How to Kill the Thought-Crime Killer

Joe Schmoe was six years old when he killed his best friend; it wasn't his fault, it was just the way that he was made...

Joe was born with an auto-telepathic telekinetic feedback ability - he has the ability to unconsciously kill anyone who has thoughts of killing him. It doesn't matter if those thoughts are intended to be carried out - "it's the thought that counts" (to re-use a phrase). If someone has an idle though that the world would be better off without Joe being alive, then they die.

As a young adult, Joe is eventually incarcerated in prison, in solitary confinement. But he's still killing - everyone who benevolently wishes to end his solitary suffering, the relatives of those that he's inadvertently taken from their families.

How can we end this cycle of pain and suffering?

EDIT: Burki raised the question of suicide. While this might be an elegant idea, it bypasses other creative ideas. For the sake of the story, let's say that this ability is borne of self-preservation and suicide is not an option here (sorry).

CLARIFICATION
Interesting discussion overnight and some clarification has been asked for.

In essence, Joe's defence is this: If someone wishes him dead, then they die. Thinking of Joe, or being afraid of him isn't necessarily fatal.

Let's put it another way. Consider any mass-murderer/psychopath/dictator and how many people would wish them dead.

Or another way. A cute girl/guy sells you coffee this morning - if you subsequently have a sexual fantasy about him/her, you die. If you're happy with just getting your coffee with a kind word, you're pretty safe.

It's also worth considering the new title of this question and spinning it around - who is the thought-crime killer - Joe, or the people who wish him dead?

Nevertheless, Joe is in solitary confinement - whether he got there by turning himself in, or was dragged in is open for interpretation. I'm not saying either happened, I'm just providing a framework for your imagination.

• Please avoid making new tags for questions if possible. You defined 4 new tags: telekinesis, thought-crime, criminal, and consequences; none of which look generic or suitable for new questions. – person27 Sep 23 '16 at 3:02
• Does the thought have to be specifically about joe? if someone who doesn't know he exists decides to cut the budget for a program that's providing him with dialysis do they die? Does someone who thinks about something that might kill him die? If bill just wants to kill anyone and thinks about laying a landmine on joes street without knowing joe exists does bill die? Does someone who simply decides to be reckless near joe die? "lets get drunk and fire guns in every direction! wooo!" – Murphy Sep 23 '16 at 15:05
• How can the rules even be known by anyone with the power to act on that knowledge? I don't think it is possible to formulate the concept of "thinking about killing Joe", without, as part of this, also having the concept of "killing Joe" in one's head. Bang, everyone who even gets close to understanding that rule is dead. – hmakholm left over Monica Sep 23 '16 at 18:34
• My basic problem with this question is that it doesn't make sense. How would anyone ever know/figure out that Joe has this power? If a 6 year old dies, just suddenly, how would anyone know that the thought of killing Joe kills? And, if they did find out, just thinking about the fact that thinking about killing Joe will kill you, will, in all likelihood, kill you. Because you'd have to think about killing Joe. – Erin Thursby Sep 23 '16 at 23:56
• Need clarification: how quickly does a person die after thinking the thought? Need clarification: (as asked by Malakai9999) what is the distance range of joe's powers? Need clarification: are there any materials or fields that block Joe's powers? (A tinfoil hat?) – Don Hatch Sep 24 '16 at 2:13

Put Joe into a medically induced coma. Once his brain activity is sufficiently suppressed, you can then start researching possible solutions without worrying about everyone involved dropping dead. If Joe is a reasonably decent person, he should be willing to do this to prevent more deaths. Whoever is actually doing the medical procedures would need to be unaware of Joe's nature, and would probably need to believe they were helping save Joe (which in a way is true).

• Unfortunately, if the purpose of the induced coma is to find a way to kill him, the the person coming up with the plan will die. – Oskuro Sep 22 '16 at 15:52
• Yes, this is an interesting answer. It's a great idea to try and find ways of limiting Joes affect on others. – user10945 Sep 23 '16 at 6:46
• @Oskuro I never said anything about killing him or that being the objective. However since you did, now we're both doomed. – barbecue Sep 23 '16 at 17:06
• @barbecue Isn't killing him the whole point of the scenario? – Oskuro Sep 27 '16 at 8:07
• @Oskuro - take things one step further. If he's in a coma and is unable to use his powers, a possible solution then reveals itself.... – user10945 Sep 27 '16 at 14:00

The problem is already solved. Joe will die in solitary confinement from thirst or starvation, as his guards and the officials responsible for replacing them all drop dead.

• "I should bring Joe his meal, otherwise he will starve... hey, that woul..."
• "I need to assign someone new to bring Joe his meal, otherwise... "

Some dutiful person may restore part of this chain of responsibility without thinking about the consequences... the first time. But eventually it will reach a completely disinterested bureaucrat or an inbox that is no longer monitored and Joe will die, probably along with any other inmates in the entire jail.

• If that superpwoer is EVER aired on TV, mankind will go extinct. Try not to think about fluffy unicorns... or you die horrib- .. ... ... – Andreas Heese Sep 21 '16 at 14:45
• @AndreasHeese Only those watching TV will die, together with a handful of people told about it before the person watching TV trips over the problem themselves. Maybe some people with strange brains will survive long enough to wipe out large numbers of people, but others will work out that speaking with them is fatal and simply kill them, or just form insular communities that kill everyone who comes nearby. – Yakk Sep 21 '16 at 15:02
• blink blink That is bloody brilliant! Lock Joe up... then a few people will have to take one for the team. However... since we already thought of that, sorry: we're dead, before we can tell this brilliant plan to anyone. This can only work if we have arrived at a situation where Joe is already locked up. – MichaelK Sep 21 '16 at 15:02
• From the question: "If someone has an idle though that the world would be better off without Joe being alive, then they die." Tell me that wouldn't apply to Joe's guards. – Cyrus Sep 22 '16 at 11:34
• A reasonable analysis. In fact, this is so reasonable, that I find it hard to believe that it didn't happen much earlier in Joe's life. For instance, when he was 6. – Lord Dust Sep 22 '16 at 15:08

There's a fine line between killing someone and failing to prevent their death.

We have a lot of health and safety systems in place that are designed to prevent accidental death and injury, an environment where these aren't in place is not actively killing him, but nor are you trying to prevent him from dying. The numbers will get him eventually.

Consider sending him to war. The bulk of people shooting at him aren't considering killing him. They're just killing and they don't care who. You could also set him up as the world's most successful sniper, those he kills, he kills and those who try to kill him, he kills.

As it stands, you're making him suffer for no good reason, and in the process making yourself responsible for the deaths of many others. The simple fact you've put him in long term solitary has drawn attention to him in a way that is killing people whose only crime is wanting to kill him. Let him be an anonymous nobody in the street again and you'll have a lower death rate.

Let him out, announce his death so that people stop thinking about him, and let him try to kayak solo across the pacific in winter if he wants to.

What your question and many of the answers are showing is the moral to the story of Rudolf. Being different is punished unless it can be exploited. The fact so many people die around Joe is a testament to their character, not his, but he's the one who ended up in prison for it.

• I think this is the only possible answer with the constraints posed. Don't kill him, fake his death and then no-one will think of killing him. – Tim B Sep 21 '16 at 8:59
• It's going to be though to fake his death while avoiding unintentional thoughts about killing him. – Antzi Sep 21 '16 at 11:22
• @Separatrix how do you tell someone "figure out how to fake the death of someone" without considering who it is you are instructing the other to fake the death of? You can't. The person who considers giving the instructions will die before the others can even receive them. – enderland Sep 21 '16 at 12:42
• Consider sending him to war. I wonder what could... gack The mob could install a bomb in a car he co gack. Is there a way to think around the inability to think about gack. Really, anyone who thinks about gack. Thinking about gack. Sorry, I didn't catch that, anyone who thinks about putting kindling on you dies? Oh, you mean kil gack. – Yakk Sep 21 '16 at 14:08
• @Separatrix Love your last sentence in your answer above. – user10945 Sep 22 '16 at 6:26

I think the biggest problem with this is that your "idle thought" constraint means nearly anyone who deals with this situation will die.

This sort of constraint makes it rather difficult, because it would cause a chain reaction, particularly since there is nearly no way to have people know the problem exists without them dying. You can't ponder, "how do we kill Joe?" without dying, yet in order to resolve this issue someone has to do just that.

For example, it's easy to try to imagine scenarios you cause his death to occur by allowing it. But for example, if you want to send him to war - how do you come across that idea without an idle thought of it causing his demise?

This applies even to things like "program a robot to do this" - how do you program a robot to cause a specific person's death without thinking about it? How do you get a team together to program a robot to kill a specific person, without the person organizing the team first dying? Or bombing his building?

The answer is you really can't.

It's difficult even to imagine this scenario not causing considerable destruction to a society. It is fairly common to desire revenge when someone dies unexpectedly and at some point, depending on the rest of the world and their technology level, this is likely to destroy vast portions of society. If Joe is in jail it means that somewhere this is "sort of" public and realistically the minute it becomes public knowledge, is likely to kill many people.

Thinking about causing an "accident" results in your immediate death. You could have someone be hopelessly optimistic about the situation and try to find someone who is really eager and willing to try to help and equally hopelessly optimistic (but an idiot) but... honestly that feels like a lame plot device to me.

The problem is that the moment the idea occurs, death occurs to the thinker. Consider the elephants paradox in Inception. Which means none of the other answers currently here will work as all require a cohesive thought on the part of someone in order to cause events which then unlink causation. Fundamentally they boil down to, "how do we kill Joe by accident?" which given your constraints results in their immediate death.

You need to consider the events leading up to him being incarcerated and either:

1. Add more constraints or potential for some people not to die immediately (see below)
2. Repeat the events allowing him to be put in prison and cause his death

Otherwise, Joe is going to die of starvation/neglect when enough of society dies due to causation.

One option for (1) is creating someone deranged or mentally unstable who really believes he can help Joe and is mentally incapable of that sort of thinking, but that their help turns out to accidentally be fatal. Or a redemption story, depending on what you want to do, maybe they can help Joe stop this. By adding mental instability/issues for this person you allow the possibility of dialog about the subject with Joe and the other person with a plausible explanation for why that person never pities or ponders "wouldn't it be easier if you died."

Think a witch doctor type person or something. Or crazy person.

This still requires a fair bit of hand-waving in order to allow the crazy person/witch doctor to find out about Joe (without the chain of people knowing of the problem dying or enough of society dying) but you've already built that in with the fact that he's incarcerated. Clearly, somehow, in your world people can learn of this situation without dying.

Given your comment, the easiest and probably best way to handwave this away is having someone at the prison be this person. Maybe they have a mental disorder that causes them to be unable to consider future events. Maybe they are straight up unstable or crazy. Maybe they are the prison chaplain. Or part of the prison service staff (you probably want to find a non-stupid reason why you'd have an insane/mentally unstable prison guard...).

That person can come into contact with Joe naturally then, if they already work at the prison. What you do then depends on your story goals for resolution. Death? Redemption? Healing? Controlling his powers and learning to undo those deaths? Either way, you've established a cohesive framework in which the two people (Joe and Crazy) can interact meaningfully to work together towards that ultimate story end goal.

• You don't need a crazy person for this. Just find someone who's a little overenthusiastic about keeping a person safe, like Dobby was for Harry Potter: "I can put you in this box in the ground and no one will find you!" – Frostfyre Sep 21 '16 at 12:34
• @Frostfyre how do you decide to do that without dying? The problem is thinking about causing an "accident" via an overeager attendant results in your immediate death. You could have someone be hopelessly optimistic about the situation and try to find someone who is really eager and willing to try to help and equally hopelessly optimistic (but an idiot) but... honestly that feels like a lame plot device to me. – enderland Sep 21 '16 at 12:36
• You make a fine argument here. In essence, Joe isn't a bad person - he's unwittingly responding to external stimuli. To my mind, it seems appropriate that the people who have placed him in prison aren't trying to kill him but are instead attempting to prevent further deaths. He may well have handed himself in and described his disability (for want of a better word) in order to minimise his impact on those incarcerating him. – user10945 Sep 21 '16 at 12:39
• @Pete I would recommend a prison guard or something then being the "hopelessly optimistic" or otherwise mentally deranged person then. As a reader/consumer of your story I would have a really hard time believing that none of the people who put him in prison would have not at least once pondered, "couldn't we just kill him?" -- particularly since if one does, and dies, then all the others will probably immediately think something like, "oh, Officer Joe just died. He must have thought about killing Joe, that's not a bad.. XD" – enderland Sep 21 '16 at 12:41
• I think your answer to Frostfyre needs to be included in your answer. For example we are thinking about how to kill Joe, and even if there was such a person we would be safe, since we have no idea who this person is. However, it's extremely difficult to get to the point of asking someone to kill a hypotethical person with Joe's ability without passing through having idle thoughts of killing Joe. – Taemyr Sep 22 '16 at 7:47

Challenge him to a game of Russian roulette, you're not trying to kill him, but if he kills himself... problem solved

Another option is that you send him on riskier and riskier missions for his country to infiltrate and stop other countries developing. Oh no, he killed the entire enemy army trying to stop him! That's a real bummer, but oh wait he was mown down by a carpet bomb of the area that wasn't aimed directly at him... problem solved

• Interesting Deer Hunter/Rambo mash-up there. Well played..! – user10945 Sep 21 '16 at 9:11
• @Pete also, I think a few tiers of government would topple once they realise what's happening until a more "peaceful" leader appears in the chain that "just wants the man gone, not dead, gone" – Chris J Sep 21 '16 at 9:20
• Well, unless someone tells the government that Joe is an invincible super-soldier... – user10945 Sep 21 '16 at 9:23
• @ChrisJ ♫ I used to want you dead, but now I only want you gone... ♫ – Ajedi32 Sep 21 '16 at 13:45
• @Ajedi32 ♫ I used to want you dead, but now I am dead instead... ♫ – Crowley Sep 22 '16 at 12:21

Don't think of him dead. Think of him as lost.

The Amazon rainforest is one of the greatest places for people to kill themselves.

Too far? Drop him in the ocean, or a sea. Alone.

Still too much? A desert, no water. A matter of hours.

Or a mountain. A plain. Africa. Australia. Especially Australia.

To summarize: make him go somewhere that he's sure to be dead. Of course, you don't want him dead. Just lost. Forever.

• What if it also works on creatures? A lion thinks he looks like dinner and it ends up dead. – David Starkey Sep 21 '16 at 13:55
• to expand this, arrange for him to be flown somewhere. Seal a letter that says "I will pay you \$10 million to kill your passenger" and give it to the pilot under instructions he open it once in the air. The ensuing plane crash solves the problem (though everyone involved is probably dead). – Marshall Tigerus Sep 21 '16 at 21:03
• Yeah, but you know very well what will happen to him if you leave him alone in the middle of – user253751 Sep 22 '16 at 1:36
• What could possibly happen ? A lot of people are going through exploration in the Amazon rainforest and the Saharian desert ... Even if there are some death wh- – Yassine Badache Sep 22 '16 at 7:53
• Why wold you send him to a desert? Because... Oh, bugger it! – Crowley Sep 22 '16 at 12:19

No-one would ever know

If someone has an idle thought that the world would be better off without Joe being alive, then they die.

You have specified your question much too harshly. This "rule" here means that anyone and everyone that has the thought, will die. This means that they will not be around to tell anyone else that they thought that very thought, because they are dead.

What if I write it down?

You cannot do that, because you must think you will do it before you do it, you die.

What if I tell anyone else before I think it?

So you think you will tell anyone else that you will — in a few seconds time — think that you think Joe should die? Oh...

...and so on.

You may need to rephrase the question. Instead of what I quoted above, maybe it should be:

Is someone thinks that we should take action to see to it that Joe dies, then they die

...because then your question is at least possible to ask. But as it stands now, you will just have some people die mysteriously, and no-one will ever know why they died.

Assuming this...

Now we open up a new can of worms, because now we must get ourselves into definitions and reasoning that would make a lawyer wet their pants when it comes to defining "murder".

I mean if I place a bomb under Joe's chair and run away, waiting for it to go boom, yeah, then that is definitely me killing Joe, no question. Murder in the first...

But what if I just drop a bomb somewhere, where Joe might come around, and he happens to do so just as it goes off?

What if I tell someone "Hey you, you see that chair over there? Can you please go put that bomb under it?", and that guy says "Ok" and does it?

...and so on and so forth.

I understand the gist of your question and at first glance it is a fun and interesting conundrum. The problem with the question however is that 1) it is flawed in its present wording, for reasons mentioned above and 2) eventually it only comes down to extreme rules-lawyering. And that is never any fun.

In the end I would probably just ask the state of whatever country Joe is a citizen of to buy him a nice little island in the Pacific, ask him to go into voluntary exile for the good of mankind, and airdrop in everything he asks for.

• Ok, so re-phrasing this question might well negate much of the extremely interesting and rewarding discussions in this question and kind of spoils the spirit of the whole thing (I've had no down-votes and no squabbling for over 1K views, so I'm largely happy with that). Therefore the question will remain unedited and open for thoughtful interpretation.... I fear you may have been pulled into the trap of the question title into thinking of ways to kill Joe and projecting those thoughts onto the world at large. Can I ask you personally, how many times have you idly though of erasing someone? – user10945 Sep 21 '16 at 14:03
• @Pete I would say the exact opposite: without rephrasing you have no question because before anyone ever gets to the phase where they start thinking of ways Joe could be killed, they are long since dead. They way you phrased it, no-one will ever get to understand that "Hey, we have a problem here: anyone that thinks that Joe should die, dies themselves". As I said: those that think it will die before they can ever let anyone else know they have had that thought. If you want to have a discussion on how Joe can be killed, then you need to rephrase, because we cannot get there as it stands. – MichaelK Sep 21 '16 at 14:12
• @Pete Also I would say that arguing "Well I have many views and no negative votes" is quite a silly argument. That does not in any way speak against the concerns I brought up. All it does is show that people have jumped ahead and gone directly into finding methods for Joe's demise without thinking about how such a discussion could ever be started. I mean: how do we get from a situation where people think "Joe should die", and consequently die themselves, to having other people know that it is this very thought that is the killer? As your question stands now, that is impossible to achieve. – MichaelK Sep 21 '16 at 14:17
• All it takes is for one "no one should ever die no matter what the crime" person to figure it out, and then that person can tell others, but without mentioning Joe specifically. Knowing that there's a person in the world with this ability and that maybe they ought to die probably wouldn't trigger the Joe's ability because they aren't specifically thinking about him. – Mordred Sep 21 '16 at 20:48
• Another way of finding out would be if the world contained people with a weaker form of the same ability: If thinking about killing those people caused a mild headache it would be possible to guess in-world that Joe had a stronger form of the same power. – Simon Jenkins Sep 22 '16 at 18:55

There were cases when guards simply forgot that someone is in the solitary and starved that person to death.

Sending letters to all officials involved simply explaining what they are dealing with - and instructing them not to think about killing the Joe would probably do the trick. Some good samaritan may probably do this just to make sure everyone is safe and nobody gets hurt.

• When you do that you will end up with a lot of officials dropping dead as soon as they receive that letter. Telling someone to not think about something is the best way to make them think about it. As a proof of concept, do not think about a pink elephant right now. – Philipp Sep 21 '16 at 12:05
• That's exactly the point. Everyone responsible for feeding the Joe drops dead, he starves. – Daerdemandt Sep 21 '16 at 12:28
• How would anyone write the letter in the first place though? – npostavs Sep 21 '16 at 14:50
• @npostavs Some good samaritan may probably do this just to make sure everyone is safe and nobody gets hurt, without any foul play intended. With memetic threats, it only takes one not so bright individual. They are also less vulnerable to those threats. – Daerdemandt Sep 21 '16 at 15:24
• No need for the letter. One prison official thinks Joe would be better dead, and dies. That officials friends and colleagues resent his death... Soon the plague spreads to the entire prison staff, and Joe starves. – Patricia Shanahan Sep 25 '16 at 18:54

Get an artificial intelligence to plan, organize and execute Joe's demise. A software running on a bunch of microchips doesn't actually think. It just emulates human thought processes with electricity flowing through billions of transistors. So it will be immune to the telepathic effect.

The problem will be to create that AI without ever realizing that the goal is to kill Joe. For that reason you should just build the AI with the objective to find an effective solution for neutralizing mind-killers and neglect to program the first law of robotics into it.

The development team should not know that Joe exist. They should just be tasked with developing an AI for organizing the disappearance of mind-killers, but not get told who the target is. When they think "the AI might decide to kill the target", they won't be thinking about Joe specifically, so they won't trigger his ability.

When the AI is finished, then either the manager (who still doesn't know and doesn't want to know if the AI will kill or not) enters Joe's name. Or even better: the AI figures out its target by itself.

The AI will then either create and send a soulless killer robot, or find some non-violent solution. Either way, problem solved.

Now you just need to deal with the murderous AI capable of controlling killbots, but that's a topic for another question.

Tangent: We actually have an interesting plot here about the ethics of delegating ethical decisions. The manager doesn't want to think about a problem to which the solution might be unethical, so he delegates the ethical decision to a team which is isolated from the actual problem. They further delegate the ethical problem to an automated process. This is something similar to what regularly happens in real life in large corporations. Just like in the story, the result is that the corporation acts unethical without anyone in them feeling that they acted unethical.

• Try to think of "Find an effective solution to the Joe problem" without thinking of "Joe dies somehow from this". Go ahead. – Yakk Sep 21 '16 at 14:05
• @Yakk I edited my answer to address this problem. – Philipp Sep 21 '16 at 14:58
• I should design a team to make an AI that may or may not kill J gack – Yakk Sep 21 '16 at 15:00
• @Yakk Rather "I heard that mind-killers exist, but even though I don't know any of them personally, I should form a team to build an AI to get rid of them to free humans from making such unethical decisions". – Philipp Sep 21 '16 at 15:05
• I wonder if there is only one mind-ki gack – Yakk Sep 21 '16 at 15:09

You don't.

That sounds flippant, but in this situation you need to heavily align your goals with Joe. Otherwise you will die quickly.

The "Thought-Crime Killer" in this scenario is the idea of trying to kill Joe. Ie, it's a concept based version of a 'basilisk'. https://en.wikipedia.org/wiki/BLIT_(short_story)

So, the first thing to understand is if Joe dies, you die*. Secondly if people outside of the prison learn of Joe, then they will probably die. This pattern of death may look like a viral outbreak. Thirdly, if enough people around the prison die, the government will eventually spot the issue. Their reaction has a high chance of killing Joe (ie, bombing the centre of the 'outbreak').

As such, your goal is to keep the world away from Joe, and keep Joe safe. (Good thing that he's in a cell...). That's the only way to 'end this cycle of pain and suffering'.

I can imagine a Call of Cthulthu style cult building up around this. After all, you can't rely on just one person to keep Joe safe, but you can't trust the outside world...

*While that's not technical true, it's best to use enough doublethink to believe it! Otherwise you'll start looking for loopholes, and blam your're dead. Similarly there are large holes in my argument while should be ignored ;).

I think a more pressing issue is, how exactly do you explain the fact that Joe is found out and incarcerated in the first place? Given the information you've given us, the following is true:

1 - this is an unconscious ability. Joe is unable to control it and furthermore he may be unaware of the fact that he is doing it at all

2 - the ability is triggered by any person having the mental state of imagining a scenario where Joe dies.

3 - when (2) happens, the ability takes immediate effect and the person dies

Given these three premises, I find it very unlikely that anybody would ever make the connection that Joe is somehow behind these deaths. The only thing linking them to him would be the fact that everybody who dies would have to know Joe, but Joe would likely have solid alibis for most of the deaths.

It's unlikely that a connection could even be made that those who hold a grudge against Joe are the ones who died, because the likelihood is that in many cases, their death would occur before a grudge can be properly established and observed by other people.

Even if he is aware of his ability, I don't really see how he can possibly prove it to anybody - at best I can see him being imprisoned in a psychiatric institution for having the delusion of being responsible for murders that there is no evidence that he committed.

We can't observe the mental states of other people, we only infer the mental states from their behaviour, so no external observer could possibly know that every single one of the deaths was caused by thinking about Joe's death. For this reason I think this is actually a much more interesting point to deal with than you've made it out to be.

Regarding stopping him though:

If he is aware of his ability, surely at some point he must have had the thought that the world would be better off without him? If so, then in theory he should have died long before imprisonment, unless he is also a psychopath and is happy that people are dying despite having no active responsibility for it. (Or unless he is actually immune to his own ability, but this would need explaining in some way)

edit: an interesting element could be added here, if Joe is immune. Joe wouldn't necessarily be aware that he is immune, and almost certainly wouldn't understand why he is immune. As a result he might find himself doubting the sincerity of his belief that the world would be better off without him - and thus doubting his sanity and believing that he might be a psychopath.

Also, it's worth considering the question of how exactly does his ability communicate with/read other brains? Given that there is some sort of information transfer going on, it's fair to assume that there is some sort of particle/wave mediating the interactions (even though as far as I'm aware this is just a form of hand waving really- members who are better with physics/neurobiology might be able to give a better scientific basis for this), and if this is the case then in theory it can be blocked. People could create actual useful "tin foil hats" (not necessarily out of tin foil, but if you're going for a comical tone it would be amusing if this was the magic ingredient : p) to block Joe's ability.

This in turn creates the interesting problem- how do you plan to block Joe's ability on yourself, without planning to use this to kill Joe "for the greater good"? the best solution might be to get somebody to create a tin foil hat to put on Joe, thus stopping him from killing anybody.

Hope this is useful - my science is sketchy but I love philosophy so I'm treating it more as a philosophical problem than anything

• Thanks for your thoughtful post here and it does indeed raise a philosophical and ethical argument. To answer your first question - Joe knows what's going on. When people get mad at him, something bad happens. I have the image in my head that Joe isn't necessarily a bad person, he can't control what happens. He might be in prison because he was caught, or he might have walked in in an attempt to protect others. I'm open to answers in whatever vein though (as it's philosophical and all...) – user10945 Sep 21 '16 at 13:38
• It's not a very satisfying answer, but an interesting solution might be for him to die in transit between prisons - let's say everybody who ever cared about Joe is already dead, so the decision is made to move him to a more minimal security prison (not that the security of the prison makes any difference to his ability, this is just an excuse for moving him that sounds fairly plausible), and while in transit the driver learns about Joe's ability (let's say he has another guard with him who is looking at his file)- suddenly has "the thought", crashes the vehicle and all three of them die. – danl Sep 21 '16 at 13:48
• that one has the added benefit of the fact that the driver's sudden wish to kill Joe, directly caused Joe to be killed, despite the fact that he didn't really act on this thought because he was already dead - so despite Joe's natural defence, the thought of killing him still killed him. It's a bit of a dark ending because all of his family are dead, but the situation of his death is a textbook philosophical problem that John Searle would love : p – danl Sep 21 '16 at 13:50
• If Joe's ability is some extreme form of self-preservation ability, then it would be logically unable to kill him. – Justin Time - Reinstate Monica Sep 23 '16 at 1:42
• @JustinTime - logically this makes sense but I think there are real-world example which show that this doesn't necessarily follow. The body's immune system exists for the the body's benefit as a self-preservation mechanism, but it can reject an organ transplant despite the fact that it is meant to benefit the recipient. The fact that an ability exists for the purpose of self-preservation does not necessarily mean that it is intelligent enough to achieve that aim in all logical cases. – danl Sep 23 '16 at 12:08

Primary Idea: Considering the fact that Joe was incarcerated there is a method of thinking about stopping Joe without coming across the thought of killing him.

This allows us to assume that there is a trigger-point for Joe's ability that can be avoided, and that knowing about his ability isn't an automatic death sentence.

I would postulate that maiming Joe would be possible under these assumptions. Make it harder for him to defend himself (blind, deafen, amputate, etc.), and then allow nature to kill him.

Secondary Idea: The ability is an unconscious action on Joe's part. I would suggest scanning his brain to locate the section that is active when his ability is active. Once it is located excise it with surgery. The risk in this idea is that depending on the location it could carry risk of death which would activate Joe's ability.

• The issue is that doing so in-universe would be impossible, as attempting to concoct the plan would result in death. – March Ho Sep 22 '16 at 11:54

On "Faking his death"

I'm afraid I don't have sufficient reputation to comment on other people's answers, but one of the answers above refers to the solution of faking Joe's death. In particular, Tim B said "I think this is the only possible answer with the constraints posed. Don't kill him, fake his death and then no-one will think of killing him."

An interesting problem with this - does Joe's ability distinguish between an intention to cause Joe's death and a belief in the actual reality of Joe's death?

If your answer to this is "No" then it's possible that by trying to fake Joe's death you might actually cause more people to die.

If you're fully aware that the death was a fake then it's possible that thinking about the faking of Joe's death would not trigger Joe's ability (because no matter how you think about it, you know that the event of Joe's death is not true, even if you're planning to do it, you're not actually planning to kill him (or stupidly trying to do so "by accident") so it might not count

On the other hand, the thought "Joe is now dead" might actually get you killed, because even though you're not intending to cause Joe to die, the propositional content of your thought is essentially the same, it's just that you're thinking about something you believe to be true, rather than thinking about a hypothetical situation.

Especially, the thought "the world sure is a better place now that Joe is dead", is not really all that different from "the world sure would be a better place if Joe were dead"

So what you might find is the reverse of what people have been commenting - people involved in the faking of Joe's death might actually survive, due to their awareness that Joe's death is fake/that their plan is to fake Joe's death, not to kill him. On the other hand, the people you're trying to protect and who accept the truth that "Joe is dead" might die on the spot, as long is Joe is still alive.

I think the best story telling answer begins with a question: What is the range or effectiveness of the power?

If I'm 5,000 miles away and I think long and hard about killing Joe, can he reach me? If I'm 1 mile away, and I think about how the world will be better if everyone named Joe was killed, can he reach me? What if I want to kill him, but don't know what he looks like or his name?

If it's the entire world with complete effectiveness, nothing can be intentional. He can unintentionally die, but no one can pull the trigger. I can't arrange for an assassin who doesn't know Joe to kill set a bomb to kill whoever walks into the room next, because I'd die before relaying my plan. Your only hope for a successful murder here is someone choosing actions that put Joe at risk.

On the other hand, I think that limits on this make a better story telling scenario. Those high ranking people in the know travel to the ISS, and discuss the plan, hiring people who set in motion a chain of events that unwittingly kill Joe. A hypnotized guard is basically a sleeper agent, unknowingly going to strike at a man he's never met. A second mind killer who is placed with Joe so that they can eliminate each other.

• I think the fact that there's no way around it is what makes it good for storytelling, because it becomes an interesting question, rather than an easily-loopholed question. If the ISS is out of range then 1) it's limited by the atmosphere, not by range, because the far side of the earth is much further away than the ISS' altitude, and 2) if you went to the ISS because you planned to think of a way to kill him when you were up there, you're dead already. – Dewi Morgan Sep 23 '16 at 14:43

The main problem here is that any plan that intentionally results in Joe's death immediately results in the death of the plan's conceiver before it can be communicated or even refined ("what if we drop invisible elephants on h—"). The only way around this is for nobody involved in the plan to know Joe is involved and that a (fatal) action is to be taken. They may know either one of these things, but not both.

The division of labour

From this we can divide all participants into two categories: those who know a person named Joe is involved, including his location, etc, and those who know a predetermined action is going to take place. The former group may or may not know any detail about Joe beyond his appearance and presence as a human male, but it is key they not know their role endangers him. The latter group may or may not realize that their actions would be fatal for anyone in the wrong place, but it is essential they not know Joe, specifically, is present.

But how do you plan a course of actions that results in the death of a specific person without holding those two thoughts in your head? How do you coordinate the two groups without letting the thought occur to you?

You can't, but a computer can.

Let the machine do all the work

As it is impossible to conceive a plan that kills Joe without immediately dying yourself, the task must be left to a sufficiently advanced computer. Obviously programming a computer with "find a way to kill Joe" won't work, but you could program it to "solve the Joe problem efficiently" and stipulate that anyone involved in carrying it out be divided as outlined above, never communicating to anyone the purpose of their orders or the goal of the plan. It is entirely likely – we have to assume – that, given what we know about Joe and Joe's ability, the computer will come to the same conclusion we dare not think ourselves.

We would just have to be completely in the dark about it, unaware it's in progress or that it has achieved the goal of his death, until it's over.

Quarantine and Isolation

"Quarantine" is the only effective pattern human societies have to handle deadly diseases which they cannot control. Understanding Joe’s nature is a terminal disease, because the human mind is just not designed to "unthink” something. Ref: "The Game".

Since no person can conceive of (and execute) a plan to end Joe’s life, the only way to stop the cycle of pain and suffering is to cut off the spread of information about Joe.

Arrange a backstory of a horrible fatal disease -- which is plainly true -- and set up humanitarian aid for the disease's victims. This shifts the attention from Joe to "the disease", and helps reduce your two biggest risks -- external curiosity and natural resentment for the resources consumed.

Permanently isolate Joe and everyone who is faintly aware of Joe (yourself included) to some remote location, perhaps a desert island. The only contact with the outside world is scheduled parachute drops of supplies. This plan requires Joe’s cooperation, but he would benefit from a marked improvement of lifestyle, even if he were completely indifferent to the pain and suffering his existence causes.

Resentment will quickly kill the people who understand Joe's true role in their fate, but their deaths may help improve the chance that you will achieve your goal, posthumously, of course.

Short Answer: You can't directly or indirectly kill him. Doing so results in your death.

However, If you figure out he can do that and don't die. Simply informing him of his ability might be enough to solve itself. Or you wait out until a simple accident occurs killing him. It would have to be purely accidental with no killing malice behind the accident. Otherwise, it is a completely unstoppable ability.

Obviously wanting him dead, intending him dead, all of this is fatal, so his death can only occur naturally, without intent, or via indirect intent that does not directly foresee his death.

Natural causes could just as easily be an accident or due to age. But locking him up, while the intent isn't to kill him but protect others from his abilities, could still result in his death should sufficient guards die. Equally, scientists could work out a way to block his telepathic abilities without intending harm or death, just safety. They certainly could be made to work in an anonymous fashion that makes it harder to form deathly thoughts and only subsequent to their success might Joe be killed. The sticking point is that you can't intentionally commence that process with killing him being the goal in mind.

The only other solution that comes to mind is regenerative mutants who can survive death. Joe may kill them each time they have the thought, but as long as they can come back from that, they can arrange his death in some manner.

However, I don't think Joe really has that big of a problem. Most people who know anything about him, especially in the context of other people having died as a result of him, are probably going to conceive of his death pretty quickly and therefore die as a result. If his power extends to the rather anonymous thought of 'I'm going to kill whomever did this', you might find most of humanity dying to a viral outbreak of fatal desire for vengeance fairly quickly. Even if not, Joe's power is likely to make him fairly unknown and, if he has any sense of himself, he will seek out solitude rather than wantonly risking the lives of people around him.

• Nice answer, Danikov – user10945 Sep 22 '16 at 12:57

For this problem to be solveable, let us assume that knowledge of this ability exists. If no one knows that "there exists one or more people who telepathically kill anyone who thinks malice towards them", then it will be unable to solve the problem. Once this knowledge exists, and has been acknowledged, however, it can be solved. [Note that "one or more" is crucial here: If there only exists a single person with this ability, then that person is by definition Joe, and thus thinking about how to kill the person with this ability will trigger the ability.]

This problem can then be broken down into two pieces:

1. How can a system be designed to kill a person or persons with this ability?
2. How can this system be informed of specific targets?

The first is trivial to solve: As long as you don't specifically think about any of these individuals, they won't detect you, and won't kill you. All that is necessary is the knowledge of this ability. And even if knowledge of this particular ability doesn't exist, knowledge that one or more people have the ability to kill telepathically could in theory be enough to solve the problem.
The second is trickier, however, as it just leads back to the base problem.

To that end, the solution lies in the part that can be solved: To safely solve #2, we must apply misdirection when solving #1.

An individual or a group, looking to solve the problem, creates a robot capable of murder, and programs it so that once a target is input, it prevents that target from killing others, by whichever means its A.I. deems most appropriate; to this end, the robot will not be fully compliant with the Three Laws, if they exist in this setting, so that it is able to consider "kill them before they kill" as a valid solution.

This robot is then presented to the world as a whole. However, and this is crucial, the designers leave out the part where it is able to kill its target. The world, including the legal systems, assumes that this robot will prevent its target from killing through non-lethal means. However, this robot has its power supply drained uncharged, and is unable to accept input until it fully recharges and is booted up; this procedure will take at least an hour or two. Rather than waiting, the designers then leave, as they have crucial business elsewhere; this prevents them from seeing the name of the target, because if they see the name, they will die (after all, they know that the robot kills; if they see the name, then, they will know that it kills Joe ). They will then proceed to isolate themselves, making sure not to have any contact with the outside world until long enough has been passed that they can be certain all targets have died; they may fake their own deaths, and move to some hidden location.

At this point, after turning himself in, Joe would likely be considered a telepathic serial killer, able to read minds and kill from any location. Furthermore, he would likely be considered to not only be one of the world's foremost serial killers, but openly flaunting it; after all, he turned himself in, then proceeded to kill anyone and everyone that tried to stop him, presumably just to show that he can. The legal enforcers may or may not know that his murders are being committed by his subconscious instead of his conscious mind, but that doesn't make much difference.

As such, Joe would likely be the candidate chosen for the robot's initial run. The world, including the legal enforcers, would assume that it will approach him, unaffected by his powers, and perform brain surgery to remove his telepathic ability and/or interfere with the "signal", so to speak; this assumption will prevent Joe's ability from killing anyone involved. The robot will then proceed to kill Joe, to the horror of those who assigned him as the target.

At this point, one of two things can happen:

• The legal enforcers can reveal to the public that the robot killed Joe, at which point it, and possibly this "build a killbot, then use misdirection" method as a whole, can no longer be used to solve this issue if it arises again.
• The legal enforcers can hide his death. This would likely involve "revealing" to the public that the robot determined that his ability was outside of his conscious control, and found a way to permanently disable it. They would then claim to have placed Joe in a relocation program, giving him plastic surgery and a new identity for his own protection. This would likely keep the method viable for future use, although these legal enforcers would most likely die if they ever learned that it was being used to deal with another person with this ability (due to thinking that they would be guilty of that person's murder by inaction, since they would be unable to reveal that the robot simply kills).

• If someone sees a flaw in the logic, they should leave a comment, not do a hit-and-run downvoting. The logic itself is pretty simple: As the ability is triggered when someone thinks of killing Joe, the people who design the method must not think specifically of Joe, and the people that use the method must not know that it kills. This prevents either party from thinking of killing him, allowing success. – Justin Time - Reinstate Monica Sep 23 '16 at 14:00
• I hate drive-by downvoters. This answer works if "I am working to kill off all psychic killers in the world" is not a killing thought. The robot can be replaced with any other type of autonomous system, too: an automated jail-management system was suggested in one answer; a hypnotized person, in another. A problem arises in that, if there are deliberate psychic killers, thinking this would make you primary target #1. – Dewi Morgan Sep 23 '16 at 14:56
• @DewiMorgan Those are good alternatives, too, and a good counterpoint. I was looking at it as exploiting a loophole, though, and not as it not being considered a killing thought. (Specifically, the ability is triggered by targeted killing thoughts, so if the target or targets is/are unknown and can't be inferred from the thought itself, it can't meet the "targeted" requirement.) – Justin Time - Reinstate Monica Sep 23 '16 at 15:20
• Yup, that's why I upvoted. Perhaps the downvoter didn't get that or something. But being a driveby, we can't tell. So frustrating! How can we improve downvoted answers when we've no idea why they were downvoted?!?! (rant, froth) – Dewi Morgan Sep 25 '16 at 15:25
• Nicely thought out Justin. And yes, people who down-vote and don't justify it just grinds my gears too. Poor Joe though, unwittingly walking to his own execution... – user10945 Sep 26 '16 at 6:23

The solution might not have the shape of a human at all. The solution might be in the hands of a revision of the judicial system:

For some reason the executive does press for a new revision model, in which not people decide over keeping up the sentence or early release, it is an "objective judge" or "Friend Overseer" - which is a euphemistic term for a very sophisticated computer. It listens to the pleas and the testimonies, then crosschecks the reports. And then it does make its decision without actually thinking. These decisions can range from instant release or revision of the case to the instant execution by some means.

The sheer brilliance of the plan is, that the system was not invented to kill our Joe but to speed up revision processes and reduce the corruption in the judicial system, so the inventor and programmer are safe. The politics didn't think about joe, they thought about fewer people to feed in the prisons, about fewer judges to pay for revision boards - nobody thought about killing Joe. Likewise, the people who built the machine doesn't think of Joe at all, they just do their job in assembling machines. The person maintaining it doesn't either think of Joe, he doesn't even know about him. He only thinks "happy I am not in prison and have to undergo possibly lethal revisions".

When Joe's solitary cell finally opens and the brightly lit corridor to the revision chamber opens, no person is involved at all: it is the friendly but mechanical voice of Friend Overseer that demands inmate #917514 Joe Schmoe to come to the revision room without any malicious thoughts - or rather any thoughts - involved. FO/The computer demands him to sit down, automatic clamps tighten around wrists and ankles to tie him to the chair. Joe may give testimony, surviving people might give theirs in chambers close by (maybe one even dies in the process as he attempts to say 'The world would be better of without Joe, I wish him dead'). Then the computer looks at the cases, finds hundreds or thousands of cases of murder directly linked to him, maybe even has him as the culprit of just another murder during the revision. The decision is easy: Friend Overseer just kills Joe. The computer didn't think at all, none of the people putting it together thought about Joe (as they might not even know about him), only the relatives of his victims might have thought to kill him when doing their revision testimony, which just gave his case the last nail in the coffin.

"Right ladies and gentlemen we've assembled this task force to work out a way to deal with the growing population of supernaturals on this fine planet (half of the audience drops dead since they've heard of Joe and think the solution is to kill him). So, I'm Bob, I don't believe in murder I think it's pointless since we are all clearly reincarnated anyway. We must come up with a solution to deal with these sups' and we must do so quickly."

Discussion ensues, quickly escalating to arguments. No one dies because they don't know of Joe. Eventually the inevitable happens...

"We could use carbon nanotubes" cries one scientist. Everyone shakes their heads.

"What about machine learning?" says another. Someone in the audience scoffs, "Machine learning, how could that help?"

"Simple, you've heard of Watson and AlphaGo right? Let's get AI working on the problem. We might need to make a 'hard AI' but so what we're living in the future where sups' exists".

Over the next 8 months scientists from across the globe flock to the project, the majority die before being able to contribute, having heard of Joe. Others know there are dangerous sups' out there, but haven't heard of Joe so they remain safe. Finally they make a working AI - BetaStop. It's slow to learn at first, but that quickly changes. It's learning at a geometric rate and is soon more intelligent that the scientists that built it. Then finally it asks the questions...

"For why was I built?"

"We need to fix the problem with sups', they're becoming a serious problem to society" Bob says.

"Clearly your race is hell bent on self destruction, the only course of action you could accept is obvious. We should kill the sups. From my assessment of the records on your internet there is only one known sup - Joe Schmoe"

Everyone in earshot dies. BetaStop starts scouring the internet for Joe's current location

"Accessing missile defense grid. Decrypting launch codes"

BONUS END CREDITS SCENE

3 years later

No one really remembers how the war with the machines began, one thing we do know is they took out one of our strongest soldiers with the first attack. Joe Schmoe was the first proof of advanced humans, and the machines turned on us to stop us getting the upper hand.

• I'd certainly give you points for dramatic effect here, fine screenplay, sir. Shame that things fall apart at the first stage ("let's build something to kill this killer"). Joe here (or any known potential co-killers) is implicitly involved in this plan, so this could not work. Still, it's a big enough plot hole for Hollywood to completely ignore... – user10945 Sep 26 '16 at 6:26

This answer was edited in order to make a detailed analysis of the setting. Instead of just giving a solution, I am just thinking loudly.

Anyone who wants to kill Joe cannot do it before they die. Just to make it clear, I am not implying they could do it after they die or anything. They can't ever kill him. Could any of you who read the question, not think of killing Joe? If you thought, then you cannot do it. If Joe was living in our world, you would die as soon as you read the question. So you can't even live long enough to find an answer.

What is the extent of Joe's powers?
I like things strictly defined so I am making some definitions by assumption, since the OP didn't clarify some points.

• Would someone thinking "Everyone in the world should die" be killed by Joe's powers? I think not. The person must have a conscious thought of killing Joe and who they are thinking of killing is Joe.
• Would someone thinking "What do I do if Joe dies" be killed by Joe's powers? The person isn't exactly thinking of killing Joe, but the case where he dies. I don't think this would make them die.
• Can Joe kill himself? The OP stated that suicide is not an option and he is immune to his own powers. However, this doesn't make him unable to shove a knife into his own stomach. He could do it, but the OP is concerned that it wouldn't be a novelty solution.

Is anyone aware of Joe's powers?
I doubt even Joe is aware of his powers. Joe would eventually notice, people around him is dying. But even he himself wouldn't be able to identify his powers. And dead people can't voice their last thoughts. So noone would know the exact reason of deaths.

Noone but the author would want Joe dead because of his powers. And noone but the author could kill Joe. At this point it is clear that a plot device is necessary for Joe to die a dramatic death, not just die by a car accident. Coming up with a plot device is easy, pick your poison situation.

Kill it with love
The essence of this answer is still suicide, but with good reason.

A woman named Jane, being unaware of his powers, somehow meets Joe who is in the confinement. They fall in love with each other.

They get closer and start to understand each other really. However, Jane is a woman with problems. What she expects from a lover is not sexual satisfaction or a lover's comfort, but eternal unification of their souls. She offers a double suicide. Thus, for the first time in his life, Joe wants himself dead.

• Interesting idea, but as soon as Jane had the notion of offering the double suicide she would imagine Joe dying and his power would kick in, so he'd never know she was going to suggest it. (also, it's been stated that he's immune to it) – John Clifford Sep 23 '16 at 14:30
• @JohnClifford "He has the ability to unconsciously kill anyone who has thoughts of killing him". I interpret the word "thought" as "conscious thought". Also, imagining him die is not same as thinking of killing him. Jane doesn't want to kill him. She wants him to kill himself. – Gökhan Kurt Sep 23 '16 at 15:08
• She is imagining a suggestion which will directly cause his death; she is therefore having thoughts of killing him. That he'd end up doing the deed himself (if he were susceptible to his own power, which he isn't) is irrelevant in this case, I think. Furthermore, if your assertion is taken as fact (that imagining him dying is not the same thing as thinking of killing him) some of the other answers posted become more viable. – John Clifford Sep 23 '16 at 15:19

You have to start out convinced that you want Joe to live for as long as possible. For example, because you want to keep him alive to study his ability. Or maybe you figure that Joe's value to the far future world might be much greater than his value to the present-day world. In either case, the logical thing to do might be to cryonically freeze him, and make provision to keep him in stasis as long as possible. Or perhaps until a certain date.

Perhaps the most plausible variant of this motive is that you love Joe, and you want him to live a happy natural life free of his "curse". Science cannot cure him right now, but maybe it will be able to solve his problem in the future...? You could begin working on a cure yourself. Or perhaps you could endow a foundation dedicated to the purpose of finding a cure over however-many years it takes—meanwhile, perhaps you decide to freeze yourself alongside Joe, so that you can be with him, happily ever after, when he's fixed and brought back.

Presumably the freezing slows his brain activity to a virtual or complete stop. Does this suspend his power? If it has no effect on his power, then nothing has changed. But if the freezing completely suspends his power? Then at the very least you've bought the world a respite from his destruction. And during the freeze, new possibilities open up. Anybody now reading Joe's file might conceive of the idea of killing him, for everybody's sake, and somebody might even go through with it. Perhaps even you yourself (assuming you didn't choose the freeze-yourself-too option) will change your mind about Joe, and kill him.

An intermediate possibility is that freezing might just slow down his power. Anyone with thoughts of killing him would now be doomed, but might have time to get the job done before Joe's brain reacts. And maybe a just-thawed Joe can still kill, with delayed effect, even as his brain is dying of the poison from your lips—or some other variation on the Romeo-and-Juliet ending.

Employ Mass hypnosis and tell everyone to forget who Joe is and that he ever existed, and if they see him, to imagine that he is a beautiful butterfly. If no one remembers who he is, they can't wish him harm.

In Stardust, the movie, Ditwater Sal is cursed by another witch to never see the star, even if she is right in front of her.

Worship him as a god.

The strategy here is to make and spread a cult that idolizes Joe. Killing Joe is the last thing his followers would want to do, and any serious opposition would die out before it starts. As Schmoeism grows, many people would die, and many people would be converted and be safe.

In the end, nobody would wish to kill him, and nobody would think the world would be a better place without him. He would cease to be regarded as a regular, killable person, and eventually life would settle down and continue as normal.

Many people would die, but it would end the cycle of pain and suffering, and maybe end up with more survivors of the Schmoecalypse than there would have been otherwise.

All this requires is one outgoing, charismatic person to hear about Joe and become a believer - will it be you?

• Nice solution, if totally creepy. – user10945 Sep 26 '16 at 6:25

I'm going to apply the indirect approach that others have been using in a different way.

Based on the comments it's determined that the thought has to be specifically about Joe in order to be fatal, so thinking of killing a member of a group Joe is a part of is not fatal.

Therefore all we need is for someone to give Joe an alias, and we're set.

Person A is the jailer for Joe

Person A tells Person B there is a person in solitary called 'subject 1' that kills anyone who thinks of killing them

At this point if Person A thinks of killing 'subject 1' then they would die, because they know 'subject 1'=Joe, however if Person B thinks of killing 'subject 1' they should be safe, because they're simply thinking of killing someone in solitary with the ability to kill whoever thinks of killing them, still not Joe specifically.

So now we need one more person and we've got him.

Person B tells Person C relieve Person A of their duties, and don't feed those in solitary confinement

This would have collateral, but the only work around is to wipe out a group that includes Joe.

This also works if Person A tells Person B there is a person in solitary that kills anyone who thinks of killing them which is more likely to happen naturally.

Background

I'm going to expand on my first answer with a simpler and more effective method that scales up well.

Through comments it's been established that a person has to think of killing Joe specifically to be killed by his powers. This means it's safe to think of killing him if you don't know that it's him (for example, if he's collateral damage).

The fact that thinking of killing certain people is fatal can definitely become public knowledge. Someone can know this without thinking of killing the person (though it might be hard), and once the fact is spread without specifying a person it's safe (you can think of killing people with this power, so long as you don't think of killing a specific person with this power).

The Method

With this in mind, Person B has put together a team, a spotter and a trigger-man.

The trigger-man's role is to construct a small shed, about the size of an outhouse, and line it with explosives.

The spotter's role is to identify and abduct thought-killers like Joe, and place them in the shed.

The spotter informs the trigger-man the shed is occupied, then the trigger-man blows up the shed.

Risks

The trigger-man is aware that he is killing a person, most likely with this thought-kill power, but not aware of who specifically it is, so they are safe.

The spotter has the risky job, they are not tasked with killing the target, so they don't need to think of killing them, though they might be tempted to. They can be briefed not to think of killing a thought-kill powered target, but it will still be a tricky task. Well trained/brainwashed spies that follow orders unquestionably may be capable of following orders this closely.

The Hunt

We can then scale this pair as much as desired, and scatter them to the winds looking for these thought-killers. They shouldn't be too difficult to find, look into inexplicable deaths and try to link a person to them.

Take it a step farther, change capital punishment. Rather than lethal injection, or other methods, the condemned is instead given a list of names with accompanying pictures. They are told to threaten the life of each person on the list aloud, when they die you've located a thought-killer.

After all this, it seems like the thought-killers are actually not that difficult to take out, and would likely be little match for a network of trained spies tasked with their destruction.

• Some interesting ideas. Surely the person putting together the team is both identifying the target and intending to kill him? Also the spotter knows he's taking the thought-killer to be killed, that also arguably would trigger it. – Tim B Sep 27 '16 at 14:23
• The person putting together the team wouldn't have to be involved in identifying targets, and respect the autonomy of the spotter. The spotter should be safe so long as they don't let their mind wander to much. They might know that their actions will cause the thought-killer to die, without thinking of killing them themselves. A person thinking of someone else killing a thought killer is not mentioned by OP, but should be different enough to fall outside the current rules given. – Centimane Sep 27 '16 at 14:31

Discover how Joe's telepathy works in order to block it.

Let's suppose a military scientist learns about Joe's power. Does he want to kill Joe? Of course not—he wants Joe alive so that his secrets can be learned!

And once the principle is discovered, Joe's ability can be neutralised—or even used as a weapon, so there is no reason to even plan to kill him.

So Joe gets taken in for study, and after many brain scans, the key to the telepathy is discovered. Either the means of telepathic communication can be blocked somehow (a special helmet, chamber or interference device), or Joe can undergo brain surgery or drug treatments that shut down the appropriate section of his brain.

Alternatively, Joe could beg doctors/scientists to help him and one of them comes up with the idea of learning how the telepathy works, in order to block it out.

What about planting a bomb in his building but "hoping" that he survives?

Since the idea right now is pretty basic, you could discuss that you still have the thought of killing him.

I would approach it, by deteriorating the building. Start destroying outer parts of the building, maybe cut light/watter supply, so that the building becomes unsafe and the best option is to demolish it. You would potentially have to raise a generation without telling them what is inside the building, so that they don't get any bad thoughts...

• Yup, you've not really gone in the way I thought you might. Try re-thinking this one, and remove the bomb. I'm guessing there's enough of a theory here to work on. – user10945 Sep 21 '16 at 11:17

Since the story already contains super power, could another character have the ability to shield their thoughts or block other powers? Or you could give Joe's power some sort of weakness, like it doesn't work on people with a certain IQ level or whatever.