Wednesday, March 21, 2012

Thought Experiment: The Stochastic Terrorist

I've created another thought experiment. I call it The Stochastic Terrorist.

Imagine that there is a political movement who's leaders have strongly expressed the opinion that someone ought to do something drastic to disrupt a given event.  A group of self-proclaimed members of the movement have chosen to carry out a terrorist deed to stop the event from occurring. They have been scouting out the location where the event is to be held for some time and, in the process, the one who is to build the bomb has gotten to know a charming young woman who works there. He is in fact so charmed that love is in the air. Yet our would-be-terrorist remains deeply committed to his political cause.

The young man is faced with a deep dilemma. Going through with the terror plot would mean the near certain death of the woman. Yet bailing would be a betrayal to his cause. To really stop the plot he would probably have to denounce his friends. He anonymously consults with the leaders of the movement to see if it's really that important that the event be disrupted. Yes, it's absolutely vital, is the answer.  Not disrupting the event could derail their entire movement! They are in a battle of apocalyptic proportions. They must struggle with every fiber in their body against the injustice of their opponents.

So, to carry out the plot and yet absolve himself of the guilt he knows he will otherwise be tormented by, our young would-be-terrorist comes up with a brilliant plan. He has just enough of a science background to pull it off. He constructs a detonator triggered by a device consisting of a radioactive material and a Geiger counter once the bomb is activated. The radioactive material has a half life such that the likelihood that the bomb will go off on the given day is 50/50. The device also has a regular chronometer that will prevent the bomb from detonating if it has failed to go off after the event.


Our young would-be-terrorist figures that it's now entirely in the hands of the power(s) that be – Deo volente. Apprehensive but at peace he delivers the bomb to the others in the group. On the day of the event, they plant the bomb and they all escape the region on trains. As the bomb maker's train pulls out of the station, he gets a text message from a very dear friend that has nothing to do with the plot. The message tells the bomb maker that his friend plans to visit the location where the event is to be held that day. Our young terrorist is paralyzed by indecision. Should he warn his friend and risk jeopardizing the plot? He assuages himself that whatever happens now is in hands of the power(s) that be...  

So, firstly, to what extent has our young bomb maker absolved himself from responsibility for the young woman's death should the bomb detonate? Secondly, does he carry less responsibility for the potential death of his friend? Thirdly, how responsible are the leaders of the political movement for any deaths that might occur? Fourthly,  if the bomb doesn't detonate, to what extent should our young man be credited for saving the young woman, or anyone else for that matter?

Lastly, and I think almost more importantly, can the stochastic process operating inside the bomb in any way be said to have caused what happens at the end of the day?

10 comments:

curious_jim said...

(Assuming there’s such a thing as moral responsibility and that’s what you mean to refer to).

So, firstly, to what extent has our young bomb maker absolved himself from responsibility for the young woman's death should the bomb detonate?

Not at all. I don’t see any relevant moral difference between taking objective chances and subjective ones. You know one deterministic bomb is a dud and the other works but don’t know which is which and you decide to ‘randomly’ plant one but not both. What’s the difference? I suppose you could say he chose to pick from the two devices of unknown functionality rather than pick a known-to-work bomb but I don’t see that this earns him any moral credit either.

Secondly, does he carry less responsibility for the potential death of his friend?

Nope. It’s exactly the same as the case with the bog-standard deterministic bomb that he doesn’t know is a dud or not.

Thirdly, how responsible are the leaders of the political movement for any deaths that might occur?

Exactly the same as if it were a bog-standard deterministic bomb.

Fourthly, if the bomb doesn't detonate, to what extent should our young man be credited for saving the young woman, or anyone else for that matter?

Not at all. It’s as if she got ‘lucky’ and it turned out to be a dud. If you toss a coin to decide whether to kill somebody you don’t deserve credit if it lands on ‘heads she lives’.

Lastly, and I think almost more importantly, can the stochastic process operating inside the bomb in any way be said to have caused what happens at the end of the day?

What’s a ‘cause’? If the bomb goes off the process was a necessary condition for the bad consequences and it seems fair to say it caused it. If the bomb doesn’t go off it was sufficient to doom the political movement. Perhaps it’s a moral tragedy and no action could have been right?

Of course, perhaps the stochastic process ‘causes’ a bomb-goes-off world and a bomb-doesn’t-go-off world split? The pre-split bomber manages to escape his dilemma by causing both outcomes to occur (in different worlds).

Ever think about over-determined events? Kennedy was hit at the same instant by a bullet from the grassy knoll and the book depository. Each bullet was sufficient for instant death but neither was necessary. Multiple causation – who killed Kennedy?

Dreas said...

A few thoughts, Jim. Couldn't the leaders of the movement claim that they said it's vital to disrupt the event, and not blow the place up? Even if they said by any means they could claim that it should have been obvious they didn't literally mean any, just any reasonable means. For example, if an absolutely crucial hub for the global movements is located less than a mile away, it wouldn't make any sense to detonate a small thermonuclear bomb where the event is to be held.

As for crediting our young bomber for saving people we can imagine another example. Imagine someone has a gun held to his head and is told to kill someone else. What if he convinces the one coercing to murder to play a game of flip the coin: heads the one being coerced commits the murder, tails the person gets to live. Do we now credit the one being coerced for potentially saving the person? At first glance maybe our bomb maker is not in the same situation. He got involved knowingly in the plot. But what if you go sailing in an area where you are told pirates roam? Are you then completely responsible for whatever you end up being coerced by pirates to do and should be punished as if you were a pirate yourself?

Introducing the word tragedy into this thought experiment is very interesting! Is our young bomb maker a Greek hero incapable of escaping the outcome of the force(s) that be because of his necessarily flawed human choices? Damned if he does damned if he doesn't? Maybe he can redeem himself by building a proper bomb, kidnapping the woman, warning his friend, and then face being exposed and carry all responsibility for what happened? Only his own death in spectacular Greek fashion can bring him redemption!

The split world interpretation seems to imply something very strange for moral responsibility. Does it mean we can all safely act irresponsibly? Or does being saved in this way require we make all our decisions with a Geiger counter and some radioactive material? Everything 50/50? Isn't that the definition of insanity? Does this mean that our bomb makers choice to build the bomb as he does is insane? So if the bomb goes off should we send him to a mental hospital rather than prison?

curious_jim said...

(I’m not working with any developed moral theory here – I’m something of a moral sceptic - just giving you my intuitions which I think would be in tune with that of most people and the provisions of most legal systems.)

- Imagine someone has a gun held to his head and is told to kill someone else. What if he convinces the one coercing to murder to play a game of flip the coin: heads the one being coerced commits the murder, tails the person gets to live. Do we now credit the one being coerced for potentially saving the person?

The threat of being murdered can ‘excuse’ most crimes but not that of murder. If he can convince the coercer to go for a coin toss that is credit-worthy - but if the coin lands the wrong way he should renege on the deal and take the bullet in his own head instead of putting it in another’s. (Kant wouldn’t approve of making the dishonest deal though – lying is never morally permissible in his account but I think that makes him a bit of a crank.)

- what if you go sailing in an area where you are told pirates roam? Are you then completely responsible for whatever you end up being coerced by pirates to do and should be punished as if you were a pirate yourself?

You’re due some blame for being reckless if you knew pirates tended to grab people and force them into criminal activity and you put yourself at risk of that, still I’d say you’d still have diminished responsibility for ‘coerced’ crimes up to the point of murder.

Moral tragedy is the (supposed) scenario where whatever you do will be morally wrong, I suppose I was thinking about the idea that setting the bomb may have been necessary to achieve a good that was equal to the bad and the same may have been true of not setting it. It was just a thought that occurred. Bernard Williams’ notion of ‘moral luck’ might interest you too – I daresay you can Google it. If you took that idea seriously the moral value of the bomb-maker’s acts would depend on the results.

I don’t know how seriously to take the many-worlds interpretation and whether I’m representing it properly. But it would only seem to apply (if at all) to quantum-entangled events. If we took it seriously that the existence of two worlds would be ‘caused’ by the bomb-maker’s actions it does raise some conceptual puzzles. You’d have two bomb-makers and obviously they can’t be numerically identical with the original one. Should each be held responsible for the both outcomes, and was the original bomb-maker not also responsible for the creation of a whole world? I think the many-worlds interpretation might be an interesting avenue to follow philosophically but I know sod all about quantum mechanics.

Dreas said...

The threat of being murdered can ‘excuse’ most crimes but not that of murder.

So are you saying, Jim, that any amount of premeditation is a clear Rubicon? Once you consider the potential of NOT killing (even for just a moment) it becomes extremely hard, if not impossible, to morally justify your act however much duress you continue to be under (including having a cocked gun held point-blank to your head). Since a bomb always implies potential deaths, our bomb maker's moral tragedy is inevitably set in motion the instance he decides to get involved with the plot. This assumes the political movement's goals are morally justified, since otherwise there would be no tragedy (the young man could just expose the whole plot). But this clear Rubicon seems to disregard that we constantly differentiate between murder, various types of manslaughter (diminished capacity, imperfect self defense, etc.) and even justified homicide. Do we need to reject most of these distinctions?

The most troubling conclusion here is for any organization backed by a military institution. Any military act carries known risks that someone is very likely to die, whether the death is a targeted killing or known or unknown collateral damage. Known collateral damage would be comparable to our young woman, unknown collateral damage to our bomb maker's dear friend. If an independent Western agent blows up a crucial Iranian facility for establishing a nuclear arsenal, there is little doubt that there will be deaths. Should the West celebrate the agent as a hero or condemn them as a morally despicable individual? Or is there room to condemn the act as geopolitically careless and yet praise the individual's intent, thereby making the act more morally acceptable?

There seems to be something to be said for whether the act is selfish or not. Our poor person being coerced to murder by a gun to their head chooses to kill to save themselves. Is their only morally acceptable decision inaction if they loose the coin flip, and consequently their likely death? But isn't this inaction more likely to result in both their own death AND the other person? And doesn't this reduce the chance that one of them will be able to escape and later seek justice? Perhaps killing the other person is less selfish than it first appears? The idea that extreme threats to your own life does not in anyway constitute justification seems to mean we should have prosecuted Filip Müller, member of the Auschwitz Sonderkomanndo, as a war criminal. Was Müller's only morally valid choice suicide? Or is he saved by the fact that he didn't directly activate the mechanism of the gas chambers, only assist in its continued operations?

Our young bomb maker, on the other hand, seems to be acting selfish if he creates a dud (i.e refuses to kill). Isn't he driven by his own desire to save the lovely young woman? Creating an actual bomb is what increases his own chance of death. This is the selfless act. He is risking his own safety for the greater cause. And frankly I can't see how it's different from signing up for the army of a modern republic except that a nation state has a clearer chain of command. In a nation state we supposedly know where the buck stops, morally exempting a soldier as long as they act according to the rules of engagement set by the state through its laws and treaties.

curious_jim said...

The reference to ‘moral tragedy’ was just a passing thought. The idea that sometimes there might be no right thing to do and that whatever you do will be morally wrong is a very contestable notion, I wouldn’t want to make anything of it other than to flag it up as an interesting notion.

I don’t think being threatened with murder by A can justify B murdering C, not in law and not according to my or most people’s intuitions. We can throw in some utilitarian thought experiment stuff e.g. what if you have the cure to cancer in your head and the person that you are supposed to murder is a thoroughly evil individual who sponsors terrorism? A reasonable person might say it’s a rule that could conceivably be justifiably broken in very extreme cases - I suppose that’s where my intuitions lie.

I don’t know exactly what to make of your Rubicon. If you don’t consider the potential of not killing if you do have time to consider at all presumably you’re thoroughly immoral or congenitally amoral. I shouldn’t want to remove the distinctions between different kinds of unlawful killing. Deliberately killing in self-defence seems justifiable to all but the most enthusiastic pacifist if it appears to be the only option. There will be cases of diminished capacity and manslaughter - I didn’t at all mean to suggest we need to reject those distinctions. Though we might want to look into whether these legal differences always capture important moral differences correctly.

The case of the bomber is, I gather, such that he didn’t aim to kill but took the risk. I didn’t think the objective chance element raised any real issue. If he did kill I think the law in most jurisdictions would call it murder, you might want to argue that it’s a lesser homicide. If, by chance, he didn’t kill (the bomb didn’t go off or didn’t kill) I suppose legally it’s in the territory of reckless endangerment or some such. I’m inclined to think luck shouldn’t come into his moral status though it does in fact do so in law – legal systems give harsher punishments to the sniper who hit his target than the one who happened to miss. I don’t see the moral difference between the two acts though there may be some practical justification for the legal distinction. I’d flag up the notions of moral luck and actual consequence consequentialism for consideration but I don’t think they fly.

Acts committed in legitimate wars aren’t something I’m arguing against, collateral damage seems inevitable. Military leaders have a duty to minimise the risk and minimise actual casualties if they know there is bound to be deaths caused to non-combatants and such risks and decisions should not be taken lightly – they need strong rationale. In principle non-governmental ‘freedom fighters’ are in the same situation. If your bomb maker’s cause is just and bombing was the only conceivable way to achieve what was needed, his planting of a deterministic non-faulty bomb would be justified. If however he’d put himself in a position that he could have avoided, so that he was left with no choice but to bomb when he could have done something else then he deserves blame for ending up in that situation. The bombing was right but him causing it to be necessary was blameworthy.

curious_jim said...

...

There does indeed seem to be something to be said for whether an act is selfish or not. B’s only morally acceptable decision is inaction if he loses the coin flip unless there is some extreme non-selfish circumstance. The case of Filip Müller is a tragic and difficult case & I shouldn’t like to rush to make a hasty judgement. The death camps could never have operated without men like him – the Germans lacked the manpower. The mind-set the Nazis depended on was ‘if I don’t do it somebody else will’ and, of course, it worked. If enough had refused to co-operate and chosen death the system would have collapsed. I like to think that I’d elect to take the gas chamber but I suspect I wouldn’t. How people act in such extraordinary cases would need to be judged carefully - perhaps it points towards moral particularism.

If the cause was just it would be selfish of the bomber to cause it to fail for the sake of a beautiful woman but wonderfully human.

Dreas said...

I don’t know exactly what to make of your Rubicon.

My point with the Rubicon is that murder implies premeditation and that we "forgive" homicides to some extent if there was no choice. Your thought is that the threat of being murdered by someone is not an excuse for killing someone other than the one threatening you under ANY circumstances. So, act A (the threat of being murdered) confers to act B (killing someone to avoid being murdered) the properties that A would have if it were carried out (i.e. premeditated homicide). I would tend to agree where there is ample opportunity to explore other valid options (like going to the police). But in the case when a cocked gun is held to the head, the options are extremely limited. We assume the one being threatened is restrained so that the only thing they can do is activate some death mechanism. Their choice is reduced to a binary ACT or NOT ACT. Your thought seems to be that ACT will mean the person committed murder (i.e. acted in a premeditated way). It could be said that they did, since they were given another option to consider: NOT ACT.

As you know, my belief is that survival plays a crucial part in concepts of morality. But not necessarily your own survival. It's my belief that choosing to die can be a morally valid option iff it helps someone else survive who in turn will be able to help others live a secure life in the future (even distant future, but let's not go there). In the gun-to-the head scenario, you don't just choose to ACT. You also, by causal extension, choose to survive. The validity of NOT ACT depends on the causal chain it sets in motion. It seems reasonable to me that it would result in the likely death of both persons under threat. Therefore I think choosing to ACT is morally acceptable. That does not mean that to NOT ACT is necessarily morally unacceptable. Nothing is that easy since the world does not seem binary in nature (at least not at higher levels of encapsulation).

You already raised a situation you seem to think might be an exception allowing one to ACT. What if you were a Tutsi being coerced to kill Georges Ruggiu – broadcaster at the Rwandan radio station RTLM – some time shortly after April 6, 1994? In contrast, I can imagine when it becomes morally preferable to NOT ACT. If either of your deaths is likely to become known and cause widespread moral outrage, then the right thing to do would be to take the bullet as you put it. For example, if the other person is Mohandas Gandhi after the Salt March on March 12, 1930, then to NOT ACT seems morally appropriate. And not because he is a Mahatma and that the rule is never kill a holy man. His sudden disappearance is bound to cause many to want to know what happened. There is no certain guarantee that anyone will. But it's reasonable to think that, given his fame and stature, someone will. Both of you become so called martyrs. Others are likely to take action so that the one going around coercing people to homicide at gun point will be sought out and prevented from continuing to do so.

Estimating how information is likely to flow is important when making moral choices.

Dreas said...

How people act in such extraordinary cases would need to be judged carefully - perhaps it points towards moral particularism.

I agree that a case like Filip Müller is extremely tricky. But moral particularism seems to me like a grain of truth unnecessarily elevated to a theory. It might be that morality is ultimately impossible to systematize into a set of universally applicable principles. Maybe vague, highly case-dependent intuitions is the best we will ever be able to achieve. But it might also be the case that quantum theory can never be unified with general relativity. So what? We couldn't fly until we flew. I'm sure there were those who claimed humans were not meant to fly. And that humans would never be able to fly. If moral particularism is just pointing out that morality is contextual and relies on a lot of guessing, then yeah. And? I'd say join the rest of the consequentialist crew.

curious-jim said...

Hi Andreas,

I don’t buy moral particularism no – but I do think morality is impossible to systematize into a set of universally applicable principles yes. Highly case-dependent intuitions is the best we will ever be able to achieve, not due to some inability to discover all the moral facts or laws, but because there are no such facts or laws ‘out there’ – there are only case-dependent subjective intuitions. Some moral theories capture aspects of our intuitions – utilitarianism with a Kantian ‘people as ends not means’ twist and a touch of virtue captures what a lot of what people tend to feel. But there is no true moral theory and there no true moral utterances. There are only moral sentiments. There is nothing out there to discover except facts about neurology, culture and evolution. We can derive no categorical ‘oughts’ from any of that. Relativism is laughable, but the belief in objective morality is, in my view, on a par with belief in God. It’s ‘natural’ and sometimes useful but entirely false.

Dreas said...

Jim, I'm not sure what you mean by saying there are no categorical "oughts". Is this similar to David Hume's following statement:

'Tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger. `Tis not contrary to reason for me to chuse my total ruin, to prevent the least uneasiness of an Indian or person wholly unknown to me. `Tis as little contrary to reason to prefer even my own acknowledge'd lesser good to my greater, and have a more ardent affection for the former than the latter.

I've been working on an idea I call Hume's Great Mistake. I haven't entirely explored and rigorously founded my idea yet, and so I'm a little uncomfortable talking about it. But I'll let myself be vulnerable and expose myself by giving you a taster of my conjecture. Essentially, it focuses on that Hume proceeded Lamarck, Darwin, Spencer, et al. He was not immersed in evolutionary ideas. He failed to realize that by rational necessity such obliterating desires as the "destruction of the whole world" are ultimately doomed to extinction. Only that which desires to exist will ultimately exist, and therefore have space to rationally express their sentiments.

Therefore there are definitive oughts that can be derived from what is. If you want to be, you must fulfill the requirements of what ought to be given what is.

I think that given how perceptive Hume was, had he lived after the emergence of evolutionism he would have rejected is own is/ought idea. As I said it's an argument that is in its infancy and therefore please critique it accordingly. Of course, nonetheless, any suggestions of where I may be faltering at the outset are greatly appreciated.