Negative Utilitarianism and Justice


B.Contestabile        First version 2005   Last version 2016





Table of Contents




1.      Introduction

2.      Rawls’ Theory of Justice

2.1  Social Contract Theory

2.2  Impartiality

2.3  Rawls’ Principles

2.4  Comparison with Bentham’s and Pigou’s Utilitarianism

2.5  Comparison with Mill’s Utilitarianism

2.6  Comparison with Harsanyi’s Utilitarianism

3.      Negative Utilitarianism (NU)

3.1  Historical Background

3.2  Definition

3.3  Modern NU

3.4  Implementation

3.5  Comparison with Rawls’ Theory

3.6  Comparison with Buddhism

4.      Conclusion











Starting point

Starting point of this paper is Popper’s controversial statement

“…from the moral point of view, pain cannot be outweighed by pleasure and especially not one man’s pain by another man’s pleasure [Popper, 284].

This citation addresses a well-known discomfort with the classical utilitarian accumulation of suffering and happiness across different people. Parfit’s claim about compensation [Parfit, 337] and Wolf’s misery principle [Wolf, 63] agree with Popper’s intuition.



Type of problem

- Does negative utilitarianism solve the problem of compensation?

- What are the differences between negative utilitarianism and Rawls’ concept of justice?



Negative utilitarianism

Negative utilitarianism (NU) is an umbrella term for ethics which models the asymmetry between suffering and happiness [Fricke, 14]. It includes concepts that

- assign a relative priority to the avoidance of suffering

- assign an absolute priority to the avoidance of suffering

- consider non-existence to be the best state of affairs


In this paper we investigate compensation across different persons (inter-personal compensation). For compensation within the same person (intra-personal compensation) see Why I’m (Not) a Negative Utilitarian.


A relative priority of (the avoidance of) suffering means that happiness has moral value, but less than suffering. The moral weight of suffering can be increased by using a "compassionate" metric, so that the result is the same as in prioritarianism. The more weight is assigned to suffering, the more difficult it becomes to compensate suffering by happiness. The intuition that “global suffering cannot be compensated by happiness” turns global well-being negative, so that the maximization of happiness turns into a minimization of suffering. The rationality of this intuition is investigated in The Denial of the World from an Impartial View.


Absolute priority represents a border case of relative priority, where the moral weight of happiness converges towards zero. Absolute priority can be approximated by relative priority. Relative priority is the general model.


The claim that non-existence is the best state of affairs is counter-intuitive for most Non-Buddhists.



Rawls’ theory versus negative utilitarianism

Priority of human rights:

- In negative utilitarianism it is theoretically possible to override human rights, if it serves the minimization of negative total welfare.

- In Rawls’ Theory of Justice it is impossible to override human rights, even if this principle causes the perpetuation of suffering.


Distribution of welfare among social classes:

Rawls’ difference principle accords well with the intentions of negative utilitarianism, if the term welfare is interpreted as life satisfaction.


Compensation among generations:

- In negative utilitarianism a generation may be obliged to sacrifice themselves, if it serves the long-term reduction of suffering.

- In Rawls’ theory the idea of a sacrifice is prevented by the principle of intergenerational impartiality.


From a negative utilitarian perspective the most urgent amendments to Rawls’ theory are the following:

- A principle for the protection of non-contractual cases (including sentient animals)

- A risk-averse population policy

- A risk-averse principle for technological progress.






1.   Introduction



Starting point

Starting point of this paper is Popper’s controversial statement

“…from the moral point of view, pain cannot be outweighed by pleasure and especially not one man’s pain by another man’s pleasure [Popper, 284].

This citation addresses a well-known discomfort with the classical utilitarian accumulation of suffering and happiness across different people. Parfit’s claim about compensation [Parfit, 337] and Wolf’s misery principle [Wolf, 63] agree with Popper’s intuition.



Type of problem

1.      Does negative utilitarianism solve the problem of compensation?

2.      What are the differences between negative utilitarianism and Rawls’ concept of justice?





2. Rawls’ Theory of Justice




2.1 Social Contract Theory


Social contract theories are theories on mutual benefit through cooperation.

Contractarians claim that moral principles derive their normative force from the idea of contract or mutual agreement. They are thus skeptical of the possibility of grounding morality or political authority in either divine will or some perfectionist ideal of the nature of humanity (contractarianism, Stanford Encyclopedia of Philosophy)


If an agreement with mutual benefit can be found, it is rational to sign a corresponding contract. But contractarians don’t claim that human behavior can be described by rationality. The claim is only that rationality has a normative force in defining ethical goals. Rationality is a possible common denominator to overcome the cultural diversity of ethical norms.


1)      Contractarianism, which stems from the Hobbesian line of social contract thought, holds that persons are primarily self-interested, and that a rational assessment of the best strategy for attaining the maximization of their self-interest will lead them to act morally (where the moral norms are determined by the maximization of joint interest) and to consent to governmental authority. Gauthier, Narveson, or Buchanan are Hobbesian contractarians.

2)      Contractualism, which stems from the Kantian line of social contract thought, holds that rationality requires that we respect persons, which in turn requires that moral principles be such that they can be justified to each person. Thus, individuals are not taken to be motivated by self-interest but rather by a commitment to publicly justify the standards of morality to which each will be held. Rawls or Scanlon are Kantian contractualists

(Contractarianism, Stanford Encyclopedia of Philosophy).


In reality social contracts may primarily be shaped by self-interest, but Rawls’ theory is normative and not descriptive [Rawls 1958, 183]. The goal is to establish an ethical ideal against which social contracts can be measured. Contractarian ideals are based on the concept of impartiality.




2.2 Impartiality




1.      Kant's categorical imperative is the best known definition of an impartial moral law. But the idea of moral impartiality is much older and can be found in many religions under the term “Golden Rule” or “ethic of reciprocity. It has been speculated that empathy lies behind the prevalence of the Golden Rule.

2.      The ancient Stoic-Skeptic tradition made a significant impact on the most prominent ethical theory of modern Europe – i.e. on Immanuel Kant’s moral philosophy. The Kantian Hellenistic ideal is originating, in my view, from the more profound and more systematic ethical teachings of the Buddhist India [Vukomanovic, 167]



The impartial observer

A contract is impartial if the principles

1)      do not depend on the specific interests of a single contractor or a group of contractors

2)      do not depend on temporary circumstances

Since politics is characterized by conflicting interests, only an impartial observer could design this kind of principles. The idea of an impartial observer was first mentioned by Adam Smith in his Theory of Moral Sentiments and later taken up by Harsanyi and Rawls.

1.      In his Theory of Justice Rawls used the following thought experiment to derive the conditions of an impartial contract: A contract is impartial, if it is derived from an original position in which rational contractors under a veil of ignorance decide how they wish to commit themselves to being governed in their actual lives (Justice as a Virtue, Stanford Encyclopedia of Philosophy). Behind such a veil of ignorance all individuals are specified as rational, free, and morally equal beings. Behind the veil of ignorance, what will the rational choice be for fundamental principles of society? The only safe principles will be impartial principles, for you do not know whether you would suffer or benefit from the structure of any biased institutions.

2.      Rawls called this concept Justice as fairness. Justice in a strict sense would include equality of opportunity in all aspects of life. Fairness is not equivalent to justice, but represents a practicable benchmark for existing political systems.

Impartiality gets support from recent discoveries in biology. We are all, regardless of race, genetically 99.9% the same (Wikipedia, Human Genome). In other words: 99.9% of the human genes are permanently being reincarnated.



Buchanan’s anarchistic equilibrium

Buchanan’s theory questions the role of the impartial observer in establishing a social contract. There is no unique interpretation of the impartial observer. Is he/she risk-tolerant or risk-averse? The concept attempts to circumvent the problem of incomparable individual utilities, but now the problem reappears in the ambiguous characteristics of the observer.


Buchanan replaces the veil of ignorance by an anarchistic equilibrium, where people with different interests overcome anarchy by a social contract. Depending on the social status of an individual, the interest to sign such a contract is stronger or weaker.





I either want less corruption or more chance to participate in it!


Ashleigh Brilliant





         The propagation of universal solidarity and compassion can be interpreted as a strategy of the infirm (as Nietzsche did).

         On the other hand, even the rich people become weak with age and the wealthy part of the population is willing to pay a price for stability.

According to Buchanan the courts are solely mediators between the parties. They are obliged to strict neutrality and have no competence in defining justice (Ökonomische Ethik, Roland Vaubel).

From the perspective of Rawls’original position, Buchanan’s vision of justice is rather descriptive than normative. It lacks the constituents to construct an ethical ideal.



Nozick’s liberalism

         Nozick, as well as Buchanan, questions the objectivity of risk-aversion and compassion. He emphasizes the importance of property rights and promotes a minimal state. The task of the minimal state is to protect property against violence and theft, but redistribution (like Rawls’ difference principle) has no moral foundation.

         From the perspective of Rawls’original position, Buchanan’s vision of justice is distorted by temporary and biased interests. It does not account for differences in talent and the contingency of life stories. Similar to Buchanan’s vision, it lacks the constituents to construct an ethical ideal.



Geuss’ historical and practical approach

         Raymond Geuss favors an approach to political philosophy in which one studies, history, social and economic institutions, and the real world of politics in a reflective way. He thinks that in using abstract methods in political philosophy, one will succeed merely in generalizing one's own local prejudices, and repackaging them as demands of reason. The study of history can help to counteract this natural human bias.

         Rawls thinks that – quite the contrary – the interpretation of history is driven by local and temporary prejudices and that abstraction is the most promising approach to overcome the natural human bias.




2.3 Rawls’ Principles


Justice has several dimensions, some of which can be influenced, whereas others can’t. Rawls Theory of Justice restricts the investigation to those dimensions that can be influenced.




1)      First principle of justice: Each person is to have an equal right to the most extensive total system of equal basic liberties compatible with a similar system of liberty for all.

2)      Second principle of justice: Social and economic inequalities are to be arranged so that they are both

a)      attached to offices and positions open to all under conditions of fair equality of opportunity and

b)      to the greatest benefit of the least advantaged, consistent with the just savings principle

 (Distributive Justice, Wikipedia)

“Just savings” is what a generation owes its descendants:

         immaterial values: just institutions, cultural values

         in economic terms: capital, factories, machines, knowledge, techniques, skills, natural resources etc.

The just savings principle follows from the principle of intergenerational impartiality (see below).




The principles are ordered in lexical priority as follows:

1)      The liberty principle: The basic liberties of citizens are, roughly speaking, political liberty (i.e., to vote and run for office); freedom of speech and assembly, liberty of conscience, freedom of property; and freedom from arbitrary arrest. It is a matter of some debate whether freedom of contract can be inferred as being included among these basic liberties.

Not mentioned in this list but naturally assumed is the protection of the citizen’s physical integrity.


2)      The arrangement of social and economic inequalities:

a)      The principle of fair equality of opportunity requires not merely that offices and positions are distributed on the basis of merit, but that all have reasonable opportunity to acquire the skills on the basis of which merit is assessed.

b)      The difference principle strives for the greatest benefit to the least-advantaged members of society.

Theory of Justice, Wikipedia



Liberty and human rights

1.      Rights theorists demand that (1) and (2a) conform to the Universal Declaration of Human Rights. By assigning lexical priority to human rights they become a side constraint for every theory that seeks a quantitative optimization of the state of affairs.

2.      Liberties are connected to duties (see rights-duty duality). Human rights balance the right to liberty (of the strong) with the right to protection (of the weak).



Liberty and redistribution

A social contract based on the liberty principle is compatible with capitalism and produces social and economic inequalities. Bernard Mandeville, an 18th century political economist and satirist, postulated in his Fable of the Bees that vicious greed leads to invisible cooperation if properly channeled. The idea was taken up by Adam Smith, who claimed that, in capitalism, egoistic behavior promotes the good of the community through a principle that he called “the invisible hand”. Experience has shown however that the “invisible hand” cannot protect many people from starving and that free markets have to be complemented by redistribution.

The essence of Rawls’ understanding of liberty can be found in the Preamble of the Federal Constitution of the Swiss Federation:

 “…only those who use their freedom remain free, and the strength of a people is measured by the well-being of its weakest members;



Fair equality of opportunity

Rawls’ concept of equality of opportunity goes beyond the one of classical utilitarianism. He denies an over-representation of upper class members in offices and positions even if it produces an increase in total welfare. The equal formal access to offices and positions is not sufficient; the statistical probability for lower class members to succeed has to be the same as for upper class members. For an example see [Clarenbach, chapt.3.2]



Intergenerational impartiality

If not mentioned otherwise in this paper, the term intergenerational impartiality is used in the sense of intergenerational moral impartiality (see Impartiality, Stanford Encyclopedia of Philosophy).

According to Rawls it is unjust to let the actual generation with its temporary and biased interests decide about the fate of future generations. The principle of impartiality should not only govern the distributive justice within the actual generation, but also among different generations.

1.      The principle says that the actual generation is not allowed to improve its situation at the cost of future generations.

2.      Inversely it says that the actual generation cannot be obliged to sacrifice itself for future generations.

According to Rawls the principle of intergenerational impartiality follows from the idea of the original position.

But the distribution of economic welfare among generations is the subject of a controversial discussion; see

1)      Should we discount future generations welfare?

2)      The Ethics of Climate Change [Broome 2008].




2.4 Comparison with Bentham’s and Pigou’s Utilitarianism


We start with a comparison of the economic aspect of justice, represented by

1.      Rawls’ difference principle

2.      Bentham’s and Pigou’s welfare functions


As long as utility functions are assumed to be linear (as in Bentham’s welfare functions) it does not matter, if the welfare of the most or the least wealthy is increased. But if we assume a diminishing marginal utility of welfare (as in Pigous’ welfare functions) then it makes sense to increase the welfare of the least wealthy. Pigou’s concept is therefore closely related to Rawls’ difference principle and – if we replace economic welfare by general welfare – to the basic idea of negative utilitarianism.




1)      In classical utilitarianism utility is a function of consumption. Consumption is measured in terms of preferences for goods and services. Utility is associated with the happiness which is created by consumption [Bentham, 120].

a)      The principle of utility is the principle that actions are to be judged by their usefulness in this sense: their tendency to produce benefit, advantage, pleasure, good, or happiness [Broome 1999, 19]

b)      Utility is the value of a function that represents a person’s preferences [Broome 1999, 21].

2)      In classical utilitarianism utility is cardinally measurable and interpersonally comparable. The term social welfare corresponds to the accumulated utilities of all individuals within the society. For details on Bentham’s utilitarianism see Short History of Welfare Economics.

3)      If not mentioned otherwise in this paper the terms happiness and (positive) welfare are used as synonyms.



Rawls’ difference principle

The difference principle holds that differences in wealth, status, etc. can be defended only if they create a system of market forces and capital accumulation whose productivity makes the lowliest members of society better off than they would be under a more egalitarian system.  The difference principle is the safest principle (Justice as Fairness, by Charles D.Kay) and corresponds to a risk-averse strategy (Rawls Unrisky Business, by Jim Holt). Its implementation though is far from trivial:


1.      Maximin interpretation:

If agents (A, B) can have incomes (5, 6) or (4, 9) then the former distribution has to be chosen.

Maximin encounters the following problem: If agents (A, B, C) can have incomes (5, 6, 9) or (5, 7, 8), then Maximin is indifferent


2.      Leximin interpretation (not to be confused with the lexical priority of human rights):

If agents (A, B, C) can have incomes (5, 6, 9) or (5, 7, 8) then the latter distribution has to be chosen. The second worst-off decides. If B’s income is the same in both distributions, then C’s income decides etc.

Leximin encounters the following problem: If agent A’s income is minimally higher in the second distribution and the income of B and C considerably lower, then the first distribution has to be chosen. This consequence is called dictatorship of the worst-off (see The Difference Principle).


3.      Prioritarian interpretation: Let’s return to Rawl’s definition of the difference principle: “Social and economic inequalities are to be arranged so that they are (…) to the greatest benefit of the least advantaged (…)”. The term greatest can be interpreted in such a way that everybody profits from inequalities, but that the least advantaged profit most. A prioritarian rule would e.g. distribute savings in such a way that the quota increases with decreasing welfare. Conversely the taxes owed would increase with increasing welfare.


For more information on this issue see

1.      Who are the least advantaged?, by Bertil Tungodden and Peter Vallentyne

2.      Rawls Differenzprinzip und seine Deutungen, by Peter Koller, Zeitschrift Erkenntnis, Vol.20, No.1, 1983


In order to compare Rawls’ difference principle with Bentham’s and Pigou’s utilitarianism, we have to map it to a welfare function. In this comparison we will use the Maximin interpretation of the difference principle. Maximin is the rule that is most frequently associated with the difference principle.



Welfare function according to Bentham

Let’s assume the society consists of two persons P1 and P2, who dispose of two goods G1 and G2. The welfare function according to Bentham would look as follows:

1.      Social welfare (W) is the total of the two individual utilities (U): W = U1 + U2

2.      The utility functions are identical for both persons. It is assumed that the utility (usefulness) of the two goods is the same for both persons.

3.      The utility depends linearly (factors a1, a2) on the quantities (q) of the goods.

U1 = (a1 x q11) + (a2 x q12)

U2 = (a1 x q21) + (a2 x q22)

where q12 = the quantity of good G2, consumed by person P1.

Under these premises social welfare should increase linearly with the Gross National Product (GNP).

[Kleinewefers, 40]



Welfare function according to Pigou

The concept of diminishing marginal utility has a long history (see Marginalism, Wikipedia). It was Gossen who found a convincing mathematical formulation and Pigou (not Bentham) who introduced it in a welfare function.

Let’s assume the society consists of two persons P1 and P2, who dispose of two goods G1 and G2. The welfare function according to Pigou would look as follows:

1.      Social welfare (W) is the total of the two individual utilities (U): W = U1 + U2

2.      The utility function is identical for both persons. It is assumed that the utility (usefulness) of the two goods is the same for both persons.

3.      The utility function U = U (q1, q2) is of the Gossen type, i.e. the marginal utility decreases with increasing consumption.

Under a given GNP the maximum social welfare can be attained, if the goods are equally distributed among the two persons [Kleinewefers, 41]. For more information on the term welfare function see Short History of Welfare Economics.




A welfare function according to Bentham sums the utility of each individual in order to obtain society's overall welfare. All people are treated the same, regardless of their initial level of utility. One extra unit of utility for a starving person is not seen to be of any greater value than an extra unit of utility for a millionaire. At the other extreme is the Maximin welfare function. According to the Maximin criterion, welfare is maximized when the utility of those society members that have the least is the greatest. No economic activity will increase social welfare unless it improves the position of the society member that is the worst off. Most economists specify social welfare functions that are intermediate between these two extremes. The social welfare function is typically translated into social indifference curves (Welfare economics, Wikipedia)

An indifference curve shows combinations of individual utilities which make no difference with regard to the total (social welfare):



Image:social indifference curves small.png




This picture was taken from the internet (author unknown)



         The Utilitarian Social Indifference Curves on the left hand side correspond to different levels of social welfare (W) in Bentham’s welfare function. Since a utilitarian social indifference curve confines an isosceles triangle, each total of individual utilities represented by a point on the curve, results in the same value. No matter how unequal the distribution is, an improvement of utility of the rich individual always compensates the loss in utility of the poor individual.

         The Max-Min (respectively Maximin) Social Indifference Curves in the middle correspond to different levels of social welfare in a Rawlsian society. The social welfare of a Maximin social indifference curve is the sum of two equal utilities and is depicted by the vertex of the curve. Increasing the utility of only one person (no matter how much) doesn’t increase social welfare.

         The Intermediate Social Indifference Curves on the right hand side can be interpreted as showing that as inequality increases, a larger improvement in the utility of relatively rich individuals is needed to compensate for the loss in utility of relatively poor individuals (Welfare economics, Wikipedia)


How does the Maximin (Rawlsian) indifference curve relate to Pigou’s welfare function? Is the difference principle compatible with the principle of diminishing marginal utility?

The neoclassical principle of diminishing marginal utility says that the greater the amount of utility a person already has the smaller will be the utility gain of any extra increment of wealth. As a consequence, total utility can be increased by a redistribution of welfare from the rich to the poor. Under these premises utilitarianism becomes egalitarian and corresponds to the Maximin principle (see Short History of Welfare Economics).

Rawls’s objection is of course that utilitarianism does not protect the claims of the most deprived from being set aside in favor of greater benefits to other, better off, social groups. But this argument is disputable: utilitarians will argue that given the diminishing marginal utility of wealth, the goal of maximizing average welfare will direct a society to prioritize improvements in the situation of the most deprived. And when one considers the detail of Rawls’s proposed policies for implementing his difference principle, it is hard to see anything incompatible with the implications of an enlightened utilitarianism (Questions of Justice).


However, if we take into account that the society’s total income is negatively affected by the redistribution (because hard working people lose the motivation to create additional income) then the optimal redistribution is rather prioritarian than egalitarian and can be associated with the prioritarian interpretation of the difference principle. This interpretation corresponds to the intermediate social indifference curve shown on the right hand side of above diagram.



The liberty principle

The two assumptions of classical utilitarianism

1.      individuals having similar utility functions

2.      diminishing marginal utility

can be interpreted as moral and political principles in a somewhat technical language (rather than psychological propositions).

One might say that this is what Bentham and others really meant by them, at least as shown by how they were used in arguments for social reform (…). But this still leave the mistaken notion that the satisfaction of desire has value in itself (…). To see the error of this idea one must give up the conception of justice as an executive decision altogether and refer to the notion of justice as fairness. Participants in the social contract have an original and equal liberty and their common practices are considered unjust unless they accord with principles which the contractors freely acknowledge before one another and so accept as fair [Rawls 1958, 191-192].

Justice cannot be the result of an executive decision which maximizes the total utility of the society; it must be a consensus of contractors with regard to liberty and solidarity. This consensus or compromise is an example of a reflective equilibrium.



Animal rights

A major point of criticism in Rawls’ concept is the lacking protection of animals. The impartial observer should protect all sentient beings according to their degree of suffering, no matter if they think rational and no matter if they are able to sign a contract. Such an impartial observer, though, would transcend the boundaries of social contract theory.

Bentham became known as one of the most influential of the utilitarians, through his own work and that of his students. These included his secretary and collaborator on the utilitarian school of philosophy, James Mill; James Mill's son John Stuart Mill; and several political leaders including Robert Owen, who later became a founder of socialism (…)

Bentham is widely recognized as one of the earliest proponents of animal rights. He argued that animal pain is very similar to human pain and that the day may come when the rest of the animal creation may acquire those rights which never could have been withholden from them but by the hand of tyranny. Bentham argued that the ability to suffer, not the ability to reason, must be the benchmark of how we treat other beings. If the ability to reason were the criterion, many human beings, including babies and disabled people, would also have to be treated as though they were things (Jeremy Bentham, Wikipedia)


Rawls was conscious that his concept doesn’t include all aspects of ethics:

Justice as fairness is not a complete contract theory. For it is clear that the contractarian idea can be extended to the choice of more or less an entire ethical system, that is, to a system including principles for all the virtues and not only for justice (…) Obviously if justice as fairness succeeds reasonably well, a next step would be to study the more general view suggested by the name "rightness as fairness." But even this wider theory fails to embrace all moral relationships, since it would seem to include only our relations with other persons and to leave out of account how we are to conduct ourselves toward animals and the rest of nature. I do not contend that the contract notion offers a way to approach these questions which are certainly of the first importance; and I shall have to put them aside. We must recognize the limited scope of justice as fairness (Animal Ethics)




2.5 Comparison with Mill’s Utilitarianism



Liberalism according to Mill

John Rawls, in his history of political philosophy [Rawls Lectures], warned to disesteem John Stuart Mill, an indication that he appreciated Mill’s work. There is, however, a profound difference between liberalism and utilitarianism:

         Utilitarianism attempts to increase the total utility of the community. The individual is subordinated to this goal.

         Liberalism claims that individual actions are only restricted by the condition not to harm others.

How could Mill claim that liberalism is compatible with utilitarianism?


The key for compatibility lies in Mills interpretation of utility [Schefczyk]:

1.      Mill stands in the tradition of Epicurus and Bentham according to which happiness (respectively the absence of suffering) is the only rational goal of human behavior. Mill’s innovation was to introduce a ranking for different kinds of happiness. The ranking is justified by competence, i.e. by the judgment of experienced persons. The judgment if football is preferable to the opera can only be made by persons who made a positive experience with both kinds of events. People who don’t enjoy the opera (or don’t even know it) cannot morally valuate the corresponding kind of happiness; they lack a specific perception, respectively knowledge.

2.      Even if all competent persons accord in ranking pleasures then this ranking is still not generally relevant. The life of an individual cannot be improved by imposing an “official” ranking upon him/her.

3.      Most philosophers agree that the life of an unhappy human is preferable to the life of a happy pig. But Mill doesn’t conclude that we should invest all our energy in top-quality actions. All kinds of happiness reach a point of satiation and we should therefore attempt to diversify our engagements.


The liberal characteristics of Mill’s concept can be summarized as follows [Schefczyk]:

1.      Tolerance, accept that perception and experiences are different

2.      Informality

3.      The denial of external and internal perfectionism

On this basis Mill claims that liberty is a consequence of the utilitarian goal. Since humans are different, the pressure for conformity and the denial of individual liberty contradicts the interests of all humans.

The point in Mill’s argument consists in not only seeing other people as competitors but also as enrichment. Diversified experiences produce empirical data about possible kinds of happiness. The ethical ideal according to Mill is not restricted to tolerance, it is open and affirmative; a liberal is willing to learn and profit from other person’s experiences.



The priority of the liberty principle

The utilitarian arguments for the priority of liberty rights are the following:

1.      Without liberty the individuals cannot articulate their preferences and the society lacks diversity. Without diversity there is no maximization of utility.

2.      A constitution which is based on liberty rights guarantees legal security. The fear of arbitrariness, torture etc. causes an enormous loss of utility.

The goods that are associated with the term liberty have a corresponding high weight within the utility function. But, from a utilitarian point of view, despite of this high weight, there is still a trade-off with other goods.


1.      A minor worsening of liberty rights would probably be tolerated for a huge gain in economic welfare. In reality the political debates prove that liberty rights don’t have an absolute priority. In the trade-off between liberty rights and economic welfare repressive regimes often find support in the population.

2.      A medic cares for two patients, one of them deadly ill and the other one curable. However he has only one dosage of the required medicament. Rawls would give the medicament to the deadly ill because he/she suffers more, a utilitarian would give it to the one with the higher potential to survive. The utilitarian position has a good chance to be accepted by the majority.

[Clarenbach, chapt.3.3]

In contrast, Rawls’ liberty principle has lexical priority and doesn’t allow any trade-off [Rawls 1958, 184-187].

While Mill recognized that reasons for justice have a special weight, he thought that it could be accounted for by the special urgency of the moral feelings which naturally support principles of such high utility. But it is a mistake to resort to the urgency of feeling; as with the appeal to intuition, it manifests a failure to pursue the question far enough [Rawls 1958, 189].



The conception of justice as fairness, when applied to the practice of slavery would not allow one to consider the advantages of the slaveholder (…). The gains accruing to the slaveholder cannot be counted as in any way mitigating the injustice of the practice [Rawls 1958, 188].

Classical utilitarianism in general (not only Mill’s utilitarianism) lacks a principle for the protection of the individual.


The example shows that in Rawls’ theory – when liberty is at stake – the means are more important than the end. In contrast to utilitarianism he supports the process-view of justice and not the end-state view (or outcome-view).



The second principle of justice

The second principle excludes the justification of inequalities on the grounds that the disadvantages of those in one position are outweighed by the greater advantages of those in another position. This rather simple restriction is the main modification I wish to make in the utilitarian principle as usually understood (…). It is a restriction of consequence, and one which utilitarians, e.g. Hume and Mill, have used in their discussions of justice without realizing apparently its significance [Rawls 1958, 168]




2.6 Comparison with Harsanyi’s Utilitarianism


Indifference curves (as depicted in chapter 2.4) are a means to compare concepts of distribution but we lack a rational criterion to select a specific distribution out of the many possible ones. Game theory is a possible means to find such a criterion. An impartial observer in the original position considers inequality as a risk, which has to be properly weighted in order to attain the best state of affairs.




1.      In Bentham’s and Pigou’s utilitarianism the term utility corresponds to the happiness which is created by consumption. In the course of history the concept of utility became more abstract and was finally interpreted as the dominant end of human behavior (no matter what that is). The term preference-satisfaction in preference utilitarianism corresponds to this level of abstraction.

2.      The replacement of the abstract concept of utility by the concrete meaning happiness is called hedonic reduction. Hedonic reduction opens the theory to empirical testing [Hirata, 24]. Examples:

a.       Quality of life (sociology)

b.      Well-being (psychology)

c.       Welfare (economics)

d.      Life satisfaction (happiness economics)

e.       Quality-adjusted life year (health care)

3.      The application of game theory (a well as the utilitarian welfare function) presupposes that utility can be measured on a cardinal scale and that it is amenable to an interpersonal comparison (see Wohlfahrtstheorie).



Game theory

The classic expositions of Harsanyi and Rawls produce a synthesis that is consistent with the modern theory of non-cooperative games, see Game Theory and the Social Contract. From a game theoretical view Harsanyi’ utilitarianism is compatible with the difference principle according to Rawls. The (risk-neutral) Bayesian maximization of utility converges towards Rawls’ (risk-averse) Maximin principle, if the weight of the worst cases increases.




Historically, hedonistic utilitarianism is the paradigmatic example of a consequentialist moral theory. This form of utilitarianism holds that what matters is the aggregate happiness, i.e. the happiness of everyone and not the happiness of any particular person (Consequentialism, Wikipedia)


Why should we subscribe to consequentialism?

An argument for consequentialism is contractarian. Harsanyi argues that all informed, rational people whose impartiality is ensured because they do not know their place in society would favor a kind of consequentialism. Broome [1991] elaborates and extends Harsanyi's argument (Consequentialism, Stanford Encyclopedia of Philosophy)

This argument is of special interest in our context because is reveals that the common denominator of Harsanyi’s utilitarianism and Rawls’ Theory of Justice is contractarian. At the same time Harsanyi opened the door to a new understanding of classical utilitarianism:

Harsanyi’s approach leads to a justification of classical utilitarianism from a remarkably new point of view (see John C.Harsanyi)

In the following we compare Rawls’ difference principle with the major competing theory of Harsanyi:



Harsanyi’s equiprobability model

Harsanyi’s equiprobability model (Gleichwahrscheinlichkeitsmodell) uses the following definition of impartial decisions:

1.      The impartial observer is an individual within society.

2.      After every decision influencing or changing society this individual can find himself with equal probability in every possible position in the changed society.

3.      Risk is defined as a product of utility and (mathematical) probability. The expected value of a decision is calculated by adding all possible risks.




Harsanyi’s concept (similarly to Rawls’) converts normative statements about compassion or solidarity into normative statements about risk.

[Harsanyi 1975] refuses Rawls risk-averse strategy, claiming that it is irrational to make behaviour dependent on some highly unlikely unfavourable contingency regardless of its low probability. He recommends not to base decisions on the worst case but on the Bayesian maximization of utility. Rawls in contrast is convinced that the rational choice of an individual behind the veil of ignorance is the Maximin strategy.



The priority of the liberty principle

Harsanyi’s concept is not restricted to the economic aspect of welfare (the difference principle). He assumes that, despite of the high weight the liberty principle has within the utility function, there is still a trade-off with other goods. Rawls however, insists on the absolute priority of the liberty principle. Classical utilitarianism can only be reconciled with the concept of fairness, if fairness is considered to be a side constraint of the utility function. But then utilitarianism loses its characteristics:

If one wants to continue using the concepts of classical utilitarianism, at least the utility functions must be so defined that no value is given to the satisfaction of interests which violated the principles of justice. In this way it is no doubt possible to include these principles within the form of the utilitarian conception; but to do so is, of course, to change its inspiration altogether as a moral conception. For it is to incorporate within it principles which cannot be understood on the basis of a higher order executive decision aiming at the greatest satisfaction of desire [Rawls 1958, 191].



Retributive justice can be implemented within a social contract of the Rawls’ type, but has no priority in the maximization of welfare. Retributive justice postulates that there should be a proportion between doing well and faring well. Not only is the result important but also the means that produced the result:

1.      Relative justice: A world where thugs fare better than decent people is morally objectionably, even if the total of the decent people’s welfare is not affected [Temkin, 354].

2.      Absolute justice: A world where thugs fare well is morally objectionable, even if decent people fare better than thugs [Temkin, 357]





3. Negative Utilitarianism (NU)




3.1 Historical Background



Ancient world

1.      The idea to formulate an ethical goal negatively originates in Buddhism and is more than 2000 years old.

2.      Greek philosopher Epicurus has sometimes been caricatured as crude hedonist. But Epicurus also maintained the puzzling doctrine that the complete absence of pain constituted "the limit and highest point of pleasure" (Epicurus, David Pearce)

3.      While Epicurus has been commonly misunderstood to advocate the rampant pursuit of pleasure, what he was really after was the absence of pain (both physical and mental, i.e., suffering) - a state of satiation and tranquility that was free of the fear of death and the retribution of the gods. When we do not suffer pain, we are no longer in need of pleasure, and we enter a state of 'perfect mental peace' (ataraxia). (Epicurus, Wikipedia)



Early utilitarianism

Historically utilitarianism was inspired by Stoicism and Epicureanism and therefore closer to negative utilitarianism than the contemporary interpretation:

Although the favored means of the term negative welfarism – a stoician-like control of the birth of one’s desires which it also calls “liberation” (moksa) – is in a sense opposed to economists’ conception, scholarly welfarism is in fact historically the direct descent of this Indian philosophy. Indeed, the 18th century founders of utilitarianism were thoroughly inspired by Stoicism and Epicureanism, whereas the influence of Buddhist and Jain thoughts on Stoician and other Hellenistic philosophies is explained in the previous reference. The oblivion of self-formation occurred with that of the Rousseau-Kant “autonomy” by some narrowminded post-Mill 19th century scholars (even Mill’s “choice of lifestyle” is a downgrading of full eudemonistic self-formation). Note that the view that utilitarianism is the necessary all-encompassing criterion was, in the West, restricted to English-language scholars influenced by Bentham who introduced this view for a political reason. This is why Rawls appeared to be much less original in other circles who acknowledged constitutional basic rights and where egalitarianism was a familiar ideal. (Macrojustice from Equal Liberty, Serge-Christoph Kolm)

Also the Stoic cosmopolitanism corresponds well to utilitarianism. Stoicism, in contrast to Buddhism, was characterized by an optimistic world view.

In contemporary negative utilitarianism we find both, optimistic and pessimistic visions of the future; see Hostility and the Minimization of Suffering.




In the 20th century, the idea to formulate an ethical goal negatively is attributed to Karl Popper:

…there are no institutional means of making a man happy, but a claim not to be made unhappy, where it can be avoided. The piecemeal engineer will, accordingly, adopt the method of searching for, and fighting against, the greatest and most urgent evils of society, rather than searching for, and fighting for, its greatest ultimate good [Popper, 158]

At this point of chapter 9, Popper added his controversial note 2:

I believe that there is, from the ethical point of view, no symmetry between suffering and happiness, or between pain and pleasure. Both the greatest happiness principle of the Utilitarians and Kant’s principle “Promote other people’s happiness…” seem to me (at least in their formulations) wrong on this point which, however, is not completely decidable by rational argument (…). In my opinion human suffering makes a direct moral appeal, namely, the appeal for help, while there is no similar call to increase the happiness of a man who is doing well anyway.

A further criticism of the Utilitarian formula “Maximize pleasure” is that it assumes, in principle, a continuous pleasure-pain scale which allows us to treat degrees of pain as negative degrees of pleasure. But, from the moral point of view, pain cannot be outweighed by pleasure and especially not one man’s pain by another man’s pleasure. Instead of the greatest happiness for the greatest number, one should demand, more modestly, the least amount of avoidable suffering for all; and further, that unavoidable suffering – such as hunger in times of unavoidable shortage of food – should be distributed as equally as possible.

There is some analogy between this view of ethics and the view of scientific methodology which I have advocated in my The Logic of Scientific Discovery. It adds to clarity in the fields of ethics, if we formulate our demands negatively, i.e. if we demand the elimination of suffering rather than the promotion of happiness. Similarly, it is helpful to formulate the task of scientific method as the elimination of false theories (from the various theories tentatively preferred) rather than the attainment of established truths [Popper, 284].


Popper’s ethics was not only influenced by his epistemological work, but also by personal and historical experience:

1)      The failure of happiness-promoting philosophies like classical utilitarianism and Marxism

2)      Sixteen of Popper’s closest relatives became victims of Nazi Germany, partially in Auschwitz, some committed suicide (from Die Erkenntnistheorie und das Problem des Friedens)


Popper was not a utilitarian and did not use the term negative utilitarianism. His notes on ethics don’t represent a theory; they rather encourage a family of ethical concepts [Fricke, 13]. In spite of that, he was confronted with the following narrow, utilitarian interpretation of his notes:




The term negative utilitarianism was introduced by R.N.Smart in a paper criticizing Popper’s approach to ethics:


….one may reply to negative utilitarianism (hereafter called NU for short) with the following example, which is admittedly fanciful, though unfortunately much less so than it might have seemed in earlier times.

Suppose that a ruler controls a weapon capable of instantly and painlessly destroying the human race. Now it is empirically certain that there would be some suffering before all those alive on any proposed destruction day were to die in the natural course of events. Consequently the use of the weapon is bound to diminish suffering, and would be the ruler’s duty on NU grounds (…).

Admittedly my example does not quite work as it stands against Professor Popper inasmuch as he propounds two other principles to set alongside NU, viz. (briefly) "Tolerate the tolerant" and "No tyranny". Presumably the benevolent world-exploder might be thought intolerant and/or tyrannical (…) …if we allow "Tolerate the tolerant" and "No tyranny" to stand as principles alongside NU, there will be a conflict between them and NU regarding our example. If we take NU seriously, surely we should over-ride the other principles (Negative Utilitarianism)

There is no indication in Popper’s text which says that the minimization of suffering should override the other principles. On the contrary, Popper’s attitude was anti-totalitarian and it is much more likely that he would have promoted human rights as a side constraint for any attempt to improve the state of affairs (see chapter 3.5).




3.2 Definition


Negative utilitarianism (NU) is an umbrella term for versions of utilitarianism which model the asymmetry between suffering and happiness [Fricke, 14]. It variously includes concepts that

1.      assign a relative priority to the avoidance of suffering

2.      assign an absolute priority to the avoidance of suffering

3.      consider non-existence to be the best state of affairs


There is no consistent usage of terms in negative utilitarianism. We use the abbreviation NU as an umbrella term in this paper, but sometimes it refers to the historically first version, as defined by R.N.Smart (see Negative Utilitarianism). R.N.Smart’s version has the following characteristics:

         It is hedonistic (as opposed to preference-based versions).

         It completely denies the moral value of happiness (see The Pinprick Argument) and therefore corresponds to the third of above concepts.

Roger Chao uses the term classical negative utilitarianism for this version [Chao, 58].


However, it turned out that R.N.Smart misinterpreted Popper’s notes on ethics. Popper later clarified that the complete devaluation of happiness would have absurd consequences and that his notes were only meant for public policy and not individual action. In the following we adopt this view and investigate

         utility functions on the group/society level and not utility functions on the individual level

         compensation across different persons (inter-personal compensation) and not compensation within the same person (intra-personal compensation)

In this context the term happiness is a synonym for life-satisfaction, well-being and general positive welfare.

The term suffering accordingly stands for uncompensated suffering [Fricke, 18] and is a synonym for negative well-being and negative welfare.


For intra-personal compensations/trade-offs (Weak NU) see Why I’m (Not) a Negative Utilitarian.



Relative priority – Modern NU

A relative priority of (the avoidance of) suffering means that happiness has moral value, but less than suffering. The moral weight of suffering can be increased by using a "compassionate" metric, so that the result is the same as in prioritarianism, see chapter 3.3



Absolute priority – Negative total Utilitarianism

The absolute priority of (the avoidance of) suffering is known under the term negative total utilitarianism [Arrhenius, 100]. Happiness has no moral value, as long as suffering exists. In the (unlikely) case that suffering can be eradicated, we need a rule for happiness [Fricke, 14, 18]. Such a rule could declare the promotion of happiness to be

1.      mandatory (Lexical Threshold NU)

2.      supererogatory

3.      neutral

In the following we will discard negative total utilitarianism, because it represents a border case of relative priority, where the moral weight of happiness converges towards zero. Absolute priority can be approximated by relative priority. Relative priority is the general model.



Non-existence as best state – Negative preference utilitarianism

Negative preference utilitarianism is based on antifrustrationism, an axiology which says that the satisfaction of a preference is not morally better than the non-existence of this preference [Fehige].

         Antifrustrationism does not prescribe a method of aggregating preferences. But preferences have to be ordered; otherwise the theory is meaningless [Broome 1999, 9-10]. If we assume that the highest order preference of human behavior is life satisfaction then the theory says that there are only lives with negative value [Stanford, chapter 2.4].

         With a utilitarian aggregation of life satisfaction across a group/society the theory becomes negative preference utilitarianism [Fricke, 20].

Apart from its affinity with Buddhism (chapter 3.6) we will discard this version of NU. The assumption that non-existence is the best state of affairs is counter-intuitive for most non-Buddhists.



Comparison with positive utilitarianism


         The classical utilitarian slogan “The greatest happiness of the greatest number” (Bentham) emphasizes the maximization of happiness

         Its negative utilitarian counterpart “The least suffering of the greatest number” emphasizes the minimization of suffering.

Happiness is something we want – a positive goal, suffering is something we do not want – a negative goal. Hence the term negative.

Negative utilitarianism fights dystopias rather than promoting utopias.



         In positive utilitarianism the promotion of happiness has the same ethical priority as the avoidance of suffering, as long as the contribution to total welfare is the same.

         In negative utilitarianism the promotion of happiness has less (or even no) priority as compared to the avoidance of suffering.

However, if ethical priorities are expressed by the metric of the hedonistic scale, then this distinction is questionable, see chapter 3.3.



In (hedonistic) negative utilitarianism the term utility corresponds to the avoidance of suffering and is considered to be cardinally measurable and interpersonally comparable, as well as in (hedonistic) positive utilitarianism. Example:

         If positive utilitarianism measures a societies’ utility in terms of the index of life satisfaction, then

         negative utilitarianism replaces this measure by an index for negative welfare, see chapter 3.4.



         In positive utilitarianism the most efficient action consists in saving as many (young) lives as possible.

         In negative utilitarianism the most efficient action consists in reducing the number of worst cases, even at the cost of a shrinking population.

For more information on cost-benefit see Negative Utilitarian Priorities.




3.3 Modern NU




The modern NU is a metric within modern hedonistic utilitarianism, which assigns a higher weight to the avoidance of suffering than to the promotion of happiness. The moral weight of suffering can be increased by using a "compassionate" metric, so that the result is the same as in prioritarianism.

Modern hedonistic utilitarianism is associated with the work of John Broome in this paper [Broome].




If the modern NU is “only a metric” within modern hedonistic utilitarianism: What is the reason for using a special term for this metric?

The reason is that utilitarianism – similar to prioritarianism – is associated with positive totals:

         In classical utilitarianism the hedonistic scale is linear and symmetric. The theory does not exclude negative totals, but in practice most utilitarians assume that – given the current state of affairs – total well-being is positive.

         The modern NU maintains that symmetric scales are a distortion of reality. The theory does not exclude positive totals, but it considers that – given the current state of affairs – total well-being might be negative. The intuition that “global suffering cannot be compensated by happiness” turns global well-being negative, so that the maximization of happiness turns into a minimization of suffering.





Intuitions with regard to global well-being are controversial:

         The majority – in particular prioritarians and classical utilitarians – think that seven billion happy people can outweigh the extreme suffering of a minority.

         Buddhists, Gnostics, Schopenhauer, Popper and many antinatalists do not share this intuition.

The majority considers the latter people to be highly risk-averse or even irrational, but possibly the majorities’ perception of risk is distorted. The rationality of the negative utilitarian intuition is investigated in The Denial of the World from an Impartial View.



Comparison with classical utilitarianism

Classical utilitarianism assumes that suffering and happiness have equal weight. Current surveys on subjective life satisfaction express this “equal weight” by symmetric point scales. Let us assume there was a survey on subjective well-being, asking the question:

What is your well-being on the following scale: Very happy, happy, neutral, suffering, severely suffering?

Classical utilitarianism transforms these ordinal values into cardinal values as follows:





 very happy = max.








 severely suffering = min.




The Modern NU suggests that the hedonistic scale is asymmetric. This thesis is justified by intuitions like the following:

         We should realize that from the moral point of view suffering and happiness must not be treated as symmetrical; that is to say, the promotion of happiness is in any case much less urgent than the rendering of help to those who suffer, and the attempt to prevent suffering [Popper, Volume I, Notes to Chapter 5, Note 6]

         It is more important to relieve suffering than to increase (already happy people’s) happiness. We can retain this important intuition … by giving more weight to negative welfare than to positive welfare [Arrhenius, 138]

The moral value of an increase in well-being matters more, the worse off people are:



         If the majority of the participants of a survey report a positive well-being, then the total is necessarily positive in classical utilitarianism.

         In the Modern NU the total may turn negative, because of the higher moral weight of negative well-being.



Comparison with Prioritarianism

The intuition of an asymmetry is also characteristic for prioritarianism, see Hostility and the Minimization of Suffering. Prioritarianism covers the functionality of the modern NU and vice-versa:

         In prioritarianism the symmetric cardinal values in above table are modified by an asymmetric (concave) weighting function.

         In modern hedonistic utilitarianism the asymmetry can be expressed by the metric of the hedonistic scale, i.e. the values in above table are asymmetric in the first place.

From a technical point of view prioritarianism may be preferable (because available surveys use linear scales) but – according to John Broome – utilitarianism has a better theoretical foundation:

Prioritarianism is not an adequate way to understand fairness [Broome, 39].

(Moral) betterness is represented by the unweighted total of people’s well-being [Broome, 137].




3.4 Implementation


In order to compare the welfare of nations one has to use indices (average level of negative welfare), so that the population size doesn’t matter. Average utilitarianism is the most popular theory among welfare economists [Arrhenius, 53].



Subjective indices

Almost all OECD countries now contain a life evaluation on a 0 to 10 rating scale, usually a question about life satisfaction [Helliwell, 15].

In other words: the OECD rating scales are linear and do not know negative values.

How can the Modern NU be implemented?

One possibility is to define a NU-index which competes with the established indices of life satisfaction.

1)      Defining a subjective suffering scale by reversing the scale of the life satisfaction index [Anderson, 5-6] does not solve the problem, because it allows compensating the minorities’ suffering with the majorities’ happiness (col.3 in Table 1). Also in Anderson’s scale suffering is expressed in economic terms and not in terms of life satisfaction.

2)      The signed linear scale (col.4 of Table 1) has the advantage that the life satisfaction of suffering people is expressed by negative numbers. This makes intuitively clear that the creation of additional lives with this quality has to be morally justified. With such a scale, however, total welfare remains positive.

3)      The NU scale should be concave (see chapter 3.3), so that the suffering of the minority cannot easily be compensated by the happiness of the majority. One possibility is to use the signed linear scale as input argument (x) for an exponential function exp(-x). This concept is functionally equivalent to prioritarianism; see Hostility and the Minimization of Suffering. If it is possible to find a consensus for progressive tax, it may also be possible – despite the limited normative force of compassion – to find a consensus for a concave scale.



Table 1: Scales for measuring welfare


Subjective Life Satisfaction



Subjective Suffering












































Objective indices

Because subjective indices include unavoidable kinds of suffering (like illnesses, aging and death) a corresponding ranking of nations cannot be related to the ethical standard of these nations. Objective indices focus on avoidable kinds of suffering. The International Human Suffering Index and Anderson’s objective suffering index [Anderson, 7] rely on the statistic of the World Bank, UN and other readily available resources. They suggest that underdeveloped nations suffer more than developed ones.

         Objective indices exclude natural disasters in order to limit the measure to preventable suffering [Anderson, 8].

         Objective indices account for factors that are closely related to the definition of underdevelopment [Anderson, 7]. The correlation with the Human Development Index [Anderson, 14] therefore doesn’t come as a surprise.

Above objective indices do not consider the ambivalence of progress as investigated in

         The Cultural Evolution of Suffering (long-term counter-productive mechanisms)

         On the Perception of Risk and Benefit (possible self-destruction of humanity)


Instead of focusing on underdevelopment, one could also concentrate on armed conflicts:

Possibly the highest rates of negative welfare (per head of population) occur in countries on war, where surveys are difficult to conduct.

If this thesis holds, then the following information could be used to guide NU priorities:





The strategy to reduce the incidence of wars is challenged by several others, claiming to have the best efficiency in the fight against suffering.

For more information about this issue see Negative Utilitarian Priorities.



Subjective versus objective

1)      In Happiness Economics, subjective suffering is related to objective data by means of factor analysis [Oswald]. Whereas above objective indices are defined by experts, factor analysis lets the majority decide about the determinants of suffering. Subjective suffering accounts for the ambivalence of progress.

2)      Ron Anderson assessed a moderately high correlation between his subjective and objective suffering index. It seems in particular that in higher developed nations subjective suffering correlates less with objective suffering than in lower developed nations [Anderson, 11]. This is reminiscent of the Easterlin paradox.



Moral killing

The most controversial strategy of NU is a violent reduction of the population-size in order to improve the state of affairs [Fricke, 18]. An extreme pursuit of this thought leads to the extermination of humanity or life as a whole [Smart]:

         A negative utilitarian believes that, if it was possible to exterminate all life in the universe instantly and painlessly and permanently, it would be correct and ethically required that we do so in order to prevent any future cases of suffering

         A classical utilitarian might decide either way, depending on his estimation of the relative amounts of future suffering and happiness.

(Introduction to utilitarianism,

There is, however, an empirical argument against such a strategy: Planning a project for the violent reduction of the population-size (e.g. by sterilization) would provoke immense distress and classify the supporters of the project among the worst kinds of criminals, terrorists and lunatics. A violent extermination of mankind is not feasible in practice. The result of such an attempt would be an increase in suffering and therefore contradict the ethical goal of NU.


Empirical arguments against moral killing are not convincing under all circumstances, but similar deficiencies can be found in positive utilitarianism and prioritarianism, see the section about Human rights and the higher purpose in chapter 3.5

For information about the moral ideal of childlessness see Hostility and the Minimization of suffering.




Whatever the future holds, negative utilitarian ethics will presumably still fail to resonate with the overwhelming majority of the population - especially after our emotional well-being increases as the adoption of enhancement technologies gathers pace. So perhaps the most effective way for a negative utilitarian to promote his/her ethical values is not to proselytize under that label at all. Instead, the negative utilitarian may find it instrumentally rational to give weight overtly to the "positive" values of ordinary classical utilitarians, preference utilitarians/preference consequentialists, and the far wider community of (mostly) benevolent non-utilitarians who share an aversion to "unnecessary" suffering. The indirect approach to NU is likely to yield the greatest payoff. Only by our striving to promote "positive" goals as well, and campaigning for greater individual well-being, is the ethic of NU ever likely to be realized in practice (Direct versus Indirect NU, in The Pinprick Argument, by David Pearce)

Acceptance can be improved, if the NU-index is used as additional information, challenging (relativizing) established indices, but not replacing them. In the established indices, the worst cases of suffering disappear in the average, suggesting that they can easily be compensated. The NU-index strengthens the awareness that the suffering minority pays (in a statistical sense) the price for the happiness of the majority. It can reveal cases in particular, where the average of all classes improves and the situation of the most suffering minority worsens.


As it was noticed by Inglehart the relation between suicide rate and subjective well-being is in general positive: “Suicide rate tends to be higher in those nations that rank high – not low – on subjective well-being.” In other words, the average life satisfaction seems to be positively correlated with the suicide rate. (…) Those cultures where suicide is most widespread have the strongest norms of describing oneself as happy. Conceivably, being deeply unhappy in a society where everybody is expected to be happy may be even more unbearable and produce a greater sense of social isolation than it would be in a society where misery is not so far from the norm [Allik]




3.5 Comparison with Rawls’ Theory


Rawls knew Popper’s normative claims [Rawls 1958, 174] and may have been influenced by Popper’s work:

1.      The liberty principle is the basis of an open society as defined by Popper.

2.      The difference principle takes account of Popper’s notes on the asymmetry between suffering and happiness.



Non-contractual cases

Philosophers, who treat morality as primarily contractual tend to discuss non-contractual cases briefly, casually and parenthetically, as though they were rather rare. The contractarian view is that those who fail to clock in as normal rational agents and make their contracts are just occasional exceptions, constituting one more minority group, but not a central concern of any society [Midgley]

In negative utilitarianism the ethical goal to minimize suffering is guided by compassion and risk-aversion. As compared to social contract theory it seems to be built on emotions. But these emotions conform to the contractor’s self-interest in the following cases:

1)      Every human being passes through stages where he/she depends on compassion. The child spends many years acquiring the competence in speech and consciousness which is typical for humans. Old people, finally, lose step by step their mental capabilities, in particular by strokes and dementia (e.g. Alzheimer’s disease).

2)      Ones own future children or grand-children could be mentally retarded, insane, handicapped or become orphans.


The protection of non-contractual cases gets support from recent discoveries in biology:

1)      Self-interest is strongly influenced by the biological utility function, see God’s Utility Function. The insight that the temporary and biased self-interests (the motives of self-realization) are in truth the ones of the biological utility function can lead to a feeling of being manipulated, to the consciousness of heteronomy and to solidarity with the victims of biological mechanisms.

2)      In negative utilitarianism – as well as in Bentham’s utilitarianism (chapter 2.4) – non-contractual cases also include sentient animals, speciesism is denied. This is reminiscent of ancient concepts or reincarnation, although genes are reincarnated rather than souls. Hinduism assumes that there is no essential difference between human souls and animal souls. Contemporary science reveals that sentient animals have a surprising genetic similarity to humans. Example: Chimpanzees and humans have 99% of coding DNA sequences in common (Humanzee, Wikipedia).



Moral killing – human rights and the higher purpose

The relation between human rights and negative utilitarianism is complex:

         Human rights are the result of historical experiences (in particular the Holocaust). They prevent some of the worst cases of suffering and insofar serve the negative utilitarian goal.

         Then again negative utilitarianism contradicts human rights, because it has a totalitarian claim. From a theoretical point of view moral killing (even the eradication of humanity) is justified under certain circumstances [Smart].

In the latter case, human rights are subordinated under a higher principle (the eradication of suffering). But note that this subordination concerns consequentialism in general [Larmore, 4]:

1)      The capital punishment in the U.S. illustrates that the idea is not far-fetched (The U.S. constitution was influenced by classical utilitarianism).

2)      Human rights may be violated in a moral dilemma called torture of the mad bomber.

3)      The arguments against moral killing in classical utilitarianism are only of empirical nature. A positive utilitarian surgeon could sacrifice an old patient (e.g. in the context of organ transplantation) in order to save the life of a young patient, if the result is a net gain in happiness, and if it were possible to do it secretly. There is even a moral obligation to eliminate lives with negative welfare, if it were possible to do it painlessly and secretly [Chao, 60-61].

4)      Classical utilitarianism (maximization of happiness) turns into negative utilitarianism (minimization of suffering), if total welfare turns negative.


An attempt to solve the problem consists in switching to a preference-based ethics. In preference-based ethics the preferences of the individuals have to be respected, so that the arguments against painless killing are of theoretical and not only empirical nature [Fricke, 20][Chao, 57-59]. However, in a utilitarian preference-aggregation, the preferences of the majority overrule the ones of the individual and the majority could theoretically decide to exterminate a minority [Hare, 121-122]. Adherents of consequentialism, who recognize the totalitarian potential as a problem, consider human rights as a side constraint and not as a subordinated issue [Wolf, 278] [Nozick]. Popper e.g. clarified, that the minimization of suffering is only one of three principles of ethics [Popper, Volume I, Notes to chapter 5, Note 6, 255-256]

1.      Tolerance toward all, who are not intolerant and who do not propagate intolerance (…) This implies, especially, that the moral decisions of others should be treated with respect, as long as such decision do not conflict with the principle of tolerance.

2.      The recognition that all moral urgency has its basis in the urgency of suffering or pain. I suggest, for this reason to replace the utilitarian formula “Aim at the greatest amount of happiness for the greatest number” or briefly “Maximize happiness”, by the formula “The least amount of avoidable suffering for all” or briefly “Minimize suffering”. Such a simple formula can, I believe, be made one the fundamental principles (admittedly not the only one) of public policy. (The principle “Maximize happiness”, in contrast, seems to be apt to produce a benevolent dictatorship) ……

3.      The fight against tyranny, or in other words, the attempt to safeguard the other principles by the institutional means of a legislation rather by the benevolence of a person’s power.


History justifies a deep skepticism opposite to ethical theories which promise improvements by restricting human rights. A major task of human rights is the protection of subjective values. The minority problem is not only disturbing in tyranny and fascism, but also in democracies. Human rights guarantee that the individual cannot be subordinated to a higher purpose of any kind. But thereby negative utilitarianism loses its original characteristics. If the higher purpose is the extermination of suffering then side constraints – like reproductive liberty – might make it impossible to ever reach this goal. Even in a perfectly fair political system, suffering could be immense. The suffering minority has to accept the state of affairs, because the happy majority accepts it.

Most people begin with the presumption that we morally ought to make the world better when we can (Consequentialism, Stanford Encyclopedia of Philosophy)

Rawls was conscious of the problem by speaking of justice as fairness and admitting that there are other meanings of justice.



Animal rights and the higher purpose

Animal rights are in conflict with both Rawls’ theory of justice and negative utilitarianism:

1)      Animal rights conflict with Rawls theory (human rights) if animals and humans compete for resources

2)      Animal rights conflict with negative utilitarianism in cases where a violation of animal rights could reduce total human suffering.


Moral dilemmas can only be avoided by retreating from life like Buddhist monks. But not every problem in the context of animal rights is an ethical dilemma:

1.      Slaughterhouses are not part of a moral dilemma because a vegetarian human population could progress as well.

2.      Animal experiments are only part of a moral dilemma, if they cannot be replaced by other research methods.



The difference principle

The absolute priority of the worst case is a possible interpretation of the difference principle (chapter 2.4), but in Rawls’ theory the worst case refers to economic welfare, whereas in negative utilitarianism it refers to life (dis)satisfaction.


According to Rawls risk-aversion follows from the idea of the original position and justifies the difference principle. But it is hard to see why distributive justice should be risk-averse and human rights risk-tolerant. Rawls obviously considered human rights to be risk-averse – consistent with the difference principle. This is plausible insofar, as human rights prevent some of the worst cases of suffering (like the torturing of opposition members in totalitarian systems). But it is implausible in other cases:


Example: If a certain speed limit causes 2% or 20% victims a year is irrelevant in Rawls’ theory, as long as it reflects the will of the majority. A moral principle which demands to lower the speed limit against the will of the majority is considered to be totalitarian.

Let us consider some justifications of risk-tolerance in the context of speed limits:

1.      Suffering belongs to life. Risk belongs to progress.

This argument does not hold because life and progress go on despite a lower speed limit.

2.      A further restricting of liberty rights asks too much from the majority and induces counterproductive results.

This argument does not hold because there are many laws which demand more self-control than a lower speed limit.

3.      To tolerate certain kinds suffering increases the competitiveness of a culture.

This is probably the essence of the matter. Road traffic is just one example of a whole collection of cruel customs with winners and losers, customs which are tolerated by the majority. The injustice concerns those who become victims in spite of denying the custom.

Conclusion: In many areas of life Rawls’ liberty principle allows applying Harsanyi’s (risk-neutral) utility maximization, although the concept of the original position argues for the rationality of risk-aversion.





We all have the strength to endure

the misfortune of others


La Rochefoucauld





Population ethics:

We shall have to make a serious effort to predict our actions’ effects on population, and to assign a value to those effects. In many cases, the factor we typically ignore—population—is likely to turn out the most important of all [Broome 2005, 411].

The number of future persons exceeds by far the ones of the actual generation:


1)      Rawls’ theory is neutral with regard to expansion and contraction, as long as the principle of intergenerational impartiality is respected, i.e. as long as the changes do not disadvantage an actual or future generation.

2)      Negative utilitarianism considers population ethics as a means to improve the quality of life.


In negative utilitarianism the future generations (as far as they can be analyzed) are treated like components of a multi-generation entity. This multi-generation entity is subject to a risk-averse principle in much the same way as a single generation is subject to the (risk-averse) difference principle. The actual generation therefore can be morally obliged to make sacrifices according to the same logic as a wealthy single person can be obliged to accept a redistribution of welfare. Possibly a small sacrifice of the actual generation could improve the situation of countless future generations. In Rawls theory there is no higher purpose like the total welfare across all generations. The principle of intergenerational impartiality assigns the same rights to each generation.


Example: If it were known that (due to a natural disaster) the world can only nourish a billion people after the next five generations then the actual world population (of about 7 billion people) would have to be reduced dramatically. But how should the responsibility be distributed among the generations?

1.      The logic of impartiality would ask for the same proportional reduction in each generation. Each single generation is a minority within the multi-generation entity. The arguments in favor of such a minority are the same as the ones in favor of minorities within a single generation.

2.      Negative utilitarianism treats a multi-generation entity like a single population. It would burden an over-proportional responsibility to the actual generation, because this policy reduces the accumulated suffering across all generations in question.




3.6 Comparison with Buddhism




In contemporary ethics the axiology which comes closest to Buddhism is negative utilitarianism [Keown, 176].

There are two different versions of Buddhism, which both share important intuitions with negative utilitarianism:


1.      A moderate version, which considers non-existence to be the lesser evil, see Negative Utilitarianism and Buddhist Intuition, pp.303-306.

The moderate version does not deny that the satisfaction of desires contributes to happiness, but assumes that the price for the joy of living in terms of risks (accidents, crimes, war etc.) and in terms of inevitable suffering (illnesses, aging, death) is too high.  


The Buddhist doctrine of reincarnation says that the only chance to permanently avoid these risks and besiege suffering (in the future) consists in the elimination of desires. The corresponding guide is the Eightfold Path, i.e. a retreat-oriented way of living and meditative exercises. Meditation leads to an absolute absence of frustration, a state which was described by terms like Sunyata or Nirwana:

         Sunyata is translated into English as emptiness, voidness, openness, spaciousness, vacuity.  In Theravada Buddhism, suññatā often refers to the not-self Pāli: anatta (Sunyata, Wikipedia)

         The word Nirwana literally means "blown out" (as in a candle) and refers, in the Buddhist context, to the blowing out of the fires of desire, aversion, and delusion, and the imperturbable stillness of mind acquired there-after (Nirwana, Wikipedia).

Both terms can be associated – if at all – with a neutral state on the hedonistic scale (a state which is given the value zero in utilitarianism). Since utilitarianism assigns the same value (zero) to the non-sentient world, the moderate version of Buddhism can be mapped to the Modern NU (applied to a pessimistic scenario). In a world, where suffering will always surpass happiness, non-existence is a rational option.


2.      A perfectionist version, which considers non-existence to be a perfect state, see Negative Utilitarianism and Buddhist Intuition, pp.306-309.

The perfectionist version is related to the Brahman concept. It postulates that there is a perfect, impersonal and spiritual form of existence, which represents the ultimate essence of material phenomena. Sentient beings come out of the Brahman and return to it after death. The Brahman concept originates in Hinduism and was later adopted by some forms of Buddhism [Fowler, 34]. Insofar as the Brahman is described as bliss [Raju, 54, 228] it is justified to assign it the maximum value on the hedonistic scale (instead of zero). If non-existence is associated with perfect happiness, then a happy life is not morally better than an unborn life, a consequence which reminds of Negative Preference Utilitarianism. However, negative preference utilitarianism assigns the value zero to non-existence, so that the hedonistic scale contains exclusively negative values.


Within such a world view it is rational to eliminate desires, as well as within the moderate version of Buddhism above. The path to go is similar, but the goal is a union with the Brahman, rather than a journey into absolute nothingness.

In Hinduism meditation is done to realize union of one's self, one's ātman, with the omnipresent and non-dual Brahman (Hinduism, Wikipedia).

Some of Buddha’s statements allow maintaining the search for perfect happiness (instead of neutrality):

….(highest level of meditation) leading to a state with neither pleasure nor pain, which the Buddha said is actually a subtle form of happiness (more sublime than pīti and sukha) …...also called a form of non-sensual happiness (Dhyana in Buddhism, Wikipedia)

The crucial question is, if meditative perfection (the non-existence of the ego) allows grasping a characteristic of the non-sentient world. Adherents of the Hindu tradition believe it does, Buddha abstains from metaphysical speculations.


In both versions the Buddhist emphasis on compassion and risk-aversion has the following consequences:

         The minimization of negative averages is considered to be morally better than the minimization of negative totals. The quality of negative welfare is more important than the quantity of negative welfare. Negative averages avoid the Negative Repugnant Conclusion [Broome 2004, 213].

         If surveys measure welfare with point scales or percentages, then an asymmetric scale has to be used (see chapter 3.3), so that the lower levels of welfare get more weight within the average.




Despite of common intuitions, there are important differences between Buddhism and negative utilitarianism:

         Buddhism developed a sophisticated virtue ethics, which is defined in the Eightfold Path.

         Negative utilitarianism is consequentialism and not virtue ethics. Moral behavior can only be defined in concrete situations.



         In negative utilitarianism the use of violence is legitimated, if the suffering produced is smaller than the suffering prevented. This allows the development of a just war theory. A negative utilitarian may be morally obliged to participate in a just war.

         Buddhism adheres to the principle of non-violence, see Ahimsa. A (disputed) interpretation of this principle goes as follows:

o   Strive for a retreat-oriented life, where conflicts are avoided in the first place (e.g. the life in a Sangha).

o   If you live in a laic community, then do not specialize on killing (i.e. do not choose military service as a profession)

o   Use violence only for self-defense, if all other options are exhausted.

The two ethics converge if it could be shown that – in the long run – the Eightfold Path is the most consequent way to minimize suffering.




4. Conclusion



Negative utilitarianism

Negative utilitarianism (NU) is an umbrella term for ethics which models the asymmetry between suffering and happiness [Fricke, 14]. It includes concepts that

         assign a relative priority to the avoidance of suffering

         assign an absolute priority to the avoidance of suffering

         consider non-existence to be the best state of affairs


In this paper we investigate compensation across different persons (inter-personal compensation). For compensation within the same person (intra-personal compensation) see Why I’m (Not) a Negative Utilitarian.


1)      A relative priority of (the avoidance of) suffering means that happiness has moral value, but less than suffering. The moral weight of suffering can be increased by using a "compassionate" metric, so that the result is the same as in prioritarianism. The more weight is assigned to suffering, the more difficult it becomes to compensate suffering by happiness. The intuition that “global suffering cannot be compensated by happiness” turns global well-being negative, so that the maximization of happiness turns into a minimization of suffering. The rationality of this intuition is investigated in The Denial of the World from an Impartial View.


2)      Absolute priority represents a border case of relative priority, where the moral weight of happiness converges towards zero. Absolute priority can be approximated by relative priority. Relative priority is the general model.


3)      The claim that non-existence is the best state of affairs is counter-intuitive for most Non-Buddhists.



Rawls’ theory versus negative utilitarianism

1)      Priority of human rights:

a)      In negative utilitarianism it is theoretically possible to override human rights, if it serves the minimization of negative total welfare.

b)      In Rawls’ Theory of Justice it is impossible to override human rights, even if this principle causes the perpetuation of suffering.


2)      Distribution of welfare among social classes:

Rawls’ difference principle accords well with the intentions of negative utilitarianism, if the term welfare is interpreted as life satisfaction.


3)      Compensation among generations:

a)      In negative utilitarianism a generation may be obliged to sacrifice themselves, if it serves the long-term reduction of suffering.

b)      In Rawls’ theory the idea of a sacrifice is prevented by the principle of intergenerational impartiality.


4)      From a negative utilitarian perspective the most urgent amendments to Rawls’ theory are the following:

a)      A principle for the protection of non-contractual cases (including sentient animals)

b)      A risk-averse population policy

c)      A risk-averse principle for technological progress.







1.      Allik Jüri and Anu Realo (1997), Psychological and Cultural Mechanisms of Suicide, in Trames, Vol.4, Tallinn Pedagogical University

2.      Anderson Ron (2012), Human Suffering and Measures of Human Progress, Presentation for a RC55 Session of the International Sociological Association Forum in Buenos Aires, Argentina

3.      Arrhenius Gustav (2000), Future Generations, A Challenge for Moral Theory, FD-Diss., Uppsala University, Dept. of Philosophy, Uppsala: University Printers

4.      Bentham, J., Value of a Pain or Pleasure (1778), in: B. Parekh (ed.): Bentham´s Political Thought, London 1973

5.      Broome John (1991), Weighing Goods, Oxford: Basil Blackwell

6.      Broome John (1999), Ethics out of Economics, Cambridge University Press, UK

7.      Broome John (2004), Weighing Lives, Oxford University Press

8.      Broome John (2005), Should We Value Population?, The Journal of Political Philosophy, Vol.13, No.4, pp.399-413

9.      Broome John (2008), The Ethics of Climate Change, Scientific American, New York, June issue

10.  Chao Roger (2012), Negative Average Preference Utilitarianism, Journal of Philosophy of Life, Vol.2, No.1, p.55-66

11.  Clarenbach Lars Albert (1999), Die Wohlfahrtstheorie auf der Grundlage kardinaler Messbarkeit und interpersoneller Vergleichbarkeit von Nutzen, Diplomarbeit, Köln

12.  Dworkin Ronald (1977), Taking Rights Seriously, Harvard University Press

13.  Everett, Daniel L. (2008), Don't Sleep, there are Snakes, Pantheon Books

14.  Fehige Christoph (1998), A Pareto Principle for Possible People, in C. Fehige and U. Wessels, eds., Preferences, Berlin: Walter de Gruyter

15.  Fowler Merv, 1999, Buddhism: Beliefs and Practices, Sussex Academic, Brighton

16.  Fricke Fabian (2002), Verschiedene Versionen des negativen Utilitarismus, Kriterion Nr.15, pp.13-27

17.  Harsanyi, John C. (1955): Cardinal Welfare, Individualistic Ethics, and Interpersonal Comparisons of Utility. In: Journal of Political Economy 63. Pp. 309-321.

18.  Harsanyi John C. (1975), Can the Maximin Principle Serve as a Basis for Morality? A Critique of John Rawls’ Theory, American Political Science Review 69: 594-606.

19.  Hare Richard Mervyn (1976), Ethical Theory and Utilitarianism, Contemporary British Philosophy, H.D. Lewis Ed.

20.  Helliwell, John, Richard Layard, and Jeffrey Sachs, eds. (2015), World Happiness Report. New Nork, The Earth Institute, Columbia University Press.

21.  Hirata Johannes (2004), Happiness and Economics

22.  Keown, Damien (1992), The Nature of Buddhist Ethics, New York, Palgrave.

23.  Kleinewefers Henner (2008), Einführung in die Wohlfahrtsökonomie, Verlag W. Kohlhammer, Stuttgart

24.  Kolm Serge-Christoph (1982), Happiness-Freedom, Deep Buddhism and Modernity, Presses Universitaires de France, Paris

25.  Larmore Charles (2010), Vernunft und Subjektivität, Frankfurter Vorlesungen, Zürcher Version

26.  Lumer Christoph (2005), Prioritarian Welfare Functions, in Daniel Schoch (ed.): Democracy and Welfare, Paderborn: Mentis

27.  Midgley Mary (1994), Duties concerning islands, an Essay in Ethics, edited by Peter Singer, Oxford University Press

28.  Nozick Robert (1994), The Rationality of Side Constraints, in Ethics, edited by Peter Singer, Oxford University Press

29.  Oswald Andrew (1999), A Non-Technical Introduction to the Economics of Happiness

30.  Parfit Derek (1984), Reasons and Persons, Clarendon Press, Oxford

31.  Popper Karl R.(1945) The Open Society and its Enemies, Volume I, London, Fifth Edition (revised), Routledge, UK, 1966

32.  Raju, P.T., 1992, Philosophical Traditions of India, Motilal Banarsidass Publishers, Delhi

33.  Rawls John (1958), Justice as Fairness, in Philosophical Review 67, pp.164-194

34.  Rawls John (1971), A Theory of Justice, Belknap Publishers, Cambridge

35.  Rawls John (2007), Lectures on the History of Political Philosophy, Cambridge, Massachusetts: Harvard University Press, edited by Samuel Freeman.

36.  Scarre Geoffrey (1996), Utilitarianism, London

37.  Schefczyk Michael (2010), Der Nutzen der Freiheit, in Neue Zürcher Zeitung Nr.12, S.69

38.  Smart J.J.C. (1973), An Outline of a System of Utilitarian Ethics, in J.J.C.Smart and Bernard Williams, Utilitarianism: For and Against, Cambridge

39.  Stanford Encyclopedia of Philosophy (2006), The Repugnant Conclusion

40.  Temkin Larry (1994), Weighing Goods: Some Questions and Comments, Philosophy and Public Affairs 23, no.4, pp.350–380.

41.  Vukomanovic Milan (1998), Schopenhauer and Wittgenstein: Assessing the Buddhist Influences, University of Belgrade

42.  Wolf Clark (2004), Repugnance, Where is Thy Sting? in The Repugnant Conclusion, Essays on Population Ethics, Kluwer Academic Publishers