Talk:unless

From Lojban
Jump to: navigation, search
Posted by pycyn on Fri 12 of Nov., 2004 10:41 GMT posts: 2388

> The short version: the English > "unless" can be translated directly > with an "or" construct and > "da'i", such as "da broda gi'a > da'i brode", if both cases can happen > at the same time. If not, you need an > "xor" construct instead > ("gi'onai" in this case).

> > The long version: In analyzing > "unless", it is helpful to have an > example. For the "both can happen at the > same time" case, we'll use "I have > blue hair, unless my eyes are bad". The > translation is "le kerfa be mi cu blanu .i > ja bo da'i le kanla be mi cu spofu". > > Call the former "BH" for blue hair, > and the latter "EB" for "eyes > are bad". > > Truth table: >

>

> BH EB BH or EB Explanation
> T T T You have blue hair. It is
> also the case that your eyes are bad, however.
> T F T You have blue hair, and
> your eyes are just fine
> F T T Your eyes are bad, and
> your hair is not, in fact, blue.
> F F F The only false case: if
> you do not have blue hair, but your eyes are
> just fine, the original statement is false.
>

> > For the "both cannot happen at the > same time" case, we'll use "I'll kill > you, unless you kill me first". Note that > because of the word "first", both > halves cannot be true at once. The translation > is "mi ba catra do .i jo nai bo da'i do > catra mi pu la'e di'u". > > Call the former "KY" for kill you, > and the latter "KMF" for "kill > me first". > > Truth table: > > || > KY | KMF| KY or KMF| Explanation > T | T | F | False case: I killed you, > and you killed me first. Patently > impossible. > T | F | T | I killed you. > F | T | T | You killed me first. > F | F | F | False case: Nobody killed > anybody, which violates the whole point of > making the statement in the first place.

What is the function of the {da'i} here? The whole is taken to be truth functional (see the tables) so movement into the subjunctive (more or less)apparently play no role. Presumably the {da'i} comes from the conceptually underlying conditional nature of "unless" ("this if not that") and (we often say) real conditionals are subjunctive. But this seems to be a genuinely truth functional use. I would welcome convincing cases to the contrary, but don't have one so far (of course, "convincing" may be the problem, since what convinces one person may not convince others; but I can't find a case which is even claimed to be subjunctive in my — woefully incomplete — files).

Score: 0.00 Vote:
1 2 3 4 5
top of page

rlpowellPosted by rlpowell on Fri 12 of Nov., 2004 10:41 GMT posts: 14214

On Thu, Nov 11, 2004 at 05:26:14PM -0800, John E Clifford wrote: > and (we often say) real conditionals are subjunctive. But this > seems to be a genuinely truth functional use.

Uhh, if "My hair is blue, unless my eyes are bad" doesn't imply that the second clause is subjunctive (the speaker doesn't actually think his eyes are bad), you and I speak very different dialects of English.

-Robin

Score: 0.00 Vote:
1 2 3 4 5
top of page

Posted by pycyn on Fri 12 of Nov., 2004 18:16 GMT posts: 2388

wrote:

> On Thu, Nov 11, 2004 at 05:26:14PM -0800, John > E Clifford wrote: > > and (we often say) real conditionals are > subjunctive. But this > > seems to be a genuinely truth functional use. > > > Uhh, if "My hair is blue, unless my eyes are > bad" doesn't imply that > the second clause is subjunctive (the speaker > doesn't actually think > his eyes are bad), you and I speak very > different dialects of > English. > > -Robin

Well, it may imply that (better implicate that, since it is a pragmatic kind of compulsion), but that doesn't make it subjunctive. It still works as a truth functional claim (as your tables show): presumably as an emphatic form of "My hair is blue." (I'm not sure it need function in that way, but am considering the case where it does.) This differs from the subjunctive cases, where we want to say that a claim is true — or false -- even though the truth functional conditions do not apply (false conditionals with false antecedents, for example). A disjunction with a (implicitly) false disjunct is non-problematically true, provided that the other disjunct is.

Score: 0.00 Vote:
1 2 3 4 5
top of page

Posted by xorxes on Fri 12 of Nov., 2004 18:16 GMT posts: 1912

In Spanish, the equivalent construct "X a menos que Y", always requires Y to be put in the subjunctive mood.

I'm not convinced "unless" can be a purely truth conditional OR. I don't want to say that "I am mortal, unless you are mortal" is true, even though "I am mortal OR you are mortal" is.

"Unless" would seem to require its complement to be either false or very unlikely/exceptional.

mu'o mi'e xorxes


'''''''''''''''''''''''''''''''''''''''''''__ Do you Yahoo!? Check out the new Yahoo! Front Page. www.yahoo.com

Score: 0.00 Vote:
1 2 3 4 5
top of page

Posted by pycyn on Fri 12 of Nov., 2004 22:37 GMT posts: 2388

wrote:

> > In Spanish, the equivalent construct "X a menos > que Y", > always requires Y to be put in the subjunctive > mood.

This is virtually automatic with a subordinate "que" clause; whether it is significant is harder to tell (as is whether that significance carries over to English, where the corresonding sentence is not subjunctive — or not unequivocally so). And, of course, that also does not speak to the logical — thus Lojban — situation.


> I'm not convinced "unless" can be a purely > truth > conditional OR. I don't want to say that > "I am mortal, unless you are mortal" is true, > even though "I am mortal OR you are mortal" is. > Yes, there does seem to be something else than simple OR going one. As noted, I suspect it is pragmatic — but that doesn't relieve us of the need to figure out what it is and display it.

> "Unless" would seem to require its complement > to be either false or very unlikely / exceptional.

This seems to be in the right direction, corresponding to the emphatic negative conditional "If he did it, I'll eat my hat." = "He definitely didn't do it" and the like.

I'm not sure what you mean by "complement," appparently the sentence after the "unless"? But that doesn't seem to fit the usual cases: "He will come unless he gets caught in traffic" which is neither obviously false nor even unlikely but merely setting up an explanation in case the other sentence turns out to be false.

What does seem to be the case is that there are two uses of "unless," one purely factual, the other with an added fillip, presumably the same one that occurs also (though in a negative presentation) in the conditional cited above, emphasizing the truth (or falsity) of one factor by the obvious falsity of the other. I think this is much less common with "unless" than "if" but the underlying mechanism is the same, truth functionality (whether MTT or MTP) and trivial falseness of one component. But none of this is subjunctive or contrary-to-fact, just a pragmatic cuteness.

Score: 0.00 Vote:
1 2 3 4 5
top of page

Posted by xorxes on Sun 14 of Nov., 2004 04:30 GMT posts: 1912

pc: > I'm not sure what you mean by "complement," > appparently the sentence after the "unless"? But > that doesn't seem to fit the usual cases: "He > will come unless he gets caught in traffic" which > is neither obviously false nor even unlikely but > merely setting up an explanation in case the > other sentence turns out to be false.

"He will come" is the default case, "he gets caught in traffic" is the exceptional case. In the default world/situation, he comes. In other than the default world, it may be the case that he gets caught in traffic and does not come.

The sentence does not mean the same as "he will get caught in traffic unless he comes", so plain OR won't do it.

> What does seem to be the case is that there are > two uses of "unless," one purely factual, the > other with an added fillip, presumably the same > one that occurs also (though in a negative > presentation) in the conditional cited above, > emphasizing the truth (or falsity) of one factor > by the obvious falsity of the other.

Yes, and both "if" and "unless" in the non-emphasis case usually carry an additional component of causality which is absent in the purely truth functional connective.

mu'o mi'e xorxes


'''''''''''''''''''''''''''''''''''''''''''__ Do you Yahoo!? Check out the new Yahoo! Front Page. www.yahoo.com

Score: 0.00 Vote:
1 2 3 4 5
top of page

Posted by pycyn on Sun 14 of Nov., 2004 15:07 GMT posts: 2388

wrote:

> > pc: > > I'm not sure what you mean by "complement," > > appparently the sentence after the "unless"? > But > > that doesn't seem to fit the usual cases: "He > > will come unless he gets caught in traffic" > which > > is neither obviously false nor even unlikely > but > > merely setting up an explanation in case the > > other sentence turns out to be false. > > "He will come" is the default case, "he gets > caught > in traffic" is the exceptional case. In the > default > world/situation, he comes. In other than the > default > world, it may be the case that he gets caught > in > traffic and does not come. > > The sentence does not mean the same as "he will > get > caught in traffic unless he comes", so plain OR > won't > do it.

Yes; some mark for the "exceptional" case (just less common or less what is desired or expected) is needed. the point is always just tht it is not {da'i}. I think the use of worlds talk here is taking things a bit too far, unless you want different occasions to be different worlds even though both may be in the real world. > > > What does seem to be the case is that there > are > > two uses of "unless," one purely factual, the > > other with an added fillip, presumably the > same > > one that occurs also (though in a negative > > presentation) in the conditional cited above, > > emphasizing the truth (or falsity) of one > factor > > by the obvious falsity of the other. > > Yes, and both "if" and "unless" in the > non-emphasis > case usually carry an additional component of > causality > which is absent in the purely truth functional > connective.

Is the causal factor semantic or merely pragmatic? In any case, something should be said to mark off the unemphatic case.

Score: 0.00 Vote:
1 2 3 4 5
top of page

clsnPosted by clsn on Mon 15 of Nov., 2004 02:29 GMT posts: 84

Jorge "Llambías" wrote:

>In Spanish, the equivalent construct "X a menos que Y", >always requires Y to be put in the subjunctive mood. > >I'm not convinced "unless" can be a purely truth >conditional OR. I don't want to say that >"I am mortal, unless you are mortal" is true, >even though "I am mortal OR you are mortal" is. > >"Unless" would seem to require its complement to be >either false or very unlikely/exceptional. > > Why? Sure, you *can* use "unless" that way. But it's also quite commonly used to mean "if not..."

There'll be a game on Thursday, unless it rains. (not necessarily very unlikely!)

Fill out form XJ912/3 unless you have already filed waiver 99234. (perfectly likely)

other examples abound. I don't see extreme unlikeliness any more implied in "unless" than it is in "if".

~mark

Score: 0.00 Vote:
1 2 3 4 5
top of page

rlpowellPosted by rlpowell on Mon 15 of Nov., 2004 02:29 GMT posts: 14214

On Sun, Nov 14, 2004 at 09:06:49PM -0500, Mark E. Shoulson wrote: > Jorge "Llamb??as" wrote: > > >In Spanish, the equivalent construct "X a menos que Y", always > >requires Y to be put in the subjunctive mood. > > > >I'm not convinced "unless" can be a purely truth conditional OR. > >I don't want to say that "I am mortal, unless you are mortal" is > >true, even though "I am mortal OR you are mortal" is. > > > >"Unless" would seem to require its complement to be either false > >or very unlikely/exceptional. > > > Why? Sure, you *can* use "unless" that way. But it's also quite > commonly used to mean "if not..." > > There'll be a game on Thursday, unless it rains. (not necessarily very > unlikely!)

Re-read what he said again: Either false *OR* exceptional.

This started because I insisted that "da'i" should be used for the second half of the clause, because it may or may not have happened or ever happen. PC argued, and xorxes got in to it with him.

-Robin

Score: 0.00 Vote:
1 2 3 4 5
top of page

Posted by pycyn on Mon 15 of Nov., 2004 20:29 GMT posts: 2388

wrote:

> On Sun, Nov 14, 2004 at 09:06:49PM -0500, Mark > E. Shoulson wrote: > > Jorge "Llamb??as" wrote: > > > > >In Spanish, the equivalent construct "X a > menos que Y", always > > >requires Y to be put in the subjunctive > mood. > > > > > >I'm not convinced "unless" can be a purely > truth conditional OR. > > >I don't want to say that "I am mortal, > unless you are mortal" is > > >true, even though "I am mortal OR you are > mortal" is. > > > > > >"Unless" would seem to require its > complement to be either false > > >or very unlikely/exceptional. > > > > > > Why? Sure, you *can* use "unless" that way. > But it's also quite > > commonly used to mean "if not..." > > > > There'll be a game on Thursday, unless it > rains. (not necessarily very > > unlikely!) > > Re-read what he said again: Either false *OR* > exceptional.

But then, it doesn't have to be false either. But it is marked off in some way to prevent conversion: "It rains Thursday unless there is a game."

> This started because I insisted that "da'i" > should be used for the > second half of the clause, because it may or > may not have happened > or ever happen. PC argued, and xorxes got in > to it with him.

We are agreed more or less that the "unless" component needs marking off, but are of different minds about how strong that mark need be. I think we agree that full world- constructing {da'i} is stronger than needed, xorxes wants it stronger than I think is necessary. I would now go with just the distinguishing mark of "but" over "and," {ku'i} — expanding its meaning somewhat, to be sure.

Score: 0.00 Vote:
1 2 3 4 5
top of page

Posted by xorxes on Mon 15 of Nov., 2004 20:29 GMT posts: 1912

> Jorge "Llambías" wrote: > >"Unless" would seem to require its complement to be > >either false or very unlikely/exceptional. > > > Why? Sure, you *can* use "unless" that way. But it's also quite > commonly used to mean "if not..."

I agree "unless" is usually interchangeable with "if not". But they both normally carry more baggage than plain logical OR. Neither of them is usually interchangeable with plain "or".

> There'll be a game on Thursday, unless it rains. (not necessarily very > unlikely!)

Default: There'll be a game on Thusday. Exception: if it rains on Thusrday, there won't/may not be a game.

It doesn't mean the same as "it'll rain on Thursday, unless there is a game", which it should if "unless" was just OR.

> Fill out form XJ912/3 unless you have already filed waiver 99234. > (perfectly likely) > > other examples abound. I don't see extreme unlikeliness any more > implied in "unless" than it is in "if".

Right, it desn't have to be extreme. It's more like an exception.

mu'o mi'e xorxes


'''''''''''''''''''''''''''''''''''''''''''__ Do you Yahoo!? Check out the new Yahoo! Front Page. www.yahoo.com

Score: 0.00 Vote:
1 2 3 4 5
top of page

Posted by xorxes on Mon 15 of Nov., 2004 20:29 GMT posts: 1912

pc: > We are agreed more or less that the "unless" > component needs marking off, but are of different > minds about how strong that mark need be. I > think we agree that full world- constructing > {da'i} is stronger than needed, xorxes wants it > stronger than I think is necessary. I would now > go with just the distinguishing mark of "but" > over "and," {ku'i} — expanding its meaning > somewhat, to be sure.

I think that's a good idea. Let's see how it would work:

mi'o klama ija ku'i carvi We go, or, in contrast, it rains. We go unless it rains.

This will also allow us to reverse the order:

ga ku'i carvi gi mi'o klama Either, in contrast, it rains, or we go. Unless it rains, we go.

This suggests that {ge ku'i ... gi ...} can also be used as a general "even though":

ge ku'i carvi gi mi'o klama Both, in contrast, it rains, and we go. Even though it rains, we go.

(Of course we also have {seki'unai}, {semu'inai}, {seri'anai}, {seni'inai} for this.)

mu'o mi'e xorxes


'''''''''''''''''''''''''''''''''''''''''''__ Do you Yahoo!? Check out the new Yahoo! Front Page. www.yahoo.com

Score: 0.00 Vote:
1 2 3 4 5
top of page

rlpowellPosted by rlpowell on Mon 15 of Nov., 2004 20:30 GMT posts: 14214

On Mon, Nov 15, 2004 at 08:54:39AM -0800, Jorge Llamb?as wrote: > > pc: > > We are agreed more or less that the "unless" component needs > > marking off, but are of different minds about how strong that > > mark need be. I think we agree that full world- constructing > > {da'i} is stronger than needed, xorxes wants it stronger than I > > think is necessary. I would now go with just the distinguishing > > mark of "but" over "and," {ku'i} — expanding its meaning > > somewhat, to be sure. > > I think that's a good idea. Let's see how it would work: > > mi'o klama ija ku'i carvi > We go, or, in contrast, it rains. > We go unless it rains. > > This will also allow us to reverse the order: > > ga ku'i carvi gi mi'o klama > Either, in contrast, it rains, or we go. > Unless it rains, we go.

That all makes sense. You gonna change the page?

-Robin

Score: 0.00 Vote:
1 2 3 4 5
top of page

Posted by xorxes on Mon 15 of Nov., 2004 20:30 GMT posts: 1912

> > I think that's a good idea. Let's see how it would work: > > > > mi'o klama ija ku'i carvi > > We go, or, in contrast, it rains. > > We go unless it rains. > > > > This will also allow us to reverse the order: > > > > ga ku'i carvi gi mi'o klama > > Either, in contrast, it rains, or we go. > > Unless it rains, we go. > > That all makes sense. You gonna change the page?

I'd rather explore it a bit more first. I see that the section on UIb (ki'u, ji'a, po'o) is still shepherdless, would you mind if I take it?

mu'o mi'e xorxes


'''''''''''''''''''''''''''''''''''''''''''__ Do you Yahoo!? The all-new My Yahoo! - Get yours free! http://my.yahoo.com

Score: 0.00 Vote:
1 2 3 4 5
top of page

rlpowellPosted by rlpowell on Mon 15 of Nov., 2004 20:30 GMT posts: 14214

On Mon, Nov 15, 2004 at 11:43:16AM -0800, Jorge Llamb?as wrote: > > --- Robin Lee Powell wrote: > > > I think that's a good idea. Let's see how it would work: > > > > > > mi'o klama ija ku'i carvi We go, or, in contrast, it rains. > > > We go unless it rains. > > > > > > This will also allow us to reverse the order: > > > > > > ga ku'i carvi gi mi'o klama Either, in contrast, it rains, > > > or we go. Unless it rains, we go. > > > > That all makes sense. You gonna change the page? > > I'd rather explore it a bit more first. I see that the section on > UIb (ki'u, ji'a, po'o) is still shepherdless, would you mind if I > take it?

I do not. As far as I'm concerned, everyone else has had *plenty* of time to offer their services; take whatever you like.

-Robin

Score: 0.00 Vote:
1 2 3 4 5
top of page

Posted by pycyn on Mon 15 of Nov., 2004 21:03 GMT posts: 2388

wrote:

> > pc: > > We are agreed more or less that the "unless" > > component needs marking off, but are of > different > > minds about how strong that mark need be. I > > think we agree that full world- constructing > > {da'i} is stronger than needed, xorxes wants > it > > stronger than I think is necessary. I would > now > > go with just the distinguishing mark of "but" > > over "and," {ku'i} — expanding its meaning > > somewhat, to be sure. > > I think that's a good idea. Let's see how it > would work: > > mi'o klama ija ku'i carvi > We go, or, in contrast, it rains. > We go unless it rains. > > This will also allow us to reverse the order: > > ga ku'i carvi gi mi'o klama > Either, in contrast, it rains, or we go. > Unless it rains, we go. > > This suggests that {ge ku'i ... gi ...} can > also be used > as a general "even though": > > ge ku'i carvi gi mi'o klama > Both, in contrast, it rains, and we go. > Even though it rains, we go. > > (Of course we also have {seki'unai}, > {semu'inai}, {seri'anai}, > {seni'inai} for this.) > Hmmm! Nice!. I'm not sure I agree that they amount to the same thing, but they are clearly related (and that sentence seems to fit in pretty well).

Playing with the origianl "but" {e ku'i}, I ran across the following, which seems different and, indeed, to actually involve subjunctives (certainly in English) "He would have come, but he was tired." This looks like "Had he not been tired, he would have come" with the usual implication (but it seems to be more in the English original) that the protasis is false, i.e., that he was tired. Question, does' {da'i} always mean that the postulated state does not hold? Surely not, since general laws are true even when we are testing them (making the condition true). So, {da'i} at most pragmatically implicates contrary-to-factness and that implication can be overridden somehow.