Oversite

A Personal View

►	Oversite Home Page
►	Philosophy Index Page

Rethinking Utilitarianism

By Bruce Barbour - Version 2.1 - April 2026, Original version 1 - February 2026

Introduction

I am interested in effectiveness, not philosophical purity.

Utilitarianism is a type of Consequentialist ethics. Consequentialism postulates that the best ethical decision is the one that produces the most favorable consequences for a group of people overall.

There are various versions of Consequentialism and also Utilitarianism. Each type of Consequentialism / Utilitarianism offers a different perspective on how to evaluate moral actions based on their consequences. I will concentrate mainly on Utilitarianism. Though some of what I discuss here may also apply more broadly to Consequentialism.

Utilitarianism claims that the morally right action is the one that generates the greatest "Good" for the greatest number of people. The 'Good' is typically understood in terms of maximizing Utility and minimizing dis-Utility for a group of people. The group can be a specific set of people or potentially the whole of society.

Initially Utility was equated to happiness. When this is done the greatest "Good" or happiness is calculated by what is called a "hedonic calculus". This basically adds up anticipated gross Utility/happiness for an impacted population and subtracts any losses of Utility for each option being considered. The option that gets the highest "score" is the preferred option. The concept of Utility has been expanded to refer to more than just happiness, referring to other "Goods" as will be discussed below. In these cases a calculus can also be done. This process could be referred to as a Utilic calculus - to make up a term - or a Utilitarian calculus if you don't like made up terms.

One of the issues is that the Utilitarianism process can lead to outcomes that may be thought of as undesirable for a section of the effected population. An example could be that a society that decides to enslave 10% of its population in order to provide free services to the other 90%. The Utility/happiness of the 90% could be improved by say 15% for each person while the lives of the slave 10% each decrease by 90%. The hedonic calculus would indicate that this is an ethically acceptable outcome. But most people would say no - this is not a satisfactory outcome even if the hedonic calculus indicates otherwise.

Another example could be having an unarmed person fight a lion to death (in 99% of the time to the person) for the entertainment of the 100s of customers attending the Colosseum. Hedonic calculus would probably indicate that this was acceptable. But most people today would say it is not.

Utilitarianism concerns itself with "ends", that is increased Utility, often meaning increased happiness, but does not concern itself with "means", how the end is achieved. Utilitarianism may allow what most people would consider horrible means to achieve an end that increases overall Utility. By ignoring means Utilitarianism does not cover all aspects of the ethics of a process or an action.

Other philosophers have noted that Utilitarianism / Consequentialism can throw up recommended actions and outcomes that have been called by some "Repugnant Conclusions"⁽¹⁾.

There is another issue with Utilitarianism. That is - why should the state, or whoever is intending to implement a change based on hedonic (or other Utilic) calculus, be allowed to decrease the happiness (or Utility) of one group to benefit the happiness (or Utility) of another group, even if the the total happiness (Utility) of the two groups is higher than before? It is unfair to the minority group.

These examples indicate that there are issues with the simplistic application of Utilitarianism and also potentially other Consequentialist approaches. How an outcome is achieved undoubtedly has an ethical dimension and an ethical theory that ignores this is deficient as a universal ethical theory⁽²⁾.

However the issues mentioned above have a simple solution.

Deontological Fix

The solution to both the Utilitarianism slavery and killing for pleasure problems is to simply say that these option are off the table. I.e. you cannot kill for pleasure and you can't enslave anybody regardless of the hedonic calculus indicating that this increases the overall pleasure / happiness of the masses.

This is putting a Deontological overlay on Utilitarianism and potentially other forms of Consequentialism - we have a moral duty not to kill and not to enslave. Hedonic calculus can still be performed but some options are not acceptable and cannot be considered⁽¹⁾.

A couple of questions that needs to be addressed straight up are: can two ethical theories, Utilitarianism and a form of Deontology, be applied at the same time? Don't they conflict?

The simple answer to the second question is that "the Deontological overlay may impact at the peripheries and also partially on how the Utilic calculus process is performed. But they do not fully contradict."⁽³⁾ Another difference between the two ethical theories is that Utilitarianism compares alternatives. Deontology does not, at least not to the same extent.

As a consequence of these differences Utilitarianism and Deontology may be able to be used to complement each other, rather than being thought of as contradictory ethical theories.

Deontology, if used in conjunction with Utilitarianism, sets the permissible boundaries and informs some of the rules of the process within which the Utilic comparison can take place. Utilitarianism, and other forms of Consequentialism, need to be done with both wisdom and virtue. A correctly chosen deontological fix plugs the manifest ethical gaps in Utilitarianism.

Deontology proposes a set of duties or rules to govern action. Kant is one of the most well known proponents of Deontology⁽⁴⁾. However Kant did not give a full list of duties/rules, which he called Categorical Imperatives or maxims, that must be adhered to. The main example he did give, an exceptionless prohibition on lying, shows that there maybe significant issues with some, if not all, exceptionless prohibitions.

The "Axeman at the Door" shows the issue with Kant's Categorical Imperative against lying. Briefly a man holding a bloodied axe knocks at your door and ask whether your friend is in the house. He is but you are certain that the Axeman intends to kill your friend once he finds him. The difficulty with Kant's Deontology is that his exceptionless maxim prohibits lying in any circumstance and says that even in this case it is the person's duty to tell the truth. In this case it means telling the Axeman that your friend is in the house. Which would probably lead to your friend's death.

Regardless of Kant's maxim I am sure that most people would not tell the Axeman the friend was in the house. This indicates a significant issue with Kantian Ethics, just as there are issues with the Utilitarian position in the earlier examples provided in this article. In some situations Kantian Ethics allows ends which cannot be thought of as ethical, subordinating ends to duties.

However Kant is not the only Deontologist. W. D. Ross proposed a form of Ethical Pluralism. Ross identifies seven fundamental prima facie duties that people have an ethical duty to adhere to. These prima facie duties⁽⁵⁾ are:

Fidelity - Keeping promises, telling the truth, avoiding deception.
Reparation - Making amends for past wrongs.
Gratitude - Returning favours and repaying kindness.
Justice - Distributing benefits and burdens according to merit.
Beneficence - Promoting others’ well-being, happiness, and improvement.
Self-Improvement - Improving one’s own character and intelligence.
Non-Maleficence - Avoiding harm or injury to others.

Ross suggests that the list may not be exhaustive – so there may be other duties as well as those listed. But he also suggests that most, if not all, other duties derive from the seven duties he listed.

He claims that the nominated duties are self-evident, similar to mathematical axioms. According to Ross they do not require proof, but are recognised through moral reflection⁽⁶⁾.

The Ethical Pluralism duties can conflict in real situations. In these real situations, the decision maker must determine the actual (or proper) duty. There is no formula or algorithm for deciding which duty wins. Ross claims moral decision-making relies on judgement. It requires the decision maker to weigh competing moral duties and come to a conclusion.

Reconsidering the axe-murderer case and the situation with lying in the light of Ross’s Ethical Pluralism duties:

Fidelity counts against lying.
Beneficence and non-maleficence count strongly in favour of lying.

The conclusion: - Lying to save a life is morally permissible—and would in most cases be required.

This example shows that some prima facie duties can override others in particular contexts. It also shows the result depends on the judgement of the decision maker, and not any formulaic evaluation⁽⁶⁾. Ethical Pluralism considers both ends and means, as shown in the axe-murderer example.

In terms of Utilitarianism I suggest that W. D. Ross's Ethical Pluralist duties are overlain on Utilitarian decision making. In particular the duties of Fidelity, Justice, Beneficence and Non-Maleficence depending on what is being considered. The other duties may also apply in specific circumstances. And, as Ross allows, there may even be duties relevant for specific cases beyond those suggested by him.

The second issue mentioned earlier was whether the State or anybody should be able to do anything that decreases the happiness (or Utility) of, or otherwise disbenefits, a minority group even if it provides a greater benefit to another group. This is not a black and white situation. Sometimes it may be necessary for the State to implement a decision that disadvantages or decreases the happiness of another group.

Take the situation where the State wants to construct a new roadway that will greatly benefit large numbers of road users. (Consider a similar case with a new train line if you are against cars and roads.) It will require the compulsory acquisition of a number of houses which will be knocked down to make way for the road. The benefits that the road will provide to the large number of users will be great. However the inconvenience and possible anguish and sorrow caused to the dispossessed home owners will be high on an individual and minority group level. They should not be forced to bear this disbenefit lightly.

There are an number of approaches to tackle this. Firstly if the proposal is to proceed the benefits should outweigh the disbenefit, not just by a small factor, but would have to outweigh the disbenefits by a reasonably large factor. I am not going to nominate what that factor is - and it might vary between cases. This is a weighted calculus approach to Utilitarianism. Weighting goes against traditional view of Utilitarianism that requires impartiality in the hedonic calculus. Weighting treats the loss of the disadvantaged as more significant than the gain of the advantaged.

The second approach - which I see as complementary rather than a replacement for the first approach - comes from the W D Ross' Deontology overlay applied earlier. One of the prime duties is "Reparation", which is "making amends for past wrongs" though in this case the wrong isn't past but current. In other words if a cost is to be borne or a disbenefit suffered by a minority group then they should be compensated for that cost and disbenefit, plus additional payment on top as a recognition of the wrong that is being done to them for the benefit of the larger group.

One other of Ross's duties could pose a challenge for Utilitarianism is the duty of Beneficence, that is the duty to promote others’ well-being, happiness, and improvement. In some Utilitarian situations some people may be worse off. However in a Utilic calculus sense overall the total sum of people effected are better off. And the duty of Reparation means that those that may be worse off should be compensated. In this case Beneficence has to be considered in Utilitarian terms. In other words, in terms of the total rather than the individual.

There are a couple of further situations to keep in mind. The proposal implemented decreases the advantage of a minority over the majority. However if this was just to fix a situation where the minority has a prior advantage and this change was bringing them back to the situation of the rest of the people then no weighted calculus or compensation should apply.

If the situation increases the benefits for all people - that is no-one suffers any loss - then even if the benefits are not spread evenly then this situation may be considered satisfactory, unless the spread of positive outcomes is significantly skewed to particular group (e.g.10% get 90% of the benefits and the remaining 90% get 10%) or most benefits go to a group that is already considered advantaged.

Welfare Utilitarianism

Utilitarianism holds that a morally right act is an act that produces the most Utility or Good, an act that produces a good outcome or good consequences. This Utility or Good is often defined as the maximisation of happiness and the minimisation of unhappiness, but can also include other forms of Utility. This type of hedonic Utilitarianism is aimed directly at a person's psychological welfare - how can overall happiness be increased directly for a person or group of people.

"Utility" or "Goodness" can be extended beyond directly considering happiness to deal with a person's or group of people physiological, psychological and social welfare. The additional "Goods" that could be considered are pleasure, social well being and connection and welfare and healthiness. Note that, as Aristotle suggests, these Goods ultimately translate into a person's happiness. For example, if you have better welfare and health you are going to be happier as well as physiologically better off. For each of the Goods if you ask a person "Why do you want X?" the answer will be, either immediately or ultimately, "Because I want to be happy." However one of the advantages of the specific Good is that they may be more targeted, more measurable and/or able to be predicted more reliably than simply happiness⁽⁷⁾.

The table below shows the associated pairs: Good/Bad. As per Utilitarianism the increase in the Good must outweigh any increase in the Bad. The proposed Deontological overlay and other aforementioned suggested modifications must also apply to these Utilic calculi.

The Good / Utility	The Bad / Dis-utility
happiness	unhappiness
pleasure	pain
social well being and connection	social disengagement / isolation
welfare and healthiness	disadvantage and diminished healthiness

There are other Goods that could also be considered but some may be more problematic. For example Eudaimonia or flourishing would be a Good. However its opposite (the Bad) could be wretchedness or something similar. You could not do a Utilic calculus with this combination. It would not be ethical in any way to increase someone's wretchedness, or make someone wretched, for the sake of increasing someone else's flourishing regardless of what Reparations could be made. Beneficence would preclude it. However what could be considered for the Utilic calculus would be pairing a Good with a lesser Good. For example: pairing Eudaimonia with happiness or with pleasure.

Other "Goods" that could be considered are preference (though this can be problematic if taken to extreme), knowledge, and justice.

Notes:
(1). The proposed Deontological fix or overlay will not have solved all issues associated with Utilitarianism / Consequentialism. E.g. Derek Parfit "Repugnant Conclusions" in "Reasons and Persons". The Repugnant Conclusions indicate to me that hedonic calculus is not suitable for assessments involving the change of population numbers, up or down, nor changing population members.
(2). The only part of Utilitarianism that has intrinsic ethical implications is whether what is defined as "Utility" or "the Good" has any ethical worth. The rest of Utilitarianism is just the process of aggregation of the consequences of actions that can just as easily be used to aggregate and maximise "the Bad" as "the Good". Similarly the ethical validity Consequentialism depends on whether the consequences are defined as ethically good. A Utility / Good such as "happiness" has been argued by many philosophers as being of ethical worth (e.g. Aristotle) but even happiness can have downsides. Say the Utility to be maximised is "Freedom". An "ethical system" that sought to maximise that could lead to an extreme form of Libertarianism - you are free to become a super rich billionaire who pays no taxes, but you are also free - and more likely - to die as a penniless pauper in the gutter. An outcome which I hope is not ethically acceptable to most people.
(3). Utilitarian purists may not agree with this proposal and say it goes against the very essence of Utilitarianism. However that does not concern me - I want an ethical system that works and covers all situations, not allowing what are clearly unethical processes or situations to get to a supposed ethical end. It could be thought of as a different ethical system, with a different name, if necessary to maintain the "purity" of standard Utilitarianism for some people, but I won't be doing that in this article.
(4). Divine Command Theory provides more widely existing Deontological ethical systems than Kantian ethics.
(5). W. D. Ross's duties could also be considered as virtues (a couple may require some minimal massaging) or duties arising from virtues.
(6). Aristotle might say the decision maker needs to be able to exercise Practical Wisdom.
(7). Aristotle suggests that happiness is not the highest Good. The highest Good is human Eudaimonia or flourishing. However the concept is not sufficiently widely understood and known about for a normal "person in the street" to answer the question "Why do you want X?" with "Because I want Eudaimonia" or "Because I want to flourish". Happiness is a generally understood Good.

Oversite Home Page
Philosophy Index Page

Top of Page

| Site Information | (C) |