Bentham: Punishment and the Utilitarian Use of Persons as Means

One of the main barriers against a Utilitarian justification of punishment is a widespread criticism that if punishment is evil justified by the good it can achieve, then the state could use persons as a means to an end in pursuing this good. This opens the door, at a theoretical level, for the potential punishment of innocents, disproportionate punishment and failure to respect persons as rational and responsible agents. Further, critics argue that any considerations of security or utility guard against the perceived risks contingently, without intrinsic commitment to respecting persons as ends in themselves. This article addresses the criticism fundamentally by returning to Bentham’s original writings and demonstrating that a principle of equality is embedded in the greatest happiness as an end of government. The principle of equality can theoretically be developed using the tools of Bentham’s political theory, including his commitments to democracy, to the elimination of pain and to the differentiation between real and fictitious entities, to ensure that a Utilitarian theory of punishment, as part of its premise, would be constrained from using persons as mere means. Further, building on the equality of happiness, the article proposes an individualistic justification of punishment that responds to the traditional accusations of innocents’ punishment and excessive punishment, and ensures the respect of persons as rational and responsible agents.

Utilitarianism characterises punishment as an evil that inflicts pain and permits the state to inflict it only if it excludes a greater evil, namely the commission of offending in the future. Although there is value in characterising punishment as pain and an evil to be used by the state as minimally as possible, one of the main barriers against a Utilitarian justification of punishment is the widespread criticism that it would permit the use of persons as a means to an end and sacrifice an individual's interests for the supposed good of a community. In addition, such a theory of punishment is commonly criticised for potentially tolerating both the punishment of innocents and excessive or cruel forms of punishment 'when doing so will advance utility', 1 as well as for failing to treat persons with the respect owed to them as responsible and rational agents. 2 In addressing possible use of persons as a means to an end, scholars have responded in various ways. Smart accepted that it is a possibility within a Utilitarian theory of punishment. 3 Binder and Smith argued that punishment of innocents is contrary to Utilitarian considerations of security, legality and transparency. 4 Hoskins argued for the compatibility of deterrent punishment with the Kantian principle of respect for persons while placing partial retributive limitations to guard against the punishment of innocents and disproportionate punishment. 5 However, for critics of a Utilitarian theory of punishment, the problem remains that Utilitarianism does not subject itself to a priori moral principles of justice, nor to the Kantian principle of respecting persons as ends in themselves. 6 Utilitarianism cannot therefore guard against sacrificing persons while seeking to maximise utility because any protection it offers is contingent and can be overridden. 7 This article aims to solve this particular problem. It argues that the greatest happiness, as the end of government, cannot be achieved without a commitment to recognise and respect every individual's happiness as equal in value. This equality is not an external moral principle, but part of the premise of Bentham's political theory; it is embedded in the greatest happiness and the principle of utility. Further, building on the principle of equality and using the tools of Bentham's political theory, including his commitment to democracy, to the elimination of pain and to the differentiation between real and fictitious entities, this article develops an individualistic justification of punishment that guards against the use of persons as mere means to an end.
The first section demonstrates how equality is embedded in the greatest happiness principle as an end of government. The second section extends the argument to the Utilitarian justification of punishment. The third to fifth sections demonstrate how the principle of equality and the individualistic justification of punishment might guard against traditional criticism of Utilitarian punishment in three respects: the punishment of innocents; excessive or disproportionate punishment of the guilty; and the reasons offered for obedience to the law.

The Greatest Happiness and Equality of Happiness
The accusation that Utilitarianism allows for the sacrifice of individuals is usually associated with the common interpretation of the greatest happiness principle as aggregative. Such an interpretation opens the door for the sacrifice of some, whether a few or a single individual, as long as the total is profit in happiness. Less common interpretations of Bentham's work offer evidence and arguments for egalitarian distribution of happiness within Utilitarianism. 8 These interpretations discuss what Bentham meant by maximisation and the dictum that 'every individual in the country tells for one; no individual for more than one', 9 famously expressed by Mill that 'everybody to count for one, nobody for more than one'. 10 This section aims to take the discussion forward by exploring what an explicit principle of equality in Bentham's political theory can offer to guard against the use of persons as mere means.
The starting point is that Bentham -towards the end of his lifecame to fear what aggregation or the 'greatest number' as an annexation to the greatest happiness principle could mean for minorities and individuals. 11 In 1828 Bentham realised that the phrase 'the greatest happiness of the greatest number' could lead to '"effects widely different from those intended -in a word mischief to an almost indefinite extent -might be produced", namely by simply ignoring the interest of the minority'. 12 In 1829 Bentham worried that the greatest number 'might be used to justify sacrificing entirely the happiness of a bare minority in the interests of a bare majority'. 13 In trying to solve this problem in the Article on Utilitarianism, Bentham proposed an example of a community of 4001 persons, in which each person is given an equal portion of happiness and equal means to happiness. The happiness, and the means to it, were then taken from 2000 individuals and given to the remaining 2001, who then enslave the 2000. In this scenario, Bentham asks whether utility is maximised given the 2001 possess the happiness and means of the entire community? His response is that '[o]n the contrary, the whole profit will have given place to loss'. 14 The reason is that by the nature of humans, 'the quantity of unhappiness' they can feel in any given time is greater than 'the quantity of happiness'. 15 Therefore, as the unhappiness of the 2000 will outweigh the happiness of the 2001, the total is a loss and thus contrary to the principle of utility.
However, a simple reply to this equation is: if the sacrifice of happiness was of only one person to give happiness to the 4000, should not Utilitarianism support this sacrifice? This is precisely the point raised by critics. The problem remains unresolved because the issue is not the balance between the quantity of pain or pleasure, but the disregard of the pain and pleasure of the minority, or in this case one individual. How can Utilitarianism, then, prevent the state from disregarding the pains and pleasures of a minority or an individual? The answer to this question can be developed from Bentham's political theory.
Bentham's political thought was underlined by a concern that 'in every political community […] the interest of the many had been sacrificed to the particular interest of those by whom supreme power was exercised'. 16 This sacrifice was seen, for example, in a partnership between parliamentarians and the legal profession in which '[p]arliament protected the lawyers in the enjoyment of their abuses, while the lawyers supported the political establishment in resisting reform'. 17 Further, parliamentarians depended on natural law and natural rights as guidance for legislation, doctrines which were abstract and 'classed by Bentham under the heading of the principle of sympathy and antipathy. We invoke them, he argued, because we approve of them and that is taken as a sufficient reason for their institution'. 18 The use of abstract notions obscured embedded subjectivism, and Bentham found that there was a need for an external ground on which to decide 'the value of laws and policies' and to hold accountable those who exercise power. 19 This ground was found in the utility principle as built on a differentiation between real and fictitious entities -a differentiation that underlies Bentham's 'whole system of thought'. 20 Real entities refer to 'physical objects', while fictitious entities refer to 'abstractions'. 21 For any principle to make sense, it must be broken down and referred back to real entities, in this case, pleasure and pain. 22 This is the heart of the advantage that Utilitarianism claims to offer, compared to other philosophical theories -namely its ability to be linked to individuals' sensational existence instead of referring to abstract notions that reflect subjective views. Such an advantage was expressed by Alexander in arguing that Bentham aimed […] to annihilate Nature as the source, legislator or measure of law. In her place, he wanted to set humanity. The utilitarian person, rational, feeling, autonomous, capable of selfgovernment and self-direction, was to be the measure and source of all things. It was through humans and their experience of pain and pleasure that the virtue or evil of law would be calculated. 23 Bentham therefore developed the principle of utility as a standard by which to censor or critically to examine the laws and actions of the state. A citizen can censor the government by assessing the extent to which it promotes pleasures and prevents pains for the greatest number of people, rather than promoting the interests of a smaller governing elite. The government and the legislature have to justify their policies and decisions in terms of the happiness experienced by people and the pains avoided, not on the basis of self-evident subjective views.
However, the principle of utility and the greatest happiness as an end of government faced two problems throughout Bentham's life. First, in pursuing happiness, individuals are under the influence of the 'universal self-preference-announcing principle', 24 providing that 'a man pursues his own happiness in preference to that of all other individuals put together: in preference to, and […] to the sacrifice of their happiness'. 25 This principle applies both to citizens and to the government. The government will not pursue the greatest happiness because those governing will always prefer pursuing their 'sinister interests' instead of the public interest. Each citizen also prefers to maximise his/her happiness at the expense of all others. Second, and to make the problem more complicated, individuals have very different and subjective views of what constitutes happiness, so that '[e]very person is not only the most proper judge, but the only proper judge of what with reference to himself is pleasure: and so in regard to pain'. 26 These problems pose two questions. First, how is it possible to ensure that the government pursues the promotion of the greatest happiness, instead of rulers' sinister interests? Second, in the face of selfinterest and the endless subjective different pleasures that people may pursue, how would the government decide what is the greatest happiness and achieve it as its end?
Bentham's solution was found in equality and democracy. In response to the problem that individuals have very different and subjective views of what constitutes pleasures and pains, Bentham's solution was to assume an explicit principle of equality in happiness. The principle of equality states that '(1) each individual has an equal right to all the happiness he is capable of experiencing; (2) individuals possess an equal capacity to experience happiness; and (3) individuals possess an equal desire for happiness'. 27 In a footnote on this principle, Bentham [Bentham's Utilitarianism] was the claim that everyone's happiness should be given equal consideration'. 29 Building on equality of happiness and trying to solve the problem of self-preference, Bentham was led to conclude that the best form of Utilitarian governance is democracy. On the one hand, periodical elections and securities against misrule create a government's dependency on the people -the government must pursue the greatest happiness or else it will be removed at the next election. On the other hand, people are given an equal chance each to pursue and express the vision of happiness they want the state to pursue. Bentham contends that when a citizen casts a vote, he/she would not find a candidate who promotes their unique sinister interests 'at the expence [sic] of all other men without distinction, and in particular at the expence [sic] of his fellow-citizens'. 30 Therefore, he/she would have to vote for the candidate who -to a greater degree -supports their interests as part of the general interest shared with others. 31 One can argue that, according to this logical sequence, the principle of equality is the means of achieving the greatest happiness. The diverse and subjective nature of happiness necessitates giving each individual an equal chance to express what happiness means for them through their vote. This precedes the government's task to serve whatever greatest happiness the voters agree that the state should pursue. It is not guaranteed that the individual's happiness will be realised, for it will depend on the extent to which it is shared with others in society. However, each individual receives an equal chance to determine, express and pursue his/ her subjective happiness through the vote.
One can develop Bentham's principle of equality further. The equality between individual votes in elections is a manifestation of a more profound idea: individual self-determination. Every individual is entitled to decide what is happiness for him/herself and to pursue it in terms of seeking pleasures and avoiding pain. Such pursuit happens through endless venues, one of which is through voting in deciding the collective greatest happiness. In this venue, the state accords equal recognition, respect and treatment to each individual vote. However, in every other venue, the state must similarly recognise, respect and treat each individual's chance to pursue happiness to be of equal value.
Such equal self-determination is inexorably connected to being a human, a real entity who feels pain and pleasure, and to the fact that each individual is 'the only proper judge of what with reference to himself is pleasure: and so in regard to pain'. 32 The state's commitment to recognise and respect individuals' pursuit of happiness follows from the principle of utility, which dictates the government's commitment to facilitate people's happiness in terms of pursuing pleasures and avoiding pain. It is a continuous process integral to understanding human life as advanced by Utilitarianism, where humans are continuously pursuing happiness, 33 under the influence 'of two sovereign masters, pain and pleasure'. 34 Equal treatment by the state means that individuals in similar comparable positions should receive similar comparable treatment. Equal recognition and respect require the state to see every individual not as a number equal to one in an aggregative equation, but as a real entity with a subjective conception of pleasures and pains, with equal capacity and desire for happiness. Postema expresses this point in what he calls an individualist conception of happiness, acknowledging that individual people suffer, people flourish, people take delight in or are distressed by the events of their lives, and these experiences cannot be abstracted from the significance they give to human lives. What gives pursuit of well-being its moral point, on this view, is that we ought to care about the people who experience it. 35 The importance of state recognition of individuality is emphasised by Bentham in arguing that it would be a deceit to talk about public interest without considering every individual interest. 36 For him, public interest 'represents nothing but the mass of individual interests […] Individual interests are the only real interests. Take care of the individuals; never molest them, never suffer any one to molest them, and you will have done enough for the public.' 37 There remains, however, a question: the whole democratic scheme is based on the prevalence, ultimately, of the choice of the majority, so what difference then does equality offer for minorities and individuals? Equality, when combined with the Utilitarian commitment to the elimination of pain -where the ideal is that 'state-produced "pain" [should be] entirely eliminated' 38 -offers protection against the use of persons as mere means. On one level, if the state must cause pain, as is the situation in all government work that imposes burdens on individuals, it should do so minimally and equally. First, the state should always seek to limit the degree of pain it imposes on individuals, given the purpose of its existence should be to facilitate individuals' pursuit of happiness. Second, in terms of equality, the state must not impose pain and restrictions on some individuals for pursuing happiness in a particular way, while giving others in similar positions unfettered freedom to pursue those same paths.
On another level, the value of happiness should not be determined abstractly and collectively as a minority to a majority, or an individual to society. Instead, the state should give equal value to each individual's chance to pursue happiness, 'taking them one by one' as real entities, humans who feel pain and pleasure. 39 The state cannot subject an individual to pain or burden just for the sake of increasing the pleasure of another individual(s), otherwise, it would be placing a higher value on the happiness of the beneficiary individual(s) over that of the sacrificed individual(s). The principle of equality effectively means that 'there […] [is] no justification for favouring a particular individual or a particular class of individuals at the expense of others'. 40 The state must offer a justification for imposing pain and it must impose it equally. Now, if we turn to the authority to punish, the question becomes: what is the state's justification for imposing pain and suffering on offenders?

Punishment and Pursuit of Happiness
Punishment responds to the problem created by the self-preference principle -that is, even if the state guarantees equal chances to all to pursue happiness, individuals themselves will interfere with each other's paths, causing pain to one another. A choice then faces the state. First, it can choose to do nothing. In this case, to honour its commitment to equality, it should refrain from interfering in anyone's path. The result would be the absence of protection from pain altogether. To borrow Bernard Williams's words, the first political question of 'securing […] protection [and] safety' would be unresolved. 41 Consequently, the result would be the inability to pose or solve any other questions related to the pursuit of happiness individually or collectively.
Second, the state can choose -as Bentham argues -to 'interfere only to prevent […] injuring each other […] it is there that the application of punishments is truly useful, because the rigour exercised upon an individual becomes in such a case the security of all'. 42 Mill later developed the harm principle, similarly finding that 'the only purpose for which power can be rightfully exercised over any member of a civilised community, against his will, is to prevent harm to others'. 43 Utilitarianism's problem, at this point, is that the justification of punishment becomes 'the security of all' 44 as an abstract fictitious entity, instead of being broken down and interpreted in terms of the real entities to which it refers. When Bentham justified punishment, he referred to a real entity, for instance to prevent one person from injuring another person. When we abstract this formula to security or to the prevention of harm generally, the link between the causer of pain and punishment is severed, creating the possibility of using persons as a means to an end, at least in some serious respects. The Utilitarian aim of punishment became the prevention of crimes, given crimes are pain, but as if pain is an abstract concept. However, in order to reflect the real entity dimension of pain as a physical sensation, we must ask who suffers pain, how much pain and who or what caused the pain. By answering, we bring the individual who caused the pain back to the centre of the discussion.
Integrating this real entity dimension with equality of happiness, we find an individualistic justification of punishment, re-establishing the link between the causer of pain and punishment. To clarify, the state is required to recognise and treat each individual's chance to pursue happiness as being of equal value. The chance to pursue happiness is the chance to pursue one's pleasures without being subjected to pain. The state, in recognising this, should refrain from causing pain to individuals as far as possible. Nonetheless, it is faced with the problem that some individuals in their pursuits cause pain to others. In this case, the authority to punish becomes justified only towards those who cause pain and injury to others, to prevent them from causing pain again. Otherwise, if the state causes pain (punishment) to an individual who caused no pain (crime), it would violate the equal recognition and treatment of this individual's chance to pursue happiness, in comparison to the chance that every other law-abiding citizen has.
This applies equally to all individuals who commit injuries or forms of pain falling within the scope of the criminal law. The state cannot therefore punish a person who has not injured another person, nor can it punish someone who has injured others while deliberately leaving another who has equally and similarly injured others without punishment. This point responds to another criticism of Utilitarianism: that the theory might advocate the punishment of one in every ten offenders if that proves to be more effective, leaving nine out of the ten unpunished. This scenario in a Utilitarian society governed by the equality of happiness is unacceptable, because punishing only one in every ten offenders means unequal recognition and treatment by the state of the one punished and the nine who retained an unfettered chance to pursue happiness despite equally causing injury. The state is therefore not allowed to use persons as a means to an end of collective happiness, because for those who are used as means, their happiness has not been recognised and treated as of equal value compared to every other individual. They have been recognised and treated as less equal, as mere means to others.
At this point, it is important to ask whether one can say inflicting pain in response to an act of the offender renders a Utilitarian theory of punishment retributive, in terms of being backward-looking? One can argue to the contrary for several reasons. First, punishment can only be justified when it is forward-looking and prevents persons from injuring one another. 45 Second, inflicting punishment is not a valuable practice in and of itself, nor is it an adequate response to injury. The evil of punishment is to be avoided, unless it is the only necessary method that responds to the evil of the offence compared to all other alternatives. Third, offenders do not deserve to suffer. Suffering is pain, and a Utilitarian state should refrain from causing pain and suffering insofar as it can, and the offender's chance to pursue happiness should be equally recognised and treated as such by the state. However, given the offender has caused pain to another, doing nothing about it is not an option for the reasons explained above. Besides, it would mean that the victim's chance of pursuing happiness has not been recognised and treated equally by the state. It therefore becomes necessary to punish the offender as minimally as possible to prevent him/her from causing pain again. How such prevention can be achieved in practice should, one thinks, be a question of empirical evidence about the programmes that succeed in helping offenders reintegrate into society to pursue a law-abiding life, in which they may pursue their happiness without causing pain to others.
It is essential to add that this view of the equality of happiness does not presuppose that individuals start from equal positions in life, or that punishment remedies inequalities caused by crimes. Rather, it admits that there are embedded distributive inequalities that have not been eradicated from most societies. Nonetheless, individuals living together in a society should be able to lead their lives in pursuit of what makes them happy without being subject to the pain of crimes from others. Now, to assess the extent of the success of the equality principle in guarding against using persons as means, the next sections analyse its application to three issues in the context of punishment: namely the punishment of innocents, the excessive punishment of the guilty and the reasons Utilitarianism offers for obedience to the law.

Using Innocents as Means
When Bentham discussed the punishment of innocents in the Rationale of Punishment, he argued that 'It is […] for the most part useless, and whenever it is not useless, it is mischievous.' 46 It is useless because none of the ends of punishment are achieved. The guilty offender has not been apprehended, nor has he/she suffered any consequences for the crime committed -there was therefore no deterrence, no reformation and no prevention of future offending, in addition to the unjustified pain suffered by the innocent. 47 The innocents' punishment thus violates the state's commitment to eliminating pain and results in 'so much evil expended in waste […] it is repugnant to utility, inconsistent with humanity'. 48 Further, even if it were proved that an innocent's punishment in some rare case would be useful to one of the ends of punishment, it would still be mischievous; 49 that is, the pain of punishment outweighs the utility sought and creates general danger and pain of insecurity in the wider community. 50 For critics of Utilitarianism, this is not a satisfactory answer because 'what is wrong with sacrificing the innocent is not that it will endanger […] others, what is wrong with it is that an innocent person is sacrificed'. 51 Thus, in this line of argument, Utilitarianism fails to grasp the intrinsic wrong of the practice. 52 However, for Bentham, '"the only objects possessed of intrinsic and independent value" were pleasure and pain', 53 and these are the bases on which he assessed right and wrong. For him, moral judgements are abstract notions, fictitious entities which can only acquire meaning when they refer to the real entities of pain or pleasure. Therefore, when Bentham says that an innocent's punishment is contrary to the utility principle, he is actually saying that it is wrong, and intrinsically wrong because it causes pain -a real entity of intrinsic value.
The problem, then, is not that Bentham did not appreciate the wrongful nature of punishing an innocent. The problem is that the interpretation of pains and pleasures as aggregative means that the only possible way to declare the act wrong is to highlight the quantity of pain resulting from an innocent's punishment. The logical solution to declare the act wrong is therefore to emphasise how it is not one individual pain only, but also a collective pain of insecurity that exceeds any intended collective happiness.
This paper proposes instead to give effect to the state's commitment to the equality of happiness, which would mean the profound rejection of the punishment of innocents without reference to the aggregation of pain and pleasure. To show how this proposition would work, I will illustrate using the famous example of framing a black person for rape: A black man has raped a white woman in a virulently racist community, and white bigots are on the verge of mass lynchings of blacks. Suppose I have the ability to frame a particular black man, whom I -and I alone -know to be innocent. If I know that my doing so would save many lives while sacrificing only one individual, consequentialism would dictate that I am duty-bound to get the innocent man convicted. 54 In this example, the problem is that the innocent's pain is measured in an aggregative way on a ratio of one to many, namely punishing and killing one person will save many more lives. Let us suppose that this community is composed of 101 individuals. The state should not combine the happiness of the 100 and compare it to the sacrificed individual as 1. Instead, it should ask why is the chance to pursue happiness for each individual in the 100 more important than the framed individual's chance?
The answer should not be about the quantity of happiness achieved, but about the recognition and quality of treatment that the individual received from the state as an equal. The framing of the innocent is a violation of the commitment to equality: it fails to recognise and treat the innocent person's chance to pursue happiness as of equal value in comparison to the chance given to every other individual -that is, every individual who committed no injury is immune from state interference and imposition of punishment. Moreover, the official representing the state's authority should not try to protect the community from pain as an abstract threat. Rather, they should use that authority to protect it by preventing persons from injuring one another. The official should therefore both prevent the real offender from causing pain again and prevent the white bigots in the said example from injuring members of the community.
In sum, a Utilitarian theory that is essentially committed to equality and its implications would not approve an innocent's punishment. This is not because the punishment of an innocent creates general insecurity for others, but because (a) the justification for punishment is tied to preventing persons from injuring one another, and the innocent by definition has committed no such injury; and (b) to ignore the fact that the innocent committed no injury and to punish him/her nonetheless means that the state is recognising and treating the innocent's chance to pursue happiness to be of much less value, compared to the chance given to every other individual.
Before leaving this point, it is worth exploring the possible critique that providing an equal chance to pursue happiness might seem similar to the arguments of Braithwaite and Pettit, who interpreted utility as the maximisation of every person's liberty or what they called 'dominion'. It may be argued that the equality to pursue happiness can be viewed as the freedom or liberty to do as one wishes subject to the limits of the law -a concept that comes very close to the idea of dominion. Dominion, according to Braithwaite and Pettit, is the social status you perfectly enjoy when you have no less a prospect of liberty than anyone else in your society and when it is common knowledge among you and others that this is so. The one qualification is that if you and your fellow-citizens are all equipped to enjoy dominion, being equal in relevant regards, you must have the largest prospects of liberty compatible with that equality. 55 Although this paper argues in a similar vein for justification of punishment within a Utilitarian theory, the idea of the equality of happiness differs from that of dominion. Dominion is composed of different liberties and can be maximised by increasing individuals' different liberties. It is a change in the good sought to be maximised from the greatest happiness to dominion, but it remains an aggregative goal. In contrast, what this paper proposes is a preliminary premise that acts as a block: that as long as a member of the society did not cause pain in violation of the law, the state is not justified in subjecting him/her to pain as part of his/her equal chance to pursue happiness as he/she wishes. As Matravers argues, 'Braithwaite's and Pettit's position is untenable because its structure remains that of identifying a good (in their case dominion) and promoting the aggregate of that good'. 56 This is the problem this article seeks to identify and overcome, by switching focus from aggregate utility to the equal chance of happiness enjoyed by every member of the society.

Using Offenders as Means
The second criticism of a Utilitarian theory of punishment is that it might use offenders as a means to achieve the end of deterrence more efficiently by punishing them excessively or disproportionately. One can argue, in the first place, that this is more a criticism of deterrence theories and by extension directed towards Utilitarianism. Deterrence theories focus on inflicting -by punishment -a level of pain that exceeds the crime's profit, thereby providing an incentive for the individual not to commit a particular crime, as well as a general threat for others to abide by the law. Individuals will calculate that committing an offence will ultimately cause them more pain, if they are caught and punished, than pleasure. There is thus no objection -in theory -to increasing the pain of punishment to an unlimited extent, so long as it exceeds the expected profit of the offence.
However, according to Bentham's Utilitarianism, the pain inflicted by the state should always be the minimum necessary to prevent the future offence. Moreover, if the justifying aim of punishment, as argued in this paper, is preventing people from injuring one another, it is not necessarily or exclusively achieved by deterrence. Instead it is a goal that should be informed by empirical evidence of what better helps offenders desist. The question remains, though, of whether Utilitarianism is compatible with the cruel, excessive or disproportionate punishment of offenders, if that could achieve more effective prevention of persons from injuring one another. One can say there are two scenarios of excessive or disproportionate punishment: first, punishing some individual offenders excessively, contingent on what better serves the prevention of offending; or second, punishing excessively all offenders, even for minor crimes that might have been prevented with less punishment.
In the first scenario it is a violation of the equality of happinessbecause a punishment system that subjects some offenders to more pain than those who committed similar crimes treats unequally the chance to pursue happiness of the individuals who were punished excessively in comparison to those punished moderately. The equality of happiness requires, in principle, equal treatment: similar, comparable punishment to similar comparable crimes.
Does this mean, then, that there can be no objection to the second scenario -that is, so long as all offenders are subjected to excessive punishment and receive equal amounts of pain? In other words, would a Utilitarian theory of punishment permit excessive punishment for minor offences or cruel punishment for serious offences? In answering these questions, I will first present Bentham's safeguards against excessive punishment and then explain how equality of happiness strengthens these to guard against excessive and disproportionate punishment.
As already noted, one of the defining features of a Utilitarian theory of punishment is that it regards punishment as pain, and an evil that the state should refrain from inflicting as much as possible. Therefore, when the state exceptionally causes pain by punishing, it is bound to prove two things. First, it must prove that punishment is a necessary tool in response to the painful act committed. This means that not all painful acts must be responded to with punishment 57 -only those which the state cannot prevent by any other means or policies. Second, that in inflicting the pain of punishment, the state must always use it as minimally as possible and only where necessary. Even in the context of deterrence, Bentham emphasised that 'whatever the mischief be, which it is proposed to prevent, to prevent it at as cheap a rate as possible', where cheap refers to the pain of punishment. 58 Thus, in principle, Utilitarianism opposes the infliction of pain -when it is necessary, then it should be minimal.
The problem, then, appears again to be located in the account of aggregate utility: what if imposing equal excessive pain on 100 people leads to better protection for thousands? Would that justify overriding these restrictions? The answer is no, because equal recognition and treatment from the state does not concern only the equality between offenders who committed similar crimes, but also applies in relation to offenders and non-offenders. As Bentham emphasises: [i]t ought not to be forgotten, although it has been too frequently forgotten, that the delinquent is a member of the community, as well as any other individual -as well as the party injured himself; and that there is just as much reason for consulting his interest as that of any other. 59 If the state were to use offenders as means to deter crime and achieve better chances to pursue happiness for law-abiding citizens, then it will have de facto established that offenders' chances to pursue happiness are of less value than those of the law-abiding. The equality of happiness should encompass every individual in society -those who pursue happiness peacefully as well as those who have caused injuries to others. The former are protected from unjust punishment. The latter have by their actions justified their receiving the pain and suffering of punishment from the state, but only as minimally necessary to prevent them from causing pain again.
However, no crime deprives the offender of his/her equal status as a person who has a capacity and desire for happiness. There is therefore no ground for using the offender as a means of less value, who can be subjected to cruel or excessive punishment to promote happiness for others.

Reasons to Obey the Law
The final criticism of a Utilitarian theory of punishment is that in using persons as a means to an end, it tries 'to affect the way people behave by threatening to make them suffer if they choose certain actions'. In so doing it offers 'prudential reasons that coerce rather than moral reasons that persuade'. 60 Furthermore, simple Utilitarian logic that individuals are self-interested and pursue their own happiness, while the law aims to aggregate happiness, leads to a situation in which the law threatens consequences that alter individuals' calculations and make disobedience costly. 61 Consequently, in this line of argument, Utilitarianism fails to respect people as rational beings who should be addressed by reasons, not by threats.
It seems relevant to start from the assumption that the purpose of law and punishment is to achieve aggregate happiness, and therefore that the law coerces the will of individuals to obey its orders to achieve the greatest happiness efficiently. Returning to Bentham's writings, it appears in his view that while the state is required to pursue the greatest happiness, law and punishment should primarily be used to eliminate pain.
To clarify, law and punishment are some of the tools of governance to be avoided wherever possible, because they restrict liberties and cause pain, contrary to the state's function of increasing happiness and eliminating pain. In describing such use, Bentham states that 'it is with government as with medicine; its only business is the choice of evils […] the evil of the offence, and the evil of the law'. 62 In choosing between these two evils What ought to be the object of the legislator? He ought to be certain of two things: first, that in every case the acts which he undertakes to prevent are really evils; and second, that these evils are greater than those which he employs to prevent them. 63 In explaining the first condition that the acts must be 'really evils', Bentham says that '[t]he power of the law need interfere only to prevent […] injuring each other'. 64 The second condition is a limitation on the use of the law -that is, if it is used to prevent pain then the law must use the lesser pain: it must produce happiness more than it causes pain. Thus the pursuit or production of happiness is a constraint on using the power of law. The real purpose of the law is that it should prevent individuals inflicting pain upon one another.
Interpreted in this new light, governed by the equality of every member of society to pursue happiness as they wish and as much as they are capable, one can find a potential non-prudential message in the criminal law, a message of peaceful co-existence. Every person is entitled equally to pursue their self-interests and happiness, which respects individuals' autonomy and diversity, as well as the fact that every person is the best judge of their preferences. The only restriction is that no individual should pursue happiness through the infliction of criminal pain on others, and this is where the state must intervene.
Therefore the message is not that an individual should obey the law because violating it causes pain when caught and punished, and pain is something the self-interested individual should avoid. Instead, the message is that an individual, just like every other member of society, is entitled equally to pursue their happiness, subject to a restriction that applies equally to everyone, namely that individuals should not injure one another. This message is one of a number of messages that the law can send to recognise every individual's equal chance to pursue happiness and express why individuals should not cause pain to one another. Most importantly, what the law is not expressing is the threat of pain for one person for the sake of others' happiness, or the deliverance of pain for collective happiness in disregard of the individual's actions as a responsible agent.
Thus when the offender is punished the message is that, in a Utilitarian theory of punishment, individuals are not punished because their punishment benefits everyone, including themselves, in terms of security. Rather, they are punished because they have caused pain to others: individuals should lead their lives without causing such pain in equal status to others refraining from causing pain to them. It aims to resonate with offenders and potential offenders by offering reasons for why it is better for his/her happiness, as well as for everyone else's, to refrain from causing pain. In so doing, the Utilitarian theory makes the end of punishment the elimination or reduction of pain; not aggregate happiness, but necessarily every person's ability to pursue their happiness equally.

Conclusion
It is essential when theorising about the justification of punishment to pay attention to the realities of actual punishment systems in modern societies. As Husak contends, '[s]ystems of penal justice in the real world are notoriously problematic: they are astronomically expensive, prone to error and mistake, and subject to enormous abuse by the officials they empower'. 65 This article argues that it is necessary to acknowledge that punishment is an evil -the evil of humans inflicting suffering on one another -and advocates adopting a doubtful, minimalist approach when theorising about the extent of state authority to inflict it. One of the main barriers against viewing punishment as an evil arguably justified by the apparent good it can achieve is the risk of using persons as mere means to such a good. This article addresses the problem by returning to Bentham's writings and proposes a theoretical development that would ensure that a Utilitarian theory of punishment, as part of its premise, would be constrained from using persons as mere means. Building on the premise that 'each individual has an equal right to all the happiness he is capable of experiencing', 66 this article claims that the state should recognise and treat as equal in value each individual's chance to pursue happiness.
Moreover, it challenges the classical Utilitarian justification of punishment as seeking security as a collective abstract goal, arguing instead for an individualistic justification of punishment. This is to say that the state is justified in punishing an individual only if he/she has interfered with the pursuit of happiness of another and has caused pain to them, within a general justifying aim of peaceful co-existence and the prevention of persons from injuring one another. This, of course, opens more questions about the definition of injuries and harms, as well as the nature of injuries that should fall within the scope of the criminal law. Nonetheless, for the purposes of justifying punishment, the answers to these questions should not affect the justification offered.