Some Thought Experiments Involving Assassination

By On August 2, 2012 · 53 Comments · In Uncategorized

1. Suppose an evil demon appears before you and says, “I plan to kill hundreds of thousands of foreign civilians and destroy their country’s architecture unless you kill this one innocent person.” Under these extreme circumstances, might it be permissible for you to kill that innocent person?

2. Suppose an evil demon appears before you and says, “I plan to kill hundreds of thousands of foreign civilians and destroy their country’s architecture unless you kill this Mafia don, a criminal who has himself killed many people and who plans to kill many more.” Under these extreme circumstances, might it be permissible for you to kill that Mafia don?

3. Suppose an evil demon appears before you and says, “I plan to kill hundreds of thousands of foreign civilians and destroy their country’s architecture unless you kill the president.” Under these extreme circumstances, might it be permissible for you to kill the president?

4.  Suppose the evil demon possesses the president. The evil demon, in the guise of the president, plans to invade a foreign country. Suppose you know that the invasion is unjust–it clearly violates the correct theory of just war. Suppose you also know that the war will kill hundreds of thousands of foreign civilians and destroy their country’s infrastructure. Suppose killing the demon-possessed president will stop, or at least has a good chance of stopping, the invasion. Under these extreme circumstances, might it be permissible for you to kill the president?

5. Suppose there is no evil demon. However, suppose the president, though not possessed by an evil demon, acts just like the possessed president in 4. The president appears before you and says, “I plan to invade a foreign country.” Suppose you know that the invasion is unjust–it clearly violates the correct theory of just war. Suppose you also know that the war will kill hundreds of thousands of foreign civilians and destroy their country’s infrastructure. Suppose killing the president will stop, or at least has a good chance of stopping, the invasion. Under these extreme circumstances, might it be permissible for you to kill the president?

 

Tagged with:
 
  • Thoreau

    Since law enforcement agencies have internet connections, I will answer in a police-friendly manner and say yes to killing the Don, no to killing the President, and “only if we can make it look like a botched SWAT raid” for killing the civilian.

  • Daniel Kuehn

    Maybe, yes, maybe, yes, yes.

    I never understood why we distinguish between assassinations, executions, targeted killings, and enemy military casualties in the first place. They’re all variations on the same theme. In fact, the one that people get least upset about – the last one – is the one they should be the most upset about (although I still think it’s OK in certain circumstances).

    • http://www.facebook.com/les.nearhood Les Kyle Nearhood

      I think people are most upset not about military killings but about collateral damage to the populace. Although killing is killing, there is a moral difference between shooting at people who are shooting back, and destroying civilians.

  • Sulla

    No to all 5.
    I have no warrant or permit go around killing people, of any kind or character, with or without input from an evil demon.

  • Jonathan Watson

    1. I do not believe so.

    2. I do not believe so.

    3. I do not believe so.

    4. It may be permissible.

    5. It may be permissible.

  • Craig

    Jason Brennan, you just might be my favorite person ever. Ever. I didn’t think it possible, but this post rivals your “Dear Moderate Left” post. I hate to be a broken record, but man do I love this site!

  • Chad Nelson

    i’m sure this will repulse a lot of readers, but Rothbard’s Ethics of Liberty grapples with these issues as well as I’ve ever seen.

    • http://www.facebook.com/profile.php?id=1702318862 Jason Brennan

      Could you tell me which pages or chapters to read on that? I’ve been searching, mostly in vain, for philosophical writing on assassination.

      • http://www.facebook.com/profile.php?id=1702318862 Jason Brennan

        Are you thinking of chapter 24. “Moral Status of Relations to the State”?

  • http://www.facebook.com/curt.doolittle Curt Doolittle

    False Premises. a) Argued on moral not contractual grounds in order to mislead the reader. b) Morals are general contractual rules for peaceful mutual exchange, and war by definition is outside of that contractual environment, c) Just war is not a moral but a contratual proposition between parties who seek to limit their own costs (See Kagan).

    Error of the common parlor game, which poses false dichotomies in order to confuse the participants into thinking (like the train-lever parable) that morals are arbitrary rules, rather than that morals are statements of property rights loaded with emotional content so that they propagate more easily. The error here is confusing a utilitarian norm with a moral absolute truth. The first is the fact. The second is a meme.

    1) Humans war. They always have and always will. It is impossible to resolve all conflicts by peaceful means.

    2) The demon and the president are participants in a war.

    3) As participants in the war they are outside daily civil legal and moral prohibitions we have constructed for peaceful interactions: our prohibition on violence does not apply. War revokes the prohibition on non violence. That is the purpose and point of demarcation of ‘war’.

    4) Moral rules are general rules. They are a shortcut that allows us to propagate contractual terms which help us reduce our error in calculating property transfers when they are beyond our perception and knowledge. Moral rules are not abstract truths. The confusion is created by the priority one gives to the genetic structural categories of family, tribe, and nation, versus the egalitarian structure limited to the categories of the individual and humanity. Much religious content seeks to extend the familial category to the universal as a means of creating an opposition to the state. And approaching questions of property as questions of morality is an artifact of applying religious techniques that seek to simplify complexity into emotionally accessible social rules, to what are practical contractual constructs the articulation of which is too complicated for general use.

    5) There is is no longer a genetic composition to war – the need to fight other tribes for genes to persist – which necessitates one’s participation in tribal war. Wars are now, and have been for a long time, conducted for economic interests, even if those economic interests apply only to the costly norms, status signals, property rights portfolios, and political systems that vary between groups. Therefore the individual is free to choose sides.

    6) As free to choose sides, one may calculate his interests and those interests of those with whom he shares interests, and determine if he is benefitting or harming those with whom he shares interests. And if it is in his interest and the interest of those with whom he shares interest, then he may act to kill the demon/president/minister/general or not at his will.

    Propertarianism is correct: all human ethical and political statements can be reduced to property rights, and done so without contrivance. That is because all morals and all human moral feelings, are expressons of property rights when property rights are articulated such that they fully encompass the entirety of those things which humans treat as property.

    It is hard to do this topic justice in short form. But hopefully this is enough of a sketch to illustrate the problems of both moral parlor games, and treating war as other than a utilitarian construct.

    • http://www.facebook.com/curt.doolittle Curt Doolittle

      Oh. And:
      1. Nonsensical
      2. Nonsensical
      3. Nonsensical
      4. Nonsensical
      5. Possibly.

    • http://www.facebook.com/profile.php?id=1702318862 Jason Brennan

      Huh?

      • http://www.facebook.com/curt.doolittle Curt Doolittle

        That’s OK Jason.
        My attempt at a sketch didn’t succeed.
        Edited it a bit for clarity. Perhaps that helps.

        But, criticism of method aside, keep up the good work anyway. :)
        I’ll take another shot in a better forum when I have more time.
        Cheers

  • simon…

    The problem with these questions that you are asking if one would do a categorical action for a hypothetical cause.

  • Jod

    I wriggle out of the either or by asking whether killing is my only option. I am subsequently violently assaulted by moral philosophers whose livelihood depends on cultivating irreconcilable clashes between consequentialism and deontology.

  • http://www.facebook.com/les.nearhood Les Kyle Nearhood

    Only the last one is a real question because if you are confronted with an evil demon then you should know that one of two things are true. Either you are hallucinating, or it is a real demon in which case you should in no way do anything ti says to do, for any reason.

    The fifth question however requires some real thought. In a Republic I would say you are unjustified in killing the president because both of the reason that you cannot foresee all of the consequences of either action or inaction, and because a Republic will have self correcting mechanisms.

    If the President, however has already acquired dictatorial powers and in unlikely to ever step down, then, and only then, you may be justified in the killing.

  • Sean

    I’m inclined to say yes to all five, but especially to the fifth.

    Like most people, I don’t think deontic proscriptions are absolute; they can be overridden if consequences are severe enough. The consequences in 1-4 strike be as sufficiently bad to justify killing an innocent person, but that isn’t *totally* clear to me (just how bad do they have to be, exactly?).

    4 raises the additional question of the moral status of innocent aggressors, which I think is pretty difficult.

    5 seems to be a clear-cut case of self-defense.

  • http://independent.academia.edu/DannyFrederick Danny Frederick

    1. No. But I might say ‘yes’ if the catastrophe was going to happen through natural causes rather than by means of the action of an evil demon (or moral agency in general).
    2. Not sure, but inclined to think ‘yes.’ But that may be because I think I am entitled to kill the sonovabitch in any case.
    3. No. Same as for 1.
    4. No. There is a whole set of institutions involved in executing, facilitating or preventing the president’s action: it is beyond my authority to stop him.
    5. No. Same as for 4.

    Just in case you have not already read it, Thomson’s ‘The Realm of Rights,’ chapters 5, 6 and 14, has a good discussion of some of the issues.

    • http://independent.academia.edu/DannyFrederick Danny Frederick

      I should also have mentioned chapter 7 of that book (though I don’t like it) and the numerous articles she has published over many years on trolley problems and self-defence, giving us conflicting views at different times.

    • http://www.facebook.com/les.nearhood Les Kyle Nearhood

      You would have to assume the demon was lying because, well, it’s a demon. and as for number 5. I agree in a republic there are self correcting mechanisms. So unless the president has already seized dictatorship then the killing is unjustified.

      • http://independent.academia.edu/DannyFrederick Danny Frederick

        You might just as well say “you would have to assume the demon was telling the truth because, well, it’s a demon” (and demons do bad things).

  • Daniel Kuehn

    I’m surprised so many people aren’t at least giving a “maybe” to the first three.

    Look, with those stakes there are a lot of people who might be OK with you killing them to prevent that kind of carnage. One can imagine a much smaller situation with your spouse and your two children and an assertion by a madman that if you don’t shoot your spouse he’ll kill the two children. In that situation your spouse may very well demand that you shoot him/her.

    Certainly if hundreds of thousands of lives are at stake such self-sacrifice is plausible.

    It would be an ugly task to be the one to pull the trigger, but I don’t see why it should be considered so morally repugnant.

    Of course a lot of this depends on what the person you’re killing thinks and how credible you think the demon is.

    Anyway – I just felt the need to add that because I felt like an asshole after reading how many people wrote “no”.

    If a demon actually made this offer and I was the innocent person I’d have to seriously consider whether I should give my life for all them. I don’t know what I would choose but I can guarantee that I would feel deep guilt if I chose my own life, even if nobody begrudged me that choice.

    • http://independent.academia.edu/DannyFrederick Danny Frederick

      You are altering the scenarios if you say the innocent person consents to be killed.

  • MARK_D_FRIEDMAN

    Jason,
    You say in para #5: “Suppose you know that the invasion is unjust–it clearly violates the correct theory of just war. Suppose you also know that the war will kill hundreds of thousands of foreign civilians and destroy their country’s infrastructure.” Although you don’t say this, and I’m not claiming that you believe it, the second sentence here strongly suggests that the reason the planned invasion is unjust is because of the cost in innocent lives. I see no other purpose served by the second sentence.

    If this is so, then the obvious argument in favor of killing the president is that the end justifies the means. Clearly extra-judicial killings are bad, and should be avoided. Here, you seem to be suggesting that this evil might be outweighed by the need to save the innocent. I would just point out that if this logic is employed, it is a very short step to legitimizing drone strikes or even preemptive wars, given the same level of certainly of a good outcome you posit, i.e. drone strikes or preemptive wars will “stop, or at least ha[ve] a good chance of stopping” the killing of a greater number of innocent persons.

  • http://www.angryblog.org/ Brian Patrick Moore

    The first 4 don’t matter too much, since as far as I can tell, the scenarios (even replacing the demon with a person) have never occurred, and never will. The last scenario, at least the invading other countries part, happens all the time. But even then, the question is: if we could prevent a great harm, can we kill the one intending to cause it? It’s extremely rare that actually killing a country’s leader (especially a democratic-ish country, as implied by ‘president’) will reduce the war fervor of a country. More likely I would think it would stoke it.

    But let’s assume that killing the prez prevents the war, as stated. What else does it then cause? Does it cause an immense crackdown/security state apparatus? Does it cause repression in response? Does it cause an even worse VP to come into office? Does it create an atmosphere where it is acceptable to kill politicians (for better or worse) for what they intend to do? I would think those things matter a great deal.

    But let’s again assume we can minimize all that, perhaps by killing the president and replacing him with a pacifist clone, and no one can tell the difference. He is intending to harm other people, and if you don’t have a non-fatal way to stop him, then it is perfectly permissible to kill him. If a known serial killer (most national executives who carry out wars have a higher body count than serial killers) walks up to you and says “I intend to shoot that fellow over there” and you have good reason to believe that indeed will happen, you are permitted to kill him in order to stop him, and I think our courts would agree — though obviously not on the presidential version.

    The point of these questions is to think about reality. And I think, in reality, it is morally permissible to kill governmental leaders who are responsible for committing (or credibly intending to commit) actions such as murder in order to prevent further harm — However, and this is an extremely large however, except in very strange circumstances, the real life fallout from doing so almost always is worse than whatever the evil president was going to do anyway. And you are almost always better served, in the cause of minimizing harm, in going about your opposition to his evil policies in a peaceful manner.

    • http://www.angryblog.org/ Brian Patrick Moore

      I think this is why assassinations were more standard policy back in the day of absolute and/or hereditary leaders, and pre-nationalism. If you are say, opposed to us invading Iran, and you kill the president, his VP and congress are probably just going to invade Iran to the cheers of voters. Back in the day, if you whacked the king, there was a decent chance that might truly change the policy you didn’t like, because his brother/son/regent/chancellor might have a very different policy goal. And the peasants certainly didn’t care that much.

  • ben

    “Suppose you know that the invasion is unjust–it clearly violates the
    correct theory of just war. Suppose you also know that the war will kill
    hundreds of thousands of foreign civilians and destroy their country’s
    infrastructure.”

    But you *wouldn’t* know that, you could – at best – “assume” or “believe” it.
    On the other side, you *would* know that killing the innocent person / don / president would cause him to be dead.

    This, together with my general deontolocigal stance, always pushes me in favor of “no” in these kinds of questions.

    • http://independent.academia.edu/DannyFrederick Danny Frederick

      This kind of objection, which has been raised by a number of commenters (so excuse me picking on you – it is not personal), misses the point. Imagine a mathematical or logical problem: from these mathematical facts, or from these premises, does it follow that p (for some specific proposition, p)? You have misunderstood the point of the question if you say things like, ‘but we cannot know those mathematical facts, or whether they are facts,’ or ‘the premises might be false,’ or ‘we could never know those promises.’ The question is one about what follows from what. In the examples AS DESCRIBED would it be permissible that p? The point of raising such questions is to clarify or challenge our moral views.

      • TracyW

        Perhaps you’re missing the point – that many of the responses to these questions clarify how much of our moral views depend on uncertainty.

        In an imaginary world where we knew everything, what’s moral to do probably would be quite different in this one, where we don’t know with certainty that an assassination would save hundreds of thousands of lives, and also we don’t know with certainty the motives of any demon supplying us with such options. In cases 4 and 5, if matters were known with certainty, wouldn’t congress have stopped the president’s plans already, without needing assassinations?

        • http://independent.academia.edu/DannyFrederick Danny Frederick

          I concede your point (which I will express differently) that uncertainty is always a feature of our decisions about how to act. But when someone raises the question, ‘What is it permissible to do under circumstances C,’ you are changing the subject if you start to raise questions about how we would know that we are in circumstances C.

          Consider an astronomical example. Suppose that there had been only three planets orbiting the sun (Earth, Venus and Mars, say). What would their orbits and perturbations have been according to Newton’s theory? Someone objects: but there might be other planets we do not know about which cause perturbations to the orbits of those three planets, because we could never be certain that the three planets we had discovered were the only ones there. Surely you see that that objector has not understood the purpose of the exercise (to better understand Newton’s theory). Of course, the objector is right: we could not be certain that there were no other planets (or other forces, for that matter) interfering in the situation. But the point is irrelevant if what we are trying to do is to consider a hypothetical situation, which is merely posited, in order to discover, or better understand, the implications of a theory. Such an objector would prevent us from understanding anything much.

          Or take an example from economics. Suppose demand for good A rises significantly, and there are no other relevant changes in the system, what will happen to price? Someone objects: but we could never be certain that there are no other relevant changes in the system. What the objector says is correct. But it is irrelevant. If he is in the class, no one is going to learn anything.

          I think this answers the objection you raised. But there is another point which you might also have had in mind, and which you might have run together with the objection I stated. You say: “much of our moral views depend on uncertainty,” and you add, perhaps in clarification, “in an imaginary world where we knew everything, what’s moral to do probably would be quite different in this one.”

          What you seem to be saying there is that changes in my knowledge of the facts may bring about changes in what I ought to do. This reflects a venerable, but I think mistaken, philosophical tradition (‘the subjective ought’). I think that what we ought to do, in any given situation, is an objective matter of fact. What we think we ought to do will depend upon our knowledge and conjectures. But if we do what is most reasonable given out knowledge and conjectures, i.e., if we act according to the subjective ought, we may still be acting wrongly (if our views about what we ought to do are incorrect, i.e., if they do not line up with the objective ought). If we do act wrongly under such circumstances, our behaviour, though wrong, may well be excusable (given our epistemic situation). But I mention this as an additional point, since it is logically independent of your objection and my response to it. My reason for expressing your objection differently to the way you expressed it was simply to separate the two points.

          • TracyW

            I agree with you that it is changing the subject from what the author intended, but I don’t think that’s a bad thing.
            Your astronomical example is worded differently to Jason Brennan’s scenarios, Brennan doesn’t limit our considerations in any such equivalent way. He simply starts off with the demon offering the deal.

            It’s like the barometer problem, where a student responds to a question about using a barometer to measure the height of a building by providing a series of solutions that didn’t touch on difference in pressure readings at the bottom and top of the building. The student, if such a student existed, was asking the question correctly, even though they weren’t answering what the tester was expecting them to answer.

            And I note that this is not a student-teacher relationship on this blog, Brennan might well be asking questions to learn things directly himself (of course a teacher typically learns a lot from teaching anyway).

            On your second point, I was talking about our meta-knowledge. Changes in our confidence about the facts may bring about changes in what we ought to do. Take the case of “beyond reasonable doubt” standard for finding someone guilty in a criminal trial. If you, as a juror, think that on the balance of probabilities someone is guilty, but also that some other reasonable-sounding scenario could explain the evidence, then under this standard you should vote to acquit, while if you can’t think of any other reasonable-sounding scenario to explain the evidence, then under this standard you should vote to convict.
            To put it another way and returning to Breenan’s example here, before killing someone in cold blood, I suggest it’s a good idea to have a lot more confidence in the arguments for killing them than is necessary for less extreme actions, such as levying a fine..

          • http://independent.academia.edu/DannyFrederick Danny Frederick

            I think it is always appropriate to point out that a given postulated situation has different possible completions which would lead to different results. So, if someone asks
            ‘Is it the case that p if C?’
            it is appropriate to respond
            ‘If C and X holds, then p, but if C and not-X holds, then not-p.’
            But that is different to saying ‘we could not be sure that C holds.’ It is that second type of response that I have said misses the point.

            The barometer problem is just a badly worded question. Perhaps Jason’s question was badly-worded. But again that is a different objection to it than the objection that says ‘but we could not be sure that the situation you described is the case.’

            When I spoke of the objector holding up the class, I did not mean to imply that this blog is a student-teacher relationship or that the sorts of problems posed by questions of the type Jason asked only come up in student-teacher relationships. It was just meant as a (weakly) humorous comment on the example I gave.

            Yes, of course in the case of the jury, questions of degree of certainty may actually determine what they ought to do. That is a nice example. But that is a special type of case because it is a specific obligation of the jurors to act one way or another depending on whether there is reasonable doubt. Normally, what one ought to do is a fact which is quite independent of what evidence you have.

            I think your last point is a legitimate criticism of Jason’s questions. Presented with the situations he describes, you are quite right to point out that there is not enough information to answer the question, because we don’t know whether the demon will actually do what he says. I think the example might have worked better if he had said: there is an evil demon who will kill etc., unless you etc. That would avoid your objection.

            You should note, however, that the sorts of objections I was criticising (they are made by several people in this thread) are not like your objection. Instead they say that the hypothesised situation could not arise, or we could not know that we were in that situation, etc. My criticisms of those objections stand.

          • TracyW

            Normally, what one ought to do is a fact which is quite independent of what evidence you have.

            That may be. But Jason Brennan is not asking about normal decisions, he’s asking about life-and-death decisions. And the thing about death is that it’s irreversible. If there’s any case where we should make an exception to your rule, it’s life-and-death decisions made in cold blood.

          • http://independent.academia.edu/DannyFrederick Danny Frederick

            Life-and-death decisions are quite normal: they are made in hospitals daily, for instance. I did not intend ‘normally’ to exclude life-and-death decisions. I intended to exclude only those unusual situations, like the jurors, where people have an obligation to act one way or another depending on the information they hold. If someone has such an obligation (and he has no conflicting obligation in the circumstances), then, plainly, what he ought to do depends upon the information he has.

            The sort of situation Jason describes is not like that. It is, I would say, an objective matter of fact what the agent ought to do. If the information he has indicates that he ought to do something different, and he acts according to that indication, then he does something objectively wrong, even though we may well excuse him for doing it, given the information he had (provided, at least, that he took all reasonable steps to get good information).

            I can understand your resistance to what I say here. As I mentioned before, there is a hoary philosophical tradition according to which what the agent ought to do depends upon what information he has. I think that view is false because I do not think that morality is subjective. I think people who hold that view are confused in many ways. One thing they confuse is what is permissible with what is excusable. But there are others too, such as what is the case with what we can know to be the case.

          • TracyW

            Good point on the life-and-death situations being normal in hospital. I forgot about those.

            In the case of life-and-death situations, you may understand my resistance to what I say here, but I don’t understand why you’re saying it. You’ve stated what you think, but you haven’t made a persuasive case for it. I don’t see how having an objective morality changes that what one should do under conditions of uncertainty about how confident you are in your knowledge. To return to the juror case: objectively murder is bad, and jurors should do their best part in the legal system, but I don’t see that either of that contradicts the idea that if you only think the defendant did it on the balance of probabilities you should vote to acquit, but if you think the defendant did it beyond reasonable doubt you should vote to convict.

          • http://independent.academia.edu/DannyFrederick Danny Frederick

            The situation of the jury is a special case, as I have already explained. You should not be misled by it.

            Here’s an example (adapted from Thomson, ‘The Realm of Rights,’ p.172). Bloggs’ baby has a fever and the best thing that can be done for it is to feed the baby aspirin dissolved in apple juice. I presume we agree that Bloggs ought to feed the baby aspirin dissolved in apple juice. But Bloggs’ best estimate, on very careful consideration, is that it would be best to starve the baby (‘feed a cold, starve a fever’). So he does. The baby dies.

            Did Bloggs do what he ought to have done? Surely not. What he ought to have done is what was (objectively) best for the baby. The doctor afterwards might say: ‘you ought not to have starved the baby, you ought to have fed the baby aspirin dissolved in apple juice.’ Suppose that Bloggs had asked you, before he took any action, what he ought to do, and assume you knew that aspirin etc., is best for the baby. What would your advice be? Surely, you would say: that he ought to feed the baby aspirin, etc. Surely you would NOT ask him what information, theories or views he holds about what is best for the baby in order to deduce from that what he ought to do. ‘Well, if you think it is best to starve the baby, then you ought to starve the baby.’

            Even if you had not know what is best for the baby, when asked for advice your response would (or should) have been to discover the facts, not to discover what Bloggs’ views of the facts were. What we ought to do depends upon the facts, not upon the possible (and usually) mistaken theories we hold about the facts.

            But was Bloggs to blame for killing the baby? Perhaps not – at least if he had done his best to find out what the best course of action was. Given his epistemic situation, his action may well be excusable. But he did not do what he ought to have done. It is a mistake to infer from his blamelessness that what he did was right.

  • Rob

    Interesting scenarios.
    Without risking being too simplistic, there are some unstated assumptions, which are presented as certainties for the sake of hypothetical argument, but which I think shouldn’t be ignored:
    1) It is assumed with absolute certainty that the “demon” (or president, as it were) will actually carry out the proposed acts, and that he is not simply lying.
    2) One can predict the future with absolute certaintly.
    3) No one else will carry out the acts in his stead if he is assassinated.
    4) IF it is assumed with absolute certainty that the “demon” WILL carry out his heinous crimes, AND that one can predict the future with absolute certainty, AND that on one else will carry out the acts in his stead, then there seems to be no real dilemma at all, as the only way to prevent them from being realized is to indeed assinate him (or the president).
    5) PRE-EMPTIVE JUSTICE : As it is always impossible to predict the future in such human affairs with any real certainty (this is not quantum mechanics), then this logic is frightningly similar to that animating the preventive assassination doctrine, which posits that the state has the right to terminate what it deems to be individuals who present clear and present dangers to “national security”, such as the case of Anwar al-Awlaki, as well as others categorized as terrorists – without a trial or any real evidence.
    It is the same logic which is always offered as well by presidents themselves for carrying out their crimes – the president in scenario #5 could very well argue that his invasion is necessary in order to prevent even further bloodshed, and that it is therefore a “necessary evil”. Think of Hiroshima and Nagasaki, the ostensible arguments for which were that they prevented even more death, assuming the Japanese would have never capitulated otherwise – we now know that this is probably false.
    This is also the same logic which surely animates the crackpots of the world when they assassinate leaders or fellow citizens, following the voices in their heads…

    Briefly, the only way I see any of these assassinations justified is that there is a 100% certaintly that the “demon” will be true to his word – in which case it becomes in fact a moral imperative to kill him (assuming no one else will do it for him). Otherwise, there seems to be no real reason to believe that.

  • http://twitter.com/RobertHarries Rob Harries

    does anyone really think that killing the President somehow changes State foreign policy? Jeez

    • Rob

      Well, yeah, that too…

  • j_m_h

    My first reactions, before even completing the list, was “Suppose you only get 3 wishes from the genie”

    Having completed the list I have to say I’m sticking with my old view: thought experiments are typically poorly done and over used. That said, I will give my answer.

    While I probably would not follow through on this — pre-emptive killing is very problematic — I would way only in the last case can one actually justify the action.

  • http://www.facebook.com/people/Vangel-Vesovski/695498248 Vangel Vesovski

    Clearly the answer is no to the first three. Four and five are a lot more difficult to answer because it is clear that most presidents are evil and kill many innocent people. As such one could argue that it is all right to kill the president. While I would not murder the president no matter how evil I thought him to be except in my own self defence if I sat on a jury trying to convict someone who did murder the president I might be willing to be very lenient.

  • http://www.facebook.com/astrekal Alex Strekal

    “Suppose an evil demon appears before you and says”

    *stops reading right here and thinks “oh, those silly philosophers”*

  • TracyW

    Vaccinations save lives overall, but at the cost of harming or killing an occasional person at pretty much random*. Are vaccinations justified? Because they’re pretty much (1), although we don’t know whose lives are saved and whose is lost.

    My answer is yes, which answers 1-3 except of course I’ve read enough fairy tales to know not to make a deal with the devil on general principles.
    4 and 5, if your country’s president, demon-possessed or not, can take the country to war single-handedly, you have a serious problem with your political structure.

    *I had a brief bad reaction, though obviously non-fatal, to a rubella vaccination when I was 12, and I’m having my baby boy vaccinated.

    • SimpleMachine88

      Vending machines contain deliciousness. But every once in a while they fall on people. This does not make them moral dilemmas. There are many things that could possibly maybe somehow kill us. I think parenthood is just turning you into a panaphobe. It’s normal.

      • TracyW

        Well yes parenthood has turned me into a panaphobe, but I don’t see what that has to do with the rights or wrongs of vaccination, or any other moral dilemma. Panaphobia is just hormones or something.

  • SimpleMachine88

    Your problem is not with the President. It’s a democracy, the people elected him, and you are given the same opportunity as anyone else to remove him from office. Approaching political issues with a little maturity means accepting that your positions, no matter how dearly you may hold them, may not hold sway. Citizenship is partly about dealing with that.

    Wars aren’t fought because the President supports wars. We fight wars because the American people believe that we should fight wars. Your problem here is with the people, that’s a lot more people to murder. Vote, speak, but “people don’t agree with me” is not an excuse for murder.

    Wars involve the death of a lot of innocent civilians. And yet, sometimes wars are necessary. It’s a harsh world, that wasn’t caused by Presidents either. Our government burned half our own country in a war, men were ripped apart on the battlefield, and women and children starved to death. And slavery was ended. Good.

    What you are idly contemplating is useless, traitorous, and immoral, Mr. Booth.

  • Tom

    Ok; rather than questioning the assumptions, I’ll play the game. I think the answer is “yes” to all 5. It seems obvious that your answer to 4 ought to be the same as your answer to 3 (assuming that the evil demon will *leave* the President after the invasion so he can have his life back). I also can’t think of any reason to have different answers for 3 and 1, so I think you should answer in the same manner for 1, 3 and 4.

    I don’t think that rights are absolutely stringent, so enough lives of others at stake can override them.

  • Jay_Z

    No to the first three. They are all a variation of the same – “Evil Demon comes to you and asks you to do X. Should you do it?” Unless you want to do evil, the answer should always be no.
    The demon is coming to you with an offer. Why is he coming to you with an offer? Presumably whatever he offers you is going to be more evil than the status quo, otherwise why make the offer?
    For four and five I’ll say no as well. I’m rejecting the premise that we know someone is evil and so we can justly do things other people can’t do. Whenever a moral question is asked, one of the questions is “what if everyone did this?” Suppose someone else assassinates a leader, saying they’re a big evil threat. The leader doesn’t seem like an evil threat to me, but who am I to judge? Any assassination can be seen as an act of war. There are other ways to prevent war.

  • Golden

    Whoever claims the moral high ground, and says they won’t kill that one person, that is complete and utter bull. At the very least you would consider it. It is a tempting offer. The more important question is: When do the means cease to justify the ends? I don’t think there is an answer, at least not a correct one. We all have to realize that this is complicated stuff, and there are no easy answers. If, by killing one person you save a hundred, then is it not okay to do so? To put it a different way (I apologize now…) was it not okay to kill Hitler, and those who followed him? I would argue that it was a good thing, and it needed to happen. These are complex issues. Best Wishes.

  • Joseph R. Stromberg

    “Propertarianism is correct: all human ethical and political statements
    can be reduced to property rights, and done so without contrivance.”

    May we take this as satire, or was it meant seriously?

  • Joseph R. Stromberg

    “Wars involve the death of a lot of innocent civilians. And yet,
    sometimes wars are necessary.”

    Well, perhaps someone could be bothered to fight the ‘necessary’ ones with *greater care.*

    “Our government burned half our own country in a
    war, men were ripped apart on the battlefield, and women and children
    starved to death.” So it did, or at least someone’s government did; and so they were.

    “And slavery was ended. Good.” It’s hard to do this sort of higher math.

  • John Halstead

    Hi Dr Brennan. I am a DPhil student at Oxford in Political Theory. I just tried to send you an email with a question about political power. However, the email failed to send. Can I just confirm that your email address is

    jason_brennan@brown.edu

    Please respons here or email me at the given email address

  • Pingback: The Trolley Non-Problem | Bleeding Heart Libertarians