When George Zimmerman shot and killed Trayvon Martin in Sanford, Florida in 2012, much of the very heated debate that followed focused on Florida’s so-called “stand your ground” law. This law permits an individual to use lethal force against a person whom she reasonably believes poses a threat of serious harm to her even if she does not believe that force is necessary for avoiding or averting that threat. Since Florida passed this law in 2005, many other states have followed suit.
Most people believe that it is at least sometimes permissible for an innocent person to use force in order to defend herself or some other innocent person from a threat of harm. But we do not usually think that one may do just anything to protect oneself from harm. Rather, we standardly claim that one may use only necessary and proportionate force in self-defence. The necessity constraint requires a defender to use the least harmful means of averting a threat. If, for example, I can avert a lethal threat that you pose by either killing you or breaking your arm, I must break your arm (assuming that either strategy will be equally effective).
It is the necessity condition that has been jettisoned by the “stand your ground” laws. Under this legislation, even a person who could simply cross to the other side of the road in order to avoid being threatened is under no legal obligation to do so. The law represents a kind of “right never yielding to wrong” approach to self-defence, where the availability of opportunities to avoid using force does not undermine the legality of a forceful response to a perceived threat.
“Stand your ground” laws raise at least two issues that have occupied the attention of philosophers working on the ethics of self-defence. The first, of course, concerns the moral status of defensively unnecessary harms. The second concerns the stipulation that it is in part the defender’s reasonable belief that determines the permissibility of using force.
Let’s start with the necessity condition: the requirement to use the least harmful means of averting a threat. One reason why this condition is not straightforward is that it’s not clear why you do anything wrong if you harm a person who is fully culpable for trying to harm you. Imagine the following case, Lucky Escape:
Murderer is shooting at innocent Victim to try to kill him because he dislikes Victim. He chases Victim to the edge of a cliff. Unbeknown to Murderer, Victim has both a gun and a parachute. He can thus save his own life by either (a) jumping to safety, using no force against Murderer, or (b) shooting and killing Murderer. Victim is a skilled parachutist, and jumping is not dangerous for him.
We can ask two distinct questions about this case: The first is whether it is wrong, all things considered, for Victim to shoot Murderer. The second is whether it wrongs Murderer if Victim shoots him. I think that most people agree that the answer to the first question is yes: it would be wrong, all things considered, to shoot Murderer instead of using the parachute. But theorists of self-defence contest the answer to the second question, because we disagree about what exactly happens to the rights of culpable attackers like Murderer.
Much of the recent self-defence literature uses the language of liability to describe a person’s rights against defensive harm. The notion of liability is tied to the idea of forfeiting one’s rights. Usually, people have a right not to be harmed. But, we can sometimes forfeit these rights by behaving in a particular way, such that we now lack a right against being harmed. Self-defence theorists summarise this by saying that the person is now liable to defensive harm – harming her would not wrong her, because she has no right not to be harmed. Ordinarily, people lack rights of defence against harms to which they are liable, since they lack rights against suffering those harms.
There are various accounts of exactly what counts as the right sort of behaviour to trigger a forfeiting of one’s rights. One popular view holds that moral responsibility for a threat of unjust harm is a necessary condition of forfeiture. A harm is unjust if it is going to befall someone who has done nothing to forfeit or otherwise lose her rights. So, in deliberately trying to kill Victim, who has done nothing to forfeit his rights, Murderer is morally responsible for threatening an unjust harm to Victim.
But philosophers disagree about whether moral responsibility for a threat of unjust harm is sufficient for liability to defensive harm. Specifically, they disagree about whether one can be liable to unnecessary defensive harms. Some writers endorse what I will call an internalist account of liability, according to which one can only be liable to necessary harms – that is, to the least harmful means of avoiding the threat for which one is responsible. So, according to the internalist, Murderer in Lucky Escape is not liable to defensive harm because Victim has a way to avoid the threat that Murderer poses without using any defensive force. This means that Murderer retains a right not to be harmed, and if Victim shoots Murderer instead of using the parachute, he will violate Murderer’s rights.
One implication of this view is that Murderer will then have a right to use defensive force against Victim, should Victim try to shoot him. If Murderer is not liable to be killed by Victim – if he has not forfeited his rights – it looks like any harm Victim inflicts upon Murderer will be unjust. But this means that Victim is now morally responsible for threatening to inflict an unjust harm on Murderer. This should make Victim liable to defensive harm – he’s now the one who has forfeited his rights. And so it looks as if the internalist must permit Murderer to kill Victim in self-defence, if that is the only way to stop Victim from killing Murderer.
In contrast to internalist accounts, externalist accounts of liability hold that whether harming Murderer is necessary is irrelevant to whether Murderer is liable to defensive harm. Rather, Murderer’s liability is determined by, for example, whether Murderer is morally responsible for an unjust threat. Whether Murderer bears such responsibility is determined independently of the necessity of harming him to avert that threat. Externalists still endorse necessity as a constraint on the overall permissibility of inflicting defensive harm. Killing Murderer would be wrong. But it would not wrong Murderer, since Murderer lacks a right against being killed. Rather, the impermissibility comes from a more general moral constraint against causing needless harm. Murderer has no special complaint in virtue of being the subject of that harm.
I think that externalist accounts that are wholly “backwards looking” – that is, accounts that only consider facts about Murderer’s moral responsibility for an unjust threat – are shown to be implausible by the following sort of case. Poison:
Murderer has been slowly poisoning Victim over the course of a week. Victim has now imbibed a lethal amount of the poison, for which there is no antidote. Victim is now certain to die.
An account that considers only what Murderer is responsible for – and pays no mind to what harming Murderer can achieve – will deem Murderer liable to harm proportionate to the threat for which he is responsible. But this seems wrong: now that nothing can be done to save Victim, there is no harm that can be inflicted on Murderer that can possibly be construed as defensive. It seems to me that liability is essentially instrumental – that is, it is about harming people in order to achieve a certain end. The harm is not an end in itself (this is what distinguishes liability from desert). Since there are no available means of saving Victim’s life in Poison, there’s nothing for Murderer to be liable to – something that we miss if our account of liability focuses only on Murderer’s responsibility.
I suggest we adopt a different kind of externalist account. According to the proportionate means account, Murderer is liable to any proportionate means of averting the threat to Victim’s life. This includes means that are not the least harmful means, and so it still holds necessity to be external to liability. But it will deem Murderer liable only to harms that are capable of averting the threat for which he is responsible. Given that in Poison there are no available means of averting the threat to Victim’s life, Murderer is not liable to be harmed to avert that threat. In the Lucky Escape case, in contrast, killing Murderer is a proportionate means of saving Victim’s life. I think Murderer is thus liable to be killed in Lucky Escape: killing him would not wrong him. But it would still be all-things-considered impermissible, since we ought not to inflict gratuitous harms.
However, I also think that not all the harm that one might inflict on Murderer would count as gratuitous. Daniel Statman has suggested that defending moral standing can ground a person’s right to inflict harm on an attacker even if this will not avert the physical threat the attacker poses. Statman argues that victims of aggression are subject to two threats: the primary threat (for example, a threatened rape or killing), and a parasitical threat to their moral standing. Victims are treated in a way that involves the attacker’s refusing to recognise their moral status – a harm that is related to, but distinct from, any physical harm being visited upon them. The moral standing justification for harming is appealing because it captures our sense that victims need not be passive in the face of attacks that they cannot avert. The scope of the justification is also likely to be restricted to culpable attackers, since those who innocently threaten, or threaten as a result of some reasonable mistake, do not threaten their victims’ standing in the same way.
An important question concerns how much harm victims may inflict on their attackers in the name of defending their moral standing. The gravity of the threat to standing is likely to increase as the seriousness of the primary threat increases, because a threat to standing is about the wrongness of thinking that one may do this sort of thing to one’s victim. The worse that one is treating one’s victim in terms of the primary threat, the more one denies her status as a person deserving of certain sort of consideration. Thinking that you are the sort of thing that I can slap because I feel like slapping you is not as significant a threat to your standing as thinking that you are the sort of thing I can rape or kill or seriously assault because I feel like raping, killing or seriously assaulting you.
But even though the threat to standing increases the more serious the primary threat becomes, this does not entail that the threat to standing is as bad as the primary threat, and that therefore whatever is proportionate in warding off the primary threat must also be proportionate in averting the threat to standing. The wrongness of being treated as the sort of thing that may be subjected to a harm is not the same (and is not as bad) as being subjected to that harm. This partly explains the difference in punishment between harms and attempts: it’s very bad when I treat you as someone whom I’m permitted to kill because I don’t like you, but it’s not as bad as actually killing you because I don’t like you. There is, therefore, a limit to what defence of moral standing alone can justify, even where the threat to standing is at the most serious end of the scale.
As always when judging proportionate harms, it’s hard to say exactly what the defence of a particular good warrants – especially when it’s an intangible good such as moral standing. But it seems to me that a moderate harm like a broken wrist falls below the threshold, and that very serious harms such as blindness, death or paralysis fall above it. Whilst moral standing is important, it is not as important as not being raped or killed, and its defence does not justify the infliction of very serious harms.
So how does all this feed into our understanding of the “stand your ground” laws that might be at play in a case like Lucky Escape? Well, if it’s true that culpable attackers simultaneously pose both physical threats and threats to their victims’ moral standing, it looks like Victim may inflict some harm on Murderer as a means of defending his standing. This means that not all the harm that Victim might inflict on Murderer in Lucky Escape is unnecessary after all. If Victim tries to break Murderer’s leg in Lucky Escape, this might be a proportionate and necessary means of defending Victim’s moral standing. On even an internalist account of liability, provided that Victim inflicts only necessary harms, Murderer will be liable to those harms, and have no right of counter-defence against Victim.
What if Victim does more harm to Murderer than is necessary to defend his moral standing? What might Murderer be permitted to do in that case? The internalist seems to again be committed to allowing Murderer a fairly permissive right of counter-defence, since these additional harms will be unjust. But the proportionate means account of liability will grant Murderer only a fairly limited right of counter-defence. Remember that on this view, Murderer is liable to be killed, because killing him is a proportionate means of averting the lethal threat to Victim. So even if Victim kills Murderer, he doesn’t inflict any unjust harm on Murderer. Nonetheless, inflicting unnecessary harm does violate a moral duty. So it might be useful to think about how much harm a third party could inflict on Victim to get him to comply with this duty. I think a bystander would be permitted to break Victim’s leg, or even both his legs, to stop him from killing Murderer. But it seems to me unlikely that she would be permitted to kill Victim, or paralyse him, or cause him to lose a couple of limbs, to save Murderer’s life. Even though Victim acts wrongly, none of the harm he is causing to Murderer is unjust, and some of it is all-things-considered permissible.
Some of what I have argued here might seem to support the use of stand your ground laws. I’ve suggested that when it comes to culpable attackers, even harms that are unnecessary for averting the physical threat can be morally permissible if they serve to defend the victim’s moral standing.
It’s important, therefore, to be mindful of two things: first, that what I have argued about defending one’s moral standing applies only to culpable attackers – to those who genuinely pose a threat, and do so not as a result of some mistake, but out of some illicit motive such as theft, anger, jealousy and so on. So nothing I’ve said here undermines the wrong done to Trayvon Martin, who posed no threat and thus could not have been culpable with respect to any threat.
Second, notice that my view does not hold that Zimmerman’s reasonable belief that Martin posed a threat would be enough to make it the case that he did not wrong Martin by killing him. Murderer in Lucky Escape is not liable to be killed because Victim believes that Murderer poses a lethal threat. Murderer is liable because he really does pose a threat, and he’s morally responsible for doing so. Even if Zimmerman’s belief that Martin posed a threat had been both genuine and reasonable, his mistaken beliefs cannot render Martin’s rights forfeit. They cannot make Martin liable to any harm. Martin had a full right of defence against Zimmerman, on my account, since all the harm Zimmerman inflicted was unjust.
This brings us to the general problem with legislating in a way that encourages people to use unnecessary force on the basis of their own reasonable beliefs. Self-defence theorists disagree about the role of belief in permissible defence. On the one hand, people acting in self-defence must invariably do so on the basis of what they believe – they have nothing else to go on. (Indeed, under British law, the relevant beliefs need not even be reasonable, but only sincerely held. If I really, honestly believe that a person is a threat to me because the person is black and I believe that all black people are terribly dangerous, it’s conceivable that the British system would find that I acted in legitimate self-defence.) But of course, on the other hand, there is something very morally worrying about the idea that my mistaken beliefs could determine the permissibility or legality of harming someone.
Some theorists bite the bullet and argue that a defender’s beliefs can affect the permissibility of using defensive force, such that victims can lack rights against being killed in these cases. Others invoke the distinction between justification and excuse to argue that a defender is excused if she harms on the basis of a mistaken belief, but what she does is nonetheless impermissible and wrongs her victim. But even those theorists who think that reasonable belief can make a mistaken killing permissible will include a necessity clause. One must believe not only that the target of force poses a threat, but that there is no way to avoid that threat without harming her. The risk of defenders getting it wrong – both with respect to whether someone poses a threat, and especially with respect to whether she is culpable – gives us overwhelming reason to not to legalise unnecessary killings. This is not a case of letting right yield to wrong, but of prioritising the rights of those who will be unjustly killed over the (alleged) rights of people to walk on their preferred side of the road or to not have to call the police to deal with potential crime. There is no plausible account of rights that supports the view that one person’s right to be in any public space at any time of her choosing outweighs the right of an innocent person not to be killed.