Lions and levels and the like
John E. Clifford
The point is that the word "lion" (and "lions") can indicate a number of different ontological levels, from the narrowest to the broadest and most abstract. There is is, though, a default level that turns up in the absence of contrary contextual clues, even though it may be easily overridden by those clues. We have words for the various levels, which we can use to explicitly set the level or change in mid discussion ("kind", "segment", "meat", "typically" and "species" roughly for the examples above). Shifting without making note of the shift or starting off at the non-default level without a flag, is a Gricean misdemeanor.
What the default level is for a given word varies from word to word: "lion" takes sort of midlevel gross physical objects, "letter" takes a highly abstracted level (there are twenty-six letters in the English alphabet). Other words probably take lower levels, Buddhist technical terms for components of a person probably somewhere around the bottom. And, as the last example indicates, each level can be expressed in a number of ways.
As far as I can figure out, the recent discussion on the {zo'e} thread (or at least one or two of those discussions) hinges on whether we have the same fluidity of levels in Lojban and whether certain moves constitute misdemeanor violation level shifting. That is, what brodas? Or, perhaps more precisely, what brodas in what way? A single thing may broda individually; a bunch may do so collectively, or conjunctively, or disjunctively, or statistically, or in many more complex ways. Also involved is the nature of some levels: are kinds just bunches of things or are the intensional objects of some sort? Are segments parts of objects or independent things to which objects may be related in a way analogous to the way kinds are related to objects? In general, no side has been very clear (at least in a single continuous statement) on any of these issues, making the whole rather difficult to follow, let alone to critique. Hopefully, this will change.
Sent from my iPad
Martin Bays
> [...]
OK then. I'll reiterate, with all the clarity I can muster.
Short version: {su'o cinfo cu broda} has to mean that some actual lion
brodas. Otherwise we have problems. This is largely independent of the
meaning of {lo cinfo cu broda}, but not of the explanation of that
meaning.
Long version:
The basic problem as I'm seeing it: if we don't specify levels, then we
don't really specify quantifier scope.
What I mean by this (i.e. by "really"): if B hears A say {su'o ctuca cu
tavla ro le tadni}, and B wants to understand what A means to say about
actual teachers and actual students, and if {ctuca} and {tadni} do not
specify levels, then B has to guess which levels A intends them to refer
to. If, for example, B guesses that A is talking about kinds of teacher
and about actual students, all B can deduce about actual teachers and
students is that every student was talked to by some teacher.
(Here I'm using 'actual' in opposition to 'kind' - I wish we had
a better word for it)
(I should also clarify that when I say "{ctuca} does not specify
a level", I mean that there are *individuals* which are e.g. kinds of
teachers and which ctuca; if a kind were implemented as being merely
a bunch of actual teachers, we wouldn't have the problems I'm talking
about.)
So I conclude that it is not befitting of a logical language for it to
have no means to specify level - where 'level' refers to whatever it is
that crossing causes these quantifier scope shifts.
This does not mean that I think lojban should only be able to discuss
actual teachers and not kinds of teachers - merely that we need to be
able to distinguish between the two.
I further note that xorlo - or rather, my understanding of xorxes'
understanding of xorlo - makes this issue less academic than it might
otherwise be. That's because it has descriptions, e.g. {lo ctuca},
habitually (though not always) referring to (bunches of) corresponding
kinds, e.g. to the kind Teacher.
So under xorxes' xorlo, kinds are not rare things summoned up only when
we specifically want to talk about them - you have to deal with them if
you want to understand any sentence using a gadri.
(Here I'm using "the kind Teacher" to refer to the whatever-it-is that
xorxes habitually refers to with {lo ctuca}; I have so far failed to
understand what this is, but it seems that whatever it is is a level up
from actual teachers as regards quantifier scope ambiguities, and that's
all we need to know about it for the present discussion)
This leaves the question of how to deal with this problem; we have
various partial answers, but perhaps I shouldn't complicate this thread
by discussing them here.
Martin
John E. Clifford
Now, I agree that I escape some problems here by taking kinds to be bunches. On the other hand, if they are not bunches, I have trouble working out what {su'o lo cinfo} means when it refers to a kind (rather than a bunch of kinds, say). Of course, it is always safer to specify you level and Lojban has at least a few devices for doing that. But it is also often long-winded to do that when it is "perfectly clear" from context. Whether either of these will help with whatever xorxes is now about remains to be seen. Past xorxes segments have been given to finding intensional content or mass content in {lo} expression (for each of which there are, alas, historical precedents). I also am a little worried about apparently focusing on the most general uses of {lo} exressions and coming up with claims that do not fit the more restricted uses.
Sent from my iPad
Jorge Llambías
>
> What I mean by this (i.e. by "really"): if B hears A say {su'o ctuca cu
> tavla ro le tadni}, and B wants to understand what A means to say about
> actual teachers and actual students, and if {ctuca} and {tadni} do not
> specify levels, then B has to guess which levels A intends them to refer
> to. If, for example, B guesses that A is talking about kinds of teacher
> and about actual students, all B can deduce about actual teachers and
> students is that every student was talked to by some teacher.
You have some hidden assumptions there, for example that there are
actual teachers of the kind that talks to every student.
And B can deduce more: that there is some kind of teacher such that
every student was talked to by some teacher of that kind.
mu'o mi'e xorxes
Martin Bays
> On Sat, Nov 12, 2011 at 2:39 PM, Martin Bays <mb...@sdf.org> wrote:
> > What I mean by this (i.e. by "really"): if B hears A say {su'o ctuca cu
> > tavla ro le tadni}, and B wants to understand what A means to say about
> > actual teachers and actual students, and if {ctuca} and {tadni} do not
> > specify levels, then B has to guess which levels A intends them to refer
> > to. If, for example, B guesses that A is talking about kinds of teacher
> > and about actual students, all B can deduce about actual teachers and
> > students is that every student was talked to by some teacher.
>
> You have some hidden assumptions there, for example that there are
> actual teachers of the kind that talks to every student.
Could there not be, assuming the kind does talk to every student?
I don't know the semantics of kinds talking to people; I was assuming
they'd just be straightforwardly disjunctive.
> And B can deduce more: that there is some kind of teacher such that
> every student was talked to by some teacher of that kind.
Right, except that the kind could (at least in theory) be the kind
Teacher.
Martin
Jorge Llambías
> * Sunday, 2011-11-13 at 10:10 -0300 - Jorge Llambías <jjlla...@gmail.com>:
>>
>> actual teachers of the kind that talks to every student.
>
> Could there not be, assuming the kind does talk to every student?
Probably not. I was just pointing out that B's trans-domain deduction
relies on more than the mere form "su'o broda cu brode ro le brodi".
It's also relying on particularities of broda, brode and brodi.
> I don't know the semantics of kinds talking to people; I was assuming
> they'd just be straightforwardly disjunctive.
Sounds reasonable to me as a first stab.
>> And B can deduce more: that there is some kind of teacher such that
>> every student was talked to by some teacher of that kind.
>
> Right, except that the kind could (at least in theory) be the kind
> Teacher.
It could, yes. It would be pragmatically unlikely because of the
"su'o", but from a strictly logical point of view it could. For B to
be able get no more than your deduction, B would have to make that
unlikely assumption.
John E Clifford
ontological commitment of things called "bunches"): {su'o lo stuci) has
essentially the same result under either interpretation, a subbunch of lo
stuci. It may, of course, not correspond to the bunches put in as kinds of
teachers, but it produces a kind of its own. Of course, there remains the issue
of how this bunch talks to all the students, but, as I have noted elsewhere, it
all works out to there being some teachers (mundanes) who talk to all the
students, even if no one teacher does.
--
You received this message because you are subscribed to the Google Groups
"lojban" group.
To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to
lojban+un...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/lojban?hl=en.
maikxlx
- (1a) Lions are ruining my garden.
- (1b) There exist some lions that are ruining my garden.
where (1a) invokes a kind and (1b) invokes a bunch or somesuch, and yet both sentences seem to have the same truth conditions or almost the same.
But yesterday as I was reading random online materials (this one - http://amor.cms.hu-berlin.de/~h2816i3x/Talks/GenericitySeattle.ho.pdf ), I found what I think is a good bunch-resisting, kind-example:
- (2a) Transistors were invented by Shockley.
One can't get the same result by referring to any bunch:
- (2b) *There exist some transistors that were invented by Shockley.
Nor does taking the biggest possible bunch of transistors help:
- (2c) *All transistors were invented by Shockley.
It seems that though transistors as a kind of thing were invented, no mundane transistor nor any extension, ensemble, or bunch of them was invented. In (2a) there does seem to be some sort of "transistor kind" (dare I say "form") above the mundane, even taking into consideration the possible worlds that Montague would have in his model.
John E. Clifford
Jorge Llambías
> Kinds don't seem to be the sort of
> things that ruin gardens, though their exemplars may. The factual
> situation, as far as transistors, etc. are concerned, is about genealogy,
> all transistors descend from something invented by Shockley. But that is at
> least as hard to express as types, so I wait a while on it.
The problem with that take is the usual one, you cannot say things
like "transistors, which were invented by Shockley (although not
quite, according to Wikipedia), are found today in practically all
electronic devices" if you don't allow that there is something both
invented by Shockley and found today in practically all electronic
devices.
I liked it better when you were of the opinion that the way in which
they were invented by Shockley may be different from the way in which
they are found all over, but that either way they are transistors, or
"the transistor" (as in English).
maikxlx
FWIW, Bunt's "ensembles" which I mentioned are a bit different than your L-sets. Ensembles are a set-like structure in which mass is the basic concept, and atomic individuals are derived or secondary. Any non-empty ensemble may contain or atomic members (essentially a count ensemble), or for lack of a better word non-atomic "stuff" (mass ensemble). It could also contain both atoms and non-atomic stuff. Count and mass ensembles may combine in predicate relationships freely as in "The five rings were gold". So they seem pretty useful for capturing human language semantics.
Pierre Abbat
> But yesterday as I was reading random online materials (this one -
> http://amor.cms.hu-berlin.de/~h2816i3x/Talks/GenericitySeattle.ho.pdf ), I
> found what I think is a good bunch-resisting, kind-example:
>
> - (2a) Transistors were invented by Shockley.
>
> One can't get the same result by referring to any bunch:
>
> - (2b) *There exist some transistors that were invented by Shockley.
>
> Nor does taking the biggest possible bunch of transistors help:
>
> - (2c) *All transistors were invented by Shockley.
>
> It seems that though transistors as a kind of thing were invented, no
> mundane transistor nor any extension, ensemble, or bunch of them was
> invented. In (2a) there does seem to be some sort of "transistor kind"
> (dare I say "form") above the mundane, even taking into consideration the
> possible worlds that Montague would have in his model.
Would it be correct to say "lo'e za'e grezunca'a cu se finti la .caklis."?
Also, does my example "lo'e .ornitorinku na fadni mabru .iki'ubo na'o se
jbena re sovda" involve a kind?
Pierre
--
I believe in Yellow when I'm in Sweden and in Black when I'm in Wales.
John E. Clifford
Sent from my iPad
John E Clifford
I'm not at all sure what this says about Lojban and {lo} expressions or about levels, come to that. But at least I am, I think finally near the page you all have been on for awhile.
Subject: Re: [lojban] Lions and levels and the like
Ah, you jarred my memory. Having studied Bunt, I think that I can make some sense of your earlier writings about Leśniewskian sets (L-sets) and Cantorian or classical sets (C-sets) e.g. http://pckipo.blogspot.com/2009/09/c-sets-and-l-sets-draft.html I take it that your "bunches" are basically L-sets as there described then? I am not quite sure how L-sets work differently than C-sets, other than they can't be nested and {a} = a. C-sets clearly handle individuals efficiently and masses poorly; how do L-sets handle masses and individuals? Or is there a difference?
FWIW, Bunt's "ensembles" which I mentioned are a bit different than your L-sets. Ensembles are a set-like structure in which mass is the basic concept, and atomic individuals are derived or secondary. Any non-empty ensemble may contain or atomic members (essentially a count ensemble), or for lack of a better word non-atomic "stuff" (mass ensemble). It could also contain both atoms and non-atomic stuff. Count and mass ensembles may combine in predicate relationships freely as in "The five rings were gold". So they seem pretty useful for capturing human language semantics.
Hmmm! Nice case! Of course, some first transistor must have been invented that all the others copied and improved upon, but that doesn't really dodge your point. At the moment, I don't know what to suggest, except to hope that Lojban still has a word for kinds. Bunches are, inter alia, Lesniewski's wholes (but xorxes doesn't like this kind of objectifying, preferring plural reference, which works the same way formally). I don't take 1a to be about kinds, but just about some unspecified bunch of lions (at least in Lojban, lo cinfo). Kinds don't seem to be the sort of things that ruin gardens, though their exemplars may. The factual situation, as far as transistors, etc. are concerned, is about genealogy, all transistors descend from something invented by Shockley. But that is at least as hard to express as types, so I wait a while on it.Sent from my iPad
I can understand the appeal of your concept of bunches -- if I understand them correctly as being something like subsets of the extensions consisting of mundanes/atoms (perhaps generalized to something like Bunt's ensemble, derivative of Leśniewski 's mereology, to cover masses). E.g.:
- (1a) Lions are ruining my garden.
- (1b) There exist some lions that are ruining my garden.
where (1a) invokes a kind and (1b) invokes a bunch or somesuch, and yet both sentences seem to have the same truth conditions or almost the same.
But yesterday as I was reading random online materials (this one - http://amor.cms.hu-berlin.de/%7Eh2816i3x/Talks/GenericitySeattle.ho.pdf ), I found what I think is a good bunch-resisting, kind-example:
maikxlx
Would it be correct to say "lo'e za'e grezunca'a cu se finti la .caklis."?
i mi na jinvi. .i ba'e lo za'e grezunca'a cu se finti la .caklis. .iku'i no da poi za'e grezunca'a zo'u da se finti la .caklis. .i ku'inai naku lo'e za'e grezunca'a cu se finti
Also, does my example "lo'e .ornitorinku na fadni mabru .iki'ubo na'o se
jbena re sovda" involve a kind?
i frica. .i lo .ornitorinku na fadni mabru .ije ja'a ro da poi .ornitorinku zo'u da na fadni mabru .i ku'inai ja'a lo'e .ornitorinku na fadni mabru
(1a) Trees are widespread. [trees as one giant mass, one "snapshot" glimpse]
(1b) Dodos are extinct. [dodos as a one species, one extinction event]
(1c) Birds evolved from dinosaurs. [(at least according to science) similar to (1b)]
(1d) Transistors were invented by Shockley et al. [the whole class of components, one invention event]
In none of these cases does it seem that ANY individual satisfies the predicate that the kind does, at least not with the same reading. Generics work differently:
(2a) Dogs are mammals. [intrinsic property for individuals: All(x): Px -> Qx]
(2b) Insects are six-legged. [characteristic near-universal property: Most(x): Px -> Qx]
(2c) Lions are ferocious. [characteristic common property: Many(x): Px -> Qx]
(2d) Humans have walked on the moon. [existential property: Exists(x): Px and Qx]
I annotated each case using (non)standard quantifiers with "x" as a variable of individuals, Px corresponding to the English subject NP and Qx to the VP, in order to compare what portion of the individuals satisfy the predicate . In these cases, it seems that some individuals at least MAY satisfy the same predicate that the generic does. In fact, generic interpretations seem to arise "squinting" at the whole class. More precisely, we are quantifying over individuals and situations.
I believe that the reason for the difference (assuming what I am saying here holds up) basically boils down to VERB ASPECT. Kinds seem to involve a perfect(ive) aspect of some sort describing episodic events that don't pertain to any individual. Generics seem to involve some sort of generic, habitual, or similar aspect describing numerous situations that may involve every individual. In Lojban, these two very different aspects need not be marked. However, by avoiding the need to mark aspect or any other grammatical category (which works fine in normal human usage I think), Lojban does incur a little bit of confusion in the formal semantic analysis.
Incidentally, IMO (as shown above) given the canonical definition of the "typical that really is", {lo'e} can refer to generics but not kinds. If the "typical" broda does something, then certainly at least one individual broada should be able to do it, which true kind-predication disallows. One can argue that {lo'e} thus can mark generics at least, but I think it would make more sense to use aspect cmavo on the bridi. What boils down to verb aspect shouldn't be marked on the gadri. That's totally my 2 cents though.
Pierre
-Mike
maikxlx
To be clear about one thing, when I said "atoms" previously I only meant "individuals" and not the things that compose molecules. In fact, masses in ensembles are a bit weird in the sense that there is absolutely no necessary commitment whatsoever to the notion of "smallest things" (except exactly when you need them). There are good reasons for this, both theoretical and practical. In the real number line, there is an infinite number of line segments and none of them are smallest. In the case of, say, water (defined as a liquid), it's really impossible to say what a smallest part actually is. A single molecule is not water, though a small group may be, obviously overlapping with other potentially smallest groups before dissolving into its surroundings. But since we're dealing with natural language semantics, there is never a practical need for "water" to encode these vanishingly small quantities, and there are other predicates to do the job.
John E Clifford
There do remain a number of cases to which this general notion does not seem to apply. Your case of the real line is one, letters seem to be another. Here the approach seems to be to start at the top, work down and then back up, I think, but I don't know just how that goes. Even with cases that fit this pattern pretty well, water, for excample, there are some problems, as you note. Water molecules don't display the characteristic behavior of water (as do not also ice and steam), since they don't flow, etc. But then, gold atoms don't shine and are not malleable, so this seems a minor problem. And for generic cases, they probably are not significant, since the far more numerous and visible sub bunches will take over the "statistics".
I find you idea of an aspect difference among the various uses of kinds (max. bunches) interesting, though I still tend to think of them in terms of different connections to predicates, a relic of the early days of plural reference. And, indeed, even with aspects, some of this will still come into play with regular bunches, that is to say, bunches which do not claim to take in all the possible "atoms" (I am used to your usage here).
I kinda like this result, since it leaves basic things basic but covers kinds and masses economically from them. Until some real snag comes along.
Martin Bays
> But yesterday as I was reading random online materials (this one -
> http://amor.cms.hu-berlin.de/~h2816i3x/Talks/GenericitySeattle.ho.pdf ), I
> found what I think is a good bunch-resisting, kind-example:
>
> - (2a) Transistors were invented by Shockley.
I don't think we should feel obliged to treat as a jbokind (whatever
they end up being) everything for which English uses a bare plural.
In this case, surely what Shockley invented was the idea (si'o) of
a transistor.
Martin
John E. Clifford
Sent from my iPad
Robert LeChevalier
> Hmmm! Nice case! Of course, some first transistor must have been
> invented that all the others copied and improved upon, but that doesn't
> really dodge your point. At the moment, I don't know what to suggest,
> except to hope that Lojban still has a word for kinds. Bunches are,
> inter alia, Lesniewski's wholes (but xorxes doesn't like this kind of
> objectifying, preferring plural reference, which works the same way
> formally). I don't take 1a to be about kinds, but just about some
> unspecified bunch of lions (at least in Lojban, lo cinfo). Kinds don't
> seem to be the sort of things that ruin gardens, though their exemplars
> may. The factual situation, as far as transistors, etc. are concerned,
> is about genealogy, all transistors descend from something invented by
> Shockley. But that is at least as hard to express as types, so I wait a
> while on it.
Actually, I just read something on this a couple days ago.
There was a patent issued in Canada for a field-effect transistor in
1925, but it isn't clear if any were built. A similar patent was issued
in Germany in 1934. Neither has much to do with the Shockley transistor
other than name and general semiconductor nature. Bardeen and Brattain
then invented the germanium transistor in 1947, but Shockley ended up
getting a share of the credit. There was an independent invention in
France in 1948. Texas Instruments produced the first silicon transistor
in 1954, and MOS-FETs came even later, and I think they are more like
the 1925 idea than like the 1947 one. It would be difficult to call any
of the reinventions prior to 1948 as being "family tree descendants".
Those afterwards would probably give some ancestral credit to Shockley's
group, but that might be a legal fiction due to patent law rather than
truly being genealogical.
The same probably goes for the brothers' Wright and the airplane. Many
others were independently inventing them when the Wrights first flew,
and it is hard to argue that airplanes are necessarily more descended
from Wright's plane than any of those others.
How to express these distinctions in Lojban is beyond anything I would
attempt to argue, especially post xorlo.
lojbab
maikxlx
As I noted, in L-sets, membership and subsets are not distinct, i.e. every member is a subset and under certain interpretations, ever subset is a member, though some members are more basic, having no members but themselves. So, a bunch of things also contains all the combinations of them. From this point of view, then, a mass is just the bunch of its lowest level (what it is a bunch of, say): gold is the maximal bunch of gold atoms, lion is the maximal bunch of lion cells and so on. More familiar object arise in the intersection of bunches: lions are in the intersection of lion and living organisms and so. And every gold thing is in the intersection of gold and its particular form. The only Lojbanic thing I see in all of this at the moment is that that maximal bunch ought to be given a separate gadri.
Well, subsetship (more generally, partship) and membership seem to work a little differently in Bunt's system. As I learned from reading here (unfortunately several preview pages missing) it turns out that ensembles are essentially C-sets and L-sets generalized under one theory and reducible to either by excluding the other; you get set theory from ensemble theory by excluding L-sets, and mereology by excluding C-sets. (IMHO this seems ideal.) As far as intersections as in "golden ring", I believe that the "overlapped" ensemble of "continuous" masses like gold and "discrete" collections like rings would necessarily be empty, but I am not 100% sure on this point. I need better documentation. At worst, I am sure you could construct a mixed ensemble that would merge all continuous masses of gold and all discrete things made of gold to get the needed result for your intersection. So maybe, just maybe, Lojban is doing the Right Thing by having predicates generally ignore the count/mass distinction.
There do remain a number of cases to which this general notion does not seem to apply. Your case of the real line is one, letters seem to be another. Here the approach seems to be to start at the top, work down and then back up, I think, but I don't know just how that goes. Even with cases that fit this pattern pretty well, water, for excample, there are some problems, as you note. Water molecules don't display the characteristic behavior of water (as do not also ice and steam), since they don't flow, etc. But then, gold atoms don't shine and are not malleable, so this seems a minor problem. And for generic cases, they probably are not significant, since the far more numerous and visible sub bunches will take over the "statistics".
I find you idea of an aspect difference among the various uses of kinds (max. bunches) interesting, though I still tend to think of them in terms of different connections to predicates, a relic of the early days of plural reference. And, indeed, even with aspects, some of this will still come into play with regular bunches, that is to say, bunches which do not claim to take in all the possible "atoms" (I am used to your usage here).
I kinda like this result, since it leaves basic things basic but covers kinds and masses economically from them. Until some real snag comes along.
I am sure there will be controversy in some of these areas at least until Lojban's semantics are much more rigorously formalized.
-Mike
John E Clifford
Of course the point of the masses I have looked at so far is that they are not continuous but ultimately discrete, as most masses are. Thus, there is nothing odd about a gold ring being at the same time a ring-shaped thing and a chunk of gold or of a pond being a bunch of water molecules. There do seem to be a few problem cases still, though.
What you are calling gnomic aspect I am calling statistical predication, more or less. Since we have no way to express any of this in Lojban that I can find, just what we call it is not important. But I am unsure about just what you now mean by kinds and generics. I don't find either "Dogs are mammals" or "Man walked on the Moon" in any way odd. But I do worry about introducing intensions into all this. To be sure, looking for a unicorn clearly takes out of the present domain of discourse to another and that move may be inherently intensional, the -- by fiat, to be sure -- the intensional part falls into {tu'a} and the like, not into the {lo}expression. In short, kinds -- if that is what is involved here and in cases like extinction or creation -- seem to me to be exactly about extensions, just maybe not this extesnsion.
Subject: Re: [lojban] Lions and levels and the like
John E Clifford
there are always folks before the glory guy. But the overall logic of the
situation for the present context seems pretty clear, even if we don't quite
know what goes where in the early part of the process (and probably never
will). In a word, we should let some ugly facts muck up a good theory.
----- Original Message ----
From: Robert LeChevalier <loj...@lojban.org>
To: loj...@googlegroups.com
Sent: Wed, November 16, 2011 4:21:00 PM
Subject: Re: [lojban] Lions and levels and the like
lojbab
-- You received this message because you are subscribed to the Google Groups
maikxlx
On Wed, Nov 16, 2011 at 8:30 AM, John E Clifford <kali9...@yahoo.com> wrote:
The only Lojbanic thing I see in all of this at the moment is that that maximal bunch ought to be given a separate gadri.
I believe that pretty much {lo si'o broda} = {lo sidbo be lo broda}, where lo broda is again is a reference to a kind with clearly _no_ actuals/mundanes/instances, either needed or wanted. Would you rewrite {lo broda cu sidbo} as {lo si'o broda cu sidbo}?
The idea that both you and John invoke is to grammaticize Lojban's kind references (and John would probabIy like to grammaticize generics as well). In all honesty, I agree we don't have to follow English, and it's worth considering the idea theoretically. It might actually be beneficial*, but aside from the arguable redundancy of kind-marking, I don't believe that anyone has cataloged the large number of bridi places with respect to kinds and generics; each place may well admit one, the other, or both. Until that happens we really don't know how big of a change this would entail -- or at least _I_ don't as a not-so-frequent user.
*For example clearing up the specific/nonspecific distinction in "mi sisku lo broda" would be valuable and should be mandatory -- assuming that both readings are allowed, as I believe they are.
-Mike
maikxlx
I haven't quite figured out yet how C-sets and L-sets can be combined in one theory, aside from just running both, with different symbols, say. But then I don't see what the interaction between them would be. I'll work on this.
Well Bunt claims to combine them in his theory, has proved consistency and equivalency with Zermelo-Fraenkel -- all this with natural language semantics aforethought. Unfortunately I don't have any good links. I have been studying Bunt's stuff off scraps I find on the Internet. I suggest the link on Google Books I gave above; it explains ensembles pretty well even if there are pages missing. I will try to write a short sketch myself on ensembles in the near future when I have time, but the literature out there is better than what I can write.
I don't find either "Dogs are mammals" or "Man walked on the Moon" in any way odd.
As I tried to convey in my reply to Pierre Abbat, it's just a little odd that that in the former case we have a necessary universal situation and in the latter case we have a marginal existential situation, and yet in Lojban both have (or are allowed to have) exactly the same logical form. Maybe not odd, but curious at least.
But I do worry about introducing intensions into all this. To be sure, looking for a unicorn clearly takes out of the present domain of discourse to another and that move may be inherently intensional, the -- by fiat, to be sure -- the intensional part falls into {tu'a} and the like, not into the {lo}expression. In short, kinds -- if that is what is involved here and in cases like extinction or creation -- seem to me to be exactly about extensions, just maybe not this extesnsion.
{tu'a} is a bit old school isn't it? We are already introducing an intension whenever {lo} refers to a kind, as in {lo grezunca'a cu se finti la .caklis.} or {lo ciprdodo cu jutmro} or {ko'a sisku lo pavyseljirna}.
Jorge Llambías
> On Wed, Nov 16, 2011 at 10:37 PM, John E Clifford <kali9...@yahoo.com>
> wrote:
>>
>> I don't find either "Dogs are mammals" or "Man walked on the Moon" in any
>> way odd.
>
> As I tried to convey in my reply to Pierre Abbat, it's just a little odd
> that that in the former case we have a necessary universal situation and in
> the latter case we have a marginal existential situation, and yet in Lojban
> both have (or are allowed to have) exactly the same logical form. Maybe not
> odd, but curious at least.
Similarly curious are "Merkel is a woman" and "Merkel had a beer",
with the same logical form even though in one case we have a universal
situation and in the other an existential one.
John E. Clifford
Sent from my iPad
John E. Clifford
Sent from my iPad
John E. Clifford
Sent from my iPad
maikxlx
Yes but following my own intuition (and possibly that of others here),
I am not sure that the similarity between the two pairs of sentences
arises from any sort of similarity between the individuals of {lo
gerku}/{lo remna} and the "stages" of {la .merkel.}.
Instead I think that in each pair, one sentence simply expresses a
proposition that happens to be true under all possible situations
(where a "situation" is a combined world/time index (w, t)), and the
other under one (or more) specific actual situation. The truth
conditions of a sentence can be strengthened if desired by adding
either a gnomic or an episodic aspect marker to the bridi; e.g.{ca}
in {la .merkel. ca pinxe lo birje} shifts the situation to present
tense, episodic aspect; {-N} in {la .merkel. -N ninmu} quantifies
over all situations gnomically (where -N means "necessarily" or
"intrinsically", I'd appreciate it if someone could tell me the
word(s) I'm looking for) .
It dawns on me in passing that in the case of {lo remna cu mabru} what
appears to be universal quantification over x1 is probably built into
the meaning {mabru} in a similar way that a kind-abstractor over x2
seems to be built into the meaning of {finti}. So maybe {X mabru}
entails {ro X mabru} automatically by predicate definition, and maybe
these "curiosities" are fewer than they appear.
To be clear, I don't see any problem using xorlo in any of these four
sentences or similar generic ones (my verdict is out on kinds). I
would like just to account for what I regard as the curiosities.
Jorge Llambías
>
> {-N} in {la .merkel. -N ninmu} quantifies
> over all situations gnomically (where -N means "necessarily" or
> "intrinsically", I'd appreciate it if someone could tell me the
> word(s) I'm looking for) .
Lojban doesn't have a word for the dual of "ka'e", so we have to make
do with "na ka'e na".
maikxlx
On Thu, Nov 17, 2011 at 3:32 PM, maikxlx <mai...@gmail.com> wrote:
> It dawns on me in passing that in the case of {lo remna cu mabru} what
> appears to be universal quantification over x1 is probably built into
> the meaning {mabru} in a similar way that a kind-abstractor over x2
> seems to be built into the meaning of {finti}. So maybe {X mabru}
> entails {ro X mabru} automatically by predicate definition, and maybe
> these "curiosities" are fewer than they appear.
>
This is totally wrong under both xorlo and CLL. Starting with:
(1a) {lo lurdzu cu remna}, which seems uncontroversial insofar as all
moon walkers have been human. We've already established:
(1a) {lo remna cu lurdzu}, despite the fact that only a small part on
humanity has walked on the moon. Just as uncontroversial as (1a) is:
(2a) {lo mabru cu danlu}. If (1a):(1b)::(2a):(2b), then:
(2b) {lo danlu cu mabru}, however curious*, must also be acceptable at
least under some interpretations. Since:
(2b') {ro lo danlu cu mabru} is always untrue, it follows that the x1
of {mabla} cannot have implicit universal quantification under xorlo
logic without contradiction with (2b).
*I say this is curious because if xorlo {lo danlu} is glorked
(contextually, say) as danlu-kind, then this is false (as it
intuitively should be), whereas {lo mabru cu danlu} is probably always
true under any domain.
maikxlx
> On Thu, Nov 17, 2011 at 5:32 PM, maikxlx <mai...@gmail.com> wrote:
> Lojban doesn't have a word for the dual of "ka'e", so we have to make
> do with "na ka'e na".
>
I'm surprised at that! I wasn't 100% sure that {ka'e}/{kakne} were the
correct words for modal possibility. Is it fair to say that the
"conditions" {te kakne} are the worlds themselves in which {da kakne},
or is x3 more general?
Jorge Llambías
> 2011/11/17 Jorge Llambías <jjlla...@gmail.com>:
>> On Thu, Nov 17, 2011 at 5:32 PM, maikxlx <mai...@gmail.com> wrote:
>> Lojban doesn't have a word for the dual of "ka'e", so we have to make
>> do with "na ka'e na".
>
> I'm surprised at that! I wasn't 100% sure that {ka'e}/{kakne} were the
> correct words for modal possibility.
"ka'e" is "fi'o se cumki". The connection with "kakne" is just mnemonic.
> Is it fair to say that the
> "conditions" {te kakne} are the worlds themselves in which {da kakne},
> or is x3 more general?
I'd say world-talk is metalinguistic, so, sort of.
maikxlx
> "ka'e" is "fi'o se cumki". The connection with "kakne" is just mnemonic.
>
Is this a newish development? {ka'e} is also a rafsi of {kakne},
often a strong sign of relatedness. Meanwhile, vlasisku, BPFK section
CAhA, cmavo.txt and the CLL say nothing about {cumki} wrt {ka'e}.
Also, while {cumki} does express possibility, {ka'e}, from the given
definitions, seems to be more about ability than possibility. In
order to say things like "it possibly brodas" and "it necessarily
brodas" I have to believe that these concepts should have their own
words, without mixing ability into it. These primitive logical
operators strike me as vastly worth assigning two disyllables from
cmavo space, especially in light of some of the other things
available. Just my 2 cents.
Jorge Llambías
> 2011/11/17 Jorge Llambías <jjlla...@gmail.com>:
>> "ka'e" is "fi'o se cumki". The connection with "kakne" is just mnemonic.
>
> Is this a newish development? {ka'e} is also a rafsi of {kakne},
> often a strong sign of relatedness.
Almost all CV(')V cmavo have the same form of a rafsi of something. My
guess is the ones that are related to that something are in the
minority. "ka'e" was obviously taken from "kakne", yes, but the
connection is kind of malglico. Similarly "pe'i" comes from "pensi",
"ti'e" from "tirna", and thare are other mnemonics that go through
malglico glosses.
> Meanwhile, vlasisku, BPFK section
> CAhA, cmavo.txt and the CLL say nothing about {cumki} wrt {ka'e}.
In jbovlaste "ka'e" is defined as "fi'o se cumki". But since I wrote
that definition I guess I can't count that as evidence. :)
> Also, while {cumki} does express possibility, {ka'e}, from the given
> definitions, seems to be more about ability than possibility.
But whose ability? Each of the arguments of the relation modified by
"ka'e"? The x1? The agent (assuming there is one)?
> In
> order to say things like "it possibly brodas" and "it necessarily
> brodas" I have to believe that these concepts should have their own
> words, without mixing ability into it.
I agree that the word "ability" should not appear in the definition of
CAhAs, since events don't really have abilities.
> These primitive logical
> operators strike me as vastly worth assigning two disyllables from
> cmavo space, especially in light of some of the other things
> available. Just my 2 cents.
I agree. I have said before that it is extremely weird that a logical
language doesn't have a word for the "necessarily" operator.
maikxlx
>
> On Thu, Nov 17, 2011 at 9:22 PM, maikxlx <mai...@gmail.com> wrote:
> > 2011/11/17 Jorge Llambías <jjlla...@gmail.com>:
> >> "ka'e" is "fi'o se cumki". The connection with "kakne" is just mnemonic.
> >
> > Is this a newish development? {ka'e} is also a rafsi of {kakne},
> > often a strong sign of relatedness.
>
> Almost all CV(')V cmavo have the same form of a rafsi of something. My
> guess is the ones that are related to that something are in the
> minority. "ka'e" was obviously taken from "kakne", yes, but the
> connection is kind of malglico. Similarly "pe'i" comes from "pensi",
> "ti'e" from "tirna", and thare are other mnemonics that go through
> malglico glosses.
>
nothing to do with pianos despite sharing {pipno}'s rafsi's form.
With {ka'e}, one would not think it was such an accident. In
principle of course no cmavo need to be related to the gismu with that
cmavo's form.
>
> > Meanwhile, vlasisku, BPFK section
> > CAhA, cmavo.txt and the CLL say nothing about {cumki} wrt {ka'e}.
>
> In jbovlaste "ka'e" is defined as "fi'o se cumki". But since I wrote
> that definition I guess I can't count that as evidence. :)
>
It's only in the Lojban record! Side note: which should I rely on
more, vlasisku or jbovlaste? I find vlasisku's cross linking and more
complete search results to be superior. If someone rolled in the
BPFK definitions and CLL sections, it would be almost ideal.
> > Also, while {cumki} does express possibility, {ka'e}, from the given
> > definitions, seems to be more about ability than possibility.
>
> But whose ability? Each of the arguments of the relation modified by
> "ka'e"? The x1? The agent (assuming there is one)?
>
You're asking me?! Well since you asked, from what I see, I would
definitely assume the x1, given the glosses, proposed keywords, and
examples in the CLL and BPFK. In particular the CLL examples indicate
very clearly that {ka'e} and related CAhA are some sort of short-scope
selbri modifiers and emphatically _not_ true modal operators with
scope over the whole bridi.
> > In
> > order to say things like "it possibly brodas" and "it necessarily
> > brodas" I have to believe that these concepts should have their own
> > words, without mixing ability into it.
>
> I agree that the word "ability" should not appear in the definition of
> CAhAs, since events don't really have abilities.
>
It's not just "ability" that seems off, it's also the ambiguous "can"
and "innate capability" as well as the conspicuous absence of "may",
"might" and above all "POSSIBLE".
> > These primitive logical
> > operators strike me as vastly worth assigning two disyllables from
> > cmavo space, especially in light of some of the other things
> > available. Just my 2 cents.
>
> I agree. I have said before that it is extremely weird that a logical
> language doesn't have a word for the "necessarily" operator.
>
The fact that there is no necessity operator strongly suggests that
the language designers did not have the foggiest notion of modal logic
when they created {ka'e}. It's clear to me from the evidence that
{ka'e} is at best roughly related, but not identical, to the
possibility operator. At the very least, it seems muddled and
contaminated with malglico. I do not read a ton of Lojban, but I find
it very doubtful that common usage is substantially better than the
flawed CLL examples. Therefore I would respectfully suggest
considering two new uncontaminated cmavo to act as true and
contaminated, wide-scope modal-logical operators:
ci'a = "it is possible that; possibly; may/might" (looks vaguely like 'cumki')
ne'e = "it is necessary that; necessarily; must" (looks vaguely like
'necessary')
> mu'o mi'e xorxes
>
mu'o mi'e .maik.
John E Clifford
a couple obvious senses, Merkel could be male. In others, not so much.
----- Original Message ----
From: Jorge Llambías <jjlla...@gmail.com>
To: loj...@googlegroups.com
Sent: Thu, November 17, 2011 4:58:28 PM
Subject: Re: [lojban] Lions and levels and the like
--
John E Clifford
danlu cu mabru} case as disjunctive predication, assuming that {lo danlu} is
here one of those superbunches. But this presents no problems (as do not the
others either) if taken as local bunches: unknown animal being investigated,
find drops of lactation in lair, so conclude above.
----- Original Message ----
From: maikxlx <mai...@gmail.com>
To: loj...@googlegroups.com
Sent: Thu, November 17, 2011 5:11:00 PM
Subject: Re: [lojban] Lions and levels and the like
--
John E Clifford
nullified.
----- Original Message ----
From: maikxlx <mai...@gmail.com>
To: loj...@googlegroups.com
Sent: Thu, November 17, 2011 5:29:34 PM
Subject: Re: [lojban] Lions and levels and the like
--
John E Clifford
However, here we do have some kinds of distinctions about what the opposing
force is (I forget whether {cunmki} has a place for conditions -- it should).
----- Original Message ----
From: maikxlx <mai...@gmail.com>
To: loj...@googlegroups.com
Sent: Thu, November 17, 2011 6:22:41 PM
Subject: Re: [lojban] Lions and levels and the like
--
John E Clifford
----- Original Message ----
From: maikxlx <mai...@gmail.com>
To: loj...@googlegroups.com
Sent: Thu, November 17, 2011 6:22:41 PM
Subject: Re: [lojban] Lions and levels and the like
--
John E Clifford
logic at all. I don't know how well he did in even that one (Lojbab does not
improve the logic input much). But in 56 years, the efforts to get necessity
operators in have come to naught -- though eventually we got something like a
necessity predicate,
----- Original Message ----
From: maikxlx <mai...@gmail.com>
To: loj...@googlegroups.com
Sent: Thu, November 17, 2011 10:23:00 PM
Subject: Re: [lojban] Lions and levels and the like
--
maikxlx
> The glorking can be made explicit in various ways. I, of course, take the {lo
> danlu cu mabru} case as disjunctive predication, assuming that {lo danlu} is
> here one of those superbunches. But this presents no problems (as do not the
> others either) if taken as local bunches: unknown animal being investigated,
> find drops of lactation in lair, so conclude above.
>
that 400+ email thread on {zo'e} which prompted my email on
truth-conditional semantics). But in the case of a specific kind of
unknown animal, say a duck-billed platypus, intuitively I would want
to say {le danlu cu mabru}. That's how I have internalized Lojban,
whether that is currently right or wrong.
Off the top of my head, the only context in which I can imagine saying
something like {lo danlu cu mabru} is if I were trying to explain to
someone what a {danlu} is, e.g. {.i lo danlu cu mabru .i lo danlu cu
cinki .i lo danlu cu se klesi so'i lo jmive}
maikxlx
> JCB had at least one course in Logic but in a school that did not favor modal
> logic at all. I don't know how well he did in even that one (Lojbab does not
> improve the logic input much). But in 56 years, the efforts to get necessity
> operators in have come to naught -- though eventually we got something like a
> necessity predicate,
>
If I may confess something, I have been studying Lojban on and off for
years, and every time I get into a learning groove I encounter some
facet of the language that strikes me as so bizarre or absurd that it
stops me in my tracks. The lack of a necessity operator and the
questionable status of {ka'e} make the current situation no exception.
I had absolutely no intention of suggesting reforms or additions
because although nearly aspect of the language screams for them, the
fact of the matter is that reform is not in the cards and the
language's foundation is pretty much set in stone. However here I
think that I will simply use {ne'e} when I need it, and possibly
{ci'a} too until I convince myself that {ka'e} does what xorxes claims
it does. I understand there is an experimental cmavo process, but I
am going to short-circuit it. Waiting 56 years for "necessarily" is
outrageous.
Michael Turniansky
On Fri, Nov 18, 2011 at 11:29 AM, John E Clifford <kali9...@yahoo.com> wrote:Is there a brivla for logical or modal necessity?
> JCB had at least one course in Logic but in a school that did not favor modal
> logic at all. I don't know how well he did in even that one (Lojbab does not
> improve the logic input much). But in 56 years, the efforts to get necessity
> operators in have come to naught -- though eventually we got something like a
> necessity predicate,
>
If I may confess something, I have been studying Lojban on and off for
years, and every time I get into a learning groove I encounter some
facet of the language that strikes me as so bizarre or absurd that it
stops me in my tracks. The lack of a necessity operator and the
questionable status of {ka'e} make the current situation no exception.
I had absolutely no intention of suggesting reforms or additions
because although nearly aspect of the language screams for them, the
fact of the matter is that reform is not in the cards and the
language's foundation is pretty much set in stone. However here I
think that I will simply use {ne'e} when I need it, and possibly
{ci'a} too until I convince myself that {ka'e} does what xorxes claims
it does. I understand there is an experimental cmavo process, but I
am going to short-circuit it. Waiting 56 years for "necessarily" is
outrageous.
mu'o mi'e .maik.
John E Clifford
specificity (I think that's the right word, technically); you know pretty much
exactly what you are talking about (even if you get the predicate wrong). I
tend not to use it until I have already talked in a more general way about the
referent or when it is also overtly marked (possessive pronoun, "here" and the
like). But in the case of a focused search of this sort, it seems appropriate.
It happens that {lo} is always appropriate -- if you have the predicate right.
So, yes, Ido use {le} when I am in doubt, or when I think mt listeners have a
different belief about the object question ("Juno was a man" sorts of caases,
The Quiet Game?)
----- Original Message ----
From: maikxlx <mai...@gmail.com>
To: loj...@googlegroups.com
Sent: Fri, November 18, 2011 11:01:54 AM
Subject: Re: [lojban] Lions and levels and the like
--
maikxlx
<mturn...@gmail.com> wrote:
>
>
> On Fri, Nov 18, 2011 at 12:39 PM, maikxlx <mai...@gmail.com> wrote:
>>
>> On Fri, Nov 18, 2011 at 11:29 AM, John E Clifford <kali9...@yahoo.com>
>> wrote:
>> > JCB had at least one course in Logic but in a school that did not favor
>> > modal
>> > logic at all. I don't know how well he did in even that one (Lojbab
>> > does not
>> > improve the logic input much). But in 56 years, the efforts to get
>> > necessity
>> > operators in have come to naught -- though eventually we got something
>> > like a
>> > necessity predicate,
>> >
>> Is there a brivla for logical or modal necessity?
>
>
> Does nibli meet your requirements?
> --gejyspa
>
work, at least not with anything close to reasonable succinctness,
unless you know a way. What we are trying to express is the
sentence-scope adverb "necessarily(S)" which means "in all possible
states of affairs, S" or more longwindedly "there is no possible state
of affairs such that it is not true that S". For example:
(1) {lo remna ne'e mabru}
= Humans are necessarily mammals.
= In all possible states of affairs, humans are mammals.
(2) {ro nanmu je se mensi ne'e bruna}
= All men with sisters are necessarily brothers.
= There is no possible state of affairs such that it is not true that
all men with sisters are brothers.
John E Clifford
----- Original Message ----
From: maikxlx <mai...@gmail.com>
To: loj...@googlegroups.com
Sent: Fri, November 18, 2011 11:39:33 AM
Subject: Re: [lojban] Lions and levels and the like
--
John E Clifford
From: Michael Turniansky <mturn...@gmail.com>
To: loj...@googlegroups.com
Sent: Fri, November 18, 2011 11:51:35 AM
Subject: Re: [lojban] Lions and levels and the like
John E Clifford
restrictions on abilities, etc.
----- Original Message ----
From: maikxlx <mai...@gmail.com>
To: loj...@googlegroups.com
Sent: Fri, November 18, 2011 12:12:28 PM
Subject: Re: [lojban] Lions and levels and the like
--
Michael Turniansky
maikxlx
> But you specifically asked for a brivla.
Right. I just figured you have more usage under your belt than I have
so I was curious what you could come up with, and you showed that
{nibli} can easily handle some cases. However, I don't think that
wanting a convenient way of saying "necessarily(S)" is hair-splitting
or counting angels on a pinhead.
maikxlx
> Amen. The line for Logjam III forms on the left.
>
has inspired dozens of reform projects over the years.
On Fri, Nov 18, 2011 at 1:16 PM, John E Clifford <kali9...@yahoo.com> wrote:
> Not really, but it can be massaged to do the work. The ideal would be "x1
> is necessary under condition/compulsion x2"
>
As an early Christmas present, I offer you and the Lojban community {zilni'i}.
Jorge Llambías
> 2011/11/17 Jorge Llambías <jjlla...@gmail.com>
>> On Thu, Nov 17, 2011 at 9:22 PM, maikxlx <mai...@gmail.com> wrote:
>
>> > Also, while {cumki} does express possibility, {ka'e}, from the given
>> > definitions, seems to be more about ability than possibility.
>>
>> But whose ability? Each of the arguments of the relation modified by
>> "ka'e"? The x1? The agent (assuming there is one)?
>>
> You're asking me?!
You seemed to think the given definitions made sense...
> Well since you asked, from what I see, I would
> definitely assume the x1, given the glosses, proposed keywords, and
> examples in the CLL and BPFK. In particular the CLL examples indicate
> very clearly that {ka'e} and related CAhA are some sort of short-scope
> selbri modifiers and emphatically _not_ true modal operators with
> scope over the whole bridi.
But CAhAs are tags, and all other tags are bridi operators. If "ka'e
citka" and "ka'e se citka" have different meanings (besides reordered
places), CAhA works nothing like other tags.
And I don't know what you would do with "ka'e na broda", or "ka'e ku
na ku broda", given that "na" has bridi scope, and "ka'e" appears to
have scope over "na" in those cases.
>> I agree that the word "ability" should not appear in the definition of
>> CAhAs, since events don't really have abilities.
>>
> It's not just "ability" that seems off, it's also the ambiguous "can"
> and "innate capability" as well as the conspicuous absence of "may",
> "might" and above all "POSSIBLE".
Right, "ability" and "capability" should not be associated with CAhAs.
> Therefore I would respectfully suggest
> considering two new uncontaminated cmavo to act as true and
> contaminated, wide-scope modal-logical operators:
>
> ci'a = "it is possible that; possibly; may/might" (looks vaguely like 'cumki')
> ne'e = "it is necessary that; necessarily; must" (looks vaguely like
> 'necessary')
In my experience, it is usually more effective to work with existing
cmavo and nudge their definitions in the right direction than propose
completely new cmavo.
Jorge Llambías
>>
> Is there a brivla for logical or modal necessity?
I guess "zilsa'u", "ko'a sarcu zi'o ko'e".
John E. Clifford
{zilni,'I} is still linked to logical necessity, which is rarely the interesting case (all S5 and the like).
Sent from my iPad
Bob LeChevalier, President and Founder - LLG
> JCB had at least one course in Logic but in a school that did not favor modal
> logic at all. I don't know how well he did in even that one (Lojbab does not
> improve the logic input much).
Actually, I had less than JCB and probably did worse. I had only one
13-week logic course, which I would have flunked if the instructor had
been rigid. He let me have an incomplete, and I got my grade up to a D
after 2 more months. Even that says very little. It was a mastery
course, so getting a D meant that I completed only something like 60-70%
of the modules and never was even exposed to the rest. For what I did
cover, I had the mechanics down (though I forgot them within a few
months), but the abstract concepts never really sunk in.
And nothing in the course, even the part I did not cover, likely dealt
with the stuff that comes up here, though I thought I followed your
explanations in our correspondence and occasional discussions back in
the 80s when this stuff was created in Lojban.
(Nora, a math major, also only had one class, but did very well in it,
so her advisor suggested that she take philosophy instead of a second
class.)
Bearing my ignorance in mind, I contribute the following based on what I
thought I understood 20 years ago.
> But in 56 years, the efforts to get necessity
> operators in have come to naught -- though eventually we got something like a
> necessity predicate,
Hey - I added what pc said we needed %^) I don't recall that JCB at the
time even had the CAhA family, but rather that I added it because of a
discussion pc provided explaining about "timeless" tense, i.e. "cu".
Examples such as whether a kid who has never been in a pool "is a
swimmer" and the meaning of "flammable" are the sorts of things I recall.
I probably still have pc's correspondence on the matter.
I associate "necessity operator" with ni'i (in BAI), and not with any of
the ka'e/CAhA family.
>>Almost all CV(')V cmavo have the same form of a rafsi of something. My
>>guess is the ones that are related to that something are in the
>>minority. "ka'e" was obviously taken from "kakne", yes, but the
>>connection is kind of malglico. Similarly "pe'i" comes from "pensi",
>>"ti'e" from "tirna", and thare are other mnemonics that go through
>>malglico glosses.
> Right, but with e.g. {pi'o}, it's a no-brainer that the cmavo has
> nothing to do with pianos despite sharing {pipno}'s rafsi's form.
> With {ka'e}, one would not think it was such an accident. In
> principle of course no cmavo need to be related to the gismu with that
> cmavo's form.
It wasn't an accident, and it is news to me that ka'e is more like cumki
than kakne. IIRC, ka'e was supposed to be the actuality neutral
combination of pu'i (can and has) and nu'o (can and has not). A
contradictory negation of a ka'e sentence would seem to give its opposite.
>>>Meanwhile, vlasisku, BPFK section
>>>CAhA, cmavo.txt and the CLL say nothing about {cumki} wrt {ka'e}.
>>
>>In jbovlaste "ka'e" is defined as "fi'o se cumki". But since I wrote
>>that definition I guess I can't count that as evidence. :)
>
> It's only in the Lojban record! Side note: which should I rely on
> more, vlasisku or jbovlaste? I find vlasisku's cross linking and more
> complete search results to be superior. If someone rolled in the
> BPFK definitions and CLL sections, it would be almost ideal.
So far as I know, jbovlaste is not official about anything. cmavo are
defined by CLL, and only by CLL, since the LogFlash cmavo lists were
deemed inadequate. One of the main reasons we did not have a dictionary
a long time ago was that I had no good idea how to create good
dictionary-style cmavo definitions.
I've never reviewed any jbovlaste definitions (not being fond of
web-interfaces in general, I want a real dictionary %^)
>>>Also, while {cumki} does express possibility, {ka'e}, from the given
>>>definitions, seems to be more about ability than possibility.
>>
>>But whose ability? Each of the arguments of the relation modified by
>>"ka'e"? The x1? The agent (assuming there is one)?
>
> You're asking me?! Well since you asked, from what I see, I would
> definitely assume the x1, given the glosses, proposed keywords, and
> examples in the CLL and BPFK.
That was the intent (or rather the "subject" rather than x1, since you
could access the x2 with "se brivla" etc), though I admit that I didn't
and still don't really understand why it couldn't apply to one of the
other places.
If I make a ka'e capability claim involving all the places of klama,
then the claim applies just as much to the place gone to as to the
go-er. If I can go to a place (from somewhere else by some route), then
that place can be gone to by me, and likewise, if I cannot, then it
cannot.
> In particular the CLL examples indicate
> very clearly that {ka'e} and related CAhA are some sort of short-scope
> selbri modifiers and emphatically _not_ true modal operators with
> scope over the whole bridi.
I won't claim to know the difference.
>>>In
>>>order to say things like "it possibly brodas" and "it necessarily
>>>brodas" I have to believe that these concepts should have their own
>>>words, without mixing ability into it.
>>
>>I agree that the word "ability" should not appear in the definition of
>>CAhAs, since events don't really have abilities.
>
> It's not just "ability" that seems off, it's also the ambiguous "can"
> and "innate capability" as well as the conspicuous absence of "may",
> "might" and above all "POSSIBLE".
"possible" (cumki) seems to ONLY be about events, whereas I thought ka'e
and CAhA was more about the sumti that participate in the events. Maybe
there isn't a lot of difference, though.
>>>These primitive logical
>>>operators strike me as vastly worth assigning two disyllables from
>>>cmavo space, especially in light of some of the other things
>>>available. Just my 2 cents.
>>
>>I agree. I have said before that it is extremely weird that a logical
>>language doesn't have a word for the "necessarily" operator.
>
> The fact that there is no necessity operator strongly suggests that
> the language designers did not have the foggiest notion of modal logic
> when they created {ka'e}.
Or maybe pc and I understood at the time that necessity was not
something covered in CAhA, (since I am pretty sure he has *at least* a
"foggiest notion".)
Clearly I do need to dig out that old correspondence, and see if this
was one of those topics that he set down in writing rather than
explained to me over the phone.
lojbab
--
Bob LeChevalier loj...@lojban.org www.lojban.org
President and Founder, The Logical Language Group, Inc.
Bob LeChevalier, President and Founder - LLG
> On Fri, Nov 18, 2011 at 11:29 AM, John E Clifford <kali9...@yahoo.com> wrote:
>
>>JCB had at least one course in Logic but in a school that did not favor modal
>>logic at all. I don't know how well he did in even that one (Lojbab does not
>>improve the logic input much). But in 56 years, the efforts to get necessity
>>operators in have come to naught -- though eventually we got something like a
>>necessity predicate,
>>
>
> Is there a brivla for logical or modal necessity?
nibli (which I think was "snola" in JCB's version)
Bob LeChevalier, President and Founder - LLG
> Not really, but it can be massaged to do the work. The ideal would be
> "x1 is necessary under condition/compulsion x2"
I think that is "sarcu"
"nibli" applies to logical necessity.
If you asked about "sufficient", I might have a harder time, because I
think that "banzu" was originally intended to cover it, but as currently
defined became a somewhat different sense than the complement of sarcu.
lojbab
Robert LeChevalier
> We're actually on about Logjam vii going back to 1955. That is the direct line with JCB or LLG in charge. I know of maybe half a dozen spinoffs, none of the very active, and I don't know what has happened over at Loglan. We're about par for the course, with Lojban as a more successful Ido. Lojban took some, but not all, of the next Loglan suggestions, added a few new ones, and threw out a few things. And has done more alteration since.
> {zilni,'I} is still linked to logical necessity, which is rarely the interesting case (all S5 and the like).
Jim Carter's guaspi was the first attempt to reform Lojban, though
Prothero threw in planb at some point.
The bottom line is that we've learned how hard it is to deisgn and
document a language. I don't think the project could complete a new
complete do-over (and remember that Lojban was itself never intended to
be a complete do-over, but an evasion of JCB's intellectual property
claims - we re-did a lot as a result, but maintaining the conceptual
status quo was a top priority).
If it ever is done, a re-do should be done by a mass of fluent
Lojbanists working solely in that language to prevent malglico and as
much malrarna as possible.
lojbab
maikxlx
> You seemed to think the given definitions made sense...
>
naturalistic point of view, and that they follow mostly from {kakne}.
I am not saying that I think that it should be this way, just that it
is.
>> Well since you asked, from what I see, I would
>> definitely assume the x1, given the glosses, proposed keywords, and
>> examples in the CLL and BPFK. In particular the CLL examples indicate
>> very clearly that {ka'e} and related CAhA are some sort of short-scope
>> selbri modifiers and emphatically _not_ true modal operators with
>> scope over the whole bridi.
>
> But CAhAs are tags, and all other tags are bridi operators. If "ka'e
> citka" and "ka'e se citka" have different meanings (besides reordered
> places), CAhA works nothing like other tags.
>
> And I don't know what you would do with "ka'e na broda", or "ka'e ku
> na ku broda", given that "na" has bridi scope, and "ka'e" appears to
> have scope over "na" in those cases.
>
If you're saying that CAhA should not stand in for {kakne}, then I
agree with you for the reasons you give. But people are going to
learn {ka'e} by what they read in the reference materials, and from
established usage, not by what the formal grammar's scope rules imply,
and the reference materials describe something closer to {kakne} than
to {cumki}, and this includes the somewhat muddled BPFK definition.
The established usage I cannot speak to due to my limited familiarity
with it. But now that Bob LeChevalier weighed in, but I would be
astounded if {ka'e} is closer to {cumki} than to {kakne}.
>> ci'a = "it is possible that; possibly; may/might" (looks vaguely like 'cumki')
>> ne'e = "it is necessary that; necessarily; must" (looks vaguely like
>> 'necessary')
>
> In my experience, it is usually more effective to work with existing
> cmavo and nudge their definitions in the right direction than propose
> completely new cmavo.
>
> mu'o mi'e xorxes
>
I didn't come here to make unwelcome suggestions, but this is an issue
that Lojban should not monkey around with. Modal-logical operators
are both extremely useful in ordinary conversation and are primitive
in logic. They are essential in Montague's program and will be needed
by Martin Bays or anyone else who is going to take a crack at a
model-theoretical formalization. At the minimum I urge the admission
of {ne'e}, which would be very useful in its own right and would fill
a major gap in Lojban (with an added benefit that its uncontaminated
counterpart can be gotten from {naku ne'eku naku} if desired). I
highly doubt that there will ever arise a better reason to assign an
unassigned CV'V cmavo . Just my 2 cents.
maikxlx
Founder - LLG <loj...@lojban.org> wrote:
>>>> Also, while {cumki} does express possibility, {ka'e}, from the given
>>>> definitions, seems to be more about ability than possibility.
>>>
>>> But whose ability? Each of the arguments of the relation modified by
>>> "ka'e"? The x1? The agent (assuming there is one)?
>>
>> You're asking me?! Well since you asked, from what I see, I would
>> definitely assume the x1, given the glosses, proposed keywords, and
>> examples in the CLL and BPFK.
>
> That was the intent (or rather the "subject" rather than x1, since you could
> access the x2 with "se brivla" etc), though I admit that I didn't and still
> don't really understand why it couldn't apply to one of the other places.
>
> If I make a ka'e capability claim involving all the places of klama, then
> the claim applies just as much to the place gone to as to the go-er. If I
> can go to a place (from somewhere else by some route), then that place can
> be gone to by me, and likewise, if I cannot, then it cannot.
>
differently among each of the bridi places. Specifically the capacity
of a goer to be a goer is expressible as something like {lo ka ka'e
klama}, while the capacity to be a place gone-to is {lo ka ka'e se
klama} -- assuming that {ka'e} carries from {kakne}, which is
something that xorxes disputes.
>>>> In
>>>> order to say things like "it possibly brodas" and "it necessarily
>>>> brodas" I have to believe that these concepts should have their own
>>>> words, without mixing ability into it.
>>>
>>> I agree that the word "ability" should not appear in the definition of
>>> CAhAs, since events don't really have abilities.
>>
>> It's not just "ability" that seems off, it's also the ambiguous "can"
>> and "innate capability" as well as the conspicuous absence of "may",
>> "might" and above all "POSSIBLE".
>
> "possible" (cumki) seems to ONLY be about events, whereas I thought ka'e and
> CAhA was more about the sumti that participate in the events. Maybe there
> isn't a lot of difference, though.
>
I agree with you here about {cumki} and {kakne}. In ordinary
conversation, often the difference is not great, as the non-existence
of purely modal-logical operators in Lojban up until now would seem to
prove. But from a logical point of view, the difference is rather
important.
> Or maybe pc and I understood at the time that necessity was not something
> covered in CAhA, (since I am pretty sure he has *at least* a "foggiest
> notion".)
>
Sounds like someone has some 'splainin to do.
maikxlx
> John E. Clifford wrote:
>>
>> We're actually on about Logjam vii going back to 1955. That is the direct
>> line with JCB or LLG in charge. I know of maybe half a dozen spinoffs,
>> none of the very active, and I don't know what has happened over at Loglan.
>> We're about par for the course, with Lojban as a more successful Ido.
>> Lojban took some, but not all, of the next Loglan suggestions, added a few
>> new ones, and threw out a few things. And has done more alteration since.
>> {zilni,'I} is still linked to logical necessity, which is rarely the
>> interesting case (all S5 and the like).
>
> Jim Carter's guaspi was the first attempt to reform Lojban, though Prothero
> threw in planb at some point.
>
none of them ever came to anything, and nothing has really popped up
since. Ceqli has is worth noting and has garnered some attention but
it's more of an IAL than a loglang. Jim Carter's guaspi is
interesting but I could get not past the tones, and it has no user
base. Plan B is a whimsical sketch that has been widely ridiculed.
There was another one that had a small planning committee but it
withered away in short time.
Bob LeChevalier, President and Founder - LLG
> On Fri, Nov 18, 2011 at 8:14 PM, Bob LeChevalier, President and
> Founder - LLG <loj...@lojban.org> wrote:
>
>>>>>Also, while {cumki} does express possibility, {ka'e}, from the given
>>>>>definitions, seems to be more about ability than possibility.
>>>>
>>>>But whose ability? Each of the arguments of the relation modified by
>>>>"ka'e"? The x1? The agent (assuming there is one)?
>>>
>>>You're asking me?! Well since you asked, from what I see, I would
>>>definitely assume the x1, given the glosses, proposed keywords, and
>>>examples in the CLL and BPFK.
>>
>>That was the intent (or rather the "subject" rather than x1, since you could
>>access the x2 with "se brivla" etc), though I admit that I didn't and still
>>don't really understand why it couldn't apply to one of the other places.
>>
>>If I make a ka'e capability claim involving all the places of klama, then
>>the claim applies just as much to the place gone to as to the go-er. If I
>>can go to a place (from somewhere else by some route), then that place can
>>be gone to by me, and likewise, if I cannot, then it cannot.
>>
>
> I have to disagree; I think that {kakne} capability manifests itself
> differently among each of the bridi places. Specifically the capacity
> of a goer to be a goer is expressible as something like {lo ka ka'e
> klama}, while the capacity to be a place gone-to is {lo ka ka'e se
> klama} -- assuming that {ka'e} carries from {kakne}, which is
> something that xorxes disputes.
But of course the capability of lo klama to be such is the capability to
klama x2 x3 x4 x5, and its capability is dependent on the values of x2,
x3, x4, and x5, and correspondingly, the claim seems evident that this
is strongly associated with the capability of that x2 to se klama x1 x3
x4 x5, and with the capability of x3 to te klama x2 x1 x4 x5, etc.
>>"possible" (cumki) seems to ONLY be about events, whereas I thought ka'e and
>>CAhA was more about the sumti that participate in the events. Maybe there
>>isn't a lot of difference, though.
>
> I agree with you here about {cumki} and {kakne}. In ordinary
> conversation, often the difference is not great,
Since that is the only sort of language use I know how to deal with, I
plead guilty %^)
> as the non-existence
> of purely modal-logical operators in Lojban up until now would seem to
> prove. But from a logical point of view, the difference is rather
> important.
I defer to the experts.
>>Or maybe pc and I understood at the time that necessity was not something
>>covered in CAhA, (since I am pretty sure he has *at least* a "foggiest
>>notion".)
>>
>
> Sounds like someone has some 'splainin to do.
Well, CAhA was certainly not intended to be the category "modal-logical
operators", and BAI was originally intended to include all of the pure
modals, since the insight from the JCB era was that linguistically the
modals and case tags/sumti tcita could be used in grammatically
interchangeable ways (we didn't think too much about semantic
differences, only grammatical ones). The intent at that point was that
ni'i used as a modal would handle logical necessity, and its possible
use as a sumti tag was consistent with this meaning. BAI has evolved
over the years, and is much more strongly associated with the place
structures of the associated gismu per the fi'o equivalence, and this
may have lost something from the intended modals that are among the set
of BAI.
Indeed, I think the current TLI language may treat ALL of the words that
comprise their current modal AND tense AND case tag complex as being a
single category to be combined willy-nilly in strings with no internal
grammatical structure, as if all of them were members of selma'o PU.
That was the case in 1987, and I doubt that it changed.
Bob LeChevalier, President and Founder - LLG
I think Plan B was whimsical sketch on purpose, intending to make fun
of the perfectionists who were perpetually proposing something new and
different to reform the language.
My insight, such as it was, is that for a language to be a LANGUAGE, the
significance of a stable and active user base is all important. A
theoretical construct that no one (or only the inventor) uses might be
more logical, but it would not really be a language. I did not win
friends in the conlang community with this attitude %^)
maikxlx
>>
>> On Fri, Nov 18, 2011 at 8:14 PM, Bob LeChevalier, President and
>> Founder - LLG <loj...@lojban.org> wrote:
>>
>>> If I make a ka'e capability claim involving all the places of klama, then
>>> the claim applies just as much to the place gone to as to the go-er. If
>>> I
>>> can go to a place (from somewhere else by some route), then that place
>>> can
>>> be gone to by me, and likewise, if I cannot, then it cannot.
>>>
>>
>> I have to disagree; I think that {kakne} capability manifests itself
>> differently among each of the bridi places. Specifically the capacity
>> of a goer to be a goer is expressible as something like {lo ka ka'e
>> klama}, while the capacity to be a place gone-to is {lo ka ka'e se
>> klama} -- assuming that {ka'e} carries from {kakne}, which is
>> something that xorxes disputes.
>
> But of course the capability of lo klama to be such is the capability to
> klama x2 x3 x4 x5, and its capability is dependent on the values of x2, x3,
> x4, and x5, and correspondingly, the claim seems evident that this is
> strongly associated with the capability of that x2 to se klama x1 x3 x4 x5,
> and with the capability of x3 to te klama x2 x1 x4 x5, etc.
>
in the context of the larger discussion is the exact relationship of
this family to {cumki}. Supposing for a moment that it could be
purified of obvious malrarna, I would still say that {kakne} makes a
stronger claim than {cumki} does, because the former imputes to an
individual in the actual world an inherent property, whereas the
latter merely claims that the overall proposition is possible, or to
put it equivalently, that it is true in some possible state of
affairs. What we want out of {ka'e} is only the latter, and if it
can't guarantee that, then something else is needed IMHO.
On Sat, Nov 19, 2011 at 7:11 AM, Bob LeChevalier, President and
Founder - LLG <loj...@lojban.org> wrote:
> I think Plan B was whimsical sketch on purpose, intending to make fun of
> the perfectionists who were perpetually proposing something new and
> different to reform the language.
>
I don't know. It was written in a pretty deadpan tone.
> My insight, such as it was, is that for a language to be a LANGUAGE, the
> significance of a stable and active user base is all important. A
> theoretical construct that no one (or only the inventor) uses might be more
> logical, but it would not really be a language. I did not win friends in
> the conlang community with this attitude %^)
> --
Well, even as a "perfectionist" that might disagree with you, I'd say
you must have done something right, because the fact of the matter is
that Lojban is the only game in town.
> Bob LeChevalier loj...@lojban.org www.lojban.org
> President and Founder, The Logical Language Group, Inc.
mu'o mi'e .maik.
Felipe Gonçalves Assis
I think the word that is missing here is volition. "x1 is Capable of x2" should
be read as "x1 will (do) x2 if he wants", implying volition per x1.
Possibility is an unrelated concept. My ability to go per se doesn't imply
anything about the possibility or necessity of my going. Conversely,
the possibility of my going doesn't imply anything about the control I
have over it.
The verb "can" in English may express Capability, Possibility, and
possibly other things:
* "I can swim." -> I'll swim if I want.
* "This can be the answer to our problems!" -> It is possible that it is.
* "Hey! You cannot smoke in here!" -> It is not allowed.
{kakne} in lojban is more useful as denoting Capability than Possibility
with a highlighted subject.
About CAhA:
I also agree with the reasons given by xorxes for it not making sense
that this selma'o denote Capability. However, the documentation has
indeed, at least, largely drifted towards the Capability interpretation.
I strongly +1 the idea that something be done to give lojban clear
modal logic tags, either by rewriting the CAhA documentation or by
devising experimental cmavo, not because I like changes, but because
this is too good a reason for one.
About selbri:
Till now, the only gismu I recognize as fit to express modal aspects is {cumki}.
meta:
Shouldn't we branch this modal discussion to another thread?
mu'o mi'e .asiz.
Jorge Llambías
Founder - LLG <loj...@lojban.org> wrote:
>
> Well, CAhA was certainly not intended to be the category "modal-logical
> operators", and BAI was originally intended to include all of the pure
> modals, since the insight from the JCB era was that linguistically the
> modals and case tags/sumti tcita could be used in grammatically
> interchangeable ways (we didn't think too much about semantic differences,
> only grammatical ones). The intent at that point was that ni'i used as a
> modal would handle logical necessity, and its possible use as a sumti tag
> was consistent with this meaning. BAI has evolved over the years, and is
> much more strongly associated with the place structures of the associated
> gismu per the fi'o equivalence, and this may have lost something from the
> intended modals that are among the set of BAI.
What were the other intended modals among the BAIs besides "ni'i"?
Consider these two sentences:
(1) ka'e ku no da klama lo tersla
(2) no da ka'e klama lo tersla
I would translate them as "it's possible nobody comes to the party"
and "nobody can come to the party" respectively. The first one is
clearly not about capability, and the second one may be about
capability but probably just circumstantial rather than innate.
If I understand your position correctly, you would understand them
both as the implausible "nobody is innately capable of coming to the
party". And in order to express my meanings with a modal you would
have to go with something like:
(1') na ku ni'i ku na ku no da klama lo tersla
(2') no da na ku ni'i ku na ku klama lo tersla
which can be simplified a bit by noting that "na ku no da" = "su'o da"
and "no da na ku" = "ro da", so:
(1'') na ku ni'i ku su'o da klama lo tersla
(2'') ro da ni'i ku na ku klama lo tersla
"it is not necessarily the case that someone comes to the party" and
"everyone is necessarily not coming to the party".
The problem of using "ni'i" for "necessarily" though is that it may
interfere with its other use for logical entailment. "te sau" is a
slightly better candidate, if it weren't for the x2 of "sarcu".
Pierre Abbat
> So, would you have us write {lo si'o [za'e] grezunca'a cu se finti la
> .caklis.}?
>
> I believe that pretty much {lo si'o broda} = {lo sidbo be lo broda}, where
> lo broda is again is a reference to a kind with clearly _no_
> actuals/mundanes/instances, either needed or wanted. Would you rewrite {lo
> broda cu sidbo} as {lo si'o broda cu sidbo}?
"si'o" is the idea abstractor; it abstracts a bridi, just like any other
abstractor. "lo si'o cmuxu'i" is different from "lo si'o selratni", even
though "lo cmuxu'i" is the same as "lo selratni" (the rest of the places are
different).
That said, I'm not sure what exactly "si'o broda" means, except for measuring
units. "lo si'o mitre" is "the meter", since "lo mitre" means not "a meter"
but "something measured in meters". "mi mitre li cifi'ure" implies "mi mitre"
even though I am longer than a meter. And lo mitre clearly exists.
Pierre
--
lo ponse be lo mruli po'o cu ga'ezga roda lo ka dinko
maikxlx
misunderstood {si'o}. I thought {ce'u} worked with {si'o} the way it
works with {ka}, but having checked the CLL, I see no evidence for
this, so the equation {lo si'o broda} = {lo sidbo be lo broda} is
probably wrong. What the difference is I am not exactly sure. And,
perhaps {si'o} is only currently defined for measuring units; I don't
know that either.
Secondly, in haste I wrote {broda} when I should have just written
{grezunca'a}. The issue being discussed was the possibility of {da
finti lo grezunca'a} being true at a moment of time when no actual
transistor yet existed (for the sake of argument let's say this is
what happened). In that state of affairs, {lo grezunca'a} has no
referent unless you posit that it can refer to transistor-kind or
something similarly intensional. But yes other {lo}-sumti such as {lo
mitre} do exist in general.
> Pierre
> --
Robert LeChevalier
> On Sat, Nov 19, 2011 at 7:05 AM, Bob LeChevalier, President and
> Founder - LLG <loj...@lojban.org> wrote:
>
>>maikxlx wrote:
>>
>>>On Fri, Nov 18, 2011 at 8:14 PM, Bob LeChevalier, President and
>>>Founder - LLG <loj...@lojban.org> wrote:
>>>
>>>
>>>>If I make a ka'e capability claim involving all the places of klama, then
>>>>the claim applies just as much to the place gone to as to the go-er. If
>>>>I
>>>>can go to a place (from somewhere else by some route), then that place
>>>>can
>>>>be gone to by me, and likewise, if I cannot, then it cannot.
>>>>
>>>
>>>I have to disagree; I think that {kakne} capability manifests itself
>>>differently among each of the bridi places. Specifically the capacity
>>>of a goer to be a goer is expressible as something like {lo ka ka'e
>>>klama}, while the capacity to be a place gone-to is {lo ka ka'e se
>>>klama} -- assuming that {ka'e} carries from {kakne}, which is
>>>something that xorxes disputes.
>>
>>But of course the capability of lo klama to be such is the capability to
>>klama x2 x3 x4 x5, and its capability is dependent on the values of x2, x3,
>>x4, and x5, and correspondingly, the claim seems evident that this is
>>strongly associated with the capability of that x2 to se klama x1 x3 x4 x5,
>>and with the capability of x3 to te klama x2 x1 x4 x5, etc.
>>
>
> I agree that there is a family of co-dependent capabilities.
Nora and I were discussing this topic yesterday, and she posed that the
explicit way of referring to just one of these capabilies would be to
mark the relevant sumti with kau, with the unmarked form technically
being nonspecific as to which of the co-dependent capabilities is being
focused on.
Thus the typical interpretation of
lo nanla cu ka'e limna
is
lo nanla kau cu ka'e limna
I like this, but it presents a possible overloading of kau if there are
more than one reason to mark a bridi, such as
mi djuno ledu'u la nanla kau ka'e djuno makau
intending
I know where the boy is capable of swimming.
The rub
> in the context of the larger discussion is the exact relationship of
> this family to {cumki}. Supposing for a moment that it could be
> purified of obvious malrarna, I would still say that {kakne} makes a
> stronger claim than {cumki} does, because the former imputes to an
> individual in the actual world an inherent property, whereas the
> latter merely claims that the overall proposition is possible, or to
> put it equivalently, that it is true in some possible state of
> affairs.
There is in my mind some difference in meaning between cumki and kakne,
and that may be it, but I will remain uncommitted.
> What we want out of {ka'e} is only the latter, and if it
> can't guarantee that, then something else is needed IMHO.
You are saying that we WANT a cumki rather than a kakne meaning for use
in the contrast between the various CAhAs?
> On Sat, Nov 19, 2011 at 7:11 AM, Bob LeChevalier, President and
> Founder - LLG <loj...@lojban.org> wrote:
>
>>I think Plan B was whimsical sketch on purpose, intending to make fun of
>>the perfectionists who were perpetually proposing something new and
>>different to reform the language.
>
> I don't know. It was written in a pretty deadpan tone.
Jeff is that way.
>>My insight, such as it was, is that for a language to be a LANGUAGE, the
>>significance of a stable and active user base is all important. A
>>theoretical construct that no one (or only the inventor) uses might be more
>>logical, but it would not really be a language. I did not win friends in
>>the conlang community with this attitude %^)
>
> Well, even as a "perfectionist" that might disagree with you, I'd say
> you must have done something right, because the fact of the matter is
> that Lojban is the only game in town.
TLI Loglan still exists with a very small rump community (some of whom
are also Lojbanists). I am not sure WHY someone would use TLI Loglan
instead of Lojban, but they do.
lojbab
Bob LeChevalier, President and Founder - LLG
> On Sat, Nov 19, 2011 at 9:05 AM, Bob LeChevalier, President and
> Founder - LLG <loj...@lojban.org> wrote:
>
>>Well, CAhA was certainly not intended to be the category "modal-logical
>>operators", and BAI was originally intended to include all of the pure
>>modals, since the insight from the JCB era was that linguistically the
>>modals and case tags/sumti tcita could be used in grammatically
>>interchangeable ways (we didn't think too much about semantic differences,
>>only grammatical ones). The intent at that point was that ni'i used as a
>>modal would handle logical necessity, and its possible use as a sumti tag
>>was consistent with this meaning. BAI has evolved over the years, and is
>>much more strongly associated with the place structures of the associated
>>gismu per the fi'o equivalence, and this may have lost something from the
>>intended modals that are among the set of BAI.
>
>
> What were the other intended modals among the BAIs besides "ni'i"?
Since I was largely oblivious of the concept of "modal" when I
reinvented the cmavo, having only recognized that JCB had made them
grammatically identical to sumti tcita and used them as such, I would
simply look at the list of modals that JCB identified in Loglan 1, and
pick out the Lojban equivalent (and there is one for each of them)
> Consider these two sentences:
>
> (1) ka'e ku no da klama lo tersla
>
> (2) no da ka'e klama lo tersla
>
> I would translate them as "it's possible nobody comes to the party"
> and "nobody can come to the party" respectively. The first one is
> clearly not about capability, and the second one may be about
> capability but probably just circumstantial rather than innate.
> If I understand your position correctly, you would understand them
> both as the implausible "nobody is innately capable of coming to the
> party".
I don't agree with your translation of the first one, which I see as
being just a rearrangement of the second one. The meaning of the second
probably is affected by xorlo, but pre-xorlo with le tersla I would have
understood it as "Nothing could have come to the party".
Which brings to mind that the glico word (which I think is a modal) that
I associate with kakne is "could have", recognizing that in English, the
distinction between can and may often is more associated with capability
vs permission rather than capability vs possibility
And in order to express my meanings with a modal you would
> have to go with something like:
>
> (1') na ku ni'i ku na ku no da klama lo tersla
>
> (2') no da na ku ni'i ku na ku klama lo tersla
I would express "it's possible nobody comes to the party" as "cumki fa
lo nu noda klama le tersla", and "nobody can come to the party" as your
(2) (noting the malrarna interpretation of noda as nobody rather than
nothing)
I don't see how logical necessity enters into the question at all, so I
cannot interpret your two prime examples into any kind of standard English.
1' It is not the case that some unspecified logic necessitates that
nothing comes to a party.
2' Nothing is not (logically necessarily not coming to a party) - the
noda and naku have become prenex-like
> which can be simplified a bit by noting that "na ku no da" = "su'o da"
> and "no da na ku" = "ro da",
I thought "naku noda" = "ro da", but I may be half asleep. I'm not sure
of your version.
so:
>
> (1'') na ku ni'i ku su'o da klama lo tersla
It is not the case that (logically) necessarily someone comes to a party.
> (2'') ro da ni'i ku na ku klama lo tersla
Everything (logically) necessarily is not coming to a party.
> "it is not necessarily the case that someone comes to the party" and
> "everyone is necessarily not coming to the party".
>
> The problem of using "ni'i" for "necessarily" though is that it may
> interfere with its other use for logical entailment.
What other meaning of necessarily are you trying to convey other than
logical entailment (with no specified logic).
Maybe I am simply failing to grasp what you mean by "necessarily"
>"te sau" is a slightly better candidate,
That is another meaning of "necessarily", I agree
if it weren't for the x2 of "sarcu".
Since tersau refers to the x3 of sarcu, I am not sure how x2 is
relevant. If I were to re-express the sentence using sarcu, x2 would
probably be the bridi that is dependent. x1 seems more problematical,
since the meat of the claim is in the x3.
But remember that BAI wasn't originally fi'o broda, at least not
strictly - that was a later insight that allowed us to clarify the use
and semantics of BAI as a class. The assignment of modals into BAI might
have suffered from that, if we chose the wrong broda for the modals in BAI.
I just (finally) looked up modal logic in Wikipedia to perhaps gain some
context. They use two sample sentences:
In a classical modal logic, each can be expressed by the other with
negation:
...
it is possible that it will rain today if and only if it is not
necessary that it will not rain today;
and
it is necessary that it will rain today if and only if it is not
possible that it will not rain today.
It seems to me that we have a greater problem in Lojban expressing that
sense of "possible" than we do "necessary", since "possible" for me has
never excluded "necessary". It certainly is the case that we did not
consider modal logic of this sort in designing Lojban (well, I cannot
say that pc did not, but I don't think he communicated it to me).
Most of the focus was on concepts that dated to Aristotle (noting that
my knowledge of what that means is somewhat less than my knowledge of
"logic"); pc often referred to how Aristotle discussed/divided matters
in his discussions.
Looking further, and without having dug into the ancient correspondence
yet, I think CAhA had more to do with:
http://en.wikipedia.org/wiki/Potentiality_and_actuality
than with modal logic, with ca'a being Aristotle's actuality and ka'e
being his potentiality, though the discussion there gets far more
complex than what we talked about with CAhA.
Jorge Llambías
Founder - LLG <loj...@lojban.org> wrote:
>
> Since I was largely oblivious of the concept of "modal" when I reinvented
> the cmavo, having only recognized that JCB had made them grammatically
> identical to sumti tcita and used them as such, I would simply look at the
> list of modals that JCB identified in Loglan 1, and pick out the Lojban
> equivalent (and there is one for each of them)
Unfortunately, it appears that JCB was equally oblivious to the
concept. Here's what he had on "modal operators":
http://www.loglan.org/Loglan1/chap5.html#sec5.6
At least he was aware that he was misusing the word "modal", as he
says in a footnote:
"3 We here use the word 'modal' in a sense that differs from its usual
meaning in logic."
I think he should have qualified that as "differs completely".
>> "na ku no da" = "su'o da" and "no da na ku" = "ro da",
>
> I thought "naku noda" = "ro da", but I may be half asleep. I'm not sure of
> your version.
"Not nothing" is "something", "Nothing not" is "everything".
> I just (finally) looked up modal logic in Wikipedia to perhaps gain some
> context. They use two sample sentences:
>
> In a classical modal logic, each can be expressed by the other with
> negation:
> ...
> it is possible that it will rain today if and only if it is not necessary
> that it will not rain today;
> and
> it is necessary that it will rain today if and only if it is not possible
> that it will not rain today.
>
> It seems to me that we have a greater problem in Lojban expressing that
> sense of "possible" than we do "necessary", since "possible" for me has
> never excluded "necessary".
Read again what Wikipedia says. "Possible" does not exclude
"necessary"! It excludes "necessary not".
The relationship between "possible" and "necessary" is the same as
that between "su'o" and "ro", which is why we have "su'o mu'ei" and
"ro mu'ei", which I should have mentioned earlier:
http://www.lojban.org/tiki/mu%27ei
maikxlx
> maikxlx wrote:
>> I agree that there is a family of co-dependent capabilities.
>
> Nora and I were discussing this topic yesterday, and she posed that the
> explicit way of referring to just one of these capabilies would be to mark
> the relevant sumti with kau, with the unmarked form technically being
> nonspecific as to which of the co-dependent capabilities is being focused
> on.
>
> Thus the typical interpretation of
> lo nanla cu ka'e limna
> is
> lo nanla kau cu ka'e limna
>
> I like this, but it presents a possible overloading of kau if there are more
> than one reason to mark a bridi, such as
>
> mi djuno ledu'u la nanla kau ka'e djuno makau
> intending
> I know where the boy is capable of swimming.
>
with {kau}'s use in bridi abstractions. Wouldn't it make the most
sense to simply understand x1 as the relevant sumti? In other words
(1) {x1 ka'e [selbri] x2 x3 [...]} would be a transformation of:
(2) {x1 kakne lo nu ce'u [selbri] x2 x3 [...]}, and vice versa.
(Note that in (1), {teka'e} could tag the {te kakne} of (2) if
desired.) There is no need for {kau} in {ka'e} bridi because you can
always use {ce'u} in the event abstraction expressed by the x2 of
{kakne} to get anything you want. {ka'e} is just there as short hand
to more conveniently express (2), which is probably the most common
{kakne} scenario.
>> What we want out of {ka'e} is only the latter, and if it
>>
>> can't guarantee that, then something else is needed IMHO.
>
> You are saying that we WANT a cumki rather than a kakne meaning for use in
> the contrast between the various CAhAs?
>
I do not necessarily* want to try to pry {ka'e} from {kakne} at this
point. It's a rather frequently used cmavo and I suspect "reforming"
it would be futile, though xorxes thinks otherwise. What I think
Lojban unequivocally needs are two new modal operators with a grammar
similar to CAhA but sensitive to scope. Lojban also needs a brivla
for modal necessity to complement {cumki}, which we probably already
have as {zilsa'u} or possibly {ziln'i'i}, I don't care which.
*Notice the modal operator usage in natural language.
> TLI Loglan still exists with a very small rump community (some of whom are
> also Lojbanists). I am not sure WHY someone would use TLI Loglan instead of
> Lojban, but they do.
>
I see no evidence of life over there. My request to join their
mailing list was never responded to.
> lojbab
Jorge Llambías
>
> Wouldn't it make the most
> sense to simply understand x1 as the relevant sumti? In other words
>
> (1) {x1 ka'e [selbri] x2 x3 [...]} would be a transformation of:
>
> (2) {x1 kakne lo nu ce'u [selbri] x2 x3 [...]}, and vice versa.
But then how would you understand:
(3) ka'e ku ge ko'a broda gi ko'e brode
I understand it as "cumki fa lo nu ge ko'a broda gi ko'e brode".
Presumably you would understand it as "ge ko'a kakne lo nu broda gi
ko'e kakne lo nu brode", yes?
And what about:
(4) ka'e ku ge no da broda gi no de brode
For me it's "cumki fa lo nu ge no da broda gi no de brode" and for you
it would be "ge no da kakne lo nu broda gi no de kakne lo nu brode"?
So "ka'e", despite all appearances to the contrary, just jumps inside
the scope of any bridi operator in sight?
What about:
(5) "ka'e ku ko'a na broda"
or:
(6) ka'e ku ge nai ko'a broda gi nai ko'e brode
I don't understand why, when the syntax provides such simple answers,
people want to complicate interpretations so much.
tijlan
> (1) {x1 ka'e [selbri] x2 x3 [...]} would be a transformation of:
>
> (2) {x1 kakne lo nu ce'u [selbri] x2 x3 [...]}, and vice versa.
I consider the following transformations as well:
ko'a ca'a broda
--> lo nu ko'a broda cu fatci
ko'a ka'e broda
--> lo nu ko'a broda cu kakne (?)
--> lo nu ko'a broda cu cumki
mu'o
Robert LeChevalier
> On Sun, Nov 20, 2011 at 7:20 AM, Robert LeChevalier <loj...@lojban.org> wrote:
>>mi djuno ledu'u la nanla kau ka'e djuno makau
>>intending
>>I know where the boy is capable of swimming.
>>
>
> I agree that this is a neat solution, and I also agree that it clashes
> with {kau}'s use in bridi abstractions. Wouldn't it make the most
> sense to simply understand x1 as the relevant sumti?
I think that is what people do as a default. The question is whether it
is allowed to use kau to mark a non-default intent, and how badly that
might confuse matters in a worst case.
For actual usage, Jorge and Robin have much more authority, since I
don't pretend that my all too rare usage is anymore standard (especially
since xorlo).
In other words
> (1) {x1 ka'e [selbri] x2 x3 [...]} would be a transformation of:
>
> (2) {x1 kakne lo nu ce'u [selbri] x2 x3 [...]}, and vice versa.
I avoid commenting on ce'u since I understand lambda calculus even less
than xorlo. I don't SEE anything wrong with this, but I doubt if I
would even if it was wrong %^)
>>>What we want out of {ka'e} is only the latter, and if it
>>>
>>>can't guarantee that, then something else is needed IMHO.
>>
>>You are saying that we WANT a cumki rather than a kakne meaning for use in
>>the contrast between the various CAhAs?
>>
>
> I do not necessarily* want to try to pry {ka'e} from {kakne} at this
> point. It's a rather frequently used cmavo and I suspect "reforming"
> it would be futile, though xorxes thinks otherwise. What I think
> Lojban unequivocally needs are two new modal operators with a grammar
> similar to CAhA but sensitive to scope. Lojban also needs a brivla
> for modal necessity to complement {cumki}, which we probably already
> have as {zilsa'u} or possibly {ziln'i'i}, I don't care which.
>
> *Notice the modal operator usage in natural language.
I note the usage. What it means in logical terms is beyond my education.
>>TLI Loglan still exists with a very small rump community (some of whom are
>>also Lojbanists). I am not sure WHY someone would use TLI Loglan instead of
>>Lojban, but they do.
>
> I see no evidence of life over there. My request to join their
> mailing list was never responded to.
You aren't missing much. Traffic averages maybe 4 messages a month,
with half or more being reminders of an apparently weekly chat session
on Second Life, or comments from people saying that they won't be at
some particular such session.
I thus suspect that the best way to talk to a TLI Loglanist is to get a
Second Life account, and be in the relevant place and time on Saturdays
(not really knowing how Second Life works).
Or just send email to xorxes (who probably uses the language better than
JCB did), or possibly Lojbanist/Loglanist Cyril Slobin %^)
lojbab
Robert LeChevalier
> On 20 November 2011 15:02, maikxlx <mai...@gmail.com> wrote:
>
>>(1) {x1 ka'e [selbri] x2 x3 [...]} would be a transformation of:
>>
>>(2) {x1 kakne lo nu ce'u [selbri] x2 x3 [...]}, and vice versa.
>
>
> I consider the following transformations as well:
>
> ko'a ca'a broda
> --> lo nu ko'a broda cu fatci
ca'a cannot equate to any sort of unspecified-tense bridi.
lojbab
maikxlx
> On Sun, Nov 20, 2011 at 12:02 PM, maikxlx <mai...@gmail.com> wrote:
>>
>> Wouldn't it make the most
>> sense to simply understand x1 as the relevant sumti? In other words
>>
>> (1) {x1 ka'e [selbri] x2 x3 [...]} would be a transformation of:
>>
>> (2) {x1 kakne lo nu ce'u [selbri] x2 x3 [...]}, and vice versa.
>
> But then how would you understand:
>
> (3) ka'e ku ge ko'a broda gi ko'e brode
>
> I understand it as "cumki fa lo nu ge ko'a broda gi ko'e brode".
> Presumably you would understand it as "ge ko'a kakne lo nu broda gi
> ko'e kakne lo nu brode", yes?
>
> And what about:
>
> (4) ka'e ku ge no da broda gi no de brode
>
> For me it's "cumki fa lo nu ge no da broda gi no de brode" and for you
> it would be "ge no da kakne lo nu broda gi no de kakne lo nu brode"?
>
> So "ka'e", despite all appearances to the contrary, just jumps inside
> the scope of any bridi operator in sight?
>
have to follow from a {kakne}-reading, which casts some doubt on it.
> What about:
>
> (5) "ka'e ku ko'a na broda"
>
Assuming I understand the standard rules correctly (tell me if I am
wrong), and extending them by positing the putative
{kakne}-transformation rule I offered earlier:
(5a) na ku ka'e ku zo'u ko'a broda
(5b) na ku zo'u ka'e ku zo'u ko'a broda
(5c) na ku zo'u ko'a kakne lo nu broda
(5d) ko'a na kakne lo nu broda
or,
(5d') na ku zo'u ko'a ka'e broda
(5e') ko'a na ka'e broda
which could have been gotten more directly by recognizing that {na}
and {ko'a} can float anywhere. However,
(7) ka'e ku ko'a na ku broda
(7a) ka'e ku zo'u na ku zo'u ko'a broda
(7b) ko'a kakne lo nu na ku zo'u broda
(7c) ko'a kakne lo nu na broda
Going backwards here, we have to be careful:
(7d) ko'a ka'e ku na ku broda
NOT (7d') ko'a ka'e na broda
No problem I can see there other than {na} acting uncooperatively as
usual. Where there seems to be a real problem is
(8) ka'e ku ro da broda
= ka'e ku zo'u ro da zo'u da broda
NOT= da kakne lo nu ro da zo'u broda
which is partially salvageable via incomplete prenex format, but only partially:
= ka'e ku zo'u ro da broda
?= ro da kakne lo nu broda
?= ro da ka'e broda
which will have a different interpretation if the relative scope of
{ka'e} and x1 is meaningful. I suspect the only solution is not to
allow {ka'e} to have scope over a quantified x1 in the first place.
Ultimately, nothing like "it is innately capable that all men swim"
makes semantic sense given the {kakne}-reading, so it probably doesn't
need to be encoded.
> or:
>
> (6) ka'e ku ge nai ko'a broda gi nai ko'e brode
>
That's another good example that I am not going to try to parse, but I
think what all this stuff shows is simply that there is an inadvertent
mismatch between the {kakne}-reading of {ka'e} and the grammar of
CAhA, which of course militates against the {kakne}-reading. On the
other hand, what it doesn't show is that {ka'e} was originally meant
or is currently used or prescribed to express the meaning of {cumki},
rather than meaning of {kakne}.
> I don't understand why, when the syntax provides such simple answers,
> people want to complicate interpretations so much.
>
I don't want to complicate anything, but at least for now I want the
Lojban that I learn to be the Lojban that is actually used and not my
own private version.
> mu'o mi'e xorxes
>
maikxlx
>>
>> (1) {x1 ka'e [selbri] x2 x3 [...]} would be a transformation of:
>>
>> (2) {x1 kakne lo nu ce'u [selbri] x2 x3 [...]}, and vice versa.
>
> I avoid commenting on ce'u since I understand lambda calculus even less than
> xorlo. I don't SEE anything wrong with this, but I doubt if I would even if
> it was wrong %^)
>
supposed to take them and which aren't. Intuitively, I want to be
able to use {ce'u} specify the role that a {kakne} has the innate
capability of playing in a {se kakne}, and unthinkingly I reach for
{ce'u}. I think the meaning is clear enough, though I don't really
know if it is allowed by the rules. I dare say it should be.
> You aren't missing much. Traffic averages maybe 4 messages a month, with
> half or more being reminders of an apparently weekly chat session on Second
> Life, or comments from people saying that they won't be at some particular
> such session.
>
> I thus suspect that the best way to talk to a TLI Loglanist is to get a
> Second Life account, and be in the relevant place and time on Saturdays (not
> really knowing how Second Life works).
>
> Or just send email to xorxes (who probably uses the language better than JCB
> did), or possibly Lojbanist/Loglanist Cyril Slobin %^)
>
Thanks for the help, but it doesn't seem worth bothering anyone over.
Michael Turniansky
Jorge Llambías
<mturn...@gmail.com> wrote:
> Of course, in many cases where xorxes wants to use ka'e for "it's
> possible",
It's not that I want to use "ka'e" for "it's possible", it's the other
way around, I want "it's possible", or "it could be", for "ka'e".
So what you need to tell us is not how you'd say "it's possible", but
how you'd understand for example "ka'e ku no da klama".
> I'd personally choose to not use a CAhA at all, but la'acu'i
Something can also be possible and extremely likely, or possible and
extremely unlikely. If "possible" corresponds to "su'o", then the la'a
scale corresponds to so'a/so'e/so'i/so'o/so'u.
>(or
> go to full-blown brivla using cumki), while recognizing that it's not
> precisely the same part of speech, and perhaps not the same logical
> implications, I think it does convey the same meaning.
That was my point, "ka'e ku broda" has basically the same meanimg as
"cumki fa lo nu broda". Or more precisely, "ka'e" is "fi'o se cumki".
> In any case, I
> wondered about the following utterance (I'm pretty sure xorxes would like it
> and understand it in the way I intend): "va'o lo nu da'i ka'e no'a kei mi ba
> co'e" (hint: it's a six syllable in English). Is it clear? (Probably less
> clear as "va'olonuda'ika'eno'akeimibaco'e" ;-) )
"If I can I will"? I count five syllables.
Bob LeChevalier, President and Founder - LLG
> On Sun, Nov 20, 2011 at 7:20 AM, Robert LeChevalier <loj...@lojban.org> wrote:
>>>What we want out of {ka'e} is only the latter, and if it
>>>
>>>can't guarantee that, then something else is needed IMHO.
>>
>>You are saying that we WANT a cumki rather than a kakne meaning for use in
>>the contrast between the various CAhAs?
>>
>
> I do not necessarily* want to try to pry {ka'e} from {kakne} at this
> point. It's a rather frequently used cmavo and I suspect "reforming"
> it would be futile, though xorxes thinks otherwise. What I think
> Lojban unequivocally needs are two new modal operators with a grammar
> similar to CAhA but sensitive to scope. Lojban also needs a brivla
> for modal necessity to complement {cumki}, which we probably already
> have as {zilsa'u} or possibly {ziln'i'i}, I don't care which.
>
> *Notice the modal operator usage in natural language.
I didn't comment at the time, needing to think about it some more, but I
am sure that this use of "not necessarily" is covered by some form of
the four or five causal cmavo in BAI, and the choice probably depends on
exactly what you mean by "not necessarily", since it isn't necessarily
(zo'o) a *logical* non-necessity.
The five in question are
ki'u
ni'i
mu'i
ri'a,
and ja'e (which does not act in parallel to the others - it was not part
of the original set, but we realized later that it somewhat overlapped
the others).
"necessarily" seems like a "therefore", which is the "se" form of the
first four and the unmodified ja'e. "Not necessarily" would then seem
to be a kind of negation of the therefore statement - not the nai form
which has been defined from the JCB era as "nevertheless", but
presumably the na form.
Whatever word you choose, it has to be used carefully. If you attach
the modal to the sumti "mi", you get
"I do not necessarily want to try to pry ka'e from kakne at this point
(but someone else might want to)."
Attached to the "at this point" would suggest that you might want to try
to do so at some other point, etc.
Both are plausible readings of what you said, but I can see a couple
more plausible readings as well (but I'd have to translate the rest of
the sentence in order to figure how to say it, and I'm lazy, and I'm not
sure that it is necessary to my point).
--
maikxlx
point" and translate it as {mi la'e de'u djica}. We want to say
(0) Not necessarily: mi la'e de'u djica
We can try
(1) na ku ni'i ku mi la'e de'u djica
How would you translate that? To me it seems to mean "not logically
because of something, I want it." In other words I do in fact want
it, but me wanting it does not logically follow from some unspecified
thing. Moving the {na ku} doesn't seem to help:
(2) ni'i ku mi la'e de'u na ku djica
What's that to you? To me it's "logically (because of something), I
don't want it" which means I actually don't want it.
> "necessarily" seems like a "therefore", which is the "se" form of the first
> four and the unmodified ja'e. "Not necessarily" would then seem to be a
> kind of negation of the therefore statement - not the nai form which has
> been defined from the JCB era as "nevertheless", but presumably the na form.
>
If you can approximate sentence (0) in Lojban using any of BAI and SE
and {na ku}, please show me. From what I can see, you can't get modal
readings from BAI.
> Whatever word you choose, it has to be used carefully. If you attach the
> modal to the sumti "mi", you get
> "I do not necessarily want to try to pry ka'e from kakne at this point (but
> someone else might want to)."
Because I do not see how the basic meaning "I do not necessarily want
it" can be gotten, I see even less how the inherence "but someone else
might want it" can be gotten. If modals and sumti interact in some
way other than by relative scope, I would definitely like to
understand how, though, preferably by Lojban examples, if anyone could
give some.
> Attached to the "at this point" would suggest that you might want to try to
> do so at some other point, etc.
Right, if for example, if xorxes convinces me that there is no other
way around the issue. Which he may do.
> Both are plausible readings of what you said, but I can see a couple more
> plausible readings as well (but I'd have to translate the rest of the
> sentence in order to figure how to say it, and I'm lazy, and I'm not sure
> that it is necessary to my point).
>
It's not; in fact it's best to use simple examples.
> Bob LeChevalier loj...@lojban.org www.lojban.org
> President and Founder, The Logical Language Group, Inc.
>
maikxlx
On Tue, Nov 22, 2011 at 8:37 PM, maikxlx <mai...@gmail.com> wrote:
>
> (1) na ku ni'i ku mi la'e de'u djica
>
> How would you translate that? To me it seems to mean "not logically
> because of something, I want it." In other words I do in fact want
> it, but me wanting it does not logically follow from some unspecified
> thing. Moving the {na ku} doesn't seem to help:
>
Sorry, (1) actually seems to me to mean "it is not the case that
logically because something I want it", which still does not mean "not
necessarily".
> it" can be gotten, I see even less how the inherence "but someone else
I meant inference, not inherence.
Michael Turniansky
On Tue, Nov 22, 2011 at 10:42 AM, Michael TurnianskyIt's not that I want to use "ka'e" for "it's possible", it's the other
<mturn...@gmail.com> wrote:
> Of course, in many cases where xorxes wants to use ka'e for "it's
> possible",
way around, I want "it's possible", or "it could be", for "ka'e".
So what you need to tell us is not how you'd say "it's possible", but
how you'd understand for example "ka'e ku no da klama".
Something can also be possible and extremely likely, or possible and
> I'd personally choose to not use a CAhA at all, but la'acu'i
extremely unlikely. If "possible" corresponds to "su'o", then the la'a
scale corresponds to so'a/so'e/so'i/so'o/so'u.
>(orThat was my point, "ka'e ku broda" has basically the same meanimg as
> go to full-blown brivla using cumki), while recognizing that it's not
> precisely the same part of speech, and perhaps not the same logical
> implications, I think it does convey the same meaning.
"cumki fa lo nu broda". Or more precisely, "ka'e" is "fi'o se cumki".
> In any case, I"If I can I will"? I count five syllables.
> wondered about the following utterance (I'm pretty sure xorxes would like it
> and understand it in the way I intend): "va'o lo nu da'i ka'e no'a kei mi ba
> co'e" (hint: it's a six syllable in English). Is it clear? (Probably less
> clear as "va'olonuda'ika'eno'akeimibaco'e" ;-) )
Martin Bays
How about: {mi ca'a gi kakne gi zukte} (9 syllables)?
Robert LeChevalier
> Let's simplify "I want to try to pry {ka'e} from {kakne} at this
> point" and translate it as {mi la'e de'u djica}. We want to say
>
> (0) Not necessarily: mi la'e de'u djica
>
> We can try
>
> (1) na ku ni'i ku mi la'e de'u djica
>
> How would you translate that?
It is not the case that: something (unspecified) logically necessitates
me wanting it.
> To me it seems to mean "not logically
> because of something, I want it."
More or less the same as mine, but moving the negation away seems to
imply that you want it, but the negation still applies to the who sentence.
Nora suggests putting a "da" after ni'i might make this more clear
(1') na ku ni'i da ku mi la'e de'u djica
It is not the case that: there exists an x such that x necessitates me
wanting it.
>In other words I do in fact want it,
In English, perhaps, but not in the Lojban.
I'm a little hazy on negation scope, but I THINK
(i.) ni'iku naku mi la'e di'u djica
and
(ii.) naku ni'iku mi la'e di'u djica
differ in whether the ni'i is included in the negation
and I think would be translated respectively.
(i.) Logically entailed by something, it is false that I want it.
(ii.) It is false that (it is logically entailed that I want it).
I think that the latter approximates to your (0).
I would normally do anything complex like this with explicit prenexes, so
(ii.') naku ni'iku zo'u mi la'e di'u djica
To indicate that you want it despite what is logically entailed, you
would use ni'inai ku, with no sentence negation.
(iii.) ni'inai ku mi la'e de'u djica
(Despite) some logic, nevertheless I want it.
It is also possible that "na'eni'i ku" would serve to negate ONLY the
entailment. But na'e is a scalar negation and we haven't formally
defined what exactly na'eni'i means. Best guess for this
(iv.) na'eni'iku mi la'e di'u djica
Other-than-logically entailed, I (still) want it.
Which is still a claim that you want it, which is not (0) as I
understand it.
> (2) ni'i ku mi la'e de'u na ku djica
>
> What's that to you?
> To me it's "logically (because of something), I
> don't want it" which means I actually don't want it.
I believe that moving the naku changes its scope with regard to
existential variables, but otherwise, it still is a negation of the
sentence as a whole.
>>"necessarily" seems like a "therefore", which is the "se" form of the first
>>four and the unmodified ja'e. "Not necessarily" would then seem to be a
>>kind of negation of the therefore statement - not the nai form which has
>>been defined from the JCB era as "nevertheless", but presumably the na form.
>>
>
> If you can approximate sentence (0) in Lojban using any of BAI and SE
> and {na ku}, please show me. From what I can see, you can't get modal
> readings from BAI.
>>Whatever word you choose, it has to be used carefully. If you attach the
>>modal to the sumti "mi", you get
>>"I do not necessarily want to try to pry ka'e from kakne at this point (but
>>someone else might want to)."
>
>
> Because I do not see how the basic meaning "I do not necessarily want
> it" can be gotten,
The basic meaning in English is ambiguous. I am inclined to think that
"I want it" would be inconsistent and "I don't want it" might or might
not be inconsistent. The truth of "I want it" isn't the essential claim.
Perhaps the real problem is that the main selbri is wrong.
(ii.'') na nibli lenu mi la'e de'u djica
seems more straightforward to achieve your (0).
Nora observes that nibli/ni'i may not be the right concept for
"necessarily" as you use it in (0).
> I see even less how the inherence "but someone else
> might want it" can be gotten.
Change the English emphasis and it becomes more obvious.
"**I** do not necessarily want it"
or
"I don't necessarily want it to rain this weekend, (but the farmer who
is facing crop loss from drought certainly does want it to rain)."
lojbab
maikxlx
> maikxlx wrote:
> > Let's simplify "I want to try to pry {ka'e} from {kakne} at this
>>
>> point" and translate it as {mi la'e de'u djica}. We want to say
>>
>> (0) Not necessarily: mi la'e de'u djica
>>
>> We can try
>>
>> (1) na ku ni'i ku mi la'e de'u djica
>>
>> How would you translate that?
>
> It is not the case that: something (unspecified) logically necessitates me
> wanting it.
>
> Nora suggests putting a "da" after ni'i might make this more clear
> (1') na ku ni'i da ku mi la'e de'u djica
> It is not the case that: there exists an x such that x necessitates me
> wanting it.
>
Yes, I think {da} helps make it clearer.
> In English, perhaps, but not in the Lojban.
>
Right. The English "I don't want it by logical necessity" and similar
sentences are often ambiguous.
> I'm a little hazy on negation scope, but I THINK
>
> (i.) ni'iku naku mi la'e di'u djica
> and
> (ii.) naku ni'iku mi la'e di'u djica
>
[i.e. basically my (2) and (1) respectively]
> differ in whether the ni'i is included in the negation
> and I think would be translated respectively.
> (i.) Logically entailed by something, it is false that I want it.
> (ii.) It is false that (it is logically entailed that I want it).
>
I agree with the gist of these, but I think that the translation that
you gave for (ii) could also be a translation of (1') with {da}:
(1') na ku ni'i da ku mi la'e de'u djica
"It is false that there is one or more things that logically entail
that I want it"
"It is false that it is logically entailed that I want it."
With implicit {zo'e} I fear you might get a different result.
> I think that the latter approximates to your (0).
>
I am unsure that (1)/(ii) with {ni'i [zo'e] ku} is a good enough
approximation. At first glance, Nora's (1') with {ni'i da ku} could
work. But there are issues. Consider this example:
(3) Necessarily: ro nanmu je se mensi cu bruna
This has "necessarily", without the "not". Dropping the {na ku} as
used in (0), the {ni'i da ku} solution gives us:
(3') ni'i da ku ro nanmu je se mensi ne'e bruna
"There exists an x such that x necessitates that all men with sisters
are brothers."
"Something necessitates that all men with sisters are brothers."
That _is_ arguably an approximation, but not really what we want to
say. It gets even clumsier when you try to express "possibly". Note
that in modal logic, "possibly" is interchangeable with "not
necessarily not". Taking advantage of this:
(4) Possibly: ko'a bruna. "He is possibly a brother".
(4') na ku ni'i da ku na ku ko'a bruna
"It is false that there exists an x such that x necessitates that he
is not a brother."
That also is not _really_ what you want to say.
> I would normally do anything complex like this with explicit prenexes, so
> (ii.') naku ni'iku zo'u mi la'e di'u djica
>
> To indicate that you want it despite what is logically entailed, you would
> use ni'inai ku, with no sentence negation.
>
> (iii.) ni'inai ku mi la'e de'u djica
> (Despite) some logic, nevertheless I want it.
>
> It is also possible that "na'eni'i ku" would serve to negate ONLY the
> entailment. But na'e is a scalar negation and we haven't formally defined
> what exactly na'eni'i means. Best guess for this
>
> (iv.) na'eni'iku mi la'e di'u djica
> Other-than-logically entailed, I (still) want it.
>
> Which is still a claim that you want it, which is not (0) as I understand
> it.
>
I follow, and agree that neither {nai} nor {na'e} get you (0).
> Perhaps the real problem is that the main selbri is wrong.
>
> (ii.'') na nibli lenu mi la'e de'u djica
>
I'd put {da} at the front to be clear, but other than that I don't see
how it's different than the {ni'i} version. "It's false that something
necessitates that I want it."
> seems more straightforward to achieve your (0).
>
> Nora observes that nibli/ni'i may not be the right concept for
> "necessarily" as you use it in (0).
>
I think Nora may be right. Entailment is not the same as modal
necessity and using {ni'i} to express the latter is at best a kludge.
>
>> I see even less how the inherence "but someone else
>>
>> might want it" can be gotten.
>
> Change the English emphasis and it becomes more obvious.
>
> "**I** do not necessarily want it"
>
Yes, but wouldn't {ba'e} be needed to achieve this kind of emphasis in Lojban?
> lojbab
maikxlx
On Wed, Nov 23, 2011 at 11:18 PM, maikxlx <mai...@gmail.com> wrote:
> (3') ni'i da ku ro nanmu je se mensi ne'e bruna
Copy-and-paste error. I meant:
(3') ni'i da ku ro nanmu je se mensi CU. bruna
selpa'i
If someone asks, out of the blue, how many lions there are and I say "About 12,000", my answer may be wrong but it is the right sort of answer. If the gotcha questioner says "no, there are four" I can righteously respond "Hey, there are five just in our Zoo". If he goes on to explain "The European (now extinct), the African, the Indian, and the Asiatic", I might exclaim "Oh, you meant *kinds* of lions!". Yet, had he begun the conversation with "There are four lions: the European, the Asiatic, the Indian, and the African", I would have understood him fine and had no complaints. I would equally have no problems with "That lion is the same as the one we saw yesterday" nor, probably, with "Lion (or lions) is (are) quite tasty, when marinated in monkey-brain sauce and roasted over an open fire", nor "Lions eat gazelles" nor "A/The lion is/Lions are the second largest cat". And so on through countless other examples
A joke I read in the newspaper today:
The teacher asks a student: "Name three animals from Africa."
The student answers: "An elephant and two lions."
mu'o mi'e la selpa'i
John E Clifford
From: selpa'i <sel...@gmx.de>
To: loj...@googlegroups.com
Sent: Monday, November 26, 2012 11:38 AM
Subject: [lojban] Re: Lions and levels and the like
You received this message because you are subscribed to the Google Groups "lojban" group.
Pierre Abbat
> If we take lo cinfoto refer to the salient sublattice of the extension of
> cinfo,
ki'asai???
mu'omi'e .pier.
--
sei do'anai mi'a djuno puze'e noroi nalselganse srera
John E Clifford
From: Pierre Abbat <ph...@bezitopo.org>
To: loj...@googlegroups.com
Sent: Tuesday, November 27, 2012 4:17 PM
Subject: Re: [lojban] Re: Lions and levels and the like
You received this message because you are subscribed to the Google Groups "lojban" group.