WikiDiscuss

As the absence of a tense (etc.) marker before the main predicate allows for any such marker to be understood contextually, so the absence of a quantifier before (or within?) a sumti allows any such marker to be understood contextually. Presumably excluding negatives in both cases. But unlike the case of tenses, there is a quantifier — or maybe two — that covers all cases: {su'o} (and maybe {pisu'o}). Viewed in this light, the difference between {lo1} and {lo2} is not about the sumti involved at all but rather about the tenses, whether the occasion is one particular one or an indefinite number of them (each under a world-creating mark of some sort — possibly implicit). That is, all the cases involving merely actual occurrences or possible actual occurrences are the same in structure and referential technique, differing only in the realms in which that technique is applied. The interesting question then is whether {lo3} and {lo4} are also cases of this sort. {lo4} looks at
first like it might be covered by fractional quantifiers, but then shows some cases where this is not obviously going to work. {lo3} presents mainly cases where we have to specify the tense — or, better, mode — to make the point clearly. But that is just the case always; clarity requires explicitness, only vaguenss can be expressed vaguely.
In short, it seems to me that old {lo} does exactly what all the various suggestions for new {lo}s seem to want to do — and without murky metaphysics and massive amounts of only vaguely intelligible argle-bargle. The effect of the argle-bargle has been mainly to set contexts where it is hard to see what the original {lo} is about, by forcing it to be viewed as more specific than it is, as different from the case being proposed rather than as embracing it. Large parts of this problem seem also to be derived from trying to bring intensional contexts up to surface level, rather than treating them where they are and from ignoring the scopes of tense/mode expressions (and, indeed, of quantifiers). I pass over the probable sources of these oversights.

pc:
> As the absence of a tense (etc.) marker before the main predicate allows for
> any such marker to be understood contextually,

Any such marker or none, yes. {li vo sumji li re li re} requires no
contextually understood tense marker.

> so the absence of a quantifier
> before (or within?) a sumti allows any such marker to be understood
> contextually.

Any such marker or none. {lo} marks the beginning of a sumti (selecting
the x1 of the following selbri), just like {cu} marks the beginning
of a selbri. {lo} adds no content of its own, just like {cu} adds
no content of its own.

> Presumably excluding negatives in both cases. But unlike the
> case of tenses, there is a quantifier — or maybe two — that covers all
> cases: {su'o} (and maybe {pisu'o}).

What about {su'oroi} for tenses? Which cases does it not cover?

....
> But that is just the case always; clarity requires
> explicitness, only vaguenss can be expressed vaguely.

Indeed.

> In short, it seems to me that old {lo} does exactly what all the various
> suggestions for new {lo}s seem to want to do — and without murky metaphysics
> and massive amounts of only vaguely intelligible argle-bargle.

Old {lo}, when understood as {su'o lo}, was not transparent to negation,
and could not be anaphorized from outside of an abstraction.

> The effect of
> the argle-bargle has been mainly to set contexts where it is hard to see what
> the original {lo} is about, by forcing it to be viewed as more specific than
> it is, as different from the case being proposed rather than as embracing it.

Probably the original idea for {lo} did not involve quantification.
Quantified {su'o lo} has lots of scope issues that a simple
selbri-to-sumti converter does not have.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

A> I am not sure that there is such a thing as "no tense/mode" but OK, let us add that. The mathematical example requires no contextual tense because it sets the context within itself: mathematics (and, indeed, at least quintal arithmetic).

B> This is less clear. {lo} apparently means at least that there are broda in the situation referrred to, ral one in that context. {cu} does not claim that there are actual occasions of the sort described.

C> {ka'e} (or the real modal that this stands in the way of), anything around {da'i} and other world-altering items from various categories.

D> And the problem with that is? If {lo} does require existence, then it is not transparent to negation; if it is totally indeterminate, then it is not anaphorizable across context.

E> I am now less sure of that than you, just as I am less sure that old {lo} really is equivalent to {su'o lo} or just {su'o}. But most of the "problems" seem, as noted, to be inherent in the notion involved, vague as it is — and in attempts to misuse it in a variety of ways.
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:

pc:
> As the absence of a tense (etc.) marker before the main predicate allows for
> any such marker to be understood contextually,

A>Any such marker or none, yes. {li vo sumji li re li re} requires no
contextually understood tense marker.

> so the absence of a quantifier
> before (or within?) a sumti allows any such marker to be understood
> contextually.

B>Any such marker or none. {lo} marks the beginning of a sumti (selecting
the x1 of the following selbri), just like {cu} marks the beginning
of a selbri. {lo} adds no content of its own, just like {cu} adds
no content of its own.

> Presumably excluding negatives in both cases. But unlike the
> case of tenses, there is a quantifier — or maybe two — that covers all
> cases: {su'o} (and maybe {pisu'o}).

C>What about {su'oroi} for tenses? Which cases does it not cover?

....
> But that is just the case always; clarity requires
> explicitness, only vaguenss can be expressed vaguely.

Indeed.

> In short, it seems to me that old {lo} does exactly what all the various
> suggestions for new {lo}s seem to want to do — and without murky metaphysics
> and massive amounts of only vaguely intelligible argle-bargle.

D>Old {lo}, when understood as {su'o lo}, was not transparent to negation,
and could not be anaphorized from outside of an abstraction.

> The effect of
> the argle-bargle has been mainly to set contexts where it is hard to see what
> the original {lo} is about, by forcing it to be viewed as more specific than
> it is, as different from the case being proposed rather than as embracing it.

E>Probably the original idea for {lo} did not involve quantification.
Quantified {su'o lo} has lots of scope issues that a simple
selbri-to-sumti converter does not have.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

pc:
> This is less clear. {lo} apparently means at least that there are broda
> in the situation referrred to, ral one in that context. {cu} does not claim
> that there are actual occasions of the sort described.

{lo broda cu brode} claims as much about brodas and brodes
as {lo brode cu broda}, doesn't it?

> {ka'e} (or the real modal that this stands in the way of), anything around
> {da'i} and other world-altering items from various categories.

{lo broda} can stand for {lo ka'e broda} to the same extent that
{cu broda} can stand for {cu ka'e broda}, can't it? Only in a very
rare context. And the same applies to {da'i} et al.

> If {lo} does require existence, then it is
> not transparent to negation; if it is totally indeterminate, then it is not
> anaphorizable across context.

You can say this with {lo}:

mi na viska lo broda i ku'i mi ja'a tirna by
I don't see brodas, but I do hear them.

That doesn't work with {su'o lo broda}.

> But most of the "problems"
> seem, as noted, to be inherent in the notion involved, vague as it is — and
> in attempts to misuse it in a variety of ways.

I don't think {lo} can be misused any more than {cu} can be misused.
It may be wrong because the syntax does not allow it, it may be
inadequate because it is not precise enough for what you want to
say, but it can't be misused for scope reasons.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

A> Yes, that there is a realm where the referred to exists/holds. So there is some interplay between them: The broda have to exist where the {cu} takes us — and {cu} takes us to a place where the broda exist.

B> With {ka'e} in its present reading, I do think this is probably raare, but with "it is possible that..." it would be fairly common. {da'i} is even more common, I suppose — if it is going to handle subjunctives.

C> What is the problem here? {mi na viska su'o broda i ku'i mi ja'a tirna by} I take it that {by} is, appropriately, a literal anaphora ({mi na viska su'o broda i ku'i mi ja'a tirna su'o broda} or a referential one that works like a variable: {mi na viska su'o da poi broda i ku'i mi ja'a tirna da} (i.e., by standard logic, {ro da poi broda zo'u mi na viska da i ku'i ja'a tirna da}).

D> So far as I know (but I woin't hold my breath) no one has proposed a weird metaphysics for {cu}, that, for example, it means the same time every occasion it is used with the same predicate,
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:

pc:
> This is less clear. {lo} apparently means at least that there are broda
> in the situation referrred to, ral one in that context. {cu} does not claim
> that there are actual occasions of the sort described.

A>{lo broda cu brode} claims as much about brodas and brodes
as {lo brode cu broda}, doesn't it?

> {ka'e} (or the real modal that this stands in the way of), anything around
> {da'i} and other world-altering items from various categories.

B>{lo broda} can stand for {lo ka'e broda} to the same extent that
{cu broda} can stand for {cu ka'e broda}, can't it? Only in a very
rare context. And the same applies to {da'i} et al.

> If {lo} does require existence, then it is
> not transparent to negation; if it is totally indeterminate, then it is not
> anaphorizable across context.

C>You can say this with {lo}:

mi na viska lo broda i ku'i mi ja'a tirna by
I don't see brodas, but I do hear them.

That doesn't work with {su'o lo broda}.

> But most of the "problems"
> seem, as noted, to be inherent in the notion involved, vague as it is — and
> in attempts to misuse it in a variety of ways.

D>I don't think {lo} can be misused any more than {cu} can be misused.
It may be wrong because the syntax does not allow it, it may be
inadequate because it is not precise enough for what you want to
say, but it can't be misused for scope reasons.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

pc:
> A> Yes, that there is a realm where the referred to exists/holds. So there
> is some interplay between them: The broda have to exist where the {cu} takes
> us — and {cu} takes us to a place where the broda exist.

So {lo} marks that which is taken, and {cu} marks the place where it
is taken. The difference between {lo broda cu brode} and {lo brode
cu broda} is one of perspective, but the end result is the same:
broda and brode coming together. {broda gi'e brode} gives a third
perspective.

> B> With {ka'e} in its present reading, I do think this is probably raare,
> but with "it is possible that..." it would be fairly common. {da'i} is even
> more common, I suppose — if it is going to handle subjunctives.

I meant it would be rare for a bare {cu} to stand for {cu ka'e}
or a bare {lo} for {lo ka'e}. I agree that explicit {ka'e} and {da'i}
are not rare.

> C> What is the problem here? {mi na viska su'o broda i ku'i mi ja'a tirna
> by} I take it that {by} is, appropriately, a literal anaphora ({mi na viska
> su'o broda i ku'i mi ja'a tirna su'o broda} or a referential one that works
> like a variable: {mi na viska su'o da poi broda i ku'i mi ja'a tirna da}
> (i.e., by standard logic, {ro da poi broda zo'u mi na viska da i ku'i ja'a
> tirna da}).

But I don't mean to say that I hear _every_ broda.

> D> So far as I know (but I woin't hold my breath) no one has proposed a
> weird metaphysics for {cu}, that, for example, it means the same time every
> occasion it is used with the same predicate,

In my view, {cu} doesn't have any meaning other than to mark the
start of the selbri. Similarly, {lo} has no meaning other than to
mark the conversion of a selbri to a sumti.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

X> Yes, though the pragmatics — and even the semantics — is going to balk at this in almost any other language. But yes, "The runner is blue" anfd "The blue thing is running" may well refer to the same event coming from different places and, probably, heading different places as well.

W> I forget what CLL says, but Loglan 1 was always clear that the basic meaning of unmarked sentences was to point to an event without any comment about whether it occurred or even could occur.

V> But every relevant one (or, more likely, the collection of the relevant ones); this is still a long way from totally pragmatically explicit, since the speaker always assumes that the hearer speaks the language and is being cooperative. Or read {by} as a literal anaphora, which may be a bit closer to what you want without going so deep into pragmatics.

U> Then whence Mr. Rabbit and all that? Is it nowhere represented in the grammatical structure but overarching the semantic one?

Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:

pc:
> A> Yes, that there is a realm where the referred to exists/holds. So there
> is some interplay between them: The broda have to exist where the {cu} takes
> us — and {cu} takes us to a place where the broda exist.

X>So {lo} marks that which is taken, and {cu} marks the place where it
is taken. The difference between {lo broda cu brode} and {lo brode
cu broda} is one of perspective, but the end result is the same:
broda and brode coming together. {broda gi'e brode} gives a third
perspective.

> B> With {ka'e} in its present reading, I do think this is probably raare,
> but with "it is possible that..." it would be fairly common. {da'i} is even
> more common, I suppose — if it is going to handle subjunctives.

W>I meant it would be rare for a bare {cu} to stand for {cu ka'e}
or a bare {lo} for {lo ka'e}. I agree that explicit {ka'e} and {da'i}
are not rare.

> C> What is the problem here? {mi na viska su'o broda i ku'i mi ja'a tirna
> by} I take it that {by} is, appropriately, a literal anaphora ({mi na viska
> su'o broda i ku'i mi ja'a tirna su'o broda} or a referential one that works
> like a variable: {mi na viska su'o da poi broda i ku'i mi ja'a tirna da}
> (i.e., by standard logic, {ro da poi broda zo'u mi na viska da i ku'i ja'a
> tirna da}).

V>But I don't mean to say that I hear _every_ broda.

> D> So far as I know (but I woin't hold my breath) no one has proposed a
> weird metaphysics for {cu}, that, for example, it means the same time every
> occasion it is used with the same predicate,

U>In my view, {cu} doesn't have any meaning other than to mark the
start of the selbri. Similarly, {lo} has no meaning other than to
mark the conversion of a selbri to a sumti.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

pc:
> I forget what CLL says, but Loglan 1 was always clear that the basic
> meaning of unmarked sentences was to point to an event without any comment
> about whether it occurred or even could occur.

I suppose the same applies to Lojban. In the absence of context,
we tend to assume we're talking about the real world, but context
may very well change that.

> Or read {by} as a literal anaphora, which may be a bit closer to
> what you want without going so deep into pragmatics.

Yes, that works in many cases. But it doesn't help with the
transparency to negation bit.

> Then whence Mr. Rabbit and all that? Is it nowhere represented in the
> grammatical structure but overarching the semantic one?

I'd say so, yes. Mr Rabbit is that which rabbits.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

S> I guess I am trying to think of a nominal that is transparent to negation. None come to mind, even Mr. Rabbit insofar as what he does depends upon instances actually doing it.

R> that is, is picked up by {lo} to distinguish Mr Rabbit from both {PA ractu} and {cu ractu}.
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
pc:
> I forget what CLL says, but Loglan 1 was always clear that the basic
> meaning of unmarked sentences was to point to an event without any comment
> about whether it occurred or even could occur.

I suppose the same applies to Lojban. In the absence of context,
we tend to assume we're talking about the real world, but context
may very well change that.

> Or read {by} as a literal anaphora, which may be a bit closer to
> what you want without going so deep into pragmatics.

S>Yes, that works in many cases. But it doesn't help with the
transparency to negation bit.

> Then whence Mr. Rabbit and all that? Is it nowhere represented in the
> grammatical structure but overarching the semantic one?

R>I'd say so, yes. Mr Rabbit is that which rabbits.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

pc:
> I guess I am trying to think of a nominal that is transparent to negation.
> None come to mind, even Mr. Rabbit insofar as what he does depends upon
> instances actually doing it.

Consider these sentences:

A1) naku la djan su'oroiku klama
It is not the case that John at least once goes.

A2) la djan su'oroiku naku klama
John at least once does not go.

A3) la djan na klama
John does not go.
(=It is not the case that John goes.)

B1) naku su'o lo ractu cu klama
It is not the case that at least one rabbit goes.

B2) su'o lo ractu naku klama
At least one rabbit does not go.

B3) lo ractu na klama
Rabbits don't go.
(=It is not the case that rabbits go.)

A3 is neither A1 nor A2. It is more vague than either A1 or A2,
because it does not go into occasions.

B3 is neither B1 nor B2. It is more vague than either B1 or B2,
because it does not go into instances.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

On Friday 04 June 2004 19:46, Jorge "Llamb=EDas" wrote:
> B1) naku su'o lo ractu cu klama
> It is not the case that at least one rabbit goes.
>
> B2) su'o lo ractu naku klama
> At least one rabbit does not go.
>
> B3) lo ractu na klama
> Rabbits don't go.
> (=3DIt is not the case that rabbits go.)
>
> A3 is neither A1 nor A2. It is more vague than either A1 or A2,
> because it does not go into occasions.
>
> B3 is neither B1 nor B2. It is more vague than either B1 or B2,
> because it does not go into instances.

As I use {lo}, B3 is equivalent to B1. To mean "Rabbits don't go", I'd say=
=20
{lo'e ractu na klama} or {lo'e ractu naku klama}.

You can't make logical deductions on sumti with {lo'e}. For instance lo'e=20
=2Eornitorinku na'o se jbena re sovda .ije lo'e .ornitorinku cu vindu se jg=
alu,=20
=2Eiku'i naku lo'e .ornitorinku na'o se jbena re sovda gi'e vindu se jgalu.=
The=20
first is Mrs. Platypus, the second is Mr. Platypus, and no platypus belongs=
=20
to both sexes but they're both typical platypuses.

phma
=2D-=20
li fi'u vu'u fi'u fi'u du li pa

On Fri, Jun 04, 2004 at 08:12:59PM -0400, Pierre Abbat wrote:
> As I use {lo}, B3 is equivalent to B1. To mean "Rabbits don't go", I'd
> say {lo'e ractu na klama} or {lo'e ractu naku klama}.

That isn't what "lo'e" means. {lo'e ractu naku klama} means that

• typical* rabbits don't go (and is trivially false). It makes no

claims, however, about rabbits in general, only the typical ones.

It *possibly* only makes claims about the *properties* of typical
rabbits, but that's not terribly important.

Regardless, a one-eared rabbit is not covered by "lo'e ractu". For
similar reasons, "lo'e" does not work for "I need a doctor".

BTW, Pierre, I'd appreciate you updating us on the exact status of your
objections to the proposal, as you are the last BPFK member with a No
vote and I'd prefer to have consensus.

-Robin

A> Assuming John names something in the (appropriate) world and is the sort of thing that can klama and has had an opportunity to — all things that are given only by context, then A1 implies A2, but is not implied by it. Notice that without those assumptions, the inference does not go through; the sentence A1 could be a way to inform someone that they have some very wrongheaded ideas about John, namely that the assumptions held. But once you let these kinds of assumptions in, all sorts of things — even {su'o} — become negation transparent, inward (as here) at least. A3 is somewhat vaguer but still requires that John be a name in the world (for something movable) before the inward passage works. It is not at all clear that the outward passage works here. It does if we assume a single occasion, but that aside — and this could be general, typical, and so on — it does not (well, it would work in a complex way for "always").

B: So, assuming there are rabbits (etc.), the inward inference hold, but even then the outward does not. B1 implies B2 (if...) but not B2 implies B1. What then of B3? I suppose it implies B1 (since the negated sentence in B1 implies the negated sentence in B3, if that means anything). Since B1 implies B2, so does B3. And that seems to mean that B3 implies {lo ractu naku klama} — assuming there are rabbits, of course. The same path will not get us back the other way, unless we are willing to reduce {lo ractu} to {su'o ractu} at least materially (i.e., they are true or false together). And even then, the next couple of steps will not work since B2 does not imply B1, nor B1 B3. And neither the vagueness of {lo} nor that of {cu} gets us through this gap, A move with a vague notion will not work if it fails for any notion within the range of its vagueness and this fails for most of the idtems in the vague range here. I know that you want to deny that {lo ractu} as such is a
vaguely quantified sumti, but every attempt to explicate — including Mr. Rabbit — makes it one overtly or implicitly. And even if you take {lo3}, species (i.d., does not go into instances), the move still only goes inward, whatever reasonable assumptions you want to make.

Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:

pc:
> I guess I am trying to think of a nominal that is transparent to negation.
> None come to mind, even Mr. Rabbit insofar as what he does depends upon
> instances actually doing it.

Consider these sentences:

A1) naku la djan su'oroiku klama
It is not the case that John at least once goes.

A2) la djan su'oroiku naku klama
John at least once does not go.

A3) la djan na klama
John does not go.
(=It is not the case that John goes.)

B1) naku su'o lo ractu cu klama
It is not the case that at least one rabbit goes.

B2) su'o lo ractu naku klama
At least one rabbit does not go.

B3) lo ractu na klama
Rabbits don't go.
(=It is not the case that rabbits go.)

A3 is neither A1 nor A2. It is more vague than either A1 or A2,
because it does not go into occasions.

B3 is neither B1 nor B2. It is more vague than either B1 or B2,
because it does not go into instances.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

On Friday 04 June 2004 20:23, Robin Lee Powell wrote:
> That isn't what "lo'e" means. {lo'e ractu naku klama} means that
> *typical* rabbits don't go (and is trivially false). It makes no
> claims, however, about rabbits in general, only the typical ones.
>
> It *possibly* only makes claims about the *properties* of typical
> rabbits, but that's not terribly important.
>
> Regardless, a one-eared rabbit is not covered by "lo'e ractu". For
> similar reasons, "lo'e" does not work for "I need a doctor".
>
> BTW, Pierre, I'd appreciate you updating us on the exact status of your
> objections to the proposal, as you are the last BPFK member with a No
> vote and I'd prefer to have consensus.

I haven't changed my vote because I don't fully understand the discussion. On
the summary Why the XS is better than the 2004 status quo:
1. What's intensionality? Why do we need it?
2. What is the difference between generic and typical?
3. Agreed, but what's extensional?
4. Not sure.

phma
--
li fi'u vu'u fi'u fi'u du li pa

On Fri, Jun 04, 2004 at 08:44:16PM -0400, Pierre Abbat wrote:
> On Friday 04 June 2004 20:23, Robin Lee Powell wrote:
> > BTW, Pierre, I'd appreciate you updating us on the exact status of
> > your objections to the proposal, as you are the last BPFK member
> > with a No vote and I'd prefer to have consensus.
>
> I haven't changed my vote because I don't fully understand the
> discussion.

So what? The discussion is irrelevant byplay; the proposal *MUST* stand
on its own.

> On the summary Why the XS is better than the 2004 status quo:

For the rest of you, I've added the text of that post.

We need intensionality and don't have a style that's agreed upon, and
this lo can cover those cases
> 1. What's intensionality? Why do we need it?

Google is your friend:

http://www.cse.buffalo.edu/~rapaport/intensional.html

We need it because it is currently *impossible* to say "I need a doctor"
in Lojban and have it mean what you want. Trust us on this. In
particular, there is no formulation of that phrase that is not

• instantly* invalidated by all doctors being dead.

As another example, it is not possible to say "Unicorns are white" in
current Lojban.

Bear in mind that, without redefinition, lo'e does *not* help.

This is, as I understand it, the reason that all of the people who
understand these things declared gadri to be broken.

In most usage it has been used as the generic
> 2. What is the difference between generic and typical?

The difference between "Unicorns are white" and "Typical unicorns are
white", where the former is _not_ "All unicorns are white". I'm not
sure how to explain it better than that.

How about this: "Typical dogs have four legs" is true, but "Dogs have
four legs" and "Dogs have three legs", in the *generic* sense, are
either both true or both neither true nor untrue, depending on how you
want to look at it. "All dogs have four legs" and "All dogs have three
legs" are both merely false.

Furthermore, we're talking about lo, not lo'e, so typical doesn't much
enter in to it.

We don't need an extensional alias for da poi; we can use su'o for
that
> 3. Agreed, but what's extensional?

www.justfuckinggoogleit.com

It will actually retain most or all of the properties which the Book
assigns it
> 4. Not sure.

What properties are you not sure about?

-Robin

Pierre:
> > B1) naku su'o lo ractu cu klama
> > B2) su'o lo ractu naku klama
> > B3) lo ractu na klama
>
> As I use {lo}, B3 is equivalent to B1.

That's the CLL-lo, right.

> To mean "Rabbits don't go", I'd say
> {lo'e ractu na klama} or {lo'e ractu naku klama}.

That's how I used to do it, too. This is not quite the standard
meaning of lo'e though, which is supposed to be used only for
properties that apply across the board. To say "rabbits don't go
into that hole but they do go into that other one", I can't make
a claim about the typical rabbit.

Another reason to prefer {lo} for the generic function is that
{lo} is the least marked gadri and this function is the one
with least content. All {lo} does is convert selbri to sumti.

> You can't make logical deductions on sumti with {lo'e}. For instance
> lo'e ornitorinku na'o se jbena re sovda .ije lo'e .ornitorinku cu
> vindu se jgalu, iku'i naku lo'e .ornitorinku na'o se jbena re sovda
> gi'e vindu se jgalu. The first is Mrs. Platypus, the second is
> Mr. Platypus, and no platypus belongs to both sexes but they're both
> typical platypuses.

With the proposed {lo} you could say:

lo ornitorinku cu se jbena re sovda fi'o fetsi ku gi'e vindu se
jgalu fi'o nakni ku iku'i no ornitorinku ge se jbena lo sovda
gi vindu se jgalu

You could leave out the fi'o fetsi and fi'o nakni, but then it
sounds like a riddle.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

Pierre Abbat:
> 1. What's intensionality? Why do we need it?

Consider for example:

le va spati cu ca'o claxu lo solri gusni
That plant is not getting enough sunlight.

We are not claiming that there is some solri gusni such that
that gusni does not get to the plant. Even if the plant got plenty
of solri gusni, there would still be some solri gusni that it doesn't
get, so that claim would be different. We don't really want to consider
instances of solri gusni here, just a relationship between le va spati
and solri gusni in general.

> 2. What is the difference between generic and typical?

Generic is when you don't go into instances/cases at all. Typical
is when you take a census of all instances/cases and get some
statistical conclusions from that.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

On Friday 04 June 2004 21:00, Robin Lee Powell wrote:
> We need it because it is currently *impossible* to say "I need a doctor"
> in Lojban and have it mean what you want. Trust us on this. In
> particular, there is no formulation of that phrase that is not
> *instantly* invalidated by all doctors being dead.

Okay, I've changed my vote. I still don't understand, though, why quantifiers
are used differently with {lo} than with {le}. To me {mu lo vo} and {mu le
vo} are equally nonsensical.

phma
--
li fi'u vu'u fi'u fi'u du li pa

Pierre Abbat:
> I still don't understand, though, why quantifiers
> are used differently with {lo} than with {le}. To me {mu lo vo} and {mu le
> vo} are equally nonsensical.

With {le}, you have a certain group. There's only one instance
of the group. The only thing to quantify over are the members.

With {lo}, you don't have a certain group. If you want to quantify
over individuals you need first to fix on one instance of the
group and then take the members of that instance. That's two steps.
On the other hand, the generic group can have many instances,
and quantifying over them is the natural option.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

Q> The typical relevant one: no one cares in this situation about the ones in Australia, say.

P> And assert there are broda.

O> Good. You have come around to using the the restricted predicate resolution. I like the restricted subject solution better, but at least you are rid of one contradiction in your system. Dropping it again in this case wuld do little harm, but you need always to have it handy.
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:

Pierre:
> > B1) naku su'o lo ractu cu klama
> > B2) su'o lo ractu naku klama
> > B3) lo ractu na klama
>
> As I use {lo}, B3 is equivalent to B1.

That's the CLL-lo, right.

> To mean "Rabbits don't go", I'd say
> {lo'e ractu na klama} or {lo'e ractu naku klama}.

Q>That's how I used to do it, too. This is not quite the standard
meaning of lo'e though, which is supposed to be used only for
properties that apply across the board. To say "rabbits don't go
into that hole but they do go into that other one", I can't make
a claim about the typical rabbit.

P>Another reason to prefer {lo} for the generic function is that
{lo} is the least marked gadri and this function is the one
with least content. All {lo} does is convert selbri to sumti.

> You can't make logical deductions on sumti with {lo'e}. For instance
> lo'e ornitorinku na'o se jbena re sovda .ije lo'e .ornitorinku cu
> vindu se jgalu, iku'i naku lo'e .ornitorinku na'o se jbena re sovda
> gi'e vindu se jgalu. The first is Mrs. Platypus, the second is
> Mr. Platypus, and no platypus belongs to both sexes but they're both
> typical platypuses.

O>With the proposed {lo} you could say:

lo ornitorinku cu se jbena re sovda fi'o fetsi ku gi'e vindu se
jgalu fi'o nakni ku iku'i no ornitorinku ge se jbena lo sovda
gi vindu se jgalu

You could leave out the fi'o fetsi and fi'o nakni, but then it
sounds like a riddle.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

N> This is not clearly an intensional case. Note that there is a perfectly good (well, only lsightly suspect) quantifier here, so it isn't even clear that this is about a relation between this individual plant and sunlight in general.

M> Well, I doubt that statistics really play a role in typical (which is why I want a definitely statistical gadri), but the sorts of things that go into gathering statstics are used on a more infomal basis (at some point — we usually get our typicals from "common knowledge"). Of course, I would say that "generic" just means we aren't/don't care what quantifier is involved (and so {su'o} is always correct).
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:

Pierre Abbat:
> 1. What's intensionality? Why do we need it?

N>Consider for example:

le va spati cu ca'o claxu lo solri gusni
That plant is not getting enough sunlight.

We are not claiming that there is some solri gusni such that
that gusni does not get to the plant. Even if the plant got plenty
of solri gusni, there would still be some solri gusni that it doesn't
get, so that claim would be different. We don't really want to consider
instances of solri gusni here, just a relationship between le va spati
and solri gusni in general.

> 2. What is the difference between generic and typical?

M>Generic is when you don't go into instances/cases at all. Typical
is when you take a census of all instances/cases and get some
statistical conclusions from that.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

L> {mi nitcu tu'a lo mikce} Note that dropping the {tu'a] makes a claim that IS falsified if all the doctors are dead (and doesn't say what is wanted — or doesn't do it very clearly — anyhow). Pierre, {tu'a} here demarcates an intensional context, another world talked about in this one.

Pierre Abbat <phma@phma.hn.org> wrote:L>On Friday 04 June 2004 21:00, Robin Lee Powell wrote:

> We need it because it is currently *impossible* to say "I need a doctor"
> in Lojban and have it mean what you want. Trust us on this. In
> particular, there is no formulation of that phrase that is not
> *instantly* invalidated by all doctors being dead.

Okay, I've changed my vote. I still don't understand, though, why quantifiers
are used differently with {lo} than with {le}. To me {mu lo vo} and {mu le
vo} are equally nonsensical.

phma
--
li fi'u vu'u fi'u fi'u du li pa

I think it's a fine and clever example of intensionality because the
selbri place in question does not demand an event, which would
grammatically justify the tu'a legerdemain. And it passes my test for
intensionality: the statement's truth is independent of the existence of
lo solri gusni. Do you have a test that it fails?

(Meta, Gripe: Look at the hash that's been made of this message. Jorge's
name is stripped; responses are scattered above and below; this sad post
makes barely any sense removed from the thread itself. Should we pass
around the hat and buy pc the url to a $free email client that doesn't
suck? Or is the problem elsewhere?)

John E Clifford wrote:

>N> This is not clearly an intensional case. Note that there is a perfectly good (well, only lsightly suspect) quantifier here, so it isn't even clear that this is about a relation between this individual plant and sunlight in general.
>
>
>
>Pierre Abbat:
>
>
>>1. What's intensionality? Why do we need it?
>>
>>
>
>N>Consider for example:
>
>le va spati cu ca'o claxu lo solri gusni
>That plant is not getting enough sunlight.
>
>We are not claiming that there is some solri gusni such that
>that gusni does not get to the plant. Even if the plant got plenty
>of solri gusni, there would still be some solri gusni that it doesn't
>get, so that claim would be different. We don't really want to consider
>instances of solri gusni here, just a relationship between le va spati
>and solri gusni in general.
>
>

--
Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."

This is probably an official point about which we should hear from
Jorge, and perhaps a specific note about Existence should go in the
proposal unless it's already there.

My understanding is no, lo does not imply existence, and "mi nitcu lo
mikce" is true when no doctors exist.

John E Clifford wrote:

>L> {mi nitcu tu'a lo mikce} Note that dropping the {tu'a] makes a claim that IS falsified if all the doctors are dead
>

--
Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."

On Sat, Jun 05, 2004 at 07:11:27PM -0400, xod wrote:
> (Meta, Gripe: Look at the hash that's been made of this message.
> Jorge's name is stripped; responses are scattered above and below;
> this sad post makes barely any sense removed from the thread itself.
> Should we pass around the hat and buy pc the url to a $free email
> client that doesn't suck? Or is the problem elsewhere?)

It is without question PC's problem. There's a list of good, free
e-mail clients at www.justfuckinggoogleit.com, but as he has rebuffed
every single attempt I've ever made to help him with computer problems
(usually in an amazingly rude fashion), I don't think such a list will
help him.

-Robin

On Sat, Jun 05, 2004 at 07:14:24PM -0400, xod wrote:
> This is probably an official point about which we should hear from
> Jorge, and perhaps a specific note about Existence should go in the
> proposal unless it's already there.
>
> My understanding is no, lo does not imply existence, and "mi nitcu lo
> mikce" is true when no doctors exist.

This is absolutely correct with the new lo.

OTOH, if you put the old default quantifiers back in (su'o lo ro mikce),
there must be at least one living doctor.

The unicorn example was intended to address this issue.

-Robin

D> I suspect the problem is in your receiver, since my version (both sent and received) has all my comments at the top and all names in place.

C>It passes the "non-existent" test only because of the negation inherent in {claxu}, you can always lack (not have) something that does not exist, not because dealing with that takes you into another realm but because the selbri applies even when the sumti predicate does not. I think, incidentally, that English {lack} is also intensional, since it seems to require not only that you don't have it but that you need — or at least want — to have it. This does not seem to be the case with {claxu} however.
xod <xod@thestonecutters.net> wrote:
C>I think it's a fine and clever example of intensionality because the
selbri place in question does not demand an event, which would
grammatically justify the tu'a legerdemain. And it passes my test for
intensionality: the statement's truth is independent of the existence of
lo solri gusni. Do you have a test that it fails?

D>(Meta, Gripe: Look at the hash that's been made of this message. Jorge's
name is stripped; responses are scattered above and below; this sad post
makes barely any sense removed from the thread itself. Should we pass
around the hat and buy pc the url to a $free email client that doesn't
suck? Or is the problem elsewhere?)

John E Clifford wrote:

>N> This is not clearly an intensional case. Note that there is a perfectly good (well, only lsightly suspect) quantifier here, so it isn't even clear that this is about a relation between this individual plant and sunlight in general.
>
>
>
>Pierre Abbat:
>
>
>>1. What's intensionality? Why do we need it?
>>
>>
>
>N>Consider for example:
>
>le va spati cu ca'o claxu lo solri gusni
>That plant is not getting enough sunlight.
>
>We are not claiming that there is some solri gusni such that
>that gusni does not get to the plant. Even if the plant got plenty
>of solri gusni, there would still be some solri gusni that it doesn't
>get, so that claim would be different. We don't really want to consider
>instances of solri gusni here, just a relationship between le va spati
>and solri gusni in general.
>
>

--
Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."

There are a bunch of different ways to go on this and — insofar as xorxes theory is consistent and regularly adhered to — {lo broda} in affirmative contexts requires that there are broda. Mr Broda exists only in his instances and if there are no instances, there is no Mr. Broda. I have offered xorxes several other ways to do it and he occasionally seems to use bits and pieces of these (which don't go well with his main thrust), but the existential import (and the concreteness) of Mr. Broda has been pretty consitently adhered to.

xod <xod@thestonecutters.net> wrote:
This is probably an official point about which we should hear from
Jorge, and perhaps a specific note about Existence should go in the
proposal unless it's already there.
My understanding is no, lo does not imply existence, and "mi nitcu lo
mikce" is true when no doctors exist.

John E Clifford wrote:

>L> {mi nitcu tu'a lo mikce} Note that dropping the {tu'a] makes a claim that IS falsified if all the doctors are dead
>

--
Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."

John E Clifford wrote:

>D> I suspect the problem is in your receiver, since my version (both sent and received) has all my comments at the top and all names in place.
>
>

Do you see Jorge's name anywhere in this mail?

Your use of tag letters is a rather unorthodox way of responding to
messages, and I though you're doing this to overcome the inability of
your mailer (Yahoo Webmail?) to correctly indicate quotations. I recall
your system adding the ">" only to the first line of the quoted
paragraph, instead of to each of its lines, which is the expected and
standard way. However, when I just tried replying on Yahoo mail, it
behaved in the standard manner.

>
>C>It passes the "non-existent" test only because of the negation inherent in {claxu}, you can always lack (not have) something that does not exist, not because dealing with that takes you into another realm but because the selbri applies even when the sumti predicate does not. I think, incidentally, that English {lack} is also intensional, since it seems to require not only that you don't have it but that you need — or at least want — to have it. This does not seem to be the case with {claxu} however.
>
>

So then you agree that this is a good example of intensionality.

Are you hinting that "moving into another realm" is your test for
intensionality which you (used to) think claxu fails?

>xod <xod@thestonecutters.net> wrote:
>C>I think it's a fine and clever example of intensionality because the
>selbri place in question does not demand an event, which would
>grammatically justify the tu'a legerdemain. And it passes my test for
>intensionality: the statement's truth is independent of the existence of
>lo solri gusni. Do you have a test that it fails?
>
>D>(Meta, Gripe: Look at the hash that's been made of this message. Jorge's
>name is stripped; responses are scattered above and below; this sad post
>makes barely any sense removed from the thread itself. Should we pass
>around the hat and buy pc the url to a $free email client that doesn't
>suck? Or is the problem elsewhere?)
>
>
>John E Clifford wrote:
>
>
>
>>N> This is not clearly an intensional case. Note that there is a perfectly good (well, only lsightly suspect) quantifier here, so it isn't even clear that this is about a relation between this individual plant and sunlight in general.
>>
>>
>>
>>Pierre Abbat:
>>
>>
>>
>>
>>>1. What's intensionality? Why do we need it?
>>>
>>>
>>>
>>>
>>N>Consider for example:
>>
>>le va spati cu ca'o claxu lo solri gusni
>>That plant is not getting enough sunlight.
>>
>>We are not claiming that there is some solri gusni such that
>>that gusni does not get to the plant. Even if the plant got plenty
>>of solri gusni, there would still be some solri gusni that it doesn't
>>get, so that claim would be different. We don't really want to consider
>>instances of solri gusni here, just a relationship between le va spati
>>and solri gusni in general.
>>
>>
>>
>>
>
>
>
>

--
Motorists honked in celebration in this Ramadi as news spread of the
assassination of the president of the Iraqi Governing Council Ezzidin Salim
Monday. "The GC is nothing," one man shouted. "They are not the Governing
Council. They are the Prostitution Council."

xorxes name is at the top of the quotes:
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
Otherwise I agree that Yahoo is rather frustrating in that it does not allow copying only parts of the given text and it does not allow inserting comments into the text in a distinctive way. So, yes, what I do is to compensate for that.

B> No. As I said, {claxu} does not seem to have the intensional feature of "lack." It apparently just means "not have" (as vaguely as that seems). It does not move us to another realm that I can see, and, yes, that is my test for intensionality — not a definition necessarily, but a handy test — hard to apply in negative contexts as you see. MOves to another realm typically involve the use of abstractions in Lojban and the fact that {claxu} does not seems a clearer reason for denying that its 2nd palce is intensional.

xod <xod@thestonecutters.net> wrote:
Do you see Jorge's name anywhere in this mail?

Your use of tag letters is a rather unorthodox way of responding to
messages, and I though you're doing this to overcome the inability of
your mailer (Yahoo Webmail?) to correctly indicate quotations. I recall
your system adding the ">" only to the first line of the quoted
paragraph, instead of to each of its lines, which is the expected and
standard way. However, when I just tried replying on Yahoo mail, it
behaved in the standard manner.

>
>C>It passes the "non-existent" test only because of the negation inherent in {claxu}, you can always lack (not have) something that does not exist, not because dealing with that takes you into another realm but because the selbri applies even when the sumti predicate does not. I think, incidentally, that English {lack} is also intensional, since it seems to require not only that you don't have it but that you need — or at least want — to have it. This does not seem to be the case with {claxu} however.
>
>

B>So then you agree that this is a good example of intensionality.

Are you hinting that "moving into another realm" is your test for
intensionality which you (used to) think claxu fails?

>xod wrote:
>C>I think it's a fine and clever example of intensionality because the
>selbri place in question does not demand an event, which would
>grammatically justify the tu'a legerdemain. And it passes my test for
>intensionality: the statement's truth is independent of the existence of
>lo solri gusni. Do you have a test that it fails?
>

pc:
> As I said, {claxu} does not seem to have the intensional feature of
> "lack." It apparently just means "not have" (as vaguely as that seems).

That's problematic though.

If you mean the "not" to have scope over the whole bridi, then that
means that {mi claxu roda} is always true (it is not the case that
I have everything) and {mi claxu su'oda} is always false
(it is not the case that there is something that I have).
Besides, it seems weird that {mi claxu roda} should mean
"I lack something" and {mi claxu su'oda} "I lack everything".
Also {mi claxu noda} comes out as "I have something".
So I doubt you mean the "not" to have scope over the whole
bridi.

If you mean it to have short scope, i.e. {claxu} is {narponse}
or something close, then {le cmana cu claxu su'o lo cidja}
means "there's some food such that the mountains don't have
that food.

{le cmana cu claxu lo cidja}, without quantifiers, simply
means that the mountains lack food.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

Yes, it is problematic. English "lack" deals with the problem in a typically fuzzy way by using "any" with lack, thereby clouding the choice between "some" and "all" (and reenforcing the claim that "lack" contains a negation. All of Lojban seems to require that the negation be short scope, i.e., just affect the brivla, but the fact that you can lack something that does not exist seems to require either that the place is intensional (even though not involving an abstraction) or that it lies within the scope of the nengation. I suppose that the intensional solution is easier — but it then requires that we have *places,* not merely structures, that are intensional, and the Lojban community has rejected that notion countless times (well, I suppose about 7, taking separate discussions rather than separate rejections as times).
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
pc:
> As I said, {claxu} does not seem to have the intensional feature of
> "lack." It apparently just means "not have" (as vaguely as that seems).

That's problematic though.

If you mean the "not" to have scope over the whole bridi, then that
means that {mi claxu roda} is always true (it is not the case that
I have everything) and {mi claxu su'oda} is always false
(it is not the case that there is something that I have).
Besides, it seems weird that {mi claxu roda} should mean
"I lack something" and {mi claxu su'oda} "I lack everything".
Also {mi claxu noda} comes out as "I have something".
So I doubt you mean the "not" to have scope over the whole
bridi.

If you mean it to have short scope, i.e. {claxu} is {narponse}
or something close, then {le cmana cu claxu su'o lo cidja}
means "there's some food such that the mountains don't have
that food.

{le cmana cu claxu lo cidja}, without quantifiers, simply
means that the mountains lack food.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

On Fri, Jun 04, 2004 at 05:23:39PM -0700, Robin Lee Powell wrote:
> On Fri, Jun 04, 2004 at 08:12:59PM -0400, Pierre Abbat wrote:
> > As I use {lo}, B3 is equivalent to B1. To mean "Rabbits don't go", I'd
> > say {lo'e ractu na klama} or {lo'e ractu naku klama}.
>
> That isn't what "lo'e" means. {lo'e ractu naku klama} means that
> *typical* rabbits don't go (and is trivially false). It makes no
> claims, however, about rabbits in general, only the typical ones.

I ended up having a long side discussion with Robin about this.

Robin's idea of {lo'e} describes existing rabbits that are typical in some way.
My idea of {lo'e} is something that has all the properties that rabbits
typically have (and this imaginary thing can be named "the typical rabbit").
Robin's {lo'e ractu} is my {lo fadni ractu}.

Both definitions can be justified by the book, but I find the {lo fadni ractu}
definition fairly useless.

--
Rob Speer

Rob Speer wrote:

>On Fri, Jun 04, 2004 at 05:23:39PM -0700, Robin Lee Powell wrote:
>
>
>>On Fri, Jun 04, 2004 at 08:12:59PM -0400, Pierre Abbat wrote:
>>
>>
>>>As I use {lo}, B3 is equivalent to B1. To mean "Rabbits don't go", I'd
>>>say {lo'e ractu na klama} or {lo'e ractu naku klama}.
>>>
>>>
>>That isn't what "lo'e" means. {lo'e ractu naku klama} means that
>>*typical* rabbits don't go (and is trivially false). It makes no
>>claims, however, about rabbits in general, only the typical ones.
>>
>>
>
>I ended up having a long side discussion with Robin about this.
>
>Robin's idea of {lo'e} describes existing rabbits that are typical in some way.
>My idea of {lo'e} is something that has all the properties that rabbits
>typically have (and this imaginary thing can be named "the typical rabbit").
>Robin's {lo'e ractu} is my {lo fadni ractu}.
>
>

When you see a typical rabbit, is it a typical rabbit? Or only just a
rabbit that closely resembles the typical rabbit? What if the rabbit
showed NO differences from the imaginary typical rabbit? If it is
identical to the typical rabbit, can't we call it The Typical Rabbit?

>Both definitions can be justified by the book, but I find the {lo fadni ractu}
>definition fairly useless.
>
>
>

--
Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."

xod scripsit:

> When you see a typical rabbit, is it a typical rabbit? Or only just a
> rabbit that closely resembles the typical rabbit? What if the rabbit
> showed NO differences from the imaginary typical rabbit? If it is
> identical to the typical rabbit, can't we call it The Typical Rabbit?

No, it's not identical, just very similar. *A* typical rabbit is one
that resembles arbitrarily closely *the* typical rabbit. But all real
rabbits must be male or female, whereas the typical rabbit is neither,
so the resemblance can't ever amount to identity. The typical rabbit
lives in a warren, but not any particular warren, just the typical rabbit
warren.

--
Dream projects long deferred John Cowan <jcowan@reutershealth.com>
usually bite the wax tadpole. http://www.ccil.org/~cowan
--James Lileks http://www.reutershealth.com

I suppose that depends upon how you translate "the typical rabbit" (or {lo'w ractu}) into semantics. If you think that it designates some particular object, then I suppose that, if you found a rabbit which had exactly the properties of the typical rabbit you would have found the particular raabbit. Of course, it is unlikely that any rabbit does this, since as has been noted here, the typical rabbit has inconsistent properties in some areas (if you take all references to the typical rabbit as referring to the same thing) and none of the choices in others (sex being the obvious one here). On the other hand, if you take talk about the typical rabbit as being like talk about the average man — simple ways to sum up complex information about a class of things — then the issue does not arise: you have pierre's fadni ractu, but not lo'e ractu.
xod <xod@thestonecutters.net> wrote:
When you see a typical rabbit, is it a typical rabbit? Or only just a
rabbit that closely resembles the typical rabbit? What if the rabbit
showed NO differences from the imaginary typical rabbit? If it is
identical to the typical rabbit, can't we call it The Typical Rabbit?

I agree with John here. And it should probably be spelled out in the
proposal, that {lo'e} refers to an abstract archetype, etc (or that it
doesn't if we decide to go that route.) We're trying to prevent future
generations from having to ask these kinds of questions; we'd best
answer them *officially* here and now.

~mark

jcowan@reutershealth.com wrote:

>xod scripsit:
>
>
>
>>When you see a typical rabbit, is it a typical rabbit? Or only just a
>>rabbit that closely resembles the typical rabbit? What if the rabbit
>>showed NO differences from the imaginary typical rabbit? If it is
>>identical to the typical rabbit, can't we call it The Typical Rabbit?
>>
>>
>
>No, it's not identical, just very similar. *A* typical rabbit is one
>that resembles arbitrarily closely *the* typical rabbit. But all real
>rabbits must be male or female, whereas the typical rabbit is neither,
>so the resemblance can't ever amount to identity. The typical rabbit
>lives in a warren, but not any particular warren, just the typical rabbit
>warren.
>
>
>

jcowan@reutershealth.com wrote:

>xod scripsit:
>
>
>
>>When you see a typical rabbit, is it a typical rabbit? Or only just a
>>rabbit that closely resembles the typical rabbit? What if the rabbit
>>showed NO differences from the imaginary typical rabbit? If it is
>>identical to the typical rabbit, can't we call it The Typical Rabbit?
>>
>>
>
>No, it's not identical, just very similar. *A* typical rabbit is one
>that resembles arbitrarily closely *the* typical rabbit. But all real
>rabbits must be male or female, whereas the typical rabbit is neither,
>so the resemblance can't ever amount to identity. The typical rabbit
>lives in a warren, but not any particular warren, just the typical rabbit
>warren.
>
>
>

If we notice differences between this rabbit and the typical rabbit,
then we can't call it a typical rabbit. But if we don't notice or don't
care about the differences, or the particular warren he inhabits, then
why can't we call him The Typical Rabbit? Anything else would be a
"distinction without a difference".

--
Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."

~mark:
> I agree with John here. And it should probably be spelled out in the
> proposal, that {lo'e} refers to an abstract archetype, etc (or that it
> doesn't if we decide to go that route.) We're trying to prevent future
> generations from having to ask these kinds of questions; we'd best
> answer them *officially* here and now.

This is what I wrote:

"The resulting expression indicates that the individuals or groups
that satisfy the selbri typically also satisfy the predicate for
which the sumti is an argument."

So {lo'e broda cu brode} is true only when brodas typically brode.

Something like {mi viska lo'e broda} would be true only if it is
typical of brodas to be seen by mi.

The way I wrote it, lo'e does not necessarily refer to an abstract
archetype but rather it hides a complex claim about brodas in
general. I understand that's what John's definition amounts to too.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

On Mon, Jun 07, 2004 at 10:37:29AM -0700, Jorge Llamb?as wrote:
snip
> "The resulting expression indicates that the individuals or groups
> that satisfy the selbri typically also satisfy the predicate for which
> the sumti is an argument."
>
snip
>
> The way I wrote it, lo'e does not necessarily refer to an abstract
> archetype but rather it hides a complex claim about brodas in general.
> I understand that's what John's definition amounts to too.

I agree, and I think xorxes' definition is both wonderful and
wonderfully subtle.

-Robni

I think that the "statistical" model is better than the archetype (which might work for "stereotypical," which floats free of what the members of the class actually do).
"Mark E. Shoulson" <mark@kli.org> wrote:I agree with John here. And it should probably be spelled out in the
proposal, that {lo'e} refers to an abstract archetype, etc (or that it
doesn't if we decide to go that route.) We're trying to prevent future
generations from having to ask these kinds of questions; we'd best
answer them *officially* here and now.

~mark

Because "the typical rabbit," when referring to a particular rabbit, has a different logic from the typical "the typical..."
xod <xod@thestonecutters.net> wrote:

If we notice differences between this rabbit and the typical rabbit,
then we can't call it a typical rabbit. But if we don't notice or don't
care about the differences, or the particular warren he inhabits, then
why can't we call him The Typical Rabbit? Anything else would be a
"distinction without a difference".

Jorge Llamb?as scripsit:

> The way I wrote it, lo'e does not necessarily refer to an abstract
> archetype but rather it hides a complex claim about brodas in
> general. I understand that's what John's definition amounts to too.

On your view, then "lo'e broda" has only contextual meaning; it does not
refer, but bridi containing it are still interpretable.

I think this is an excellent formulation, because it avoids crap about
whether the typical rabbit has two eyes, or only two typical eyes.

--
John Cowan <jcowan@reutershealth.com> http://www.ccil.org/~cowan
"One time I called in to the central system and started working on a big
thick 'sed' and 'awk' heavy duty data bashing script. One of the geologists
came by, looked over my shoulder and said 'Oh, that happens to me too.
Try hanging up and phoning in again.'" --Beverly Erlebacher

Robin:
<<Assuming for the moment that {zo'e} cannot be quantified over (and I
am, indeed, assuming that, or past usage gets Really Really Wierd), this
is potentially a drastic change to past usage, in as much as it is easy
to construct sentences that change drastically in meaning.
For example, in CLL {lo broda cu na brode} means that there are *NO*
brodas, anywhere, that brode. None at all. This is a consequence of
what "da" and "na" mean.

In xorlo, the meaning of {lo broda cu na brode} is, of course,
determined by context. There is nothing requiring it to mean the same thing as
the CLL version, since xorlo is not quantified (i.e. does not imply
"da").>>

{zo'e} in a verso context (inside the scope of an odd number of {na(ku/i)}) is to be replaced by {ro da} rather than {da}. Alternatively, it is to be replaced (at each occurrence) by a (different) {da} repeating one prenexed to the entire sentence (that is outsied the scope of everything possible). xorxes does not agree with this.

On Tue, Jun 15, 2004 at 08:56:37AM -0700, John E Clifford wrote:
> {zo'e} in a verso context (inside the scope of an odd number of
> {na(ku/i)}) is to be replaced by {ro da} rather than {da}.
> Alternatively, it is to be replaced (at each occurrence) by a
> (different) {da} repeating one prenexed to the entire sentence (that
> is outsied the scope of everything possible). xorxes does not agree
> with this.

I see *no* evidence of an equivalence between {zo'e} and anything
involving {da} in any Lojbanic teaching texts (all two of them).

CLL:

Only the presence of an explicit x1 (other than ``zo'e'', which is
equivalent to omission) can force the goer and the walker to be
identical.

Furthermore, such an equivalence would absolutely destroy past usage:

mi na klama

Currently this means "I don't go <somewhere>, <from somewhere>, etc".

If all the empty places are "da", it means:

The does not exist a place I have gone to, nor a place I have come
from, nor a route I have used for going, nor a means I have used for
going.

This is preposterous beyond all bounds, and I'm stunned that even you
would suggest it.

-Robin

I've put a rough draft of a paper about species on the wiki under the heading Species. Comments appreciated. Imitation for other proposals encouraged.

xorxes:

lo broda = zo'e noi ke'a broda

Given this, {lo broda cu brode} would be the claim that some unspecified thing

is brode and, incidentally broda. This claim is false, of course, when the thing

is not a broda as much as when it is not a brode. What we will want is that

the thing, however unspecified otherwise, is clearly marked as a broda; then

only its non-brodehood will make the claim false. Actually, as later moves

make somewhat clear, {zo’e} probably does not stand for a concrete individual,

but rather for a group of such individuals or an abstraction – both individuals

within Lojban norms. So, at least in the case of distributive groups, it is not the

group, but each member of the group, that is broda (or the group is, but that

just means that every member is). From earlier discussions, I imagine that xorxes

is using {zo’e} as a logical proper name, something transparent to negation

and quantifier order. That will fail if it is a distributive group, since either the

negation allows that some members have and other lack the property in question

– a situation that can no longer be summarized as a distributive group – or it

means that all the members lack the property – in which case the intermediate

position (some do, some don’t) cannot be expressed. But this is speculation

on my part.

xorxes:

PA lo broda = PA mupli be lo broda

PA le broda = PA cmima be le brode

PA la broda = PA cmima be la broda

Assuming that {zo’e} is totally unspecified but is an individual within Lojban norms (a real individual, a collective, a set, a distributive group or an abstraction — property, proposition, event, …), the definition of {PA lo broda}, using {mupli}, forces the abstraction-property reading, especially in contrast with the set-group-collective reading for {le} and {la} at this point. {PA lo broda} is PA examples (instances?, specimens? loci?) of the property zo’e, which is broda. Not, note, the property of being a broda, but some – unspecified – property that is a broda. But for most properties, properties don’t have them. And when they do, their having the property is not what saying {PA lo broda cu brode} intends, namely that PA things which are brodas are brode (i.e., not PA things which have a property, which incidentally – not even definingly — has the property of being broda, are brode). But we have not gotten to the longer context yet, so suffice it is to say that {PA lo
broda} refers in some way (probably distributively, I suppose) to PA things which have a property, which property happens to have brodaness.

(of course, PA le broda are things that are in some set which I happen to describe as having brodaness).

Xorxes:

lo PA broda = lo PAmei be fi lo broda

le PA broda = le PAmei be fi lo broda

So, {lo PA broda} is a PAad whose members are lo broda, some unspecified thing which is, incidentally, a broda. In this case, the unspecified thing seems to have to be a distributive group, since it has to come up with PA things to be members of the PAad. I see why I thought that xorxes had a species in mind (even if not so named), for he needs {zo’e} to stand for something that is sometimes a group (or even a set) and sometimes for a property. But it turns out:

xorxes:

PA1 lo PA2 broda = PA1 mupli be lo PA2 broda

PA1 le PA2 broda = PA1 cmima be le PA2 broda

PA1 la PA2 broda = PA1 cmima be la PA2 broda

that what {lo PA broda} creates is not a PAad but a property that has, incidentally, the property of being a broda PAad. I suppose that this – and the previous – problem could be met by modifying the definition of {mupli} (and there may be other reasons for doing this) to include samples drawn from sets and even groups (apparently one of the current – last time used – meanings of {me}). But then we would be back essentially to {cmima} and that we do not want.
These questions (reducible to “What the Hell does it mean?”) afflict all the subsequent definitions of gadri phrases in isolation. Whatever the problems with the definitions of {lo’e} and {le’e} and of the {lVi}’s — and they seem to be numerous – they at least take the terms in context and explain the whole, rather than trying to explain each part, with the danger that that part will have to mean something else in some other place, as happens regularly with these definitions of the basic gadri.

pc:
> Actually, as later moves
> make somewhat clear, {zo’e} probably does not stand for a concrete
> individual,
> but rather for a group of such individuals or an abstraction – both
> individuals
> within Lojban norms.

You cannot pinpoint what {zo'e} stands for without first giving a
context. In the absence of context, any of your choices are
valid possibilities.

> [From earlier discussions, I imagine that xorxes
> is using {zo’e} as a logical proper name, something transparent to negation
> and quantifier order.

Yes, I hold that {zo'e naku zo'e broda} is equivalent to
{naku zo'e zo'e broda} and to {zo'e zo'e naku broda}.
Also, {roda zo'e broda} is equivalent to {zo'e roda se broda}.
I don't see how it could be otherwise, because when {zo'e} is
elided we have no way of telling in which order it was meant
to fall with respect to other terms or naku.

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail

But all this gets you into difficulties later on: {zo'e} takes on a variety of characters and for some of them, at least, the transparency does not work. You are quite right to insist upon context, but by defining things without a context, you give contradictory claims. Now, of course, contradiction does have the advantage that you can prove anything from it, but it is not very helpful for understanding. The contextless definitions is not the only problem you have, of course, but it tends to be what leads to the others.
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:pc:
> Actually, as later moves
> make somewhat clear, {zo’e} probably does not stand for a concrete
> individual,
> but rather for a group of such individuals or an abstraction – both
> individuals
> within Lojban norms.

You cannot pinpoint what {zo'e} stands for without first giving a
context. In the absence of context, any of your choices are
valid possibilities.

> [From earlier discussions, I imagine that xorxes
> is using {zo’e} as a logical proper name, something transparent to negation
> and quantifier order.

Yes, I hold that {zo'e naku zo'e broda} is equivalent to
{naku zo'e zo'e broda} and to {zo'e zo'e naku broda}.
Also, {roda zo'e broda} is equivalent to {zo'e roda se broda}.
I don't see how it could be otherwise, because when {zo'e} is
elided we have no way of telling in which order it was meant
to fall with respect to other terms or naku.

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail

pc:
> But all this gets you into difficulties later on: {zo'e} takes on a variety
> of characters and for some of them, at least, the transparency does not work.

Could you give an example of zo'e where transparency does not work?

> You are quite right to insist upon context, but by defining things without a
> context, you give contradictory claims.

The context dependency of {lo broda} is akin to that of {zo'e},
that's all that the formal definition says. You can replace {lo broda}
with {zo'e noi ke'a broda} in any context. I don't see what
contradictory claims I give.

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail is new and improved - Check it out!
http://promotions.yahoo.com/new_mail

A> If zo'e is a distributive group, then {zo'e broda} presumably means that each of the members broda. {zo'e na broda} denies this, so either it means that some of the members do not broda — which is not expressible using {zo'e} any more (in particular is not {zo'e naku broda} — or it means all the member lack broda — in which case 1) {na} is something stronger than simple negation and 2) the case of real negation (some do, some don't) is not expressible. In either case, the move from {zo'e na broda} to {zo'e naku broda} in the usal interpretation fails.

B> lo PA broda = lo PAmei be fi lo broda. So, to fit into {PAmei} {lo broda} stand for a property. Since it expands into {zo'e noi broda}, {zo'e} must be a proeprty and, further, a property that is a broda — something that properties generally are not. But what is intended is that you get something other than properties (usually — and in any case — things that have the defining property not that have one of its properties.) You come in a level too high.

Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
pc:
> But all this gets you into difficulties later on: {zo'e} takes on a variety
> of characters and for some of them, at least, the transparency does not work.
A>Could you give an example of zo'e where transparency does not work?

> You are quite right to insist upon context, but by defining things without a
> context, you give contradictory claims.

B>The context dependency of {lo broda} is akin to that of {zo'e},
that's all that the formal definition says. You can replace {lo broda}
with {zo'e noi ke'a broda} in any context. I don't see what
contradictory claims I give.

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail is new and improved - Check it out!
http://promotions.yahoo.com/new_mail

pc:
> If zo'e is a distributive group, then {zo'e broda} presumably means that
> each of the members broda. {zo'e na broda} denies this, so either it means
> that some of the members do not broda — which is not expressible using
> {zo'e} any more (in particular is not {zo'e naku broda} — or it means all
> the member lack broda — in which case 1) {na} is something stronger than
> simple negation and 2) the case of real negation (some do, some don't) is not
> expressible. In either case, the move from {zo'e na broda} to {zo'e naku
> broda} in the usal interpretation fails.

{zo'e na broda} = {zo'e naku broda} = {naku zo'e broda}
= {na broda} = {naku broda}, because by definition {zo'e}
is elidable in that case.

I don't know how you can make any interpretation of zo'e without
a context. How do you get it ot be a distributive group?

> lo PA broda = lo PAmei be fi lo broda. So, to fit into {PAmei} {lo broda}
> stand for a property.

x3 of PAmei is for the members of the group.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

A> There seems to be two unrelated matters here 1) {zo'e} is lpn and so by definition transparent to negation. There is precious little evidence that this is true; nothing else in Lojban is, with the possible exception of bound variables inside the scope of their binding quantifiers. And you have denied that {zo'e} is one of those. 2) {zo'e} is elidable. True but that just means that we always know where to put it if it isn't there, by some syntactic rule. In the case of {naku broda}, we don't know where to put it, before or after the {naku}. There may be no semantic difference between the two, but that is irrelevant to elidability, which is syntactic. By the way, that there is no differnce between the two is what you are out to prove, not a given you can use in the proof.

B>See C

C> Oops, wrong example. I meant to use {PA lo broda} which require {ze'o} to be a property which brodas. {lo pa broda} requires {ze'o} to be a distributive set, with the consequences noted in the first quote.
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
pc:
> If zo'e is a distributive group, then {zo'e broda} presumably means that
> each of the members broda. {zo'e na broda} denies this, so either it means
> that some of the members do not broda — which is not expressible using
> {zo'e} any more (in particular is not {zo'e naku broda} — or it means all
> the member lack broda — in which case 1) {na} is something stronger than
> simple negation and 2) the case of real negation (some do, some don't) is not
> expressible. In either case, the move from {zo'e na broda} to {zo'e naku
> broda} in the usal interpretation fails.

{zo'e na broda} = {zo'e naku broda} = {naku zo'e broda}
= {na broda} = {naku broda}, because by definition {zo'e}
is elidable in that case.

B>I don't know how you can make any interpretation of zo'e without
a context. How do you get it ot be a distributive group?

> lo PA broda = lo PAmei be fi lo broda. So, to fit into {PAmei} {lo broda}
> stand for a property.

x3 of PAmei is for the members of the group.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

pc:
> There seems to be two unrelated matters here 1) {zo'e} is lpn and so by
> definition transparent to negation. There is precious little evidence that
> this is true; nothing else in Lojban is, with the possible exception of bound
> variables inside the scope of their binding quantifiers. And you have denied
> that {zo'e} is one of those.

What's lpn?

The evidence is that you can use {zo'e naku broda}, {naku zo'e broda}
and {naku broda} to mean the same thing.

I don't see a difference between {la djan naku klama le zarci}
and {le zarci naku se klama la djan} either. Both are ~K(d,z).

> I meant to use {PA lo broda} which require {ze'o} to
> be a property which brodas. {lo pa broda} requires {ze'o} to be a
> distributive set, with the consequences noted in the first quote.

Think of it as the locus/focus of the property.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - Send 10MB messages!
http://promotions.yahoo.com/new_mail

A> logical proper name
B> And the evidence for that claim is? If I see {naku broda} I believe it has to mean {zo'e naku brode}, since that is the place where the {zo'e} goes syntactically Or, if it is not that, if I have the rule backward, then it has to be {naku zo'e broda}. "Elidable" means we always know where to put it in.

C> Yes, that seems to work, but is not the hard case of negation transfer, i.e., {naku la djan cu klama le zarci} to {la djan naku clama le zarci}, which goes from true to false at least when there is no John (and in a few other cases less certainly — John is a mountain, say).

D> But that is not what it means — according to you. (I think that is always what {lo broda} means in transparent contexts. Well, not a locus but a distributive group of loci. Its meaning does not hop around and so most of your expansions are wrong — according to me,) It has to be (because of {mupli3} a property (not the locus of one) and according to its earlier definition something that is — incidentally --a broda.

Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
pc:
> There seems to be two unrelated matters here 1) {zo'e} is lpn and so by
> definition transparent to negation. There is precious little evidence that
> this is true; nothing else in Lojban is, with the possible exception of bound
> variables inside the scope of their binding quantifiers. And you have denied
> that {zo'e} is one of those.

A>What's lpn?

B>The evidence is that you can use {zo'e naku broda}, {naku zo'e broda}
and {naku broda} to mean the same thing.

C>I don't see a difference between {la djan naku klama le zarci}
and {le zarci naku se klama la djan} either. Both are ~K(d,z).

> I meant to use {PA lo broda} which require {ze'o} to
> be a property which brodas. {lo pa broda} requires {ze'o} to be a
> distributive set, with the consequences noted in the first quote.

D>Think of it as the locus/focus of the property.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - Send 10MB messages!
http://promotions.yahoo.com/new_mail

pc:
> If I see {naku broda} I believe it
> has to mean {zo'e naku brode}, since that is the place where the {zo'e} goes
> syntactically Or, if it is not that, if I have the rule backward, then it has
> to be {naku zo'e broda}. "Elidable" means we always know where to put it in.

I don't recall ever seeing such a rule. In my case, I don't need a
rule because I get the same meaning no matter where zo'e goes.

> Yes, that seems to work, but is not the hard case of negation transfer,
> i.e., {naku la djan cu klama le zarci} to {la djan naku clama le zarci},

It's the same case as far as I can tell. {la djan} appears one
side of {naku} one time and the other side the other time.

> which goes from true to false at least when there is no John

How can there be no John if {la djan} is {zo'e noi zo djan ke'a
mi cmene}? I'm talking about that which I call John, so there has
to be that which I call John. (Whether it is a real, imaginary,
abstract, whatever thing is another matter, but there has to be
that which I call John if I'm saying something meaningful.)

>(and in a few
> other cases less certainly — John is a mountain, say).

If John is a mountain, I don't see how it can go to the market,
whether in front or after naku.

> It has to be (because of
> {mupli3} a property (not the locus of one) and according to its earlier
> definition something that is — incidentally --a broda.

If you don't like my (ab)use of {mupli}, change to {PA mupli be lo ka ce'u
broda}, or {PA xxxxx be lo broda} where xxxxx stands for "x1 is an
instance of x2".

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail

On Thu, Jun 24, 2004 at 09:20:21AM -0700, John E Clifford wrote:
> xorxes:
>
> lo broda = zo'e noi ke'a broda

Though this spawned a large thread about the applicability of {zo'e}, I find
the {noi} to be the part that can't be right.

Why not use poi and voi, like it's explained in many places, including, I
assume, in the book?

I believe you've answered this at one point, but the answer was pretty
confusing.

--
Rob Speer

A> I don't recall seeing the rule either; there just has to be one. As for getting the same meaning 1) you have not shown that yet and 2) it is irrelevant because "elidable" is a syntactic, not a semantic notion.

B> This is not the place to argue about what variously placed negations mean. If {la djan} is one time inside the scope of {naku} and one time out, then it is essentially the same. And the transformation is not an equivalence.

C> But you are the one that says that is what it is and are trying to defend that claim against objections. I assume that {la crlakolmz} is — by you — {zo'e noi lu crlakolmz li'u cmene ke'a mi}, but that doesn't mean that Sherlock Holmes or that particular zo'e exists. You can go over to the outer domain if you want, but Lojban has fairly regulary nixed that idea when it is presented. Or you could say that sentences involving the names of non-existents are always false or always true or never have a truth value. None of these things make the referent of the expression exist (in this world now). None of these suggestions has clearly been adopted by the Lojban community, but the suggestion that atomic sentences with the names of non-existent things are always false has a considerable following among logicians (followed by no truth value, with true very far behind — making all names universals, usually: a case where Aristotle or his followers was less than brilliant). So, I am
using that one. If you want the True-version, then it is the inference the pother way that doesn't work. On the Gap-version, they seem to be equally truth-valueless but consequently equally uninformative (the pair seems to amount to the claim that the thing named does not exist). This is not quite the same as being meaningless, more like being pointless.

D> Just the point. If John is a mountain, then {naku la djan ...} is true since John doesn't have the appropriate property, but {la djan naku ...} is false, because John the Mountain also doesn't have the propeprty of not going to the store; it's not the sort of thing that comes and goes (along a route). As I say, this is not a clear cut case like the non-existent one, but has a lot going for it (it is back to the issue of whether -P is the contradictory or the contrary opposite of P).

E> This fixes that one rather nicely, so I accept your correction. It is not me that doesn't like your definitions, it is that the definitions don't define what they are supposed (and are distinctly odd in some cases).

Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
pc:
> If I see {naku broda} I believe it
> has to mean {zo'e naku brode}, since that is the place where the {zo'e} goes
> syntactically Or, if it is not that, if I have the rule backward, then it has
> to be {naku zo'e broda}. "Elidable" means we always know where to put it in.

A>I don't recall ever seeing such a rule. In my case, I don't need a
rule because I get the same meaning no matter where zo'e goes.

> Yes, that seems to work, but is not the hard case of negation transfer,
> i.e., {naku la djan cu klama le zarci} to {la djan naku klama le zarci},

B>It's the same case as far as I can tell. {la djan} appears one
side of {naku} one time and the other side the other time.

> which goes from true to false at least when there is no John

C>How can there be no John if {la djan} is {zo'e noi zo djan ke'a
mi cmene}? I'm talking about that which I call John, so there has
to be that which I call John. (Whether it is a real, imaginary,
abstract, whatever thing is another matter, but there has to be
that which I call John if I'm saying something meaningful.)

>(and in a few
> other cases less certainly — John is a mountain, say).

D>If John is a mountain, I don't see how it can go to the market,
whether in front or after naku.

> It has to be (because of
> {mupli3} a property (not the locus of one) and according to its earlier
> definition something that is — incidentally --a broda.

E>If you don't like my (ab)use of {mupli}, change to {PA mupli be lo ka ce'u
broda}, or {PA xxxxx be lo broda} where xxxxx stands for "x1 is an
instance of x2".

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail

Well, I agree that it should be {poi} — or where applicable, {voi}. xorxes defense is to claim that {poi} and {noi} have logics which there is no reason that he has presented to think they have.
Rob Speer <rspeer@MIT.EDU> wrote:On Thu, Jun 24, 2004 at 09:20:21AM -0700, John E Clifford wrote:
> xorxes:
>
> lo broda = zo'e noi ke'a broda

Though this spawned a large thread about the applicability of {zo'e}, I find
the {noi} to be the part that can't be right.

Why not use poi and voi, like it's explained in many places, including, I
assume, in the book?

I believe you've answered this at one point, but the answer was pretty
confusing.

--
Rob Speer

Rob:
> Why not use poi and voi, like it's explained in many places, including, I
> assume, in the book?

Because poi (at least, I'm not sure about voi) is restrictive.
{zo'e poi ke'a broda} says that you restrict the values of zo'e
to those that fit the x1 of broda, but that's not what {lo broda}
does. In {lo broda} we don't start with a larger meaning that
gets restricted to brodas, we only talk about brodas.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

pc:
> I assume that {la crlakolmz} is — by you --
> {zo'e noi lu crlakolmz li'u cmene ke'a mi}, but that doesn't mean that
> Sherlock Holmes or that particular zo'e exists.

"Exists" as in is a real person in the real world, or "exists" as in
it can be the value of a variable? If the former, then he doesn't
exist, but which predicates it satisfies is not really relevant here.
If the latter, then yes, he exists. Otherwise, what would you be
talking about?

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail

On Thu, Jun 24, 2004 at 02:51:12PM -0700, Jorge Llambías wrote:
> pc:
> > If I see {naku broda} I believe it
> > has to mean {zo'e naku brode}, since that is the place where the {zo'e} goes
> > syntactically Or, if it is not that, if I have the rule backward, then it has
> > to be {naku zo'e broda}. "Elidable" means we always know where to put it in.
>
> I don't recall ever seeing such a rule. In my case, I don't need a
> rule because I get the same meaning no matter where zo'e goes.
What's he's saying makes sense, though, Xorxes. If you take examples
where people don't elide the zo'e, it's clear that moving naku
through it changes the meaning of the sentence.

So for an example:
A- mi kalte lo pavyseljirna
B- mi jinvi ledu'u zo'e na zasti

Now, if the last sentence is interpreted with zo'e == lo pavyseljirna,
moving naku very much changes the meaning.

For example, B might've replied:

B'- mi jinvi ledu'u zo'e naku zasti

which has a different meaning (and would of course make no sense).

Perhaps a better example:
A- mi xebni ro bradi
B- zo'e na xebni do
B'- zo'e naku xebni do

The first means "not all of them hate you", the second means "all
of them don't hate you".

--
Jordan DeLong
fracture@allusion.net

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---
Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFA26kUDrrilS51AZ8RAh7XAJ9IDM28rqUJ6OTyhzPCAx9kfgGNJQCeMNHK
9GynO4QMd1Msd3MpgHQsBGs=
=JyHq
---END PGP SIGNATURE---

Huh? But that is exactly what {poi} does: it means we only talk about brodas. And, of course, {zo'e noi broda} starts with "a larger meaning," the totally unspecified {zo'e} and then rather than restricting us to broda, keeps the unrestricted base and adds brodaness in incidentally.
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:Rob:
> Why not use poi and voi, like it's explained in many places, including, I
> assume, in the book?

Because poi (at least, I'm not sure about voi) is restrictive.
{zo'e poi ke'a broda} says that you restrict the values of zo'e
to those that fit the x1 of broda, but that's not what {lo broda}
does. In {lo broda} we don't start with a larger meaning that
gets restricted to brodas, we only talk about brodas.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

Jordan:
> If you take examples
> where people don't elide the zo'e, it's clear that moving naku
> through it changes the meaning of the sentence.
>
> So for an example:
> A- mi kalte lo pavyseljirna
> B- mi jinvi ledu'u zo'e na zasti
>
> Now, if the last sentence is interpreted with zo'e == lo pavyseljirna,
> moving naku very much changes the meaning.

I don't think so. {mi jinvi lo du'u zo'e na zasti}, {mi jinvi
lo du'u na zasti}, {mi jinvi lo du'u naku zasti},
{mi jinvi lo du'u zo'e naku zasti}, {mi jinvi lo du'u zo'e zo'e zo'e
naku zasti} all mean the same to me. No rule tells you how to fill
the blanks. If it is not clear from context, you need to ask the
speaker to be more precise, you cannot get from one form anything
that you cannot get from another.

> For example, B might've replied:
>
> B'- mi jinvi ledu'u zo'e naku zasti
>
> which has a different meaning (and would of course make no sense).

I can make as much sense of it as of the naku zo'e version. I don't
know how you can talk about unicorns if you can't refer to them.

> Perhaps a better example:
> A- mi xebni ro bradi
> B- zo'e na xebni do
> B'- zo'e naku xebni do
>
> The first means "not all of them hate you", the second means "all
> of them don't hate you".

How can you tell? Why can't the first one be understood as
"not one of them hates you"? Why can't the second one be "some of
them don't hate you"? zo'e does not repeat a previous sumti.
Neither {na xebni do} nor {naku xebni do} are very clear with
the context you provide, and I don't get different meanings from
them.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

pc:
> Huh? But that is exactly what {poi} does: it means we only talk about brodas.
> And, of course, {zo'e noi broda} starts with "a larger meaning," the totally
> unspecified {zo'e} and then rather than restricting us to broda, keeps the
> unrestricted base and adds brodaness in incidentally.

Would you say the claim with {zo'e poi} is materially equivalent to
the one with {zo'e noi}?

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail is new and improved - Check it out!
http://promotions.yahoo.com/new_mail

Well, I am susceptible to this muddle myself, so I can't complain too much. There are, however, two types of responses standard in logic. 1. So called names are actually disguised descriptions, which are then expanded into longer expressions in which no description occurs, only quantifiers: Fthe xGx = Some x both fx and all y as Gy so y=x ("as...so..." is {go....gi...}). This presents no problems for denotationless names, since the name disappears before evaluation. All atomic sentences with denotionless names turn out to be false (there is some possibility of futzing with what constitutes an atomic sentence, e.g, the scope of negation and modals — the latter the de dicto-de re distinction). 2). Flat out: atomic sentences (with the same futzability) containing denotationless names have no truth value and this is inherited upward to complex sentences (with the several sorts of exceptions, mainly for sentences which recieve the same value in all worlds in which the names do
name, i.e., tautologies). The third possibility is to allow naming to cover the outer domain, reserving the inner for what exists here and now (typically, the extension of {zasti}). The issue then is about the extensions of predicates: do we interpret them over the inner and outer domains or only over the inner (and add "and exists" when we are interested in the latter matter)? By and large, logicians have tended to use only the inner domain for predicates and quantifiers, which, again, makes atomic sentences with denotationless names false, even though there be (but not "are" or, at least not "exist") refernts for them. Option 2 is the only one with signicant support that does not make external negations of denotationless naming sentences true and ones with internal negation false. We can adopt this option — making both truth valueless (but not thereby equivalent even then) but the move requires some explicit notifcation before being taken for granted.
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
pc:
> I assume that {la crlakolmz} is — by you --
> {zo'e noi lu crlakolmz li'u cmene ke'a mi}, but that doesn't mean that
> Sherlock Holmes or that particular zo'e exists.

"Exists" as in is a real person in the real world, or "exists" as in
it can be the value of a variable? If the former, then he doesn't
exist, but which predicates it satisfies is not really relevant here.
If the latter, then yes, he exists. Otherwise, what would you be
talking about?

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail

On Fri, Jun 25, 2004 at 06:03:01AM -0700, Jorge Llambías wrote:
> Jordan:
...
> > Perhaps a better example:
> > A- mi xebni ro bradi
> > B- zo'e na xebni do
> > B'- zo'e naku xebni do
> >
> > The first means "not all of them hate you", the second means "all
> > of them don't hate you".
>
> How can you tell? Why can't the first one be understood as
> "not one of them hates you"? Why can't the second one be "some of
> them don't hate you"? zo'e does not repeat a previous sumti.
> Neither {na xebni do} nor {naku xebni do} are very clear with
> the context you provide, and I don't get different meanings from
> them.
We're assuming that zo'e in this case is the referent of {le se
go'i}. {zo'e} is probably most frequently explicitly specified as
a super-generic sumti anaphora---someone's going to hear this, and
think "what's zo'e-oh something for which xebni makes sense-oh
probably ro bradi, since that's what we're talking about".

In realistic communication, B and B' would likely be interpreted
with a different meaning---but with the same binding for zo'e. Just
because zo'e *could* be bound to different things in each of those
two sentences doesn't mean that moving naku through zo'e has no
effect. For a given binding of the zo'e, moving naku can very much
change the meaning of the sentence.

--
Jordan DeLong
fracture@allusion.net

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---
Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFA3CkCDrrilS51AZ8RAixGAJ4lCnjAuvYACwRUHgx/b1Ir/rIVXgCgikMY
/zy5FqjZjGeW2WdKDt+RlwA=
=zjJc
---END PGP SIGNATURE---

A>Although this is dangerous to say about the star Lojban translator, xorxes seems not to understand an admittedly rather picky point of grammar and semantics.

B> All the contexts in this case are opaque, so we are here talking ablut unicornity, which exists whether or not unicorns do. Note that the references to what I can talk about also are opaque, i.e. about the property, not the loci.

C> See A. {zo'e} in this context means something like "the obvious sumti," which in this context pretty much has to be {ro bradi}, the only sumti available other than {mi}, which being unlikely and shorter than {zo'e} would be sued if intended.
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
Jordan:
> If you take examples
> where people don't elide the zo'e, it's clear that moving naku
> through it changes the meaning of the sentence.
>
> So for an example:
> A- mi kalte lo pavyseljirna
> B- mi jinvi ledu'u zo'e na zasti
>
> Now, if the last sentence is interpreted with zo'e == lo pavyseljirna,
> moving naku very much changes the meaning.

A>I don't think so. {mi jinvi lo du'u zo'e na zasti}, {mi jinvi
lo du'u na zasti}, {mi jinvi lo du'u naku zasti},
{mi jinvi lo du'u zo'e naku zasti}, {mi jinvi lo du'u zo'e zo'e zo'e
naku zasti} all mean the same to me. No rule tells you how to fill
the blanks. If it is not clear from context, you need to ask the
speaker to be more precise, you cannot get from one form anything
that you cannot get from another.

> For example, B might've replied:
>
> B'- mi jinvi ledu'u zo'e naku zasti
>
> which has a different meaning (and would of course make no sense).

B>I can make as much sense of it as of the naku zo'e version. I don't
know how you can talk about unicorns if you can't refer to them.

> Perhaps a better example:
> A- mi xebni ro bradi
> B- zo'e na xebni do
> B'- zo'e naku xebni do
>
> The first means "not all of them hate you", the second means "all
> of them don't hate you".

C>How can you tell? Why can't the first one be understood as
"not one of them hates you"? Why can't the second one be "some of
them don't hate you"? zo'e does not repeat a previous sumti.
Neither {na xebni do} nor {naku xebni do} are very clear with
the context you provide, and I don't get different meanings from
them.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

Jordan:
> We're assuming that zo'e in this case is the referent of {le se
> go'i}. {zo'e} is probably most frequently explicitly specified as
> a super-generic sumti anaphora---someone's going to hear this, and
> think "what's zo'e-oh something for which xebni makes sense-oh
> probably ro bradi, since that's what we're talking about".

I dispute zo'e works like that.

A - mi klama le zarci
B - mi go'i

"Ah, so B must be going from the same place, via the same route
and by the same means as A"? Even at the same times? I don't think so.

> In realistic communication, B and B' would likely be interpreted
> with a different meaning---but with the same binding for zo'e.

I don't see why.

> Just
> because zo'e *could* be bound to different things in each of those
> two sentences doesn't mean that moving naku through zo'e has no
> effect. For a given binding of the zo'e, moving naku can very much
> change the meaning of the sentence.

I don't think zo'e ever gets fixed so that you are forced into one
interpretation or the other. A speaker saying {na broda} or
{naku broda} provides exactly the same amount of information to
the audience.

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail - 50x more storage than other providers!
http://promotions.yahoo.com/new_mail

Not even that. {zo'e noi broda cu brode} is false for any zo'e that is not a broda, whether ort not it is a brode (Its negation is false, too, in this case — {noi} is a context-leaper "and"). {zo'e poi broda cu brode} simply does not apply to any zo'e that is not a broda (so, it might be false if there are no broda, but otherwise the truth of {zo'e broda} is a presupposition of the sentence as a whole.
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:pc:
> Huh? But that is exactly what {poi} does: it means we only talk about brodas.
> And, of course, {zo'e noi broda} starts with "a larger meaning," the totally
> unspecified {zo'e} and then rather than restricting us to broda, keeps the
> unrestricted base and adds brodaness in incidentally.

Would you say the claim with {zo'e poi} is materially equivalent to
the one with {zo'e noi}?

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail is new and improved - Check it out!
http://promotions.yahoo.com/new_mail

A> Your evidence? Jordan's comment seems a reasonable interpretation of CLL 392 (152): {zo'e} stands for "You know who."

B> But note that none of the su'ocimoic terms is in the original. In that case they are unspecified and so may be taken as such in the repeat.

C> Can't argue with that.

D> As you so often say, context? If "you know who" is fixed in the context, then, while the two sentences may provide pretty much the same *amount* of information, the information is different (I suspect that Shannon would insist that the amount is different as well: {na broda} is much less informative than {naku broda}, generally).
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
Jordan:
> We're assuming that zo'e in this case is the referent of {le se
> go'i}. {zo'e} is probably most frequently explicitly specified as
> a super-generic sumti anaphora---someone's going to hear this, and
> think "what's zo'e-oh something for which xebni makes sense-oh
> probably ro bradi, since that's what we're talking about".

A>I dispute zo'e works like that.

A - mi klama le zarci
B - mi go'i

B>"Ah, so B must be going from the same place, via the same route
and by the same means as A"? Even at the same times? I don't think so.

> In realistic communication, B and B' would likely be interpreted
> with a different meaning---but with the same binding for zo'e.

C>I don't see why.

> Just
> because zo'e *could* be bound to different things in each of those
> two sentences doesn't mean that moving naku through zo'e has no
> effect. For a given binding of the zo'e, moving naku can very much
> change the meaning of the sentence.

D>I don't think zo'e ever gets fixed so that you are forced into one
interpretation or the other. A speaker saying {na broda} or
{naku broda} provides exactly the same amount of information to
the audience.

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail - 50x more storage than other providers!
http://promotions.yahoo.com/new_mail

pc:
> The third possibility is to allow naming to cover
> the outer domain, reserving the inner for what exists here and now
> (typically, the extension of {zasti}).

Surely this is what Lojban does. Otherwise {ro da zasti} and
{no da xanri} would be tautologically true, which is not the
case. I'm not sure whether things like lo namcu you take to be
in the inner or the outer domain.

> The issue then is about the
> extensions of predicates: do we interpret them over the inner and outer
> domains or only over the inner (and add "and exists" when we are interested
> in the latter matter)?

I would say it depends on the context.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - Send 10MB messages!
http://promotions.yahoo.com/new_mail

On Fri, Jun 25, 2004 at 06:25:38AM -0700, Jorge Llambías wrote:
> Jordan:
> > We're assuming that zo'e in this case is the referent of {le se
> > go'i}. {zo'e} is probably most frequently explicitly specified as
> > a super-generic sumti anaphora---someone's going to hear this, and
> > think "what's zo'e-oh something for which xebni makes sense-oh
> > probably ro bradi, since that's what we're talking about".
>
> I dispute zo'e works like that.
>
> A - mi klama le zarci
> B - mi go'i
>
> "Ah, so B must be going from the same place, via the same route
> and by the same means as A"? Even at the same times? I don't think so.
Those are elided {zo'e}.

When {zo'e} is not-elided, because of Grice the listener is going
to assume it is a relevenant zo'e, or a zo'e whose value they should
be able to actually guess. Whereas elided zo'e implies that its
value is not currently important.

> > In realistic communication, B and B' would likely be interpreted
> > with a different meaning---but with the same binding for zo'e.
>
> I don't see why.

• shrug*. Do you dispute that that is the case though?

Show up on irc and try saying sentences with na and explicit {zo'e},
and see how they get interpreted (only works if the victims are
unsuspecting).

> > Just
> > because zo'e *could* be bound to different things in each of those
> > two sentences doesn't mean that moving naku through zo'e has no
> > effect. For a given binding of the zo'e, moving naku can very much
> > change the meaning of the sentence.
>
> I don't think zo'e ever gets fixed so that you are forced into one
> interpretation or the other.

What does that mean?

Any sentence with naku boundaries moved must have the same meaning.
Agreed?

> A speaker saying {na broda} or
> {naku broda} provides exactly the same amount of information to
> the audience.

I'd agree, zo'e-wise, since all the zo'e positions are not specified.

--
Jordan DeLong
fracture@allusion.net

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---
Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFA3DLVDrrilS51AZ8RAsPXAKCK4YNqOd2OAaLGrusLi1LhJW6f0QCcC7Ac
8xFSbr4Ym79F95s9jmnA9R4=
=CvF8
---END PGP SIGNATURE---

On Fri, Jun 25, 2004 at 06:29:40AM -0700, John E Clifford wrote:
> > > Huh? But that is exactly what {poi} does: it means we only talk about brodas.
> > > And, of course, {zo'e noi broda} starts with "a larger meaning," the totally
> > > unspecified {zo'e} and then rather than restricting us to broda, keeps the
> > > unrestricted base and adds brodaness in incidentally.
> >
> > Would you say the claim with {zo'e poi} is materially equivalent to
> > the one with {zo'e noi}?
>
> Not even that. {zo'e noi broda cu brode} is false for any zo'e
> that is not a broda, whether ort not it is a brode (Its negation
> is false, too, in this case — {noi} is a context-leaper "and").
> {zo'e poi broda cu brode} simply does not apply to any zo'e that
> is not a broda (so, it might be false if there are no broda, but
> otherwise the truth of {zo'e broda} is a presupposition of the
> sentence as a whole.
{poi} is also an "and".

noi and poi are redundant, logically. They have the same structure.

da poi brode cu broda == da zo'u ge da broda gi brode
da noi brode cu broda == da zo'u ge da broda gi brode

The only difference is that poi implies that the listener should
understand which thingie is being talked about based on the information
in the poi clause. The noi implies that the listener should already
know what is being talked about. This means noi is definitely correct
for xorxes' definitions.

Your comment about negation is unfounded.

--
Jordan DeLong
fracture@allusion.net

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---
Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFA3DPqDrrilS51AZ8RAoRiAKCmnqzNl5JpvGTE5+wQySis4BBA+ACdFpxP

+n4aAVWo07ZPIAPZy7GG/Nk=

=6w4Y
---END PGP SIGNATURE---

pc:
> All the contexts in this case are opaque, so we are here talking ablut
> unicornity, which exists whether or not unicorns do.

To me {tavla fi lo ka pavyseljirna} is not the same as
{tavla fi lo pavyseljirna}, and the latter does not claim
unicorns are real. But I can accept it that you cannot fathom
the second one. We simply take referents from different domains.

> {zo'e} in this context means something like "the obvious sumti,"
> which in this context pretty much has to be {ro bradi}, the only sumti
> available other than {mi}, which being unlikely and shorter than {zo'e} would
> be sued if intended.

I guess when we get to {zo'e} we will have to discuss all these
rules I didn't know existed that so limit the possible referents of
zo'e.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

Jordan:
> When {zo'e} is not-elided, because of Grice the listener is going
> to assume it is a relevenant zo'e, or a zo'e whose value they should
> be able to actually guess. Whereas elided zo'e implies that its
> value is not currently important.

In {zo'e noi ke'a broda}, zo'e cannot be elided. The definition
does not imply that zo'e has to pick its referent from a recent
sumti.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - Send 10MB messages!
http://promotions.yahoo.com/new_mail

A> This suggests that even quantifiers extend over the outer domain. I don't think that is right. so, it is hard to say whether those sentences are tautologies or not. We certainly don't want to infer from {lo pavyseljirna cu xanri danlu} to {da xanri danlu} — or do we, at least sometimes?

B> Maybe. But then we can't assume that it is one way rather than the other. But to attack validity it is suffiecient to find one contrary case — that is not too outre'.
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
pc:
> The third possibility is to allow naming to cover
> the outer domain, reserving the inner for what exists here and now
> (typically, the extension of {zasti}).

A>Surely this is what Lojban does. Otherwise {ro da zasti} and
{no da xanri} would be tautologically true, which is not the
case. I'm not sure whether things like lo namcu you take to be
in the inner or the outer domain.

> The issue then is about the
> extensions of predicates: do we interpret them over the inner and outer
> domains or only over the inner (and add "and exists" when we are interested
> in the latter matter)?

B>I would say it depends on the context.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - Send 10MB messages!
http://promotions.yahoo.com/new_mail

A> these are certainly different cases, we can talk about lo ka pavyseljirna without ever mentioning unicorns as such, but propbably not about lo pavyseljirna. I did not say they were the same claim, I said (well, meant, anyhow) that what plays a role in these cases is the property not the object itself. The role is also different (overlap vs the usual mingle, in that awful terminology) So, strictly speaking, what is used in the first cases is lo ka ce'u ka ce'u pavyseljirna, and lo ka ce'u pavyseljirna in the second. Neither uses directly the thing mentioned ("the extensions in intensional contexts are the intensions of the ordinary extensions").

B> In a lot of contexts, {zo'e} also means "I don't give a damn which sumti," and those are what you want (and can use out of context, as in definitions and the like).
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
pc:
> All the contexts in this case are opaque, so we are here talking ablut
> unicornity, which exists whether or not unicorns do.

A>To me {tavla fi lo ka pavyseljirna} is not the same as
{tavla fi lo pavyseljirna}, and the latter does not claim
unicorns are real. But I can accept it that you cannot fathom
the second one. We simply take referents from different domains.

> {zo'e} in this context means something like "the obvious sumti,"
> which in this context pretty much has to be {ro bradi}, the only sumti
> available other than {mi}, which being unlikely and shorter than {zo'e} would
> be sued if intended.

B>I guess when we get to {zo'e} we will have to discuss all these
rules I didn't know existed that so limit the possible referents of
zo'e.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

pc:
> This suggests that even quantifiers extend over the outer domain. I
> don't think that is right. so, it is hard to say whether those sentences are
> tautologies or not. We certainly don't want to infer from {lo pavyseljirna
> cu xanri danlu} to {da xanri danlu} — or do we, at least sometimes?

I think we do, at least sometimes.

What we don't want to infer is {su'o da poi zasti le fatci
munje cu xanri danlu}, but the restriction of {da} to {da poi
zasti le fatci munje} is not a given in every context.

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail

A>Well, we disagree and CLL does not seem to help much. I take {da poi broda cu brode} to be ExFxGx and {da noi broda cu brode} to be Ex(Fx & Gx). To be sure, these two cases are materially equivalent, but in more complex cases — even negation — they differ: AxFx~Gx and Ex(Fx & ~Gx). Oh, but that depends upon how you stand on quantifiers. Never mind.
Why redundant. What there covers their information?
B> But this is just what is not happening in these definitions. We do not know what zo'e is and it is the clause immediately after that sets us looking in the right direction. {noi} might make sense in a definition of {le}, but not of {lo} (even {zo'e} is suspect for (lo)).

C> You mean about {noi} being a leaper? I take that to be what saying it introduces incidental information means: that it is parallel to but not part of the central sentence, the thing likely to be denied or approved of — and so on.

Jordan DeLong <fracture@allusion.net> wrote:
On Fri, Jun 25, 2004 at 06:29:40AM -0700, John E Clifford wrote:
> > > Huh? But that is exactly what {poi} does: it means we only talk about brodas.
> > > And, of course, {zo'e noi broda} starts with "a larger meaning," the totally
> > > unspecified {zo'e} and then rather than restricting us to broda, keeps the
> > > unrestricted base and adds brodaness in incidentally.
> >
> > Would you say the claim with {zo'e poi} is materially equivalent to
> > the one with {zo'e noi}?
>
> Not even that. {zo'e noi broda cu brode} is false for any zo'e
> that is not a broda, whether ort not it is a brode (Its negation
> is false, too, in this case — {noi} is a context-leaper "and").
> {zo'e poi broda cu brode} simply does not apply to any zo'e that
> is not a broda (so, it might be false if there are no broda, but
> otherwise the truth of {zo'e broda} is a presupposition of the
> sentence as a whole.
A>{poi} is also an "and".

noi and poi are redundant, logically. They have the same structure.

da poi brode cu broda == da zo'u ge da broda gi brode
da noi brode cu broda == da zo'u ge da broda gi brode

B>The only difference is that poi implies that the listener should
understand which thingie is being talked about based on the information
in the poi clause. The noi implies that the listener should already
know what is being talked about. This means noi is definitely correct
for xorxes' definitions.

C>Your comment about negation is unfounded.

--
Jordan DeLong
fracture@allusion.net

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---
Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFA3DPqDrrilS51AZ8RAoRiAKCmnqzNl5JpvGTE5+wQySis4BBA+ACdFpxP

+n4aAVWo07ZPIAPZy7GG/Nk=

=6w4Y
---END PGP SIGNATURE---

Yes, because it has no context.
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:Jordan:
> When {zo'e} is not-elided, because of Grice the listener is going
> to assume it is a relevenant zo'e, or a zo'e whose value they should
> be able to actually guess. Whereas elided zo'e implies that its
> value is not currently important.

In {zo'e noi ke'a broda}, zo'e cannot be elided. The definition
does not imply that zo'e has to pick its referent from a recent
sumti.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - Send 10MB messages!
http://promotions.yahoo.com/new_mail

Probably true. But for practical reasons, not from any theoretical problems like needing referents for certain phrases. To be sure, it solves those problems true and that counts in its favor. Now we need to specify the occasions of use. And I bet an honest "You'll know 'em when you see 'em" won't pass.
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:pc:
> This suggests that even quantifiers extend over the outer domain. I
> don't think that is right. so, it is hard to say whether those sentences are
> tautologies or not. We certainly don't want to infer from {lo pavyseljirna
> cu xanri danlu} to {da xanri danlu} — or do we, at least sometimes?

I think we do, at least sometimes.

What we don't want to infer is {su'o da poi zasti le fatci
munje cu xanri danlu}, but the restriction of {da} to {da poi
zasti le fatci munje} is not a given in every context.

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail

On Fri, Jun 25, 2004 at 05:48:11AM -0700, John E Clifford wrote:
> Huh? But that is exactly what {poi} does: it means we only talk about brodas. And, of course, {zo'e noi broda} starts with "a larger meaning," the totally unspecified {zo'e} and then rather than restricting us to broda, keeps the unrestricted base and adds brodaness in incidentally.

I actually agree with PC here.

Perhaps the poi/noi distinction is irrelevant. I'd still prefer to see {voi}
used for {le} instead of the {mi skicu...} thing; that's most of the reason
that {voi} exists.
--
Rob Speer

Jordan DeLong scripsit:

> Those are elided {zo'e}.
>
> When {zo'e} is not-elided, because of Grice the listener is going
> to assume it is a relevenant zo'e, or a zo'e whose value they should
> be able to actually guess. Whereas elided zo'e implies that its
> value is not currently important.

zo'e is semantically the same as elision; it is a speaking silence, intro
duced to keep the place count correct. Lojban makes no distinction between
broda and broda zo'e.
--
Some people open all the Windows; John Cowan
wise wives welcome the spring jcowan@reutershealth.com
by moving the Unix. http://www.reutershealth.com
--ad for Unix Book Units (U.K.) http://www.ccil.org/~cowan
(see http://cm.bell-labs.com/cm/cs/who/dmr/unix3image.gif)
John Cowan jcowan@reutershealth.com www.ccil.org/~cowan www.reutershealth.com
Any sufficiently-complicated C or Fortran program contains an ad-hoc,
informally-specified bug-ridden slow implementation of half of Common Lisp.
--Greenspun's Tenth Rule of Programming (rules 1-9 are unknown)

Jordan DeLong scripsit:

> noi and poi are redundant, logically. They have the same structure.
>
> da poi brode cu broda == da zo'u ge da broda gi brode
> da noi brode cu broda == da zo'u ge da broda gi brode

The latter is correct but the former is not. "da poi brode cu broda"
asserts only that those da's that brode also broda; it is equivalent
to "da zo'u ganai da brode gi da brode".

--
John Cowan jcowan@reutershealth.com www.reutershealth.com www.ccil.org/~cowan
Assent may be registered by a signature, a handshake, or a click of a computer
mouse transmitted across the invisible ether of the Internet. Formality
is not a requisite; any sign, symbol or action, or even willful inaction,
as long as it is unequivocally referable to the promise, may create a contract.
--Specht v. Netscape

Rob:
> Perhaps the poi/noi distinction is irrelevant. I'd still prefer to see {voi}
> used for {le} instead of the {mi skicu...} thing; that's most of the reason
> that {voi} exists.

If {voi} means {noi mi skicu ...}, then that works. Probably
that's what it should mean, but we didn't get there yet.
(poi wouldn't be right for le broda because that would mean that
given the referents of zo'e, we restrict to those that I describe
as broda. But zo'e here is not supposed to have a more ample
referent than what I describe as broda.

The {noi mi skicu...} thing is no worse than the {noi cmene fi mi}
thing, and I think it is not a bad thing to show explicitly how
similar le and la are.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

John Cowan:
> Jordan DeLong scripsit:
> > noi and poi are redundant, logically. They have the same structure.
> >
> > da poi brode cu broda == da zo'u ge da broda gi brode
> > da noi brode cu broda == da zo'u ge da broda gi brode
>
> The latter is correct but the former is not. "da poi brode cu broda"
> asserts only that those da's that brode also broda; it is equivalent
> to "da zo'u ganai da brode gi da brode".

You are describing {ro da poi broda}, and Jordan was thinking
of {su'o da poi broda}. This is the curse of implicit quantifiers.

I would recommend asking {xo da poi broda}, just to make sure,
if someone uses {da poi broda} without an explicit quantifier.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

A> Don't sound so surprised; we all have ourt good days.
Rob Speer <rspeer@MIT.EDU> wrote:On Fri, Jun 25, 2004 at 05:48:11AM -0700, John E Clifford wrote:
> Huh? But that is exactly what {poi} does: it means we only talk about brodas. And, of course, {zo'e noi broda} starts with "a larger meaning," the totally unspecified {zo'e} and then rather than restricting us to broda, keeps the unrestricted base and adds brodaness in incidentally.

A>I actually agree with PC here.

Perhaps the poi/noi distinction is irrelevant. I'd still prefer to see {voi}
used for {le} instead of the {mi skicu...} thing; that's most of the reason
that {voi} exists.
--
Rob Speer

Does this mean that xorxes "definitions" are argle-bargle even before he gets to the parrts we disagree with?
John Cowan <jcowan@reutershealth.com> wrote:Jordan DeLong scripsit:

> Those are elided {zo'e}.
>
> When {zo'e} is not-elided, because of Grice the listener is going
> to assume it is a relevenant zo'e, or a zo'e whose value they should
> be able to actually guess. Whereas elided zo'e implies that its
> value is not currently important.

zo'e is semantically the same as elision; it is a speaking silence, intro
duced to keep the place count correct. Lojban makes no distinction between
broda and broda zo'e.
--
Some people open all the Windows; John Cowan
wise wives welcome the spring jcowan@reutershealth.com
by moving the Unix. http://www.reutershealth.com
--ad for Unix Book Units (U.K.) http://www.ccil.org/~cowan
(see http://cm.bell-labs.com/cm/cs/who/dmr/unix3image.gif)
John Cowan jcowan@reutershealth.com www.ccil.org/~cowan www.reutershealth.com
Any sufficiently-complicated C or Fortran program contains an ad-hoc,
informally-specified bug-ridden slow implementation of half of Common Lisp.
--Greenspun's Tenth Rule of Programming (rules 1-9 are unknown)

A> tugni (note what happens to all the {zo'e}s here)
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:
Rob:
> Perhaps the poi/noi distinction is irrelevant. I'd still prefer to see {voi}
> used for {le} instead of the {mi skicu...} thing; that's most of the reason
> that {voi} exists.

If {voi} means {noi mi skicu ...}, then that works. Probably
that's what it should mean, but we didn't get there yet.
(poi wouldn't be right for le broda because that would mean that
given the referents of zo'e, we restrict to those that I describe
as broda. But zo'e here is not supposed to have a more ample
referent than what I describe as broda.

A>The {noi mi skicu...} thing is no worse than the {noi cmene fi mi}
thing, and I think it is not a bad thing to show explicitly how
similar le and la are.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

Well, the rule is clear; but apparently the practice is not (and the "all" reading has a long histiry, too — perhaps longer than that for the "some" reading; "all" seems to be Aristotle's usage.)
Jorge Llambías <jjllambias2000@yahoo.com.ar> wrote:John Cowan:
> Jordan DeLong scripsit:
> > noi and poi are redundant, logically. They have the same structure.
> >
> > da poi brode cu broda == da zo'u ge da broda gi brode
> > da noi brode cu broda == da zo'u ge da broda gi brode
>
> The latter is correct but the former is not. "da poi brode cu broda"
> asserts only that those da's that brode also broda; it is equivalent
> to "da zo'u ganai da brode gi da brode".

You are describing {ro da poi broda}, and Jordan was thinking
of {su'o da poi broda}. This is the curse of implicit quantifiers.

I would recommend asking {xo da poi broda}, just to make sure,
if someone uses {da poi broda} without an explicit quantifier.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

On Friday 25 June 2004 10:12, Jordan DeLong wrote:
> When {zo'e} is not-elided, because of Grice the listener is going
> to assume it is a relevenant zo'e, or a zo'e whose value they should
> be able to actually guess. Whereas elided zo'e implies that its
> value is not currently important.

Not necessarily. I might say {mi klama zo'e le zarci} and mean no more than
{mi klama fi le zarci}. Or I might put a {zo'e} in a NU-clause to preempt a
{ce'u} that would otherwise be assumed to be in that place.

phma
--
li fi'u vu'u fi'u fi'u du li pa

---

> On Fri, Aug 06, 2004 at 06:45:29AM -0700, Jorge Llamb?as wrote:
>
> > ti noi djacu cu klani li re lo kabri
> > This, which is water, amounts to 2 measured in cups.
> >
> > But {lo djacu} is not a sumti there. As a sumti:
> >
> > lo djacu je klani be li re bei lo kabri
> > Water and amount 2 in cups.
>
> Err, what?
>
> lo djacu cu klani li re lo kabri

That's not a sumti, it's a bridi, and it's practically what I
wrote above. We were looking for a way to say "two cups of water".

> Read klani a bit more carefully.

How does any of what I wrote conflict with the definition of klani?

> > Another possibility is to use the number "2 cups" as a quantifier
> > of {djacu}:
> >
> > vei mo'e re kabri ve'o djacu
>
> That works too.

Unfortunately, we can't get rid of the bracketing.

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail - 50x more storage than other providers!
http://promotions.yahoo.com/new_mail

On Mon, Aug 09, 2004 at 02:02:51PM -0700, Jorge Llamb?as wrote:
>
> --- Robin Lee Powell <rlpowell@digitalkingdom.org> wrote:
> > On Fri, Aug 06, 2004 at 06:45:29AM -0700, Jorge Llamb?as wrote:
> >
> > > ti noi djacu cu klani li re lo kabri This, which is water,
> > > amounts to 2 measured in cups.
> > >
> > > But {lo djacu} is not a sumti there. As a sumti:
> > >
> > > lo djacu je klani be li re bei lo kabri Water and amount 2 in
> > > cups.
> >
> > Err, what?
> >
> > lo djacu cu klani li re lo kabri
>
> That's not a sumti, it's a bridi, and it's practically what I wrote
> above. We were looking for a way to say "two cups of water".
>
> > Read klani a bit more carefully.
>
> How does any of what I wrote conflict with the definition of klani?

No, I was confused.

-Robin

--
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/
Reason #237 To Learn Lojban: "Homonyms: Their Grate!"

NIce, but in keeping with what I said before, I would probably use {loi djacu} (with natural masses it does not seem to make much difference).

Robin Lee Powell <rlpowell@digitalkingdom.org> wrote:On Fri, Aug 06, 2004 at 06:45:29AM -0700, Jorge Llamb?as wrote:
>
> pc:
> > I suggest the same might be said for goo, which can, in Lojban, be
> > viewed as a kind of collective, and one that can take numeration:
> > "three teaspoons of salt" or, more clearly, "two cups of water." Of
> > course it is not clear where these numerations go, but the "this is
> > in cups number two of water" does not seem the best solution.
>
> It's not even clear how to say that. Probably something like:
>
> ti noi djacu cu klani li re lo kabri
> This, which is water, amounts to 2 measured in cups.
>
> But {lo djacu} is not a sumti there. As a sumti:
>
> lo djacu je klani be li re bei lo kabri
> Water and amount 2 in cups.

Err, what?

lo djacu cu klani li re lo kabri

Read klani a bit more carefully.

> Another possibility is to use the number "2 cups" as a quantifier
> of {djacu}:
>
> vei mo'e re kabri ve'o djacu

That works too.

-Robin

--
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/
Reason #237 To Learn Lojban: "Homonyms: Their Grate!"

On Fri, Jul 30, 2004 at 07:04:40AM -0700, wikidiscuss@lojban.org wrote:
> Re: Robin's gadri Proposal
>
> > Personal Specificity ??? sa'e The specific thing(s) I have in mind
> > (i.e. "le", but without implication of veridicality).
>
> I insist that {le} has no implication of veridicality, so the comment
> '"le", but without implication of veridicality' does not make sense.

Fair.

> > le broda ~= lo sa'e broda == su'o da poi mi pensi ke'a zi'e poi ke'a
> > broda
>
> "~=" here means approximately equal, right? (Sometimes "~" is used for
> negation.) Does this mean that you are dropping CLL's default {[ro]
> le}?

No, I just forgot about it.

> Consider:
>
> le prenu cu klama le zarci
>
> CLL: Each of the people I have in mind goes to each of the
> markets I have in mind.
> RGP: At least one of the people I have in mind goes to at
> least one of the markets I have in mind.
> XS: The people I have in mind go to the markets I have in mind.
>
> I believe XS is closest to usage, and CLL is the canonical
> prescription, but your version is neither.

RGP? You meant RLP, yes? In which case, where did you find a keyboard
where G is close to L?

-Robin

Sure, but you have yto know where to look; neither {drayckei} nor {sibvidru} are on the generial lists and role-playing games are so remote from my life (since I played chaturanga and some derivatives in college) that they would not occur to me — until I went to the original site and then looked up {xarpre}. u'u though

Robin Lee Powell <rlpowell@digitalkingdom.org> wrote: On Mon, Aug 09, 2004 at 02:18:35PM -0700, John E Clifford wrote:
> Of what? Our patience at decrypting nonsense.

For sending a test message.

> As far as I could make out {dracykei} is just {xe draci} and
> {sibvidru} is a terrible metaphor for "germ of an idea"

jbovlaste is out there for a reason, you know.

http://www.lojban.org/jbovlaste/wiki/Role%20Playing%20Terms

http://www.lojban.org/jbovlaste/wiki/Orion's%20Arm%20Story%20Terms

-Robin

RGP is presumably "Robin's Gadri Proposal"

Robin Lee Powell <rlpowell@digitalkingdom.org> wrote:On Fri, Jul 30, 2004 at 07:04:40AM -0700, wikidiscuss@lojban.org wrote:
> Re: Robin's gadri Proposal
>
> > Personal Specificity ??? sa'e The specific thing(s) I have in mind
> > (i.e. "le", but without implication of veridicality).
>
> I insist that {le} has no implication of veridicality, so the comment
> '"le", but without implication of veridicality' does not make sense.

Fair.

> > le broda ~= lo sa'e broda == su'o da poi mi pensi ke'a zi'e poi ke'a
> > broda
>
> "~=" here means approximately equal, right? (Sometimes "~" is used for
> negation.) Does this mean that you are dropping CLL's default {[ro]
> le}?

No, I just forgot about it.

> Consider:
>
> le prenu cu klama le zarci
>
> CLL: Each of the people I have in mind goes to each of the
> markets I have in mind.
> RGP: At least one of the people I have in mind goes to at
> least one of the markets I have in mind.
> XS: The people I have in mind go to the markets I have in mind.
>
> I believe XS is closest to usage, and CLL is the canonical
> prescription, but your version is neither.

RGP? You meant RLP, yes? In which case, where did you find a keyboard
where G is close to L?

-Robin

> > Re: Robin's gadri Proposal

> > Consider:
> > le prenu cu klama le zarci
> >
> > CLL: Each of the people I have in mind goes to each of the
> > markets I have in mind.
> > RGP: At least one of the people I have in mind goes to at
> > least one of the markets I have in mind.
> > XS: The people I have in mind go to the markets I have in mind.
> >
> > I believe XS is closest to usage, and CLL is the canonical
> > prescription, but your version is neither.
>
> RGP? You meant RLP, yes? In which case, where did you find a keyboard
> where G is close to L?

See this thread's subject line.

mu'o mi'e xorxes

__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

On Mon, Aug 09, 2004 at 03:45:58PM -0700, John E Clifford wrote:
> Sure, but you have yto know where to look; neither {drayckei} nor
> {sibvidru} are on the generial lists

I have no idea what a "general list" is in this context, but:

http://www.lojban.org/jbovlaste/dict/sibvidru

dracykei is my bad; I'll fix it momentarily.

-Robin