1. The Other Bloodguilt Passages

In this section, I quote the passages in full and explain them only briefly. I do further exegetical work in the following section as I construct my argument that bloodguilt is a symbol.

At the end of Metaphysics of Morals, Kant claims that when we think about justice, we ‘personify’ it as though it were an entity that acted of its own accord (MM, 6: 489). He explains that we think of it ‘as if it were a substance (otherwise called eternal justice), which, like fate (destiny) of the ancient poets, is above even Jupiter – that pronounces on rights in accordance with an iron, inevitable necessity which we cannot penetrate further’ (ibid., emphasis original). Kant then provides examples of this personified justice:

This bloodguilt bears some similarities to the bloodguilt in the Doctrine of Right. In the same way that bloodguilt clings to the members of the dissolving society, this bloodguilt can be transported from place to place – the wise ruler ought not allow bloodguilt to come upon his land. Here Kant claims that crime must be avenged no matter what, just as the last murderer must be executed even if society is dissolving. Unlike the main bloodguilt passage, however, he suggests that the criminal’s descendants or innocent parties must suffer if the criminal does not. If someone who spills innocent blood is not punished, then either his descendants must pay or he must pay in the next life.

The two shorter bloodguilt passages from the Metaphysics of Morals drafts also contain ideas that are found in the published passage. The first claims that granting pardon to an evil person is contrary to justice and will result in a ‘blood debt’:

Here Kant attributes the views that he lists in the published passage to theologians specifically – they are the ones who think innocents should atone for sins. The second passage from the Metaphysics of Morals drafts contains the idea of justice being personified: ‘The virtue of the highest commander as such is justice. Beneficence can only be exercised at the expense of subjects. Of the idea of a justice that can be personified. A blood-guilt (Blutschuld) lies upon a country. In any event it must be removed by innocent successors, etc.’ (MMDr, 23: 354). Like the published passage, bloodguilt is Kant’s example of what personified justice might look like. Bloodguilt is like a stain or an entity that tarnishes a land, which must be removed. Included in this passage is also the claim that innocent successors might have to settle the blood debt.

These three passages contain many similar themes. Bloodguilt is a moral stain acquired by spilling innocent blood. That stain can attach to people or to a land, and it can be passed down to future generations. We feel so strongly that innocent blood should be avenged that we might be willing to spill the blood of people who are associated with the author of the original wrong even if they themselves are innocent. Kant’s talk of bloodguilt in the Doctrine of Right is disturbing enough on its own, so these additional passages – some of which are darker than the original – are unlikely to be attractive to many of Kant’s readers. Nevertheless, I want to propose that they are not as bizarre as they seem. All three are psychological descriptions of how human moral agents think about justice. In all of the passages, bloodguilt is symbolic, and I suggest this is precisely what Kant intends. My proposal is this: bloodguilt is a symbol that helps us explain the demandingness of justice.

2. Bloodguilt as Analogy or Symbol

It is important first to define symbols in Kant’s work.Footnote ⁸ The clearest description of symbols appears in section 59 of the Critique of the Power of Judgement. Kant introduces ‘hypotyposis’, which is a way of ‘making something sensible’ (CPJ, 5: 351). Symbolic hypotyposis occurs when the power of judgement uses an analogy to attribute an intuition to a concept ‘which only reason can think and to which no sensible intuition can be adequate’ (ibid.). In other words, symbols help represent concepts that cannot be represented in other ways. Kant’s famous examples are representations of states: ‘Thus a monarchical state is represented by a body with a soul if it is ruled in accordance with laws internal to the people, but by a mere machine (like a handmill) if it is ruled by a single absolute will’ (5: 352).Footnote ⁹ Of course, the relationship between the despotic state and the handmill is not one of straightforward resemblance. Instead, Kant claims that the similarity is ‘between the rule for reflecting on both and their causality’ (ibid.). When we reflect on the state ruled by a single absolute will and the handmill, we find similarities between the way they operate that allow us to come to further clarity about the despotic state. The handmill symbol ‘gives flesh to our understanding of the abstract notion “despotism”’ (Pillow Reference Pillow2001: 194).

In the same passage where Kant introduces the handmill symbol, he argues that our cognition of God is likewise symbolic:

Echoed here is Kant’s familiar language from the first Critique about the limits of speculative theology. If asked ‘whether we may not at least think this being [God] different from the world in accordance with an analogy with objects of experience’, Kant claims that the answer is ‘by all means’ (A697/B725, emphasis original). He claims that ‘we can allow certain anthropomorphisms … without fear or blame’, provided that we use them for practical rather than theoretical purposes (ibid.; cf. CPJ, 5: 456, 458). To use an example, Kant thinks we often analogize God to a ‘wise being ruling the world according to moral laws’ (5: 458). There is no harm in thinking symbolically about God; in fact, we are welcome to do so because it helps us gain clarity about the practical implications of the idea of God.

It is significant that Kant uses the example of God as symbolic thinking because it is in exactly this context where the bloodguilt passage at the end of the Metaphysics of Morals appears. In it, Kant is considering whether religion properly belongs to a metaphysics of morals. He argues that the ‘formal aspect of all religion’ belongs to morality because ‘this definition expresses only the relation of reason to the idea of God which reason makes for itself’ (MM, 6: 487, emphasis original).Footnote ¹⁰ Kant claims that the idea of God is what allows us to ‘make obligation (moral constraint) intuitive for ourselves’ because we cannot think of obligation without ‘thinking of another’s will’ (ibid., emphasis original). Kant is clear, however, that the idea of God does not entail duties to God. Instead the duty is ‘a duty of the human being to himself … for the sake of strengthening the moral incentive of our own law-giving reason’ (ibid.). The idea of God is something that reason ‘makes for itself’ in order to help it think about moral obligation.Footnote ¹¹

After these remarks about God, Kant introduces justice as ‘the principle of God’s right’ (MM, 6: 488). He then claims that, ‘in the judgment of our own reason, the claim that divine justice makes upon us … is that of punitive justice’ (6: 489). That is, reason is led to think of divine justice – the kind that God would seek on His own behalf – as punitive. Kant points out that humans have no way to conceptualize this sort of justice. He writes, ‘This concept is transcendent … and involves extravagant principles that cannot be brought into accord with those we would use in cases of experience’ (ibid., emphasis original). As Kant points out, however, divine justice is not a mere idea: it ‘makes demands on us’, which are ‘not only as great but even greater’ than God’s demands that we love others (ibid.). Since we cannot fully understand this sort of divine justice, we instead personify or symbolize it. Bloodguilt is then one form that this symbolization takes.

This bloodguilt passage, the passage on symbols from the third Critique and the passage on analogy from the first describe similar phenomena. Reason encounters a concept or idea that it cannot make sense of without reaching beyond the boundaries of experience. Nevertheless, that concept or idea makes practical demands on us. As a result, we must use symbolic thinking in order to better understand the idea or concept so that we can meet its practical implications. Just as we use analogies to think about God and a handmill to think about the despotic state, we use bloodguilt to think about justice. Kant argues that when we claim that crime must be avenged or that innocent blood seems to cry out for vengeance, we are thinking of ‘justice by itself, as a transcendent principle ascribed to a supersensible object’ (MM, 6: 490, emphasis original). As Kant points out, however, a transcendent principle of this kind is not something we can think clearly about because it ‘cannot be brought into accord with [principles] we would use in cases of experience’ (6: 489). Since ‘justice itself’ is not something we can make clear sense of, we think of it in other terms – as some sort of entity (ibid.). This is why Kant talks about bloodguilt lying upon a land: the unavenged crime is like a stain or a pool of blood. Likewise, in the longer draft passage, Kant claims that the theologians appeal to bloodguilt as a way of expressing their anger when crimes go unpunished (MMDr, 23: 248). The theologians treat bloodguilt as a hereditary curse or they analogize it to original sin – something that can be passed on even to our offspring. These sorts of descriptions match Kant’s explanation of how symbols function in the third Critique. Since we cannot represent ‘justice itself’ in other ways, we must use symbols to help us reflect upon it.

It is tempting to think that bloodguilt is so bizarre and gruesome that there is no way Kant could have reasonably thought it would help people make sense of justice. But the sentiment that bloodguilt represents is not so mysterious that we cannot see evidence of this sort of idea in our own lives. The notion that someone’s crimes will come back to haunt her even if she is not immediately punished for them is a common theme in films, television shows and novels. We use idioms like ‘what goes around comes around’ and ‘you reap what you sow’ to express this very idea albeit less colourfully. Again, Kant is not claiming that bloodguilt is a real entity or that it actually can be passed down to subsequent generations. He is also clearly not claiming that we ought to punish an innocent party in order to make up for innocently shed blood. Instead, the bloodguilt passages seem to be describing how people think about justice. Bloodguilt allows us to (as Pillow puts it) give flesh to an abstract notion of justice.

So far, I have argued that the three bloodguilt passages treat bloodguilt as a symbol for justice, but there are two unanswered questions. First, Kant is clear that our other kinds of symbolic cognition are supposed to help us gain clarity about one of the partners in the analogy (e.g. thinking about the handmill helps us thinking about despotism). What exactly is bloodguilt helping us make sense of? Second, symbols are supposed to have a practical purpose. Symbolic cognition of the idea God helps us clarify the ‘purposive use of it’ (CPJ, 5: 353). What is the ‘purposive use’ that bloodguilt is supposed to illuminate? I will answer the first question in section 3 and the second in section 4.

3. The Demandingness of Justice

To answer the first question, we need to know what Kant means by ‘justice itself’. First, given the similarity between the bloodguilt passage and other passages on symbolic thinking, I think Kant treats ‘justice itself’ as an idea of reason. Kantians are of course most familiar with ideas of reason in the context of the Transcendental Dialectic where Kant introduces the three transcendental ideas: the soul, the world-whole and God (A335/B392). But, as Rohlf (Reference Rohlf and Paul2010: 202–3) points out, there are many ideas that function in similar ways, such as the idea of a just constitution (A316–17/B372–4) and the ideas of pure earth, water and air (A646/B674). Typically, justice is not on the list of examples, but, as I will argue, several passages about justice suggest that it should be. Second, we must be clear what kind of justice bloodguilt represents. Bloodguilt appears always in discussions of punishment (usually for shedding innocent blood), but only in the Doctrine of Right does it refer to legal punishment. Everywhere else, bloodguilt appears in conjunction with a much broader conception of punitive justice. My suggestion here is that our idea of justice is the conviction that wickedness should be paid back with ill or harm.Footnote ¹² Kant has no systematic discussion of this idea, but he uses the phrase ‘idea of justice’ or ‘justice itself’, and we can see how it operates in some key passages.

In the Lectures on Ethics, we see a lengthy discussion of punishment where Kant switches back and forth between legal and moral punishment (L-Eth, 27: 551–6). Punishment in this section is defined broadly as ‘a physical evil apportioned to someone because of a moral evil’ (27: 547). Kant does not mean legal punishments here; he means the rewards and punishments that come from obeying or disobeying the moral laws. Kant puts it this way: ‘Our idea of justice requires that the moral worth of the action be recognized. We think it quite contrary to the order of things that a morally bad action should by its nature be coupled with impunity, and punishment depend merely on arbitrary chance; reason at all times connects the rectitude of moral conduct with worthiness for happiness’ (27: 552).

Similar remarks appear in the second and third Critiques. In the third Critique, Kant writes: ‘As soon as human beings began to reflect on right and wrong … the judgment must inevitably have occurred to them that it could not in the end make no difference if a person has conducted himself honestly or falsely, fairly or violently even if to the end of his life he has found no visible reward for his virtues or punishment for his crimes’ (CPJ, 5: 458). Kant describes it as hearing ‘an inner voice’ that says vice should not go unpunished (5: 459). In the second Critique, he claims that ‘there is in the idea of our practical reason something further that accompanies the transgression of a moral law, namely its deserving punishment’ (CPrR, 5: 37, emphasis original). Kant argues that this connection holds even when punishment is not meant to benefit the person being punished. Even if the transgressor ‘could see no kindness hidden behind this harshness’ he would nonetheless ‘admit that justice was done to him and that what was allotted him was perfectly suited to his conduct’ (ibid.). In this way, the idea of justice – deserving harsh treatment for a transgression – is prior to punishment. Kant writes: ‘In every punishment as such there must first be justice, and this constitutes what is essential in this concept’ (ibid.).Footnote ¹³

Later in the second Critique, Kant claims that ‘if someone who likes to vex and disturb peace-loving people finally gets a sound thrashing for one of his provocations … everyone would approve of it and take it as good in itself even if nothing further resulted from it’ (CPrR, 5: 61). He insists that even the troublemaker himself ‘must in his reason recognize that justice was done to him because he sees the proportion between well-being and acting well, which reason unavoidably holds before him, here put into practice exactly’ (ibid.). Even the offender’s own reason makes the connection between harsh treatment and his bad actions: he can think of himself as deserving what he gets even if he does not enjoy it. Further, Kant thinks we are particularly distressed when the wicked do not suffer for their wickedness. As he puts it in ‘On the Miscarriage of All Philosophical Trials in Theodicy’, we ‘witness with indignation a life led with crying injustice and yet happy to the end’ (MPT, 8: 261). But when ‘an unjust and especially violent villain does not escape unpunished from the world, the impartial spectator rejoices’ (8: 260 n.).

From these passages, we can reconstruct the following claim: on Kant’s view, moral agents have a conviction that wickedness ought to be paid back with ill or harm.Footnote ¹⁴ The precise nature of this conviction is difficult to specify, and Kant does not provide a clear definition.Footnote ¹⁵ I will use the term ‘practical demand’ because, as will become clear shortly, Kant thinks we experience this conviction as making demands on our reason or judgement. In part, the practical demand seems to be a felt need or desire: we want to see harm visited upon the wicked. Yet Kant also suggests that it goes beyond a desire. From our perspective, the idea that bad actions deserve ill seems like a conceptual claim: it ‘accompanies’ our understanding of transgression and is ‘combined with the concept of punishment as such’ (CPrR, 5: 37). In other words, the conviction that the wicked should suffer seems to us as though it is built into the very concept of punishment itself. Kant is clear, however, that we can provide no proof that wicked people will necessarily suffer for their misdeeds. As he is reported as saying, ‘That this link is a necessary one, and physical evil a direct consequence of moral badness … cannot be discerned through reason nor proved either, and yet it is contained in the concept of punishment’ (L-Eth, 27: 552). We can see why this would be the case by considering Kant’s remarks on the necessary link (or lack thereof) between virtue and happiness. Kant recognizes that ‘no necessary connection of happiness with virtue in the world … can be expected from the most meticulous observance of moral laws’ (CPrR, 5: 113). In the same way, we also cannot assume the opposite: we cannot expect that vice will necessarily cause unhappiness. As he puts it:

On Kant’s view, the notion that moral transgressions deserve to be punished is something we cannot help but think in spite of the fact that we know the wicked will not necessarily get their comeuppance. We are thus left with the problem of reconciling two conflicting things: (a) our inability to prove that good conduct will be met with happiness or that wickedness will be met with harm, and (b) the practical demand that we experience, which tells us that wickedness deserves punishment. In light of this conflict, we try to gain a clearer or more refined grasp of the ‘inner voice’ that tells us wickedness must be punished. Why do we hear this inner voice? What exactly are we demanding? We experience the conviction that wickedness deserves punishment as particularly forceful or stringent and we need a way to clarify this conviction for ourselves.

As Kant argues in the first and third Critiques, symbols play this clarifying role. The bloodguilt passage from the end of the Metaphysics of Morals explains how we personify justice in order to make it thinkable for us. In this passage, Kant claims that punitive justice makes demands on us ‘in the judgment of our own reason’ (MM, 6: 489). Yet in spite of these demands, any theoretical principles we try to find to ground them are ‘quite empty for our practical reason’ (ibid.). Representing justice symbolically as bloodguilt captures the way we experience the demand that wickedness be met with punishment. Thinking of this demand as though it is an otherworldly entity or a moral stain thus helps us make clear for ourselves the ‘iron, inevitable necessity’ we think that it has (ibid.). Justice seems to us to be part of the order of the world; it seems like a force that lies beyond us. The ‘ancient philosophical poets’ talk about justice as fate or destiny because they share similar characteristics (ibid.). Fate, as the ancient poets understand it, is inescapable, and justice seems likewise inescapable. We think of justice as making demands or claims on us, so we endow it with agential characteristics. The innocent blood that is shed takes on those characteristics – it ‘cries out for vengeance’ (ibid.).

We also treat bloodguilt like a stain or a hereditary trait because these ways of thinking capture its apparent permanence. It ‘lies upon a land’ (MMDr, 23: 248) or ‘lies upon a country’ (23: 254) as though it inhabits the very ground under our feet. Bloodguilt attaches to the criminal’s descendants, which is why people sometimes feel as though ‘children must atone for their parents’ guilt’ (23: 248). The demands of justice are unyielding, which is why we sometimes claim that justice must be satisfied even if the next blood spilled is the otherwise innocent descendant of the original murderer. Because we experience the demands of justice as strict, we imagine that the moral stain of an unavenged crime could be passed down to future generations. Kant is not claiming that we ought to punish the innocent just so blood can be washed away with blood. Bloodguilt is not a literal recommendation; it is a symbol. Just as the handmill represents the despotic state, bloodguilt represents the tight connection we make between wickedness and punishment.

4. The Purposive Use: The Original Bloodguilt Passage

The second question I need to answer is about the practical implications of bloodguilt as a symbol of justice. For example, symbolic cognition of God, on Kant’s view, partly helps us determine ‘what the idea ought to be for us’ (CPJ, 5: 353). If I am right that bloodguilt helps us clarify the demandingness of justice, to what end does it do this? To answer this question, I will re-examine the bloodguilt passage from the Doctrine of Right because I think it is where we can see how the practical implications of bloodguilt might work.

First, it is important to note that some of Kant’s language in the section in question (section E of the General Remark following §49) resembles his language in the other bloodguilt passages: it is often dramatic and heavy-handed. Kant talks about justice as something we must ‘satisfy’ (MM, 6: 333). It is also ‘pure and strict’ (6: 332). A justice that can be ‘bought for any price whatsoever’ actually ‘ceases to be justice’ (6: 333). These descriptions sound like the ‘iron, inevitable necessity’ that justice has in the bloodguilt passage at the end of the Metaphysics of Morals (6: 489). If the Doctrine of Right bloodguilt is the same as the others, then what role is it playing? I suggest that one of Kant’s concerns in section E, although he does not state this explicitly, is how to make rightful punishment (that is, punishment doled out by the sovereign in the civil condition) consistent with our basic conviction about justice.Footnote ¹⁶ As Fletcher puts it, ‘formal principles of Right do not mix readily with substantive criteria for justice’ (Reference Fletcher1989: 214). In other words, moral agents do not approach the question of how to implement legal punishment as blank slates. The belief that people who do bad things should have bad things happen to them is prior to our thinking about the penal system.Footnote ¹⁷ Kant recognizes that the state cannot concern itself with punishing vice and can only punish crimes (6: 331). Yet he also recognizes that there will be some overlap between the demands of justice and legal punishments. If this is so, we are faced with the practical problem reconciling our basic convictions about justice with the justice that the state is able to dole out. I do not mean to suggest that Kant thinks we ought to make legal punishments as close to moral punishments as we can. Instead, his concern is that the penal system ought not contradict our idea of justice.

Many of Kant’s remarks in section E seem to involve concerns about how to reconcile these two things. First, proportionality is central to our concept of justice: the more wicked someone is, the more he should suffer. Since justice demands that wickedness be punished proportionately but the state cannot properly determine someone’s wickedness, the best we can do is to try to punish people using the ‘law of retribution’ because only it can ‘specify definitely the quality and the quantity of punishment’ (MM, 6: 332). Only lex talionis is suitable ‘for a sentence of pure and strict justice’ (ibid.). Justice is pure and strict because we experience the connection between wickedness and punishment as a conceptual link. On Kant’s view, the law of retribution captures perfectly this conviction: whatever harm or ill you do to someone, you deserve equally in return. Thus only the law of retribution can ensure (as best we can) that the criminal is punished in the proper proportion. Further, when wicked people escape punishment, we think of it as a grave injustice. Our convictions about punitive justice can explain why Kant is wary of the sovereign’s right to grant clemency. In the longer bloodguilt passage from the drafts, he claims that we find pardon ‘as contrary to justice as the highest holiness’ (MMDr, 23: 248). The idea that the sovereign would grant pardon to a murderer defies our most basic conviction about what justice requires. In the Doctrine of Right, Kant warns that the sovereign who grants clemency to a criminal does ‘injustice in the highest degree’ because the sovereign offends his subjects’ ideas about the demandingness of justice (6: 337).

Likewise, if Kant is concerned about the consistency between legal punishment and our idea of justice, it can explain why he worries about punishment and honour. Right after the bloodguilt passage, he claims that the death penalty punishes the murderer ‘in proportion to his inner wickedness’ (MM, 6: 333, emphasis original). Kant imagines two people involved in the Scottish rebellion: a man of honour who believed that he was ‘performing a duty’ he owed to the House and a scoundrel who was only looking out for ‘private interests’ (ibid.). If we were to allow both rebels to choose their punishment, Kant thinks ‘the man of honour would choose death, and the scoundrel convict labour’ (6: 334). The death penalty in a case like this punishes in proportion to each one’s inner wickedness. Kant explains, ‘Since the man of honour is undeniably less deserving of punishment than the other, both would be punished quite proportionately if all alike were sentenced to death; the man of honour would be punished mildly in terms of his sensibilities and the scoundrel severely in terms of his’ (ibid.). We need not defend Kant’s views about the death penalty to see how he arrives at this example. Justice demands that wickedness be punished, but the state cannot fully determine the inner wickedness of criminals: it only punishes actions and not hearts. Yet we are convinced that the scoundrel deserves worse than the man of honour because less wicked people deserve a less harsh punishment. In Kant’s telling, giving each man the death penalty mitigates our worries that the man of honour is punished too harshly.

The same concern arises in Kant’s discussion of the unwed mother and the military duellist.Footnote ¹⁸ Kant claims that murders committed for the sake of honour (infanticide of an illegitimate child and military duels) are less deserving of the death penalty than other murders (MM, 6: 336). If the state decides to pursue capital punishment in these cases, Kant warns that ‘the public justice arising from the state becomes an injustice from the perspective of the justice arising from the people’ (6: 337, emphasis original). For there to be a conflict between the state and the people, there must be some standard of justice to which the people are appealing to support the claim that the death penalty is unjust in these cases – that standard is our conviction that wicked people deserve worse than honourable people. We thus encounter a tension between what we think justice requires and what the state can implement. As Kant puts it, legislation cannot ‘wipe away the stain’ of a bastard child or military cowardice (6: 336). In these cases, our idea of justice leads us to conflicting conclusions. On one hand, both the unwed mother and the duellist commit murder. On the other hand, they do so for (on Kant’s view) legitimate reasons. Because the state cannot legislate away someone’s dishonour, the death penalty in these cases seems too harsh.

Additionally, my interpretation of bloodguilt can explain Kant remarks in section E about the temptations we face in implementing legal punishments. We think of the demands of justice as unconditional, which is why bloodguilt helps us to make sense of them. At the same time, the demands of justice can seem obscure and hard to articulate, which is why we think of them by means of a symbol in the first place. Because ‘justice itself’ is vague and difficult for us to grasp, doubts about fulfilling its demands will easily creep in, and it will be easy for us to try to rationalize them away. We face these rationalizations when instituting policies about punishment, especially when that punishment seems to require that the state should end people’s lives. In the beginning of section E, Kant describes these rationalizations: the person who ‘crawls through the windings of eudaimonism’ is seeking some way out of punishing the offender (MM, 6: 331). The suggestion that we ought to ‘preserve the life’ of an offender and do ‘dangerous experiments’ on him so that ‘physicians learn something new of benefit to the commonwealth’ is yet another bargain Kant thinks we might be tempted to make (6: 332). Because the demands of justice seem mysterious in the face of pragmatic considerations, we might start thinking that punishment ought to serve some further beneficial goal. Why not, the rationalizations will say, get some good out of this burdensome punishment business by doing experiments on prisoners and putting them to good use? What does ‘doing what justice requires’ get us at the end of the day?

This is why Kant uses the inhabitants of the dissolving society as an example. The inhabitants of the island assume that if society dissolves there is no point in executing the last murderer. But this is to assume that the point of executing murderers is for some benefit to society and not because it is simply what justice demands. The members of dissolving society are in this way ‘collaborators’ in a violation of justice (MM, 6: 333). The justice they violate is not the justice of the civil condition (they already violate that by disbanding in the first place), but rather the demands of ‘justice itself’. If the bloodguilt that clings to the island’s inhabitants for not executing the murderer is the same bloodguilt that appears elsewhere in Kant’s writings, then we can see why he would employ it here. Bloodguilt is a symbol for the strict demands of justice, which entails that wickedness should not be met with impunity. Kant thinks that leaving innocent blood unavenged because we do not see what benefit can come from it is contrary to the very idea of justice. As he puts it, ‘justice ceases to be justice if it can be bought for any price whatsoever’ (6: 332). On Kant’s view, human beings who are willing to let innocent blood be spilled with no consequence are abandoning their basic convictions about justice.