Talk:methods of resolving mismatches between place structures and number of overt sumti

From Lojban
Jump to navigation Jump to search

Posted by daltong on Sat 02 of Apr., 2005 00:24 UTC posts: 4

This I want to see. Because to me it looks utterly bogus --- nitcion.

  • If 5 years ago I had (by some sort of timewarp) seen this discussion with the attributions removed, I'd have bet that the bits written by you were written by me and the bits written by me were written by you. You're now the one taking an ultra-hard line. Conceptually, the hardline is nicer than what I've proposed on these pages. But (a) I think gismu blotation is a big problem and one for which there is no other solution in sight, (b) I very strongly expect that your hardline will not survive usage. And it's not just because of slack usage, either: if we learn a brivla from usage, and its final sumti place is always left implicit (-- which is not implausible), we never learn that final sumti place, and end up having learnt a different brivla. So all in all, yes, I accept the principle that the brivla meaning is defined by the placestructure, but all in all, the way this is implemented in Lojban (e.g. by dependent-marking rather than head-marking; with much blotation; etc.) makes me feel fairly strongly that that principle cannot survive. --And

Okay. Here goes.

  • In usage, if people said (in standard grammar) what they mean, then lots of implicit sumti would have to be filled with overt zi'o.
    • Usually this is inadvertent, though I expect that most of those who do this inadvertently don't really care about this sort of error, because it so rarely impairs communication.
    • Sometimes it is advertent, with places of bloated gismu being left unfilled in the hope that through desuetude they will wither and die. But mere desuetude is insufficient to annul a sumti place; we would also need some sort of evidence that the speaker was actually expressing a meaning equivalent to the empty place being zihoed, yet this sort of evidence must be pretty hard to come by.
  • Many gismu are bloated. We would like to ziho off many of their sumti places, but are all pretty much in agreement that filling them with overt zi'o is unworkable.
  • The range of meanings covered by zi'o broda is a superset of the range covered by zo'e broda, so the proposed convention does not reduce the range of meanings expressed by the current convention.
  • The downside to the proposal is that each selbri potentially has as many polysemes as it has permutations of filled and unfilled places. But I don't see this as a problem: for example, even if mamta in ko'a mamta, which would meanko'a mamta zi'o, could in principle mean something different from mamta in ko'a mamta fo'e, in practise they would mean the same thing, because the most natural interpretation of ko'a mamta (zi'o) is ko'a mamta da. The effects of the convention would only bite in the case of predicates where it makes good semantic sense to ziho off a place, and in these cases the effect of the proposal is to require the presence of an overt zo'e so as to express the intended polyseme.
  • The polysemy, then, would be limited to cases where it actually makes good sense, and as usage develops, the polysemy could be recorded in future dictionaries.
  • This is not so distinct from the situation that will arise from the status quo, except that because zo'e is not obligatory in places that could sensically be annulled, it is much harder to discern what is going on in place-structure usage.
  • The proposal would not totally eradicate bloated gismu syndrome, because someone could still learn the bloated places from documentation and then use them, but I think that force of usage would tend to kill that off (-- Why use a sumti place that nobody else can remember the meaning of?). But bloated gismu syndrome would be much less of a problem, because one would only need to learn the places that actually get used.

--And

Treasonous balderash.

  • Lojban is pro-drop. I damn well know that if I say ko'a mamta, I don't mean ko'a mamta zi'o, but ko'a mamta zo'e.
    1. Yes, in the sense that you know the Insert-Zohe rule. But the actual logical formula you intend to express is almost certainly going to be ko'a mamta da or ko'a mamta fo'e. Now, the only sensible interpretation for ko'a mamta would be ko'a mamta da. (This is less the case when zi'o can never be implicit, because nobody would say ko'a mamta zi'o when they mean ko'a mamta da, since it's antigricean.) But what if you want to express the meaning ko'a mamta fo'e? Well, I think that in context, da broda can implicate fo'e broda, especially if the da is implicit. (If the da were explicit then it would be antigricean, raising the question "Why say da when you mean fo'e?") Here's an example. In English, I read. and I ate. mean "mi cilre/citka da". But I picked up the magazine and read pretty strongly implicates "read it". Even in a context where I'm on the train, catch someone's eye, point to the newspaper and say "Can I read?" — which is of course not a normal thing to say — it still seems to clearly implicate "Can I read it?".
    2. A better testcase is whether when you say da vi pixra you (Nick) are always claiming that da actually depicts something, or whether you do not intend to exclude pictures that don't depict anything. Even if you yourself are scrupulous about this, are you at confident that the average comparatively fluent Lojban speaker is? I'm not. I think it would be preferable to have da vi pixra mean "Here's a picture", and to claim that the picture actually depicts something you'd need to say da vi pixra zo'e or da vi pixra de.
    • ko'a mamta entails ko'a mamta zo'e. Logically, ko'a mamta zo'e does indeed mean ko'a mamta da. The implicature ko'a mamta fo'a can be generated; but it's just an implicature, and zo'e includes da and fo'a in its denotation. In fact, da includes fo'a. So big deal. --n
    • If a Lojbanist can say pixra for things that pixra noda, they might as well say citka for people that citka noda.pixra does not mean 'picture'. It means 'relation of depiction between x1, a picture, and x2, a depicted. If the average fluent Lojbanist is using pixra to refer to nondepicting entities, then he's misusing the predicate pure and simple. Or what do you think? That Lojban does not have a surfeit of people asking broda ma every time they talk? i thought the discussion between xorxes and xod on djuno was pointless, but I now remember gleefully that every time xorxes said jetnu, xod screamed jetnu ma. (And the answer was always le se djuno cu jetnu le ve djunoxorxes) As well he should. Because in that instance, xod grokked Lojban. You don't like the fact that pixra involves depiction? Tough. You should have protested the x2 of pixra when you had the chance. Now, pixra does not mean picture, dammit. It means depiction. n
  • If the most natural interpretation of ko'a mamta zi'o is ko'a mamta da, then why is the natural interpretation of ko'a botpi fo zi'o not ko'a mamta fo da? If zi'o does not undefine places everywhere, but only where it semantically 'makes sense', then you're translating polysemy and pragmatic loading from the gismu to the cmavo. Don't. We now know what zi'o means. If this gets through (which it won't), then zi'o will be blanched of meaning; it will end up meaning zo'e much of the time anyway. So you're back where you started.
    • On the mamta/botpi issue, the difference is that it's hard to conceive of a mother without progeny, but easy to think of a lidless bottle. Under my proposals, zi'o would be pragmatically distinct from zo'e if and only if the distinction makes sense for the predicate in question. My proposals do not necessarily constitute a revised conception of zi'o per se, since the definitions of predicates can often include existentially quantified variables that do not get expressed syntactically by an overt sumti. For example, 'denti'/'tooth' involves as part of its definition "If x is denti then Ey x consists of material y", regardless of the fact that 'denti' rightly has no x3 place for material. So jubme fe zi'o ("table of material zi'o") should similarly not entail that materiality is not a necessary property of the predicate's sense.
      • Congratulations on turning me into xod. A botpi isn't a bottle, man. It's a lidded bottle. By definition. The problem is in the place structure of botpi --- and in your assumption that it must must must match the denotation of 'bottle'. To say a botpi is a bottle, and potentially lidless, but a mamta is a mother, and not potentially without progeny is defining English, not Lojban. You know very well what a mamta be zi'o is: it's a fertile woman. Where do you get off using English keywords to describe botpi be zi'o as plausible and mamta be zi'o as implausible?
        • I (kreig.daniyl.) have taught myself botpi/patxu as "lidded container/any container". Since doing so, I've swhly noticed that jars really do have more in common with bottles (similar liddedness) than with pots (similar shape) - so the distinction makes more sense to me than the English one. The keywords simply have nothing to do with the meaning - in fact, my family has several lidded pots which do in fact patxu but which also botpi. For gricean reasons, I believe that any jbojbe would refer to them as botpi.
        • Craig, normally I take issue with everything you say :-) , but here you're spot on. Yes, the distinction between botpi and patxu is different from that between bottle and jar.
      • If jubme fe zi'o does not entail that materiality is necessary to tablehood, then you're humpty-dumptying zi'o (and mamta, and botpi). And the teeth thing is a logical fallacy. An unmarked place may still be essential to the definition; that does not mean a place marked explicitly as inapplicable can also be inapplicable. If jubme fe zi'o does not entail that materiality is necessary, why are you putting zi'o in there at all? And why is it different from botpi be zi'o, the whole point of which is to make lids unnecessary to bottles?
  • You're breaking the understanding of 1000 functioning gismu, to remedy a bloated 300, as an active tinkering programme. That's intolerable. If you're going to set up contexts in which zi'o does and does not actually remove places from the definition, why don't you be honest about it and state that information where it belongs: in the lexicon.
    • The change would validate a lot of usage, would invalidate little, would remedy the bloated gismu, would not make anything unsayable (since overt zo'e or da or ko'a would still be an option), and would remedy the problem of zi'o signally failing to be suitable for the job is was supposed to do.
  • Even if bloated places were skimmed off the gismu list tomorrow, zi'o as default interpertation would still leave fear and uncertainty as to what people mean when they are being elliptical. Things shouldn't be decided when people ellipse. They should remain undecided. zo'e is the more conservative interpretation in that case: it leaves the place contents undecided, it doesn't decide they are empty. (Unless you redefine zi'o - as you do - to make it decide they're empty only conditionally. Which is rank adhoccery.)
    1. I can't actually think of any examples, off the top of my head, where fear and uncertainty should arise. Part of what I'm proposing is tantamount to abolishing syntactic elliptizability of sumti: the only syntactic sumti are the ones overtly present, but the effects on the meaning of the selbri vary from brivla to brivla, being part of the lexical entry for the brivla, in particular part of the specification of the sense of the brivla.
    2. What do you mean by "decide that [place contents] are empty"? I'm not suggesting a redefinition of zi'o, so much as a revision in our understanding of it. (Not mere sophistry: I really think those two things are different.) I don't think zi'o ever meant that a place was empty; it meant that it needn't figure in the definition of the selbri. Now, if every zi'o has to be overt, then the only point in using it is to signal that a place doesn'tfigure in the definition of the selbri. But if zi'o could be implicit, then that contingency ("needn't figure...") becomes significant, and would effectively mean "doesn't figure, if it makes sense for it not to figure, and otherwise, if it doesn't make sense for it not to figure, does figure".
    • So, when it's explicit, it means one thing, and when it's implicit, it means something else. And this is not casuistry. Riight --n.
  • The range of means is a superset?! doi And, we're not talking denotation. We're talking sense. Just because the denotation of leaner gismu is a superset of the fatter ones, doesn't mean you should go ahead and assume the leaner gismu: the meaning *does* change, in a not backwards-compatible fashion.
    • It's not backwards-compatible with Correct usage, but is arguably more compatible with actual usage than the current rule is.
  • If Jorge is saying stuff like this, then he deserves a lot of the flak that you have been copping. And I'm starting to believe he does anyway.
    • Since he is embarrassed by my bouquets, he'd probably welcome the odd brickbat. Plus, flak from you is quite fun flak...
      • I do deserve any flak that And gets, as I tend to agree with him about most things. Also, I am embarrased and very flattered by And's bouquets, and as usual welcome any fun flak from Nick.
  • CLL is baselined. Feel free to do this; and expect me to misunderstand you and follow the established zo'e convention. CLL conventions should only be jettisoned when demonstrably bogus; the motivations you have presented are not enough to do the trick.
    • True. I bet, though, that if we could silently put the proposed convention into operation — e.g. if people started using it without advertising the fact — then nobody would notice (except that the occasional overt zo'e might be a telltale sign). It might even improve comprehension...
      • This was actually the gist of the comment of mine that started this off. There were some adamant claims that lu'i no da cu se cmima was true, meaning that the empty set is a set, so I suggested that the rule for empty places in practice would become that they are filled by zi'o. I also have the impression that the bloating of gismu happened as a result of an implicit assumption that this would be the rule. How do we say stripes? We don't have a word for it? Let's add a place to tirxu, and we can also use it for zebras in te tirxu xirma , where the x3 of tirxu is the only place that intervenes. --xorxes
        • Then whoever was insisting on this was a damn fool, and you're another. :-) se cmima does not mean 'set'. It means 'thing with membership'. Which means 'non-empty set'. You want a general term for 'set', you know where to find zi'o. A mathematical set is zilselcmi, pure and simple. And I will not have the interpretive conventions of Lojban destroyed because people can't be arsed using zi'o where they should. --n
          • Well, I was defending the same position you're defending now, with the same argument, that se cmima is a thing with members, so that must make you a fool too. :-) --xorxes
          • Nononono, Xorxes. Then, you were sane to insist that se cmima is a thing with members. Now, you're just coming around again. :-)
        • And I doubt severely that gismu bloating had anything to do with this. The x3 of tirxu is insane, but its insanity is far likelier to be "duh, tigers, stripes, intrinsic" than the psychopathy of "oh, your shirt is looking very be-tigered today." --n
          • I have a faint memory of reading somewhere that the US flag was be-tigered. In some oldju'i lobypli maybe. --xorxes
          • *shrug* I could do the supplication and find out, but this really doesn't sound like the way gismu places were coined back in the day... But then, I've blocked all memory of my life before this year. I always do. :-)

(... meaning, I presume,.i mi djica sisti lenu morji roda pe lemi nunji'e zi'epe tu'o bi'i ke'o le nanca be le nu morji . How do we anaphorise the latter tense again? Clearly not nau...)


  • If you sisti le nu morji ro da, you could still morji su'o da. You want roda outside.
  • Is ke'o meant to be ke'i? It still seems ungrammatical. Between tu'o and what?
  • The x2 of nanca is a number. le nu morji can't be a number.
  • Maybe ro da pe le nu mi jmive pu le ca nanca zo'u mi djica sisti le nu morji da i mi roroi go'i. I think ca should fall within the scope of roroi as wanted. --xorxes

I (mi'e maikl.) can't help but see this page in the context of the style I have been developing (lesi'o sumykuntybri(external link)) , & accepting the proposal would turn all utterances on the order of ".i broda brode" into a mass of nonsensical zi'o's. For me, unfilled sumti positions are norje'u, like Schroedinger's Cat. For those ZI'OISTS who can't tolerate ambiguity, there is indeed a way to write in the style which best fits your personality. But it(external link) doesn't have to be made the rule for everyone.


Posted by JohnCowan on Sat 02 of Apr., 2005 18:43 UTC posts: 149

The notion that zihoed places are more algorithmically defined than zohed ones is just false. It's true that zo'e appeals to the shared context, but zi'o appeals to metaphysics that may or may not be shared. Interpreting be zi'o directly as a projection (in the sense of relational algebra) doesn't capture it: in that sense, nothing is broda be zi'o except what is broda.

mi'e djan


Posted by Anonymous on Sat 02 of Apr., 2005 21:08 UTC

> The notion that zihoed places are more algorithmically defined than zohed ones is just false. It's true that zo'e appeals to the shared context, but zi'o appeals to metaphysics that may or may not be shared. Interpreting be zi'o directly as a projection (in the sense of relational algebra) doesn't capture it: in that sense, nothing is broda be zi'o except what is broda. > > mi'e djan

{lo'i broda be zi'o} is a superset, not a subset, of {lo'i broda}. For example, {lo se cmima be zi'o}, a "bemembered thing" but without the place for the members, can be things (one thing actually) without any members, whereas {lo se cmima} can't. {broda} could be a projection of {broda be zi'o}, not the other way around.

mu'o mi'e xorxes


Posted by pycyn on Sat 02 of Apr., 2005 22:24 UTC posts: 2388

I find this discussion baffling, perhaps because I do not see a problem with putting {no da} in any place where it might fit to reflect the way the world is, so {no da}ing the "lid" place of {botpi} seems perfectly meaningful and often involved in true descriptions of situations. On the other hand, the mess in "the {zo'e} series" probably should be cleaned up (more a problem about what is said about it than about usage). The chief problem is not {zi'o} but {zo'e} (and also blank, the lack of overt sumti in a place). Well, {zu'i} needs the vagueness of "typical" cleared up: in this place?, in this place with (some particular ones of?) these other arguments? of skipped sumti altogether? and so on. {zi'o} plugs up a place in the predicate so that judgments about whether the predicate applies to the remaining arguments (explcit or implicit) is made without reference to what would have been in that place. Thus, ordered n-1 tuples might be admitted in the extension of this new predicate that are not part of any n tuple in the extension of the old, though the appropriate n-1 part of each the acceptable n tuples is included. But the descriptions of{zo'e} (and blank, if you take that as an abbreviation for {zo'e}) are a mess. At various times and places {zo'e} is said to stand for the obvious value (in the context, I suppose), to stand for anything other than {no da} (and its equivalents) and {zi'o}, to indicate that it doesn't matter what you put in there (?so long as it makes the sentence true?), and probably a few other things as well. This is nmot much of a set of guidelines, being contradictory as a whole (the obvious thing is clearly not just anything — and, I would add, may be nothing). Before it was talked about as an abbreviation for {zo'e}, the blank was just that: no sumti given, no arguemnt specified. What that amounted to was to be worked out pragmatically: obvious, indifferent, unimportant, unknown, secret — all these interpretations had their uses. To be sure, if what was skipped was known to be {no da}, a good Gricean might object to a speaker not mentioning that, if anything might hang on it, but in the case of indifference -- and certainly of ignorance — {no da} is a possible value to be discovered. If blank is really an abbreviation for {zo'e}, then (or {zo'e} the blank made visible) surely it has this value (pragmatic, not semantic) "For some reason, I am not telling you about things that go in here" and usually this causes no problems. And when it does cause problems, well, everyone knows where to find {ma}, though a lot of metaanswers are possible: "I never even thought about that," for example. Now, insofar as {zo'e} — and blank — are used out of indifference, that is, that anything reasonable can go in the slot without affecting the outcome, that slot is being disregard and so, in effect, we have dropped that slot from consideration. So, in some cases, even {zi'o} can be what blank and {zo'e} "mean." Of course, the problem with all of this is that, unless we ask — which is pretty pointless in most cases — we do not know why the place is zeroed out and so, if we guess wrongly, we may go astray, picking "don't give a damn" when "obvious" was intended, for example. Sorting these possibilities out seems to me more important than any other ways of screwing around with the series (which, properly used, is already doing most of what it appears that various people want it to do).


> > The notion that zihoed places are more > algorithmically defined than zohed ones is just > false. It's true that zo'e appeals to the > shared context, but appeals to metaphysics > that may or may not be shared. Interpreting be > zi'o directly as a projection (in the sense of > relational algebra) doesn't capture it: in that > sense, nothing is broda be zi'o except what is > broda. > > > > mi'e djan > > {lo'i broda be zi'o} is a superset, not a > subset, of {lo'i broda}. > For example, > {lo se cmima be zi'o}, a "bemembered thing" but > without the place for the > members, can be things (one thing actually) > without any members, > whereas {lo se cmima} can't. {broda} could be a > projection of {broda be zi'o}, > not the other way around. > > mu'o mi'e xorxes > > > >


Posted by Anonymous on Sat 02 of Apr., 2005 22:50 UTC

On Apr 2, 2005 7:23 PM, John E Clifford wrote: > I find this discussion baffling, perhaps because > I do not see a problem with putting {no da} in > any place where it might fit to reflect the way > the world is,

Hopefully nobody sees a problem with that.

> so {no da}ing the "lid" place of > {botpi} seems perfectly meaningful and often > involved in true descriptions of situations.

Of course. Does anyone oppose using {no da} there?

> But > the descriptions of{zo'e} (and blank, if you take > that as an abbreviation for {zo'e}) are a mess.

I think it's the other way around: zo'e is a way of indicating blank when leaving it blank would change the meaning: {broda zo'e ko'a} is not the same as {broda ko'a}.

> Before it was talked about as > an abbreviation for {zo'e}, the blank was just > that: no sumti given, no arguemnt specified.

That's what both zo'e and blank are, as far as I understand.

> What that amounted to was to be worked out > pragmatically: obvious, indifferent, unimportant, > unknown, secret — all these interpretations had > their uses.

Yes.

> To be sure, if what was skipped was > known to be {no da}, a good Gricean might object > to a speaker not mentioning that, if anything > might hang on it, but in the case of indifference > — and certainly of ignorance — {no da} is a > possible value to be discovered.

I don't think {no da} can be skipped, either by blank or by {zo'e}. That amounts to skipping {na ku}.

> If blank is really an abbreviation for {zo'e}, > then (or {zo'e} the blank made visible) surely it > has this value (pragmatic, not semantic) "For > some reason, I am not telling you about things > that go in here" and usually this causes no > problems.

Right. "What things" in the sense of what referents, not in the sense of what words. zo'e or blank don't stand for omitted words, they are the omision of any mention of the arguments, the things, that play the corresponding role in the corresponding relationship with the other (mentioned) things.

mu'o mi'e xorxes


Posted by pycyn on Sun 03 of Apr., 2005 01:54 UTC posts: 2388

> On Apr 2, 2005 7:23 PM, John E Clifford > wrote: > > I find this discussion baffling, perhaps > because > > I do not see a problem with putting {no da} > in > > any place where it might fit to reflect the > way > > the world is, > > Hopefully nobody sees a problem with that.

Well, the presentation is pretty muddled, but both & and n seem to say this at one point or another. Hopefully it is just unclarity.

> > so {no da}ing the "lid" place of > > {botpi} seems perfectly meaningful and often > > involved in true descriptions of situations. > > Of course. Does anyone oppose using {no da} > there?

Well again this seems to turn up even with exactly this example.

> > But > > the descriptions of{zo'e} (and blank, if you > take > > that as an abbreviation for {zo'e}) are a > mess. > > I think it's the other way around: zo'e is a > way of > indicating blank when leaving it blank would > change > the meaning: {broda zo'e ko'a} is not the same > as {broda ko'a}.

But surely not from {broda fe ko'a} (or whatever place it is that {ko'a} occupies).

> > Before it was talked about as > > an abbreviation for {zo'e}, the blank was > just > > that: no sumti given, no arguemnt specified. > > That's what both zo'e and blank are, as far as > I understand.

Well, then we do not disagree at all so far, but both of us seem to be at odds with both & and n.

> > What that amounted to was to be worked out > > pragmatically: obvious, indifferent, > unimportant, > > unknown, secret — all these interpretations > had > > their uses. > > Yes. > > > To be sure, if what was skipped was > > known to be {no da}, a good Gricean might > object > > to a speaker not mentioning that, if anything > > might hang on it, but in the case of > indifference > > — and certainly of ignorance — {no da} is a > > possible value to be discovered. > > I don't think {no da} can be skipped, either by > blank > or by {zo'e}. That amounts to skipping {na ku}.

Well, that is debatable, of course; it depends on what you mean by "skip" or whatever word you use. Would we really say of a person who did not know that he lied when he left out a reference that proved to be {no da}? Wold we even insist that what he said was false? And what about the cases where you don't give a damn with what or whether the place is filled? As noted, it is nice to say it if you know it and it makes a difference.


> > If blank is really an abbreviation for > {zo'e}, > > then (or {zo'e} the blank made visible) > surely it > > has this value (pragmatic, not semantic) > "For > > some reason, I am not telling you about > things > > that go in here" and usually this causes no > > problems. > > Right. "What things" in the sense of what > referents, > not in the sense of what words. zo'e or blank > don't > stand for omitted words, they are the omision > of any mention of the arguments, the things, > that > play the corresponding role in the > corresponding > relationship with the other (mentioned) things. > Well, I would say (indeed just did) they don't stand for anything though they convey that I am not going to tell you what goes there — whether word or referent. If they stood for something then I would have defeated my purpose in using them (alrthough you would not get much more information than you do now — what they stand for, in particular). I am not clear just how "omission of any mention" is different from "omission of words for;" it is surely the words that are omitted, not the things.


Posted by Anonymous on Sun 03 of Apr., 2005 03:34 UTC

On Apr 2, 2005 10:52 PM, John E Clifford wrote: > > --- Jorge Llambías wrote: > > I think it's the other way around: zo'e is a > > way of > > indicating blank when leaving it blank would > > change > > the meaning: {broda zo'e ko'a} is not the same > > as {broda ko'a}. > > But surely not from {broda fe ko'a} (or whatever > place it is that {ko'a} occupies).

{fi} in that example. Yes, {broda fi ko'a} is the same as {broda zo'e ko'a}, or the same as {se te se broda ko'a}. In general, {zo'e} can be avoided if you really want to.

> > > To be sure, if what was skipped was > > > known to be {no da}, a good Gricean might > > object > > > to a speaker not mentioning that, if anything > > > might hang on it, but in the case of > > indifference > > > — and certainly of ignorance — {no da} is a > > > possible value to be discovered. > > > > I don't think {no da} can be skipped, either by > > blank > > or by {zo'e}. That amounts to skipping {na ku}. > > Well, that is debatable, of course; it depends on > what you mean by "skip" or whatever word you > use.

I mean leaving the place blank or filling it with {zo'e} when the intended meaning corresponds to {noda}.

> Would we really say of a person who did not > know that he lied when he left out a reference > that proved to be {no da}?

Probably not. Lying requires intent to deceive so it is hard to lie unknowingly.

> Wold we even insist > that what he said was false?

Well, yes, if what he said was false, it was false. If he says {broda} and the fact is that {naku broda}, then {broda} is false.

> And what about the > cases where you don't give a damn with what or > whether the place is filled?

If the argument in question is not related to the other arguments, we are simply using the wrong selbri for the relationship we want to express. (The bloating of some gismu means that we sometimes end up doing this unfortunately.)

> > > If blank is really an abbreviation for > > {zo'e}, > > > then (or {zo'e} the blank made visible) > > surely it > > > has this value (pragmatic, not semantic) > > "For > > > some reason, I am not telling you about > > things > > > that go in here" and usually this causes no > > > problems. > > > > Right. "What things" in the sense of what > > referents, > > not in the sense of what words. zo'e or blank > > don't > > stand for omitted words, they are the omision > > of any mention of the arguments, the things, > > that > > play the corresponding role in the > > corresponding > > relationship with the other (mentioned) things. > > > Well, I would say (indeed just did) they don't > stand for anything though they convey that I am > not going to tell you what goes there — whether > word or referent.

I don't think it does that if you allow words without referents among the possible omissions.

> If they stood for something > then I would have defeated my purpose in using > them (alrthough you would not get much more > information than you do now — what they stand > for, in particular). I am not clear just how > "omission of any mention" is different from > "omission of words for;" it is surely the words > that are omitted, not the things.

"Words for", yes. Words that don't refer (such as {no da} or {zi'o}), no.

mu'o mi'e xorxes


Posted by Anonymous on Sun 03 of Apr., 2005 07:48 UTC

John E Clifford scripsit:

> I find this discussion baffling, perhaps because I do not see a problem > with putting {no da} in any place where it might fit to reflect the way > the world is, so {no da}ing the "lid" place of {botpi} seems perfectly > meaningful and often involved in true descriptions of situations.

That doesn't work. ta botpi no da means that nothing stands in a botpi relationship with that object; this sentence affirms nothing.

> Thus, ordered n-1 tuples might be admitted in the extension of this new > predicate that are not part of any n tuple in the extension of the old,

Yes, but which novel (n-1)-tuples are allowed and which are not? There is no simple universal answer to this, which is what makes default zihoing such a bad idea.

> At various times and places {zo'e} is said to stand for the obvious > value (in the context, I suppose), to stand for anything other than > {no da} (and its equivalents) and {zi'o}, to indicate that it doesn't > matter what you put in there (?so long as it makes the sentence true?)

These are different ways of formulating the same thing. When I assert "ta gerku", there is some dog breed (se gerku) to which that object belongs, I'm just not bothering to specify it. I simply exclude, on Gricean grounds, the notion that you're allowed to say "broda" when in fact what you mean is "noda broda".

If "ta gerku" is to be interpreted as "ta gerku zi'o", then we don't know what that means until we nail down just which objects are gerku be zi'o but not gerku be da.

> To be sure, if what was skipped was known to be {no da}, a good Gricean > might object to a speaker not mentioning that, if anything might hang > on it, but in the case of indifference — and certainly of ignorance -- > {no da} is a possible value to be discovered.

Not so. We do not want a world-3 in which seemingly affirmative claims turn out to be disguised versions of negative ones.

> If blank is really an abbreviation for {zo'e}, then (or {zo'e} the > blank made visible) surely it has this value (pragmatic, not semantic) > "For some reason, I am not telling you about things that go in here" > and usually this causes no problems. And when it does cause problems, > well, everyone knows where to find {ma}, though a lot of metaanswers > are possible: "I never even thought about that," for example.

Just so.

-- Don't be so humble. You're not that great. John Cowan --Golda Meir [email protected]


Posted by Anonymous on Sun 03 of Apr., 2005 07:51 UTC

Jorge Llambías scripsit:

> {lo'i broda be zi'o} is a superset, not a subset, of {lo'i broda}.

Indeed. You mistook my strawman for a real argument: zi'o cannot be defined as a mere projection.

> For example, > {lo se cmima be zi'o}, a "bemembered thing" but without the place for the > members, can be things (one thing actually) without any members, > whereas {lo se cmima} can't. {broda} could be a projection of {broda be zi'o}, > not the other way around.

But it's only in specialized cases that we definitely know what things are broda be zi'o but not broda be da, like this one. Most of the time, it's a huge semantic defining effort that hasn't even been begun.

-- And it was said that ever after, if any John Cowan man looked in that Stone, unless he had a [email protected] great strength of will to turn it to other www.ccil.org/~cowan purpose, he saw only two aged hands withering www.reutershealth.com in flame. --"The Pyre of Denethor"


Posted by pycyn on Sun 03 of Apr., 2005 13:02 UTC posts: 2388

> On Apr 2, 2005 10:52 PM, John E Clifford > wrote: > > > > --- Jorge Llambías > wrote: > > > I think it's the other way around: zo'e is > a > > > way of > > > indicating blank when leaving it blank > would > > > change > > > the meaning: {broda zo'e ko'a} is not the > same > > > as {broda ko'a}. > > > > But surely not from {broda fe ko'a} (or > whatever > > place it is that {ko'a} occupies). > > {fi} in that example. Yes, {broda fi ko'a} is > the same as {broda zo'e ko'a}, > or the same as {se te se broda ko'a}. In > general, {zo'e} can be avoided > if you really want to.

Good. I see that you intended your citation only aas a bridi tail, not the whole thing.

> > > > To be sure, if what was skipped was > > > > known to be {no da}, a good Gricean might > > > object > > > > to a speaker not mentioning that, if > anything > > > > might hang on it, but in the case of > > > indifference > > > > — and certainly of ignorance — {no da} > is a > > > > possible value to be discovered. > > > > > > I don't think {no da} can be skipped, > either by > > > blank > > > or by {zo'e}. That amounts to skipping {na > ku}. > > > > Well, that is debatable, of course; it > depends on > > what you mean by "skip" or whatever word you > > use. > > I mean leaving the place blank or filling it > with {zo'e} > when the intended meaning corresponds to > {noda}.

The point of blank is that it has NO intended meaning. At most you can say — and even this is often doubtful — that the intention is to not say what, if anything, goes in the slot. I see that what I said was somewhat less clear than it should be. I meant that, while {broda no da} is incompatible with {broda da (or ko'a or lo brode, etc.)} it need not be incmpatible with {broda - }, which may just indicate total lack of interest about what or whether goies there. that is, the blank may stand for {zi'o} as well as {zo'e} or {zu'i} and probably a few other things as well, all just about degrees of non-commital-ness.

> > Would we really say of a person who did not > > know that he lied when he left out a > reference > > that proved to be {no da}? > > Probably not. Lying requires intent to deceive > so it > is hard to lie unknowingly. > > > Wold we even insist > > that what he said was false? > > Well, yes, if what he said was false, it was > false. > If he says {broda} and the fact is that {naku > broda}, > then {broda} is false.

The point again is that {broda no da} does not entail {naku broda --}.

> > And what about the > > cases where you don't give a damn with what > or > > whether the place is filled? > > If the argument in question is not related to > the other arguments, > we are simply using the wrong selbri for the > relationship we want to > express. (The bloating of some gismu means that > we sometimes > end up doing this unfortunately.)

The point here is that this move on our part is not a mistake but a legitimate use of nonreference.

> > > > If blank is really an abbreviation for > > > {zo'e}, > > > > then (or {zo'e} the blank made visible) > > > surely it > > > > has this value (pragmatic, not semantic) > > > "For > > > > some reason, I am not telling you about > > > things > > > > that go in here" and usually this causes > no > > > > problems. > > > > > > Right. "What things" in the sense of what > > > referents, > > > not in the sense of what words. zo'e or > blank > > > don't > > > stand for omitted words, they are the > omision > > > of any mention of the arguments, the > things, > > > that > > > play the corresponding role in the > > > corresponding > > > relationship with the other (mentioned) > things. > > > > > Well, I would say (indeed just did) they > don't > > stand for anything though they convey that I > am > > not going to tell you what goes there -- > whether > > word or referent. > > I don't think it does that if you allow words > without > referents among the possible omissions.

But of course {da} has no referent and is clearly an acceptable word there; ditto then {no da}.

> > If they stood for something > > then I would have defeated my purpose in > using > > them (alrthough you would not get much more > > information than you do now — what they > stand > > for, in particular). I am not clear just how > > "omission of any mention" is different from > > "omission of words for;" it is surely the > words > > that are omitted, not the things. > > "Words for", yes. Words that don't refer (such > as {no da} or > {zi'o}), no. > I'm sure you means something here, but it is not clear what. {zo'e}, the official replacement for blank, has no referent (at least in most of the stories about what it means), it is just a pragmatic refusal to commit, yet — even if you don't think that {zi'o} and {zu'i} can replace blank — it clearly can replace blank.


Posted by pycyn on Sun 03 of Apr., 2005 13:17 UTC posts: 2388

> John E Clifford scripsit: > > > I find this discussion baffling, perhaps > because I do not see a problem > > with putting {no da} in any place where it > might fit to reflect the way > > the world is, so {no da}ing the "lid" place > of {botpi} seems perfectly > > meaningful and often involved in true > descriptions of situations. > > That doesn't work. ta botpi no da means that > nothing stands in a botpi > relationship with that object; this sentence > affirms nothing.

But you just said it affirms that nothing stands in that relationship to ta. I suppose you mean it is equivalent to no affirmative sentence involving just these sumti and brivla. I didn't say it did (or didn't mean it if that is how you take what I said). I was rejecting the view that {ta botpi fo no da} is somehow illegitimate altogether and in particular is incompatible with {ta botpi} tout court.

> > Thus, ordered n-1 tuples might be admitted in > the extension of this new > > predicate that are not part of any n tuple in > the extension of the old, > > Yes, but which novel (n-1)-tuples are allowed > and which are not? There > is no simple universal answer to this, which is > what makes default > zihoing such a bad idea.

Well, the most that would be claimed is that the one presented fits, which is in effect what is claimed by any omission of places.

> > At various times and places {zo'e} is said to > stand for the obvious > > value (in the context, I suppose), to stand > for anything other than > > {no da} (and its equivalents) and {zi'o}, to > indicate that it doesn't > > matter what you put in there (?so long as it > makes the sentence true?) > > These are different ways of formulating the > same thing. When I assert > "ta gerku", there is some dog breed (se gerku) > to which that object belongs, > I'm just not bothering to specify it. I simply > exclude, on Gricean > grounds, the notion that you're allowed to say > "broda" when in fact > what you mean is "noda broda".

Bad example. Every dog must be of some breed -- however broadly "breed" is defined. but many places ({botpi4} the usual example, though I suppose it could be argued it is not a good one) are not essential and so {no da} does not affectr some sense here. If you want to argue that all places are essential by definition or some such, then {no da} can never be correct, which seems extreme. The point of leaving a blank is often -- not always of course — that what goes in that place makes no difference and thus even if nothing goes there it should not affect the rest of the sentence.

> If "ta gerku" is to be interpreted as "ta gerku > zi'o", then we don't > know what that means until we nail down just > which objects are > gerku be zi'o but not gerku be da.

See above. Note we have exactly the same problem if we assert {ta gerku zi'o}, but the issue seems never to be raised in that case.


> > To be sure, if what was skipped was known to > be {no da}, a good Gricean > > might object to a speaker not mentioning > that, if anything might hang > > on it, but in the case of indifference — and > certainly of ignorance -- > > {no da} is a possible value to be discovered. > > Not so. We do not want a world-3 in which > seemingly affirmative claims > turn out to be disguised versions of negative > ones. Whoa, Nelly. I didn't say {ta broda --} *means* {ta broda no da}, i.e., makes a negative claim. It merely (if this is what is going on) makes a claim to which {ta broda no da} is irrelevant or, at least, is not contradictory.

> > If blank is really an abbreviation for > {zo'e}, then (or {zo'e} the > > blank made visible) surely it has this value > (pragmatic, not semantic) > > "For some reason, I am not telling you about > things that go in here" > > and usually this causes no problems. And > when it does cause problems, > > well, everyone knows where to find {ma}, > though a lot of metaanswers > > are possible: "I never even thought about > that," for example. > > Just so.

Well, then, what is your problem?


Posted by pycyn on Sun 03 of Apr., 2005 13:25 UTC posts: 2388

> Jorge Llambías scripsit: > > > {lo'i broda be zi'o} is a superset, not a > subset, of {lo'i broda}. > > Indeed. You mistook my strawman for a real > argument: zi'o cannot be > defined as a mere projection. > > > For example, > > {lo se cmima be zi'o}, a "bemembered thing" > but without the place for the > > members, can be things (one thing actually) > without any members, > > whereas {lo se cmima} can't. {broda} could be > a projection of {broda be zi'o}, > > not the other way around. > > But it's only in specialized cases that we > definitely know what things are > broda be zi'o but not broda be da, like this > one. Most of the time, it's > a huge semantic defining effort that hasn't > even been begun. > But one that we are not generally called to solve, even for the case at hand. If we do not want to accept that the present case is true, that the ta mentioned in {ta broda zi'o} really is a broda zi'o, then we can get into a discussion with the claimant on that issue, if it matters enough. And he either comes up with a (totally uninteresting) something that goes in the slot or he makes a case that ta at least is enough like things that broda da to admit into the sorta-broda class even though they broda no da. Or you come up with a killer difference (not just about what goes into the slot). Or you end up just disagreeing on the question.


Posted by pycyn on Sun 03 of Apr., 2005 14:33 UTC posts: 2388

> --- John Cowan > wrote: > > John E Clifford scripsit:

> > > At various times and places {zo'e} is said > > > to stand for the obvious > > > value (in the context, I suppose), to stand > > > for anything other than {no da} (and its > > >equivalents) and {zi'o}, to indicate that it > > > doesn't matter what you put in there (?so > > > long as it makes the sentence true?) > > > > These are different ways of formulating the > > same thing.

I pass3ed over this remark before, taking it to be too ludicrous to justify a comment, but I see now that it might be taken seriously. If, perhaps, it means that were we to take all of these formulations and add them together we would get a formulation of th use of blank or {zo'e} that fit the case, I would agree. However, it seems to be the claim that each of these taken alone amounts to the same thing as any other of these taken alone, and that is clearly false. To be sure, if, for example, the speaker intends to say that it doesn't matter what you choose for the slot and the hearer takes this to indicate the obvious choice, the hearer will not err in extension, though he will in intension. On the other hand, if the positions are reversed, the hearer is in real danger of erring in extension as well, choosing a non-obvious candidate. And, of course, taking the gap for anything other than {no da} and {zi'o} is clearly talking about terms, not things, and so would allow in any sort of referential sumti expression — with any sort of referents — or any non-referential sumti, {da, roda, cida...}, which clearly goes beyond the intention of the last alternative (which does seem to have a relevance restriction at least). these are various things that omitting a place may mean, but they are both semantically and pragmatically very different — and there are probably a few more that also need to be dealt with — but which no one takes to be a possibility for {zo'e}, even though they are often the force of blank.


Posted by Anonymous on Sun 03 of Apr., 2005 15:26 UTC

On Apr 3, 2005 10:16 AM, John E Clifford wrote:

> I was rejecting the view that > {ta botpi fo no da} is somehow illegitimate > altogether and in particular is incompatible with > {ta botpi} tout court.

Nobody has ever argued that it is illegitimate altogether.

Many have argued that it is incompatible with {ta botpi}. One way of making it compatible is to say that {ta botpi} really is {ta botpi zi'o zi'o zi'o} rather than {ta botpi zo'e zo'e zo'e}.

> Every dog must be of some breed -- > however broadly "breed" is defined. but many > places ({botpi4} the usual example, though I > suppose it could be argued it is not a good one) > are not essential and so {no da} does not affectr > some sense here.

Then we at least agree that {no da} is not compatible with blank/zo'e in essential places? If so, we have identified where our disagreement lies, namely:

> If you want to argue that all > places are essential by definition or some such,

That's what I want to argue. That's why I dislike bloating, because bloated relationships have a much more restricted application than intended.

> then {no da} can never be correct, which seems > extreme.

{no da} can never be omitted. Obviously it can often be correct.

> The point of leaving a blank is often > — not always of course — that what goes in that > place makes no difference and thus even if > nothing goes there it should not affect the rest > of the sentence.

That would be the blank = zi'o position.

mu'o mi'e xorxes


Posted by pycyn on Sun 03 of Apr., 2005 22:34 UTC posts: 2388

> On Apr 3, 2005 10:16 AM, John E Clifford > wrote: > > > I was rejecting the view that > > {ta botpi fo no da} is somehow illegitimate > > altogether and in particular is incompatible > with > > {ta botpi} tout court. > > Nobody has ever argued that it is illegitimate > altogether.

Well, to be precise, it has been argued that nothing that is a bottle can be botpi no da and that often gets formulated to make {no da} appear to be always illegitmate. Both positions seem to me just wrongheaded.

> Many have argued that it is incompatible with > {ta botpi}. > One way of making it compatible is to say that > {ta botpi} > really is {ta botpi zi'o zi'o zi'o} rather than > > {ta botpi zo'e zo'e zo'e}.

This seems to me to be almost as extreme as some of the other positions and so as unacceptable. For the usual cases, the most that seems to be required is {zo'e zo'e zi'o} (or maybe {zu'i} for {zo'e} and even that seems overprecise for the real situation.

> > Every dog must be of some breed -- > > however broadly "breed" is defined. but many > > places ({botpi4} the usual example, though I > > suppose it could be argued it is not a good > one) > > are not essential and so {no da} does not > affectr > > some sense here. > > Then we at least agree that {no da} is not > compatible > with blank/zo'e in essential places? If so, we > have identified > where our disagreement lies, namely:

Not quite what I said, which was that a case using an obviously always filled place is not a good example for making the point that {no da} can never be true when blank is true (or is it the other way round?).

> > If you want to argue that all > > places are essential by definition or some > such, > > That's what I want to argue. That's why I > dislike bloating, > because bloated relationships have a much more > restricted > application than intended.

But of coudrse, calling it bloating is just saying that some listed places are not really essential, which is at least one thing that is often going on when someone "omits" a place.

> > then {no da} can never be correct, which > seems > > extreme. > > {no da} can never be omitted. Obviously it can > often be correct.

Well, again, the point here is that it can often be omitted without affecting the rest of the sentence.

> > The point of leaving a blank is often > > — not always of course — that what goes in > that > > place makes no difference and thus even if > > nothing goes there it should not affect the > rest > > of the sentence. > > That would be the blank = zi'o position.

No, because often what goes in there — the obvious thing, for example, does make a difference and {zi'o} would possibly get it wrong. Blank is simply refusing to specify at all, with no indication — overtly in this place anyhow -- of what, why, or whether is glossed over.


Posted by Anonymous on Mon 04 of Apr., 2005 00:52 UTC

On Apr 3, 2005 7:33 PM, John E Clifford wrote: > --- Jorge Llambías wrote: > > On Apr 3, 2005 10:16 AM, John E Clifford > > wrote: > > > > > I was rejecting the view that > > > {ta botpi fo no da} is somehow illegitimate > > > altogether and in particular is incompatible > > with > > > {ta botpi} tout court. > > > > Nobody has ever argued that it is illegitimate > > altogether. > > Well, to be precise, it has been argued that > nothing that is a bottle can be botpi no da

The argument is that nothing that is a _botpi_ can be a botpi fo no da. Things that are bottles can be botpi fo no da, and therefore not botpi. In other words, the English "bottle" is not coextensive with the Lojban {botpi}.

> and > that often gets formulated to make {no da} appear > to be always illegitmate.

I never saw that formulation other than in your characterization.

> > > If you want to argue that all > > > places are essential by definition or some > > such, > > > > That's what I want to argue. That's why I > > dislike bloating, > > because bloated relationships have a much more > > restricted > > application than intended. > > But of coudrse, calling it bloating is just > saying that some listed places are not really > essential,

More precisely: they would not be essential with a sane meaning of the gismu. The place structures of the listed gismu, in some cases, are insane. So the alternatives are to pay no attention to the listed place structures, and use the gismu as if they had more usable place structures (a very popular choice in practice) or take the place structures seriously, with the resulting complications in the expressions. If we take the listed place structures seriously, every argument takes as much a part in the relationship as every other.

> > {no da} can never be omitted. Obviously it can > > often be correct. > > Well, again, the point here is that it can often > be omitted without affecting the rest of the > sentence.

{ta botpi fo no da} says "that doesn't botpi a single thing", which cannot be equivalent to {ta botpi}, which means "that does botpi". So at least in my understanding of Lojban, omitting {noda} is like omitting {naku}, it always affects the rest of the sentence rather crucially.

mu'o mi'e xorxes


Posted by pycyn on Mon 04 of Apr., 2005 13:35 UTC posts: 2388

> On Apr 3, 2005 7:33 PM, John E Clifford > wrote: > > --- Jorge Llambías > wrote: > > > On Apr 3, 2005 10:16 AM, John E Clifford > > > wrote: > > > > > > > I was rejecting the view that > > > > {ta botpi fo no da} is somehow > illegitimate > > > > altogether and in particular is > incompatible > > > with > > > > {ta botpi} tout court. > > > > > > Nobody has ever argued that it is > illegitimate > > > altogether. > > > > Well, to be precise, it has been argued that > > nothing that is a bottle can be botpi no da > > The argument is that nothing that is a _botpi_ > can be a botpi fo no da. > Things that are bottles can be botpi fo no da, > and therefore not botpi. > In other words, the English "bottle" is not > coextensive with the Lojban > {botpi}.

Right; but at the same time they call the place structure of {botpi} bloated, that is, containing unnecessary places. If {botpi} is not "bottle" because of the lid place, then it cannot be called bloated, since every place in its structure is essential to it. You can't have it both ways and the thrust seems to be generally in the direction of calling {botpi} bloated. Hence, {botpi} ought to be compatible with current {botpi fo zi'o}.

> > and > > that often gets formulated to make {no da} > appear > > to be always illegitmate. > > I never saw that formulation other than in your > characterization.

See above on the implications of the positions. I don't think anyone else has pointed out this consequence.

> > > > If you want to argue that all > > > > places are essential by definition or > some > > > such, > > > > > > That's what I want to argue. That's why I > > > dislike bloating, > > > because bloated relationships have a much > more > > > restricted > > > application than intended. > > > > But of coudrse, calling it bloating is just > > saying that some listed places are not really > > essential, > > More precisely: they would not be essential > with a sane meaning > of the gismu.

Sorry, if you change the place structure, you change the meaning. With the same meaning, you are stuck with the same place structure.

> The place structures of the > listed gismu, in some > cases, are insane.

This simply means that you do not — for whatever reason — like that place structure. So change it — that seems to be what bpfk is all about. The place structure — by your claim — is (by definition) exactly right for the concept it represents. Anything else would be insane (well, wrong any way).

> So the alternatives are to > pay no attention > to the listed place structures, and use the > gismu as if they > had more usable place structures (a very > popular choice in > practice) or take the place structures > seriously, with the resulting > complications in the expressions. If we take > the listed place structures > seriously, every argument takes as much a part > in the relationship > as every other.

I take it that the popular response to this pseudo problem is to leave places empty, with various treatments of that response when asked to fill in the gaps (I suspect, "I have no idea" and "I don't care" would be the most usual responses). In short, just what I have been reading as the interpretation of blank.

> > > {no da} can never be omitted. Obviously it > can > > > often be correct. > > > > Well, again, the point here is that it can > often > > be omitted without affecting the rest of the > > sentence. > > {ta botpi fo no da} says "that doesn't botpi a > single thing", > which cannot be equivalent to {ta botpi}, which > means > "that does botpi". So at least in my > understanding of Lojban, > omitting {noda} is like omitting {naku}, it > always affects the > rest of the sentence rather crucially.

And yet there are many sentences that are accepted into the corpus and that are even thought to be true but that have or may well have for all we know just that omission. What I am saying is just what people are doing all the time.


Posted by JohnCowan on Mon 04 of Apr., 2005 14:01 UTC posts: 149

John E Clifford scripsit:

> Right; but at the same time they call the place > structure of {botpi} bloated, that is, containing > unnecessary places. If {botpi} is not "bottle" > because of the lid place, then it cannot be > called bloated, since every place in its > structure is essential to it. You can't have it > both ways and the thrust seems to be generally in > the direction of calling {botpi} bloated. Hence, > {botpi} ought to be compatible with current > {botpi fo zi'o}.

I agree with all but the last sentence, which should be "Hence people who talk of bloated place structures should put up or shut up."

> Sorry, if you change the place structure, you > change the meaning. With the same meaning, you > are stuck with the same place structure.

+1


> This simply means that you do not — for whatever > reason — like that place structure. So change > it — that seems to be what bpfk is all about. > The place structure — by your claim — is (by > definition) exactly right for the concept it > represents. Anything else would be insane (well, > wrong any way).

+1


My take is that zi'o semantics are never obvious or clear-cut: using zi'o generates a novel predicate just as much as making tanru does, and its meaning cannot be determined by fiat, any more than tanru meaning can. There may be things that mamta be zi'o, or gerku be zi'o, or botpi be fo zi'o, but only speaker consensus can say what they may be.

-- John Cowan [email protected] www.ccil.org/~cowan www.reutershealth.com I must confess that I have very little notion of what [http://www.lojban.org/tiki/s.%204%20of%20the%20British%3Cbr%20/%3ETrade%20Marks%20Act,%201938 s. 4 of the British Trade Marks Act, 1938] is intended to convey, and particularly the sentence of 253 words, as I make them, which constitutes sub-section 1. I doubt if the entire statute book could be successfully searched for a sentence of equal length which is of more fuliginous obscurity. --MacKinnon LJ, 1940


Posted by JohnCowan on Mon 04 of Apr., 2005 14:40 UTC posts: 149

Jorge Llamb?as scripsit:

> {ta botpi fo no da} says "that doesn't botpi a single thing", > which cannot be equivalent to {ta botpi}, which means > "that does botpi". So at least in my understanding of Lojban, > omitting {noda} is like omitting {naku}, it always affects the > rest of the sentence rather crucially.

+1


-- So that's the tune they play on John Cowan their fascist banjos, is it? [email protected] --Great-Souled Sam http://www.ccil.org/~cowan


Posted by Anonymous on Mon 04 of Apr., 2005 14:41 UTC

On Apr 4, 2005 10:34 AM, John E Clifford wrote: > If {botpi} is not "bottle" > because of the lid place, then it cannot be > called bloated, since every place in its > structure is essential to it.

For someone who accepts the gismu list as is, no place structure is bloated. Each place structure defines exactly the meaning of each gismu.

For someone who would like a less arbitrary choice of basic words, some place structures, as included in the gi'uste, are bloated, because they correspond to somewhat abstruse concepts leaving no basic word for some more basic concept.

> You can't have it > both ways and the thrust seems to be generally in > the direction of calling {botpi} bloated.

Calling a place structure bloated is simply saying that a different concept would have been a better use for the word form in question.

> Hence, > {botpi} ought to be compatible with current > {botpi fo zi'o}.

{broda f* zi'o} covers in general more ground than {broda}. That's a rule that works for every broda.

{botpi} without an x4 would have been a more general concept than {botpi} as defined.

> Sorry, if you change the place structure, you > change the meaning.

Indeed.

> With the same meaning, you > are stuck with the same place structure.

Indeed.

> > The place structures of the > > listed gismu, in some > > cases, are insane. > > This simply means that you do not — for whatever > reason — like that place structure.

Exactly.

> So change > it — that seems to be what bpfk is all about.

I would very much like to do a rationalization of place structures. Unfortunately (or fortunately rather) it is not only up to me.

> The place structure — by your claim — is (by > definition) exactly right for the concept it > represents. Anything else would be insane (well, > wrong any way).

Right. The choice of concept for the word is the problem. Once the concept is fixed, we are stuck with it (more or less anyway, we always have the choice of ignoring the gi'uste).

The claim is that as long as we take the place structure of {botpi} seriously, then {ta botpi} is incompatible with {ta botpi fo no da}, just as {ta citka} is incompatible with {ta citka no da}, and {ta broda} is incompatible with {ta broda f* noda}. If we make them compatible, we are changing the sense of the relationship.

> I take it that the popular response to this > pseudo problem is to leave places empty, with > various treatments of that response when asked to > fill in the gaps (I suspect, "I have no idea" and > "I don't care" would be the most usual > responses). In short, just what I have been > reading as the interpretation of blank.

Yes, concentrating on keywords and ignoring place structures is very common.

> > {ta botpi fo no da} says "that doesn't botpi a > > single thing", > > which cannot be equivalent to {ta botpi}, which > > means > > "that does botpi". So at least in my > > understanding of Lojban, > > omitting {noda} is like omitting {naku}, it > > always affects the > > rest of the sentence rather crucially. > > And yet there are many sentences that are > accepted into the corpus and that are even > thought to be true but that have or may well have > for all we know just that omission. What I am > saying is just what people are doing all the time.

What sentences from what corpus? I don't think we have any officially approved corpus, do we? I certainly don't deny that people do this sort of thing all the time, but I'm not sure that they would claim it's correct when pointed out. That's one of the reasons I proposed at some point to re-interpret blank in the prescription as {zi'o} rather than as {zo'e}, to bring the prescription more in line with usage.

mu'o mi'e xorxes


Posted by Anonymous on Mon 04 of Apr., 2005 14:43 UTC

On Apr 4, 2005 10:59 AM, John Cowan wrote: > John E Clifford scripsit: > > > Right; but at the same time they call the place > > structure of {botpi} bloated, that is, containing > > unnecessary places. If {botpi} is not "bottle" > > because of the lid place, then it cannot be > > called bloated, since every place in its > > structure is essential to it. You can't have it > > both ways and the thrust seems to be generally in > > the direction of calling {botpi} bloated. Hence, > > {botpi} ought to be compatible with current > > {botpi fo zi'o}. > > I agree with all but the last sentence, which should be > "Hence people who talk of bloated place structures > should put up or shut up."

Put what up exactly? A list of bloated place structures?

Obviously what counts as bloated is very subjective, but it is clear that the more places a gismu has, the more restrictive its meaning is. Some gismu have very specific senses. Some useful more general senses don't have any gismu associated with it. But this is all very opinable.

mu'o mi'e xorxes


Posted by pycyn on Mon 04 of Apr., 2005 14:46 UTC posts: 2388

> John E Clifford scripsit: > > > Right; but at the same time they call the > place > > structure of {botpi} bloated, that is, > containing > > unnecessary places. If {botpi} is not > "bottle" > > because of the lid place, then it cannot be > > called bloated, since every place in its > > structure is essential to it. You can't have > it > > both ways and the thrust seems to be > generally in > > the direction of calling {botpi} bloated. > Hence, > > {botpi} ought to be compatible with current > > {botpi fo zi'o}. > > I agree with all but the last sentence, which > should be > "Hence people who talk of bloated place > structures > should put up or shut up."

Okay, scratch the "ought", or rather extend it to "if they are serious about what they say" or some such. It occurs to me that there is an unexplored possibility here (which has the same effect as other suggestions but comes at it in a different way): not all places in the place structure or a predicate need to be essential. that is, some places can be {no da} without that spilling over to negate the whole, {no da} then becomes — somewhat against apparent logic -- just another choice among many for that slot, a breed along side saluki and pit bull for {gerku}, for example.

> > Sorry, if you change the place structure, you > > change the meaning. With the same meaning, > you > > are stuck with the same place structure. > > +1 > > > This simply means that you do not — for > whatever > > reason — like that place structure. So > change > > it — that seems to be what bpfk is all > about. > > The place structure — by your claim — is > (by > > definition) exactly right for the concept it > > represents. Anything else would be insane > (well, > > wrong any way). > > +1 > > My take is that zi'o semantics are never > obvious or clear-cut: > using zi'o generates a novel predicate just as > much as making > tanru does, and its meaning cannot be > determined by fiat, > any more than tanru meaning can. There may be > things that > mamta be zi'o, or gerku be zi'o, or botpi be fo > zi'o, but > only speaker consensus can say what they may > be. > I suppose that the *idea behind* {zi'o} is that it takes in those tuples that satisfy the predicate except for this one place: classically things that are just like {botpi} but for the lid -- or that would be botpi had they lids. We have in mind something we already know to be a bottle (or a container meant to have a lid) but that lacks a lid and we want to get it into the botpi -- or something very like. Clearly this works in a lot of cases — dogs without breeds (taking that in a strict technical sense), containers without lids, and the like. And so we see these places as being dispense-with-able and the concept therefore boated. We would not (could not) do the same for other cases — a trip without a route or a means of conveyance seems impossible to us (space warps not being in our experience — assuming "route" is about intervening points and sequence). The stripe place of {tirxu} seems to be of the first sort because we know of semi-albino tigers and totally albino other big cats — and also because we have NO idea what sorts of expressions to put into that place. (Of course, the fact that this is the only way to bring in stripes, regardless of what we are talking about, is another problem -- one of omissions from the vocabulary altogether.) But the issue here at the moment is not about what {zi'o}d expressions mean but rather about what triggers this modification of meaning. MY position is that plain omission is (and always has been) used in a way that corresponds not only to overt {zo'e} (whatever that may be) but also to overt {zu'i} and {zi'o} (in the sense at least that the claim would not be withdrawn were it shown that nothing fit in the omitted place) and several other versions of "I don't say" as well. >From my point of view, what is needed is only to clarify and perhaps expand on the {zo'e} series to provide clear overt markers for the various grounds for omission, to meet inquiries about the omitted places.


Posted by JohnCowan on Mon 04 of Apr., 2005 14:47 UTC posts: 149

Jorge Llamb?as scripsit:

> I would very much like to do a rationalization of place > structures. Unfortunately (or fortunately rather) it is > not only up to me.

There is certainly no reason why you can't create a proposal with place structure changes and the reasons for them under the byfy aegis.

> That's one of the reasons I proposed at some point to > re-interpret blank in the prescription as {zi'o} rather than > as {zo'e}, to bring the prescription more in line with usage.

This evades the issue, though, because it brings in novel semantics under the table. If you want new semantics for gismu, figure out what they should be, rather than claiming that "da broda" and "da broda de" have different possible values for da. Doing that splits each n-place gismu into 2^n-1 new predicates, only one of which has a formal definition. (It also forces observatives to be treated as even more of a special case than they already are.)

-- A: "Spiro conjectures Ex-Lax." John Cowan Q: "What does Pat Nixon frost her cakes with?" [email protected] --"Jeopardy" for generative semanticists http://www.ccil.org/~cowan


Posted by Anonymous on Mon 04 of Apr., 2005 15:42 UTC

On Apr 4, 2005 11:44 AM, John E Clifford wrote: > It occurs to me that there is an > unexplored possibility here (which has the same > effect as other suggestions but comes at it in a > different way): not all places in the place > structure or a predicate need to be essential. > that is, some places can be {no da} without that > spilling over to negate the whole, {no da} then > becomes — somewhat against apparent logic -- > just another choice among many for that slot, a > breed along side saluki and pit bull for {gerku}, > for example.

A "Nobody" that the messanger can pass on the road, as it were.

So if ta is a cat, {ta gerku noda} would be false?

mu'o mi'e xorxes


Posted by pycyn on Mon 04 of Apr., 2005 16:07 UTC posts: 2388

> Jorge Llamb?as scripsit: > > > I would very much like to do a > rationalization of place > > structures. Unfortunately (or fortunately > rather) it is > > not only up to me. > > There is certainly no reason why you can't > create a proposal with > place structure changes and the reasons for > them under the byfy > aegis. > > > That's one of the reasons I proposed at some > point to > > re-interpret blank in the prescription as > {zi'o} rather than > > as {zo'e}, to bring the prescription more in > line with usage. > > This evades the issue, though, because it > brings in novel semantics > under the table. If you want new semantics for > gismu, figure > out what they should be, rather than claiming > that "da broda" > and "da broda de" have different possible > values for da. Doing > that splits each n-place gismu into 2^n-1 new > predicates, only > one of which has a formal definition. (It also > forces observatives > to be treated as even more of a special case > than they already are.) > To summarize the situation "briefly." Many (most?) predicates have places that are often (usually?, always?) left unfilled in use. This includes use in forming lujvo, where some places never appear in the resulting new predicate. >From this it might be inferred that such places are not really essential to the concepts intended for the predicates — or, rather, that the concepts that should have been intended for the predicates were ones that did not involve these places. This latter claim is reinforced by (and reinforces) notions about what are basic and useful concepts to have, in light of which it is possible to say that some predicates are for concpts that are narrower, less basic, than other concepts closely related to the given ones and not represented (typically generic concpts to which the given are specific). The straightforward response to this perceived problem would be to propose redefinition of some (probably, if everyone got into the act, virtually all) predicates, eliminating "unnecessary" places and moving to more general terms where "appropriate." For a variety of reasons, this has not seemed to be the way to proceed (it involves a lot of work on each narrow individual case and so is much harder than making sweeping changes in broad general areas). So the discussion has turned to finding some way to achieve the desired results within the existing defintions of predicates. The obvious solution -- within the existing frame — for solving the "extra" place problem is to use {zi'o}, which is designed for just that purpose. However, since the "bloated" predicates are so numerous and the "extra" places even more so, this would result in an overabundance of {zi'o}s in the langauge, several in almost every sentence — at least an esthetic (and practical) disaster. So the suggestion comes along to make the omission of a sumti in a place the equivalent of the occurrence of {zi'o}. (There appears to be another suggestion in the &-n discussion, but I can't figure it out.) Unfortunately, omission of sumti is already a common occurrence (again in virtually every sentence — this is what started the process) and it is clear that it does not always mean {zi'o} -- although it is equally clear that it often does or at least might well. So, to change omissions to {zi'o} the suggestion arises that other omissions should instead be replaced by overt (unclear) {zo'e} or some other appropriate marker (the only one currently available is the vague {zu'i} but more could be added if need be). This, of course, means that we are relieved of a plethora of {zi'o} only at the cost of a plethora of {zo'e}s; not an obvious gain. Indeed, a loss (according to Cowan) since we now have unmarked a change in the meaning of a sentence which is also in some sense uncontrolled — we don't know what is the result of removing one place from a predicate, especially if the removal allows that that place might have no filler in the larger predicate. (Of course, we very often do have in mind a particular new predicate for that new expression, but there is no guarantee that that is what we will get — at least in the eyes of our conversational partners.) If we look at actual omissions in use, we find that (so far as we can reconstruct or remember or notice them) they arise from a variety of reasons: those represented by {zo'e} (whatever they may be) as well as {zu'i} and even {zi'o} and several others besides. The blank = {zi'o} proposal would require a massive reworking of the existing text — much more than redefining predicates, if the claim that certain places are never used is correct. Formulating and formalizing the existing practice might be simpler, if less satisfying in some intellectual sense. So far as I can tell, the reasons for omission can be divided into three broad categories (to match the three available markers -- each category has obvious subdivisons): a) place filler significant, speaker knows what fills it, hearer can supply the filler correctly (several senses of {zo'e}and {zu'i} too), b) place filler insignificant, sppeaker doesn't care what, if anything, goes there, hearer shouldn't bother trying to find out ({zi'o} plus a bit) c) place significant but speaker is not going to say -- can't or won't — what goes there, hearer can not certainly identify the filler (ignorance, secrets, nastiness and so on). If we move {zu'i} to the last of these, we can briefly explain away any challenged omission, though in the blank = {zi'o}'s defense a hearer who cannot identify the filler immediately is probably wisest to assume it is insignificant.


Posted by Anonymous on Mon 04 of Apr., 2005 16:12 UTC

On Apr 4, 2005 11:46 AM, John Cowan wrote: > > There is certainly no reason why you can't create a proposal with > place structure changes and the reasons for them under the byfy > aegis.

I'll think about it. One reason for not doing it would be that it's a lot of work and it's not clear that it has much of a chance of being accepted, but maybe it's worth at least starting it.

> > That's one of the reasons I proposed at some point to > > re-interpret blank in the prescription as {zi'o} rather than > > as {zo'e}, to bring the prescription more in line with usage. > > This evades the issue, though, because it brings in novel semantics > under the table. If you want new semantics for gismu, figure > out what they should be, rather than claiming that "da broda" > and "da broda de" have different possible values for da. Doing > that splits each n-place gismu into 2^n-1 new predicates, only > one of which has a formal definition.

Yes, you would need definitions for all of those, but most of them can be read directly from the existing definitions. This is because the existing definitions are written in English, and in most cases the English words don't have the restrictions imposed be the argument structures of gismu.

> (It also forces observatives > to be treated as even more of a special case than they already are.)

I don't treat omitted x1 in any special way.

mu'o mi'e xorxes


Posted by pycyn on Mon 04 of Apr., 2005 16:13 UTC posts: 2388

> On Apr 4, 2005 11:44 AM, John E Clifford > wrote: > > It occurs to me that there is an > > unexplored possibility here (which has the > same > > effect as other suggestions but comes at it > in a > > different way): not all places in the place > > structure or a predicate need to be > essential. > > that is, some places can be {no da} without > that > > spilling over to negate the whole, {no da} > then > > becomes — somewhat against apparent logic -- > > just another choice among many for that slot, > a > > breed along side saluki and pit bull for > {gerku}, > > for example. > > A "Nobody" that the messanger can pass on the > road, > as it were. > > So if ta is a cat, {ta gerku noda} would be > false? > Well, yes. Fredegisus had his point after all. More to the point, if ta is a mutt, {ta gerku no da} would be true (taking "breed" in a technical sense).


Posted by pycyn on Mon 04 of Apr., 2005 16:16 UTC posts: 2388

wrote: > > --- Jorge Llambías > wrote: > > On Apr 4, 2005 11:44 AM, John E Clifford > > wrote: > > > It occurs to me that there is an > > > unexplored possibility here (which has the > > same > > > effect as other suggestions but comes at it > > in a > > > different way): not all places in the place > > > structure or a predicate need to be > > essential. > > > that is, some places can be {no da} without > > that > > > spilling over to negate the whole, {no da} > > then > > > becomes — somewhat against apparent logic > -- > > > just another choice among many for that > slot, > > a > > > breed along side saluki and pit bull for > > {gerku}, > > > for example. > > > > A "Nobody" that the messanger can pass on the > > road, > > as it were. > > > > So if ta is a cat, {ta gerku noda} would be > > false? > > > Well, yes. Fredegisus had his point after all. > > More to the point, if ta is a mutt, {ta gerku > no > da} would be true (taking "breed" in a > technical sense). And still more to the point, {ta gerku} would be false of the cat and true of the dog.


Posted by pycyn on Tue 05 of Apr., 2005 16:11 UTC posts: 2388

wrote:

> So far as I can tell, the reasons for > omission can be divided into three broad > categories (to match the three available > markers > — each category has obvious subdivisons): a) > place filler significant, speaker knows what > fills it, hearer can supply the filler > correctly > (several senses of {zo'e}and {zu'i} too), b) > place filler insignificant, sppeaker doesn't > care > what, if anything, goes there, hearer shouldn't > bother trying to find out ({zi'o} plus a bit) > c) > place significant but speaker is not going to > say > — can't or won't — what goes there, hearer > can > not certainly identify the filler (ignorance, > secrets, nastiness and so on). If we move > {zu'i} > to the last of these, we can briefly explain > away > any challenged omission, though in the blank = > {zi'o}'s defense a hearer who cannot identify > the > filler immediately is probably wisest to > assume > it is insignificant.

A bit more systematically, when a blank is challenged, the speaker — who probably left it blank witout thinking about it, maybe even with realizing there was a place to fill — should consider these questions and answer accordingly. Is the place significant? A. No {zi'o} B. Yes. Does the speaker (s) know the filler? 1. No. Does he at least know that it is filled? a. No. X (probably {zi'o}, ignorance} b. Yes. {da} 2. Yes. Does s want hearer (h) to know the filler? a. No. Y, secrecy (probably not {zi'o}) b. Yes. Can h come to an adequate identification on the basis of the information so far? i. No. s tells h what the filler is. ii. Yes. {zo'e} (either the obvious one or any one within a contextually defined range).


Posted by Anonymous on Tue 05 of Apr., 2005 18:14 UTC

On Apr 5, 2005 1:09 PM, John E Clifford wrote: > A bit more systematically, when a blank is > challenged, the speaker — who probably left it > blank witout thinking about it, maybe even with > realizing there was a place to fill — should > consider these questions and answer accordingly.

In my view, we must distinguish two situations: 1) The speaker is an ideal fluent speaker with full command of the language. 2) The speaker is a current less than fluent speaker with struggling command of the language.

I don't dispute that your recipee probably applies for 2. But a fluent speaker with full command of the langusge will not leave a blank without realizing there was a place to fill. If fluent speakers do that, then in fact there is no such place to fill: the predicate has lost that place if it ever had it. Of course even fluent speakers will sometimes misuse a word, but in that case when challenged they will retract the word and replace it with a more appropriate one, maybe even replace {broda} with {broda be zi'o}, but knowing that in that case {broda} by itself was inappropriate.

Let me illustrate with an example. {jbena} in the gi'uste is defined as:

x1 is born to x2 at time x3 birthday and place x4 birthplace;

Someone who has not incorporated the full place structure might naively say something like: {le mi selfamti pu jbena ca le prulamdei} intending to say "my niece/nephew was born yesterday". But taking the place structure of {jbena} seriously, the tenses there make very little sense. If the niece/nephew was born on April 4, 2005, that fact was true yesterday, it is true today and will be true at any other time. So:

le selfamti cu jbena fi li 2005;04;04 ca le prulamdei le selfamti cu jbena fi li 2005;04;04 ca le cabdei le selfamti cu jbena fi li 2005;04;04 ca le bavlamdei le selfamti cu jbena fi li 2005;04;04 ze'e ku

The tense gives no useful information to someone who takes the place structure seriously. Of course what was meant was to use some other predicate that refers to the act of birth, which did happen yesterday and not today or tomorrow, and not to some abstract relationship between two people, a date and a place. {jbena} is bloated. Without an x3 and x4 it would be a more useful relationship, that could be used for the actual act of birth (and also tagged with a date {de'i} or site {tu'i} if so desired.)

A fluent speaker that accepts {jbena} to have four places will simply not say {le mi selfamti pu jbena ca le prulamdei}.

mu'o mi'e xorxes


Posted by pycyn on Tue 05 of Apr., 2005 19:53 UTC posts: 2388

> On Apr 5, 2005 1:09 PM, John E Clifford > wrote: > > A bit more systematically, when a blank is > > challenged, the speaker — who probably left > it > > blank witout thinking about it, maybe even > with > > realizing there was a place to fill — should > > consider these questions and answer > accordingly. > > In my view, we must distinguish two situations: > 1) The speaker is an ideal fluent speaker with > full > command of the language. > 2) The speaker is a current less than fluent > speaker > with struggling command of the language. > > I don't dispute that your recipee probably > applies for 2. > But a fluent speaker with full command of the > langusge > will not leave a blank without realizing there > was a place > to fill. If fluent speakers do that, then in > fact there is no > such place to fill: the predicate has lost that > place if it ever > had it.

I think that this amounts to saying that there has almost never been a fluent speaker, since I think this happens all the time. Alternatively, it amounts to saying that most places after the second are not real. In that case, bpfk has a serious problem in revising the whole gismu list (at least). I suppose they are up to it, since they have not stumbled in mucking about with various cmavo subsets.

Of course even fluent speakers will > sometimes > misuse a word, but in that case when challenged > they > will retract the word and replace it with a > more appropriate > one, maybe even replace {broda} with {broda be > zi'o}, but > knowing that in that case {broda} by itself was > inappropriate. > > Let me illustrate with an example. {jbena} in > the gi'uste is > defined as: > > x1 is born to x2 at time x3 birthday and > place x4 birthplace; > > Someone who has not incorporated the full place > structure > might naively say something like: {le mi > selfamti pu jbena > ca le prulamdei} intending to say "my > niece/nephew was born > yesterday". But taking the place structure of > {jbena} seriously, > the tenses there make very little sense. If the > niece/nephew was > born on April 4, 2005, that fact was true > yesterday, it is true > today and will be true at any other time. So: > > le selfamti cu jbena fi li 2005;04;04 ca le > prulamdei > le selfamti cu jbena fi li 2005;04;04 ca le > cabdei > le selfamti cu jbena fi li 2005;04;04 ca le > bavlamdei > le selfamti cu jbena fi li 2005;04;04 ze'e > ku

So the knowledgable person would say today (050405) {le sefamti cu jbena fi le prulamdei}

> The tense gives no useful information to > someone who takes > the place structure seriously. Of course what > was meant was > to use some other predicate that refers to the > act of birth, which > did happen yesterday and not today or tomorrow, > and not to > some abstract relationship between two people, > a date and > a place. {jbena} is bloated.

It probably was picked for the line on all thiose documents that ask exactly those questions. In that context it is not bloated. It is probably not about the event at all (which needs another gismu perhaps — or, if we use jbena for this, then we need a lujvo to get those usual places back in).

Without an x3 and > x4 it would be > a more useful relationship, that could be used > for the actual act > of birth (and also tagged with a date {de'i} or > site {tu'i} if so > desired.) > > A fluent speaker that accepts {jbena} to have > four places > will simply not say {le mi selfamti pu jbena ca > le prulamdei}. > Right, see above.


Posted by Anonymous on Tue 05 of Apr., 2005 20:15 UTC

On Apr 5, 2005 4:51 PM, John E Clifford wrote: > I think that this amounts to saying that there > has almost never been a fluent speaker, since I > think this happens all the time.

Why "almost"? :-)

There are certainly people with much knowledge of the language, who can maintain a conversation and understand each other, so if that's the measure of fluency there are some. But if we mean people who speak Lojban at a similar level to their native language, I'm quite sure there are none. A speaker at this level will practically never hesitate about the place structures of words, just like we don't hesitate about using the right verbs transitively or intransitively in our natlangs.

Alternatively, > it amounts to saying that most places after the > second are not real.

I don't know about most, but I would bet that a good many will disappear if we ever reach the stage of having truly fluent speakers.

In that case, bpfk has a > serious problem in revising the whole gismu list > (at least). I suppose they are up to it, since > they have not stumbled in mucking about with > various cmavo subsets.

Well, I'm not so sure. Some people feel that the gi'uste should not be touched, warts and all.

> > le selfamti cu jbena fi li 2005;04;04 ca le > > prulamdei > > le selfamti cu jbena fi li 2005;04;04 ca le > > cabdei > > le selfamti cu jbena fi li 2005;04;04 ca le > > bavlamdei > > le selfamti cu jbena fi li 2005;04;04 ze'e > > ku > > So the knowledgable person would say today > (050405) {le sefamti cu jbena fi le prulamdei}

Right. But he can't use {jbena} to say that the birth lasted for two hours, for example.

> It is probably > not about the event at all (which needs another > gismu perhaps — or, if we use jbena for this, > then we need a lujvo to get those usual places > back in).

Right, but it is always easier to add places than to remove them. {jbedetri} for "birth-date" for example, of {jbena de'i ...}.

mu'o mi'e xorxes


Posted by Anonymous on Tue 05 of Apr., 2005 22:15 UTC

On Monday 04 April 2005 11:40, Jorge Llambías wrote: > On Apr 4, 2005 11:44 AM, John E Clifford wrote: > > It occurs to me that there is an > > unexplored possibility here (which has the same > > effect as other suggestions but comes at it in a > > different way): not all places in the place > > structure or a predicate need to be essential. > > that is, some places can be {no da} without that > > spilling over to negate the whole, {no da} then > > becomes — somewhat against apparent logic -- > > just another choice among many for that slot, a > > breed along side saluki and pit bull for {gerku}, > > for example. > > A "Nobody" that the messanger can pass on the road, > as it were.

That "Nobody" is not {noda}, it's {lo nomei}, or {lo nondza}, or something like that.

> So if ta is a cat, {ta gerku noda} would be false?

If ta is a cat, {ta gerku noda} is true, but {ta gerku lo nomei} is false. Whereas if ta is a mutt, {ta gerku noda} is false, but {ta gerku lo nomei} may be true (or you could say {ta gerku la ginxre}).

phma -- le xruki le ginxre xrixruba xu xrula cu xrani?


Posted by pycyn on Tue 05 of Apr., 2005 22:58 UTC posts: 2388

> On Monday 04 April 2005 11:40, Jorge Llambías > wrote: > > On Apr 4, 2005 11:44 AM, John E Clifford > wrote: > > > It occurs to me that there is an > > > unexplored possibility here (which has the > same > > > effect as other suggestions but comes at it > in a > > > different way): not all places in the place > > > structure or a predicate need to be > essential. > > > that is, some places can be {no da} without > that > > > spilling over to negate the whole, {no da} > then > > > becomes — somewhat against apparent logic > -- > > > just another choice among many for that > slot, a > > > breed along side saluki and pit bull for > {gerku}, > > > for example. > > > > A "Nobody" that the messanger can pass on the > road, > > as it were. > > That "Nobody" is not {noda}, it's {lo nomei}, > or {lo nondza}, or something > like that.

Well, for the story to be funny, it pretty much has to be {no da}. An empty set/group.bunch won't do since at least some of these are somethings and I confess I can't make any sense out of {nondza}, ?x1 is nothing, not something?

> > So if ta is a cat, {ta gerku noda} would be > false? > > If ta is a cat, {ta gerku noda} is true, but > {ta gerku lo nomei} is false. > Whereas if ta is a mutt, {ta gerku noda} is > false, but {ta gerku lo nomei} > may be true (or you could say {ta gerku la > ginxre}). > The interesting question is whether {ta gerku} period is true or false. It is not clear to me why an empty breed works here, though the mixed genetic background does. Of course, it also gets rid of the possibility tht there are dogs of no breed whatsoever, that is, takes us away from "breed" in a technical sense.


Posted by pycyn on Wed 06 of Apr., 2005 16:43 UTC posts: 2388

Okay, aside from the remark about whether the speaker realized there was a place to fill or thought about what to fill it with, this discussion seems beside the point. Our native speaker of Lojban will (I think) still leave blanks in what he says for a variety of reasons -- at least the one listed and perhaps others. I think even assuming that he always knows that and why he is leaving a gap is pressing ones luck a bit (I leave gaps in English wihtout thinking about it, after all, and with no discernable reason). So, I think this divsion will continue to apply even for totally competent users.


> On Apr 5, 2005 1:09 PM, John E Clifford > wrote: > > A bit more systematically, when a blank is > > challenged, the speaker — who probably left > it > > blank witout thinking about it, maybe even > with > > realizing there was a place to fill — should > > consider these questions and answer > accordingly. > > In my view, we must distinguish two situations: > 1) The speaker is an ideal fluent speaker with > full > command of the language. > 2) The speaker is a current less than fluent > speaker > with struggling command of the language. > > I don't dispute that your recipee probably > applies for 2. > But a fluent speaker with full command of the > langusge > will not leave a blank without realizing there > was a place > to fill. If fluent speakers do that, then in > fact there is no > such place to fill: the predicate has lost that > place if it ever > had it. Of course even fluent speakers will > sometimes > misuse a word, but in that case when challenged > they > will retract the word and replace it with a > more appropriate > one, maybe even replace {broda} with {broda be > zi'o}, but > knowing that in that case {broda} by itself was > inappropriate. > > Let me illustrate with an example. {jbena} in > the gi'uste is > defined as: > > x1 is born to x2 at time x3 birthday and > place x4 birthplace; > > Someone who has not incorporated the full place > structure > might naively say something like: {le mi > selfamti pu jbena > ca le prulamdei} intending to say "my > niece/nephew was born > yesterday". But taking the place structure of > {jbena} seriously, > the tenses there make very little sense. If the > niece/nephew was > born on April 4, 2005, that fact was true > yesterday, it is true > today and will be true at any other time. So: > > le selfamti cu jbena fi li 2005;04;04 ca le > prulamdei > le selfamti cu jbena fi li 2005;04;04 ca le > cabdei > le selfamti cu jbena fi li 2005;04;04 ca le > bavlamdei > le selfamti cu jbena fi li 2005;04;04 ze'e > ku > > The tense gives no useful information to > someone who takes > the place structure seriously. Of course what > was meant was > to use some other predicate that refers to the > act of birth, which > did happen yesterday and not today or tomorrow, > and not to > some abstract relationship between two people, > a date and > a place. {jbena} is bloated. Without an x3 and > x4 it would be > a more useful relationship, that could be used > for the actual act > of birth (and also tagged with a date {de'i} or > site {tu'i} if so > desired.) > > A fluent speaker that accepts {jbena} to have > four places > will simply not say {le mi selfamti pu jbena ca > le prulamdei}. > > mu'o mi'e xorxes > > > >


Posted by Anonymous on Wed 06 of Apr., 2005 18:22 UTC

On Apr 6, 2005 1:41 PM, John E Clifford wrote: > Our native > speaker of Lojban will (I think) still leave > blanks in what he says for a variety of reasons

Indeed.

> — at least the one listed and perhaps others. I > think even assuming that he always knows that and > why he is leaving a gap is pressing ones luck a > bit

I don't think a fluent speaker necessarily knows *why* he says what he says. He just knows *how* to say it.

(I leave gaps in English wihtout thinking > about it, after all, and with no discernable > reason).

Yes, but surely you can't omit any word at all. There are rules you follow, even if you don't know them explicitly. You can't omit "nothing" in "he ate nothing" and get anything like the original meaning, for example.

> So, I think this divsion will continue > to apply even for totally competent users.

I think some of the possible reasons you mention for omission (those that involve zi'o) won't apply to someone who has incorporated the full meaning of a brivla.

mu'o mi'e xorxes


Posted by pycyn on Wed 06 of Apr., 2005 21:38 UTC posts: 2388

> On Apr 6, 2005 1:41 PM, John E Clifford > wrote: > > Our native > > speaker of Lojban will (I think) still leave > > blanks in what he says for a variety of > reasons > > Indeed. > > > — at least the one listed and perhaps > others. I > > think even assuming that he always knows that > and > > why he is leaving a gap is pressing ones luck > a > > bit > > I don't think a fluent speaker necessarily > knows *why* he > says what he says. He just knows *how* to say > it. > > (I leave gaps in English wihtout thinking > > about it, after all, and with no discernable > > reason). > > Yes, but surely you can't omit any word at all. > There are rules you follow, even if you don't > know > them explicitly. You can't omit "nothing" in > "he ate nothing" and get anything like the > original > meaning, for example.

I can't omit "nothing" but the case Ihave in mind are not omissions in that sense: I do not know what goes there, I just don't care for the present purpose. So I say nothing about it. With predicates with clearly defined places, this amounts to {zi'o}, if you will, in some occasions and to {da} or {zo'e} even in others.

> > So, I think this divsion will continue > > to apply even for totally competent users. > > I think some of the possible reasons you > mention for omission > (those that involve zi'o) won't apply to > someone who > has incorporated the full meaning of a brivla. > I guess we just disagree about that. I have no trouble in someone finding a bottle and saying {ti botpi} or even {ti botpi la kok lo blaci} and meaning (correctly it turns out) that this is just like a bottle but for its lack of a lid.


Posted by Anonymous on Fri 08 of Apr., 2005 03:31 UTC

On Wednesday 06 April 2005 17:37, John E Clifford wrote: > I guess we just disagree about that. I have no > trouble in someone finding a bottle and saying > {ti botpi} or even {ti botpi la kok lo blaci} and > meaning (correctly it turns out) that this is > just like a bottle but for its lack of a lid.

Yes, but {ti caca'a botpi la kok} and {ti caka'e botpi} are false. The bottle has a lid, but the lid is somewhere else, and without a lid it can't contain something in a botpi way.

Of course if you find a bottle top, you can say {ti ve botpi}, and the same tense applies.

phma -- ..i le babzba ba zbasu lo jbazbabu lo babjba


Posted by pycyn on Fri 08 of Apr., 2005 14:00 UTC posts: 2388

> On Wednesday 06 April 2005 17:37, John E > Clifford wrote: > > I guess we just disagree about that. I have > no > > trouble in someone finding a bottle and > saying > > {ti botpi} or even {ti botpi la kok lo blaci} > and > > meaning (correctly it turns out) that this is > > just like a bottle but for its lack of a lid. > > Yes, but {ti caca'a botpi la kok} and {ti > caka'e botpi} are false. The bottle > has a lid, but the lid is somewhere else, and > without a lid it can't contain > something in a botpi way. > > Of course if you find a bottle top, you can say > {ti ve botpi}, and the same > tense applies. > Here again, we just disagree. My point is just that a blanked place may represent an explicit {zi'o} as well as {zo'e} or {da} or.... and that, even if its lid has been reduced to basic physical components and scattered across the universe (or never created at all for that matter) {ti botpi} — even with {caca'a} still applies correctly.