The -gua!spi dictionary, but not the Lojban dictionary, prescribes this. dimna: (abstract) x1 is the destiny of x2. The common property Z is having the destiny Y (de). K. I hope our bladderlings will rise into the upper reaches of the brightest past. .a'o loi gapydakcifnu poi se rirni mi'o cu nitkla (I)hope the* bladderling RSC parent by we(you+I) ascend le pagbu poi galtu be le selgusrai lo'i purci part RSC high brightest set past Diklujvo: gapydakcifnu - gapci dakli cifnu - gas bag baby - bladderling: Two nested transitive diklujvo. It seems to me that cifnu-baby will be used terminally in many compounds such as kitten, colt, tadpole, anythingling. Therefore it needs a terminal rafsi. nitkla - ascend: See sentence I. selgusrai - se gusni traji - illuminated superlative - brightest. Abstract diklujvo using definition discussed below. Probably bogus: poi se rirni mi'o. The English sentence says "our bladderlings", which literally translated means "which you and I are the parents of". But most likely "our" means "all Yiks'", in which case "the* bladderling" (minus "our") is quite sufficient. So the RSC should be omitted entirely. Definitely bogus: le pagbu poi galtu be le... The clauses are required to be after the s-bridi with its arguments. However, the result in this instance is atrocious in style; required is G SSSS C (G = s-selbri, SSSS = long sumti, C = clause) while you much prefer to have the longer element after the shorter: G C SSSS. The real solution is to use le galtu pagbu be le... (creative tanru whose meaning you can't be sure of) or le galpau be le... (matching lujvo). But pagbu = "x1 is a part of whole x2" with no place for a selecting property. galpau could be considered an abstract diklujvo were the definition like this: "x1 is a part of x2 with common property x3" -- in this case, the upper part. A gismu, similarly defined, is also needed for "subset". I have a converse problem with traji-superlative. It is "x1 is superlative in property x2" but it should be defined as "x1 is (one of) the extreme member(s) of set x2 in (abstract property) x3". "One of" is needed because ties are possible. L. I believe the shadows of our downtime bladderlings fell across us upperday. .ia le ctino be le ma'a ni'a gapydakcifnu (I)believe shadow our below bladderling cu te no'egu'i ma'a ga'u pa djedi x3<->x1 unilluminate we above one day This sentence is very philosophical, and so I have elected to translate "our" less anally retentively this time, as "ma'a"-"you + I + unspecified other(s)". Expressing the riptime connections is equally successful (or obscure) using human time 's or Bladder Slug spatial 's. I suddenly notice that no'e is not listed as a rafsi; hence the usage no'egu'i is questionable; perhaps it should be no'egusni. Is there an actual morphological reason to suppress no'e as a rafsi representing no'e? (And similarly for other .) --------------- Conclusion: Lojban temporal and spatial tenses are highly parallel and so it was a non-issue to deal with the Yik spatial tense metaphors. This text has a number of diklujvo. I would judge that even a person unfamiliar with their guidelines for interpretation can accept their meanings as plausible, no more uncertain than with any other unofficial lujvo. But the listener with access to the guidelines can interpret them as surface structures representing a deep structure in which each bridi or s-bridi is headed by a gismu; thus he can interpret these lujvo with the same certainty (solipsist or uniform) that he can expect for gismu. James F. Carter (213) 825-2897 UCLA-Mathnet; 6221 MSA; 405 Hilgard Ave.; Los Angeles, CA, USA 90024-1555 Internet: jimc@math.ucla.edu BITNET: jimc%math.ucla.edu@INTERBIT UUCP:...!{ucsd,ames,ncar,gatech,purdue,rutgers,decvax,uunet}!math.ucla.edu!jimc