Smullyan's Paradox

[Flying Pig]

Well, it's not a paradox; you just simplified it in a way that's not a paradox either, it's just nonsensical. It's akin to saying 'is my pet a cat or an animal?' - it's not clever, interesting or paradoxical; it's meaningless and trivial.

 

[Gori the Grey]

The definition of a paradox is a situation that includes a contradiction with no apparent solution.

Not even my definition. Leoreth's in post 42.

I agree that it's trivial. I made my second intervention in this thread when it seemed to me that people were taking this as a prompt for serious legal and ethical reflection, when I regarded it merely as a little game, a bit of nonsense, as you say.

I don't agree that it's uninteresting. It's sustained our collective interest for four days now.

 

[Leoreth]

Don't put too much stock into this definition because I basically formulated it ad hoc.

 

[El_Machinae]

I've really enjoyed this thread, and I've gone back to thinking about it. It's just so tough for me (now) to call B a murderer. This is because B did not deprive C of water. He deprived C of poison.

If A had filled C's canteen with sand, and then, in the middle of the night B stole C's canteen and ran away ... did B cause C to die of thirst? Well, no? There was no water in the canteen when it was stolen. In a similar way, there was no water in the canteen when B punched a hole in it. Yes, the poison was diluted in water, but the canteen was not capable of preventing dehydration anyway.

If A hypnotised C into thinking that canteen sand was water, so C would drink a bit of sand every once in awhile until he died ... B still would not have stolen water

 

[Gori the Grey]

Don't put too much stock into this definition because I basically formulated it ad hoc.

You put enough stock in it, at the time, to add the (what I regarded as snide) "That's pretty straightforward."

El Machinae, you're experiencing Smullyan's Paradox as a paradox. I think that going back and forth is what Smullyan designed. I keep hearing that its not a paradox, though I do everything in my power to frame it up as one.

 

[Flying Pig]

I've really enjoyed this thread, and I've gone back to thinking about it. It's just so tough for me (now) to call B a murderer. This is because B did not deprive C of water. He deprived C of poison.

He deprived C of poisoned water; had A put sugar in the water instead of poison, he'd still be a murderer.

 

[Mise]

I've really enjoyed this thread, and I've gone back to thinking about it. It's just so tough for me (now) to call B a murderer. This is because B did not deprive C of water. He deprived C of poison.

If A had filled C's canteen with sand, and then, in the middle of the night B stole C's canteen and ran away ... did B cause C to die of thirst? Well, no? There was no water in the canteen when it was stolen.

In this case, I agree that A killed C, while B did not. C died due to dehydration, which is due to the water in his canteen being removed, which was caused by A.

In a similar way, there was no water in the canteen when B punched a hole in it. Yes, the poison was diluted in water, but the canteen was not capable of preventing dehydration anyway.

I'm not buying this part. C dies because of dehydration. Poison does not cause dehydration. Poisoned water does not cause dehydration (it causes death by poisoning). A cannot be the cause of C's death. There being no water in the canteen is what caused dehydration. This was caused by B, therefore B caused C's death.

Poisoned water may not be able to prevent dehydration, but neither can a map. If person D had given C a map before the trip, would D be responsible for C's death? No. "Not capable of preventing dehydration" isn't enough to assign causation.

 

[lovett]

@Lovett: I thought consequentialism was about the "expected" or "intended" consequence of the action, rather than the actual, factual consequence. So someone who attempts but fails to murder someone is still in the wrong morally -- just that his "wrongness" was due to the expected/intended/potential consequence of his act, rather than because it defies some rule of ethics for one's behaviour, or because it makes him an unvirtuous person, or whatever. In other words, consequentialists would say that what is wrong about attempted murder is that somebody could have died; deontologists would say that was is wrong about attempted murder is that it is carrying out an act that violates an ethical rule of theirs; virtue ethicists would say that it is incompatible with being a virtuous person or whatever. But they would all agree that attempting to murder someone is equally wrong, whether or not they were successful. I don't think there's an ethical system that distinguishes between an attempted murderer and an actual murderer; indeed, I think that if there were such a system, it would be rightly criticised for doing so!

I could be wrong about consequentialism, but that was my understanding of it.

Well ,that's one form of consequentialism. There are many. What you're talking about is known as motive-consequentialism (or, one form thereof). It assess actions (as right or wrong) on the basis of the motives which lead to them (usually: were those motives the motives that normally lead to the best consequence). But motive-consequentialism probably isn't the most widely known form of consequentialism. Act-consequentialism is the most widely known form of consequentialism. Act-consequentialism assesses acts on the basis of their consequences alone. So an act which has good consequences is right and one which doesn't is wrong. This is the form of consequentialism to which Bentham and Mill (actually, Mill is hard to place) ascribed (and for that matter, Sidgwick and Moore). This sort of consequentialist say an attempted murder is wrong if and only it leads to bad consequences. If an attempted murder fails, and actually brings about world peace or something, it is right.

But what we are talking about here, remember, is not standards of rightness. It is standards of responsibility. The two are indubitably linked, but the former does not imply the latter (conceptually). The former determines when an action is right or wrong (obligatory/not obligatory), the latter determines when someone is responsible for a wrong - when someone is blameworthy. Consequentialists have historically had a bit of trouble with standards of responsibility. The classical utilitarian response is to say that people should be held responsible for an action -blamed- if and only if blaming them is the thing that will produce the best consequences - because blaming itself is an action an should be assessed by the standard of rightness of all consequences. This, to many, has seen like a shallow account of moral responsibility.

What arises from the consideration in this thread is an account of the proper standards of responsibility. Those standards can't be based on whether one has caused a particular state of affairs to obtain, because this makes them arbitrary. So, killing someone is not the thing that makes one morally responsible - blameworthy. What makes one morally responsible -guilty- is one's attempt to kill. One's intention. Or, more likely, what makes one morally responsibility is something more holistic about one's character; perhaps one's propensity to respond to to immoral reasons and not moral one's (there's some very interesting account like this). In any case, what makes one responsible for an event isn't located in one's causal relation to the event itself. It is located in something about you; your character, your intentions or whatever. The locus of repsonsibility is the agent.

One might also get some very serious results from this. A lot of people don't think the standards of rightness and responsibility are different. If they're not different, then these considerations amount to arguments against any sort of act-based moral theory.

I'll tell you what I think is the problem there. As soon as the ethical consideration causes us to change our focus from "death by dehydration" to "the fact that C ended up dead," the causal situation becomes exactly like your window-smashing analogy: A is responsible because he rendered B's actions moot.

edit, afterthought: It's tantamount to asking did C die or did he die of dehydration? Do you see a paradox, or enigma, or conundrum, or mind-bender, or brain-teaser in that?

I don't see what your trying to get at here, your going to have to explain. Nor do I see the paradox in your question. As Flying Pig said the answer is quite simply; yes. C both died and died of dehydration.

 

[Gori the Grey]

What is the "great wrongdoing" of which A is guilty that is "equally bad" to B's killing C?

 

[Flying Pig]

Poisoned water may not be able to prevent dehydration, but neither can a map. If person D had given C a map before the trip, would D be responsible for C's death? No. "Not capable of preventing dehydration" isn't enough to assign causation.

And yet if we're saying that he 'gave C poisoned water', we must also say that he took away C's unpoisoned water, since he turned one into the other - had he replaced the contents of his flask with a map, he would have caused his death by dehydration. And yet I maintain that C died because he could not drink, and C would have drank poisoned water, and so C died through B's actions, not A's.

 

[Mise]

And yet if we're saying that he 'gave C poisoned water', we must also say that he took away C's unpoisoned water, since he turned one into the other - had he replaced the contents of his flask with a map, he would have caused his death by dehydration. And yet I maintain that C died because he could not drink, and C would have drank poisoned water, and so C died through B's actions, not A's.

Well you just split the cause again don't you. If A replaced all the water in the canteen with a map, then C would die due to dehydration caused by there being no water in the canteen. If B then replaces the map-filled canteen with an empty canteen, then C dies due to dehydration caused by the new canteen being empty.

I admit that in this situation, it is now easier to say that B attempting murder while A is committing it. 50/50 maybe. In the case with the poisoned water, I am more confident that B caused C's death, while A attempted but failed to murder C.

 

[Gori the Grey]

Nor do I see the paradox in your question. As Flying Pig said the answer is quite simply; yes. C both died and died of dehydration.

I spelled it out more in a later post. You can't just say yes to both.

If C died, A is a murderer and B an attempted murderer. If C died of dehydration, B is a murderer and A an attempted murderer.

If (under the ethical consideration what is bad is that) C died (no concern to cause), then A is a murderer (on your glass breaking analogy) and B is an attempted murderer (his attempt preempted by A's getting there first). The alternate modes of death are removed, so it becomes just like which brick shattered the glass.

If (under the causal consideration), C died of dehydration, B is a murderer (as you've shown elaborately through the thread) and A is an attempted murderer.

If you say "In the case of C's death: A is the murderer and B an attempted murderer AND B is the murderer and A is an attempted murderer," you are saying something paradoxical.

 

[Flying Pig]

Well you just split the cause again don't you. If A replaced all the water in the canteen with a map, then C would die due to dehydration caused by there being no water in the canteen. If B then replaces the map-filled canteen with an empty canteen, then C dies due to dehydration caused by the new canteen being empty.

I admit that in this situation, it is now easier to say that B attempting murder while A is committing it. 50/50 maybe. In the case with the poisoned water, I am more confident that B caused C's death, while A attempted but failed to murder C.

I agree with your conclusion, but only because I reject the analogy - that hinges, however, on the belief that poisoned water hydrates.

 

[El_Machinae]

In this case, I agree that A killed C, while B did not. C died due to dehydration, which is due to the water in his canteen being removed, which was caused by A.

I agree that B caused the dehydration death scenario, I just don't agree that B took away the water, since there was nothing in the canteen that could have prevented dehydration

 

[Cheezy the Wiz]

Which it may very well. The poison may be neurotoxic, and kill by asphyxiation due to paralysis. Puffer fish poison works in precisely this manner.

 

[Mise]

There's dehydration due to poisoning*, and there's dehydration due to the canteen being empty. C died of the latter, not the former.

*-either because the water is not potable or because the poison literally causes dehydration in the same manner as a Puffer fish.

 

[Flying Pig]

In which case you'll die from something other than dehydration, as Cheezy points out - but then, that might be right: except that then we have to accept that there's also dehydration due to not having milk, dehydration due to not having tea, and so on.

 

[Mise]

Well the more you have to split the cause into narrower and narrower distinctions, the more it resembles throwing a brick through a window that someone else has already broken.

 

[Leoreth]

You put enough stock in it, at the time, to add the (what I regarded as snide) "That's pretty straightforward."

The definition was serviceable for the point I was making at that moment. Had you contested it, I would have done some research to revise it; I was certain that my argument didn't depend on that.

 

[Flying Pig]

Well the more you have to split the cause into narrower and narrower distinctions, the more it resembles throwing a brick through a window that someone else has already broken.

On that note, it seems to me more like seeing a brick flying towards a window and shooting a bullet through the brick, into the window. Would it have broken had you done nothing? Yes. Did you prevent it being broken by that method? Certainly. Did you break the window? Absolutely!

That analogy certainly seems to make it clear that B is the murderer, and A the attempted murderer.