This conclusion is derived from modal logic, one axiom of which goes like this: □P -> ◇P = If necessary that P then possible that P. Duties are necessary (obligatory) only if it's possible for them to be fulfilled. — TheMadFool

In deontic logic the diamond operator means “permissible”, not “possible”, just like the box operator means “obligatory” rather than “necessary”. So it follows that if something is obligatory it is permissible, and if it’s not permissible it’s not obligatory, but that doesn’t say anything at all about alethic possibility: it might be that morally obligatory things are impossible so we’re just fucked.


I do like that you’ve noted the connection between dialethism and morally intractable situations though. I’ve noted that before myself, in my earlier thread on mood operators and their implications on nonclassical logics, where I wrote:

The use of these mood functions also facilitates something superficially resembling the motivations for non-classical types of logic such as paraconsistent logics and intuitionist logics, without actually abandoning the principle that differentiates classical logic from them: the principle of bivalence. The principle of bivalence is the principle that every statement must be assigned exactly one of two truth values, "true" or "false", no more and no less. Intuitionist logics allow for statements to be assigned neither of those truth values, while paraconsistent logics allow for statements to be assigned both of them at the same time.

With these mood functions, similar things can be constructed without actually violating the principle of bivalance, because there is nothing strictly logically prohibiting it being the case that neither is(P) nor is(not-P), if for example P were some kind of descriptively meaningless statement; it is merely necessary, to preserve bivalance, that either is(P) or not(is(P)), but not(is(P)) doesn't have to entail that is(not-P).

Similarly, there is nothing strictly prohibiting it being the case that be(P) and be(not-P), if for example there were some morally intractable situation where both P and not-P were required, and so any outcome was unacceptable; it is merely necessary, to preserve bivalence, that either be(P) or not(be(P)), and be(not-P) doesn't have to entail not(be(P)), so could be compatible with be(P).

Fleshing out the philosophical implications of things like descriptively meaningless claims and morally intractable situations is beyond the scope of this particular essay on logic, other than to point out that a logic of this form is in principle capable of discussing things that are, in a sense, "both true and false" or "neither true nor false", without technically violating the principle of bivalence.


But I don’t think any of this is really necessary to make sense of Kant. He was just being inconsistent. Surely his categorical imperative would generally prohibit killing, but Kant was fine with capital punishment, so sometimes killing must be okay, by his reasoning. If killing can sometimes be excused even though it’s generally wrong, surely the same must apply to lying.

You tell the truth. Your friend dies but you've told the truth. — TheMadFool

This is not Kant's example. If you think it is, then you need to read it. Kant's point, among several, is that in replying to the murderer at the door, you do not know where your friend is, and consequently if you lie, you may actually then have killed your fried, except in this latter case, you bear responsibility! And other points as well. Read the thing!

In deontic logic the diamond operator means “permissible”, not “possible”, just like the box operator means “obligatory” rather than “necessary”. So it follows that if something is obligatory it is permissible, and if it’s not permissible it’s not obligatory, but that doesn’t say anything at all about alethic possibility: it might be that morally obligatory things are impossible so we’re just fucked — Pfhorrest

□p (necessarily p) is equivalent to ¬◇¬p ("not possible that not-p")
◇p (possibly p) is equivalent to ¬□¬p ("not necessarily not-p") — Wikipedia

??? :chin:

You're talking about deontic logic but it's an offshoot of modal logic, the latter subsuming the former as it were. Ergo, if something goes wrong at the level of modal logic, it wouldn't work in deontic logic too.

Think of it.

Modal Logic: If necessary P then possible P

Deontic Logic: If necessary P then permissible P

If deontic logic were independent of modal logic then it should be possible for a proposition P to be impossible and yet permissible. That doesn't make sense, right? The permissible supervenes on the possible.

But I don’t think any of this is really necessary to make sense of Kant. He was just being inconsistent. Surely his categorical imperative would generally prohibit killing, but Kant was fine with capital punishment, so sometimes killing must be okay, by his reasoning. If killing can sometimes be excused even though it’s generally wrong, surely the same must apply to lying. — Pfhorrest

I didn't know Kant supported the death penalty and Wikipedia describes it as an "extreme position". Noted! However, does it follow from his ethics? I'd like to see how it does if it does? Any ideas?

The point I'm making though is that there's a contradiction that follows from Kant's ethics in the lying to a murderer thought experiment and I'm sure this is just one case of many conflicts of duty scenarios that are possible and also actual but the contradiction can be denied by positing a third option - amorality - and one does that, we're not in the moral sphere anymore - good/bad no longer apply to a person who can't, in Kantian terms, perform faer duty. That it's necessary implies that it's possible. This is where @tim wood comes in - we have to have freedom to perform our duties, Kantian or otherwise.

You're talking about deontic logic but it's an offshoot of modal logic, the latter subsuming the former as it were. Ergo, if something goes wrong at the level of modal logic, it wouldn't work in deontic logic too.

Think of it.

Modal Logic: If necessary P then possible P

Deontic Logic: If necessary P then permissible P

If deontic logic were independent of modal logic then it should be possible for a proposition P to be impossible and yet permissible. That doesn't make sense, right? The permissible supervenes on the possible. — TheMadFool

Deontic logic is a type of modal logic. The usual type is called alethic; there's also temporal, doxastic, and probably other kinds. The formal structures within them are all basically the same, []P iff ~<>~P, and <>P iff ~[]~P, but what exactly [] and <> mean differs between them.

In alethic modal logic, [] is necessity and <> is possibility.

In deontic modal logic, [] is obligation and <> is permission.

There's no modal logic I'm aware of that has [] as obligation and <> as possibility, and it would be really weird if there were. But that only means that possibility is not logically entailed by obligation. It might still be the case that "ought implies can", and Kant certainly thought so, but it would just have to be for other reasons than logical entailment.

(I don't think that's the case myself. I think it's totally possible that things that ought to be might not be possible, and that'd just mean we're stuck with a shitty situation. The practical upshot of "ought implies can", that there's no point blaming or punishing people for things beyond their power to avoid, still holds on my view, but not because of any relationship between obligation and possibility.)

I didn't know Kant supported the death penalty and Wikipedia describes it as an "extreme position". Noted! However, does it follow from his ethics? I'd like to see how it does if it does? Any ideas? — TheMadFool

I don't know off the top of my head of Kant's own justification for his pro-capital-punishment position in light of his broader ethics, but I would expect it would be something along the lines of the full context of an act mattering for the general duty you're following. Instead of "never do X", a duty could be "never do X when Y unless Z". So he might have thought the general rule was "never kill someone who's not actively trying to kill someone else unless as punishment for attempted murder" or something like that. If so, he could just as readily have endorsed a more sophisticated duty regarding lying, and it seems irrational of him to have instead bit the bullet and just insisted that all lying is always wrong all the time no matter what.

There's no modal logic I'm aware of that has [] as obligation and <> as possibility, and it would be really weird if there were — Pfhorrest

IPlease visit Stanford Encyclopedia Of Philosophy. Also why would it be "weird"? Really why? What makes it weird?

I get that the diamond operator is read as permissible in deontic logic but my point is what would be really weird in the truest sense of that word is if someone told me that something impossible is permissible. Take a moment to consider the situation if someone did tell you that a certain action x is impossible but that x permissible. Impossible means you can't do it and permissible is you may do it. That you may or may not do something makes sense only if it's possible to do that thing, right? In the simplest sense, permissible implies that you have an option but impossible implies that you have none!

I don't know off the top of my head of Kant's own justification for his pro-capital-punishment position in light of his broader ethics, but I would expect it would be something along the lines of the full context of an act mattering for the general duty you're following. Instead of "never do X", a duty could be "never do X when Y unless Z". So he might have thought the general rule was "never kill someone who's not actively trying to kill someone else unless as punishment for attempted murder" or something like that. If so, he could just as readily have endorsed a more sophisticated duty regarding lying, and it seems irrational of him to have instead bit the bullet and just insisted that all lying is always wrong all the time no matter what. — Pfhorrest

So, you're of the opinion that Kant's notion of moral duties came with caveats, conditions that made room for his pro-death stance? The only way Kant's pro-death views make sense is if it's a moral duty to execute murderers. Any ideas how we may arrive at that conclusion within the framework of Kant's ethics? Can we reason backwards from the position that the death penalty is justified to a foundational premise in Kant's moral theory? :chin: You may ignore this part of my post as I think it's only incidental unless of course you had a good reason to bring it up in which case I'd like you to give me more to go on. Thanks.

there's a contradiction that follows from Kant's ethics in the lying to a murderer thought experiment — TheMadFool

What contradiction? You haven't read it!

Please visit Stanford Encyclopedia Of Philosophy. — TheMadFool

I have, many times. Do you have a particular part in mind?

Also why would it be "weird"? Really why? What makes it weird? — TheMadFool

On a broad level, it mixes "is" with "ought" in a way that doesn't normally fly. For a narrower example, it would imply that nothing that can (possibly) happen is wrong (forbidden), since "P is forbidden" = "[]~P" (if [] is deontic) and "P is possible" = "<>P" (if <> is alethic), and <>P iff ~[]~P (in all forms of modal logic), which would read as "it is possible that P if and only if it is not forbidden that P", if <> were alethic and [] were deontic.

Take a moment to consider the situation if someone did tell you that a certain action x is impossible but that x permissible. Impossible means you can't do it and permissible is you may do it. That you may or may not do something makes sense only if it's possible to do that thing, right? In the simplest sense, permissible implies that you have an option but impossible implies that you have none! — TheMadFool

But in very different senses of "have an option". It makes perfect sense to me to say that it would be morally okay to do something (that if someone did it there would be nothing morally wrong with that), but also, that nobody is actually able to do it. I think this is very clear when we're talking about practical rather than in-principle possibilities: say someone is about to drown or get hit by a car or something else deadly, and nobody is in a position to act quickly enough to save their life. That doesn't mean that it would be wrong (impermissible, forbidden) to save their life, just because nobody (possibly) can. It just means that a morally okay thing (even a positively morally good thing) is beyond our capabilities.

So, you're of the opinion that Kant's notion of moral duties came with caveats, conditions that made room for his pro-death stance? — TheMadFool

I see no reason why moral duties couldn't be context-sensitive. That wouldn't make them merely conditional imperatives rather than categorical, because a conditional imperative is conditional on the end in mind, not on the context: "(if you want) to achieve X, do Y, otherwise do Z", not just something like "(regardless of what you want) do Y whenever X is the case, otherwise do Z", which as I understand it would still be categorical.

And supposing that Kant had a context-sensitive duty in mind regarding killing seems the simplest way to explain his pro-capital-punishment stance. "Kill people whenever [they deserve it some way or another], otherwise don't kill anyone."

That just raises the question of why the duty not to lie can't depend on context too. "Lie if necessary to save a life, otherwise don't lie", for example. If everyone followed that maxim, it wouldn't destroy language the way "lie whenever you feel like it" would, and so wouldn't result in the contradiction of will that supposedly makes lying categorically impermissible.

The only way Kant's pro-death views make sense is if it's a moral duty to execute murderers. — TheMadFool

Or at least, if it's not a moral duty not to execute murderers.

I have, many times. Do you have a particular part in mind? — Pfhorrest

Well, look up modal logic in SEP and read the section explaining the diamond and box operators. Box = Necessary and Diamond = Possible

On a broad level, it mixes "is" with "ought" in a way that doesn't normally fly. For a narrower example, it would imply that nothing that can (possibly) happen is wrong (forbidden), since "P is forbidden" = "[]~P" (if [] is deontic) and "P is possible" = "<>P" (if <> is alethic), and <>P iff ~[]~P (in all forms of modal logic), which would read as "it is possible that P if and only if it is not forbidden that P", if <> were alethic and [] were deontic. — Pfhorrest

You're all over the place at least that's how it seems to me. Can you break up what you said into two sections 1) Modal logic and 2) Deontic logic and then explain how it's weird that diamond operator should refer to possibility? Please read the modal logic page in Stanford Encyclopedia Of Philosophy and clear your doubt - there's nothing weird about the diamond operator signifying possibility, in fact it's meant to do so.

Deontic logic is a sub-field of modal logic and may have its own peculiar interpretations of what the operators mean but that's beside the point. Read belowl

Taking a different approach, it must be that x is permissible only if x is possible, right? If something is impossible then there's no point in talking about permissibility. If something is impossible then surely the question of whether it's permissible doesn't arise. It's a mistake to think of permissibility in the impossible for the simple reason that the former offers no options and the latter depends on options, a contradiction: no option (impossible) & option (permissible).

Coming to what I said: []P -> <>P, it's listed as an axiom in the Wikipedia modal logic entry. See here

If necessary that P then possible that P. In the context of the lying to a murderer thought experiment in Kant's ethics

1. If it's necessary that (you shouldn't lie AND you should save your friend) then possible that (you shouldn't lie AND you should save your friend)

2. It's not possible that (you shouldn't lie AND you should save your friend)

Ergo,

3. It's not necessary that (you should lie and you should save your friend)

And 3 is where Kantian ethics ends - there are no duties to fulfill for the reason that it's not possible to fulfill them.

context-sensitive duty — Pfhorrest

One hint, One word, universalizability.

1. You tell the truth. Your friend dies but you've told the truth. You're good (you fulfilled your duty to truth) AND you're bad (you failed your duty to value life)

2. You lie. You save your friend but now you've lied. You're good (duty to value life fulfilled) AND you're bad (you fail your duty to truth) — TheMadFool

3. You tell the murderer that you’re not going to tell him where your friend is because you don’t want him to kill anyone. You’re good - you’ve fulfilled your duty to truth AND to value life.

It’s not that hard.

I think that part of the problem with applying Kant's system of thought to practical ethical situations is that his whole emphasis was on moral duties, in terms of motivations rather than ends. In the present time, we are more concerned with the effects of our behaviour. The example of the conflict of duty that you give is the most extreme. However, for many ethical decisions we make on a daily basis it is so much more complicated applying his whole system of logic, especially the principle of universalisation because we are often confronted by the particulars and the unique.

You're all over the place at least that's how it seems to me. Can you break up what you said into two sections 1) Modal logic and 2) Deontic logic and then explain how it's weird that diamond operator should refer to possibility? — TheMadFool

There's nothing weird about the diamond operator meaning possibility, in an alethic modal logical, wherein box means necessity. What's weird is if you mix alethic and deontic modes like you suggest.

If diamond means possibility, then box has to mean necessity -- alethic necessity, not deontic obligation -- or else you get problems like the example I gave, where nothing you can do is forbidden.

Conversely, if box means obligation, then diamond has to mean permission -- deontic permission, not alethic possibility -- or else you get the same problem.

It's still possible to argue about whether or not obligation requires possibility, but you can't just mix together deontic and alethic modes like that to "prove" it without causing strange consequences like "nothing you can do is forbidden".

It's not possible that (you shouldn't lie AND you should save your friend) — TheMadFool

The way you've phrased it, that's not so. It can be the case both that you shouldn't lie, and that you should save your friend; intuitively we'd usually agree it is, both of those "should" claims are true, since those are both good things, and since they are both true they can be both true.

But it simultaneously might be the case that you can't both not lie and also save your friend. Note the absence of "should"s here. Maybe it can't be that you don't lie and you do save your friend. "Do", rather than "should".

It's possible for both states of affairs to be moral (for both "should" statements to be correct), but it might not be possible for both states of affairs to be real (for both of the corresponding "do" statements to be correct).

There is still room here to argue about whether or not the impossibility of that combined state of affairs means it must be an omissible (non-obligatory) state of affairs too, but you don't get the conclusion that it does for free just out of the structure of the logic.

One hint, One word, universalizability. — TheMadFool

Universalizability means that it applies for all similarly-situated moral agents.

Or to quote Wikipedia, "the most common interpretation is that the categorical imperative asks whether the maxim of your action could become one that everyone could act upon in similar circumstances".

The circumstances can be accounted for without compromising the universalizability.

There's nothing weird about the diamond operator meaning possibility, in an alethic modal logical, wherein box means necessity. What's weird is if you mix alethic and deontic modes like you suggest. — Pfhorrest

Oh. I get what you mean. However, deontic logic supervenes, if not in entirety at least in the part that's got to do with relationship between possibility and permissibility, over modal logic, right? Modal logic is about raw possibility and necessity, deontic logic is about moral possibility and necessity. If something is impossible in modal logic, there's no point in discussing it's permissibility in deontic logic. Deontic claims must first run the gauntlet of modal logic before deontic logic can even get its hands on them. All I'm saying is we must first deal with possibility before we start to work on moral possibility. Reasonable?

Universalizability means that it applies for all similarly-situated moral agents.

Or to quote Wikipedia, "the most common interpretation is that the categorical imperative asks whether the maxim of your action could become one that everyone could act upon in similar circumstances".

The circumstances can be accounted for without compromising the universalizability. — Pfhorrest

Kant used the example of lying as an application of his ethics: because there is a perfect duty to tell the truth, we must never lie, even if it seems that lying would bring about better consequences than telling the truth — Wikipedia

I don't see and context-sensitive assertions being made. :chin:

However, deontic logic supervenes, if not in entirety at least in the part that's got to do with relationship between possibility and permissibility, over modal logic, right? — TheMadFool

"Modal logic" isn't just alethic modal logic, the logic of necessity and possibility. Alethic modalities are just one kind of modality. There are different kinds of modalities, and they don't necessarily have to have any relationship to each other; as in, it's not baked into the logic itself.

I don't see and context-sensitive assertions being made. :chin: — TheMadFool

Yes, and that's the perplexing thing about Kant on lying, because his system generally seems to permit maxims that take context into account, and in other opinions (such as about capital punishment) he seems to take context into account, so why not on lying?

"Modal logic" isn't just alethic modal logic, the logic of necessity and possibility. Alethic modalities are just one kind of modality. There are different kinds of modalities, and they don't necessarily have to have any relationship to each other; as in, it's not baked into the logic itself — Pfhorrest

I don't get it at all but that probably says more about my ignorance anything else. My reasoning is simple:

1. Before we can consider whether a person is permitted to buy a drink at a bar, we must consider whether that's even possible, right? It doesn't make sense to discuss whether your pet dog, assuming it's not an undercover MIB agent, is permitted to buy a drink at the local bar for the simple reason that it's impossible for it to do so.

2. Before we consider moral possibility (permissibility), we must first deal with possibility itself

Permissibility supervenes on possibility, Without the latter, the former just doesn't make sense.

Yes, and that's the perplexing thing about Kant on lying, because his system generally seems to permit maxims that take context into account, and in other opinions (such as about capital punishment) he seems to take context into account, so why not on lying? — Pfhorrest

My hunch is Kant's opinion on capital punishment - his pro-death stance - is totally unrelated to his ethical theory. Kant was merely expressing an opinion that had nothing to do with his deontological ethics. If you think otherwise, kindly show me which specific proposition(s) in his moral theory lead to the conclusion that the death penalty is necessarily just.