| Thinking hard|
or hardly thinking?
|Major trains of thought|
|The good, the bad|
and the brain fart
|Come to think of it|
Utilitarianism is an ethical system that bases the rightness or wrongness of an action on whether that action is net beneficial or harmful. In addition, an action with greater benefit than another is more right, and vice versa. Utilitarianism can roughly be summed up as a kind of moral arithmetic -- if an action causes more good than bad, then the action is good, and vice versa.
Utilitarians typically do have one single moral absolute, usually pleasure or satisfaction of self-interest, which they consider intrinsically good and the basis for morality.
The trouble with utilitarianism is that it assumes humans have utility functions. It's a tempting idea because people and society actually use something like utilitarianism in the non-extreme ranges. But it's brittle in obvious ways, leading easily to absurd results.
The version of utilitarianism promoted by Mill and Bentham holds pleasure to be the only intrinsic good. Bentham summarized classical utilitarianism by stating that "nature has placed mankind under the governance of two sovereign masters, pain and pleasure."
This is often criticized in modern times for four serious problems: there is no difference between "base" pleasure and more sophisticated or intellectual pleasure (except Stuart Mill differentiates between them in his book Utilitarianism); equality is not taken into account (if rich people get richer while the poor continue in misery, this doesn't increase the net pleasure of everyone on the planet. Studies show that there's a certain limit on income that prevents additional happiness. That amount is around 75,000 dollars annually. Annual earnings over 75,000 dollars stifles the accumulation of additional happiness.); pleasure is too abstract to be measured or quantified accurately (different people find different things pleasurable); and the concept of utilitarianism ignores recent research on the notion of the "hedonic treadmill" and the habituation of pleasurable experiences. (For example, listening to the same song doesn't maximize happiness any further. It may eventually drive the person crazy after a certain point).
Preference utilitarianism is an alternative to classical utilitarianism which defines satisfaction of personal preference or interests as "good," with pleasure being only one of these interests. The most vocal proponents of this type of utilitarianism are R. M. Hare and Peter Singer. Preference utilitarianism avoids many of the issues with classical utilitarianism because it recognizes other forms of "good" which might otherwise be overlooked in the name of pleasure. For example, classical utilitarianism cannot explain why adultery is wrong if one's partner is not made to suffer by it. In preference utilitarianism, the preference of the adulterer's wife or husband matters even if the frustration of that preference wouldn't increase pain or minimize pleasure.
Utilitarianism is sometimes divided into "positive"--which seeks to maximize net happiness--and "negative"--which seeks to minimize net suffering. Negative utilitarianism is often criticized for implying that it would be morally good to destroy all life painlessly 
Positive utilitarianism has criticisms too. It implies that overpopulation (of sentient beings) is mandatory, unless the suffering caused by this entirely "cancels it out" to the point that a low-population alternative would've had more net happiness anyway.
Average utilitarianism is an alternative that essentially states that population doesn't matter as much as "average" happiness or satisfaction per sentient being. A world with 4 sentient beings who are all 90% "happy" is better than a world with 8 sentient beings at 70% (this ratio is after taking loneliness into consideration). The 90% world is also better than a world with one person at 70%. The 90% world has neither the most net happiness nor the least net suffering (assuming the function used to compare is simple addition, and that each being's percentage scale is the same), but it is preferred under average utilitarianism.
Confusion is still caused when averaging lower sentient beings, such as rodents, with humans. Perhaps this can be solved by averaging the "goodness" of an individual's experience, and then multiplying that ratio with the individual's "amount of sentience," to determine the weighted average. Perhaps a scale can add weight to beings that suffer more, mathematically implying a prioritarian concept of equality.
Average utilitarianism sorta implies that the ideal world is a monoculture of the happiest (or least frustrated) beings available (perhaps managed by non-feeling AI and accomplished through eugenics), though this is arguably better than nonexistence, or stuffing as many slightly-happy beings into the world as possible.
Act vs. rule utilitarianism
Act utilitarianism evaluates every individual act, declaring it "good" if it produces the most utility out of all possible acts in the circumstances. Critics charge, however, that we cannot go through life judging every single circumstance individually - we need easy to follow rules to guide our moral conduct. We often cannot predict the outcomes of our actions beforehand - so we need to follow uniform, easily-understandable rules. Thus, rule utilitarians take a different approach - they seek to establish moral rules, choosing moral rules based on which rules will produce the most utility out of all possible moral rules. However, having so chosen those rules, the rule utilitarian then judges individual acts, not by the amount of utility they produce individually, but on conformance to those rules.
A good example: stealing generally has bad consequences so rule utilitarians view stealing as unethical in all scenarios, even those when doing so would have overall good consequences. An act utilitarian can justify stealing medicine to save a child's life because stealing has the best overall consequences in that specific scenario, even though it normally has overall bad consequences. Rule 'utilitarianism' is, actually, deontology.
Problems and criticisms
Many criticisms are on the basis of absolute moral standards other than those proposed in utilitarian thought. These may be religious ones, or maybe secular ones based on human rights. Most forms of utilitarianism consider only the effects of actions and place no judgement on a person's motivations and whether they have "good intentions".
Hypothetically, classical utilitarianism could be used to justify the enslavement of a small minority in order to bring greater benefit or pleasure to the majority and hence a net gain in happiness. Therefore, utilitarianism is not as "egalitarian" as it might seem. This ignores the suffering involved and the fact that this would seem morally dubious to most people.
As soon as they are introduced to utilitarianism in first-year philosophy pretty much everyone realizes the mentioned absurd results that can potentially result from utilitarianism in the real world.
Although utilitarianism seems very logical, it can lead to some questionable moral statements, and many see the need to correct it. Some argue that the philosophy lacks attention to the individual and have proposed ethical egoism as an alternative. Others see equality as the problem and have proposed Prioritarianism and strict egalitarianism as alternatives. Yet others say that Utilitarianism doesn't have enough Goddidit, but that is probably just bullshit.
And, empirically, making utilitarian assertions in trolley car style dilemmas is not associated with better utilitarian performance in the real world, and has a negative correlation to impartial altruist concern for others.
Utilitarianism proposes that pain and pleasure can be systematically quantized, measured, and weighed. However, there is, as yet, no standard unit of suffering or bliss, without which it is exceedingly difficult to compare harms and benefits in the form of pain or pleasure. Though utilitarian truths may be intuitively educed, these truths are correspondingly vague. For example, while it is clear that factory farming causes intense, prolonged suffering, it is not clear whether, all considered, every last factory should be closed; it is possible that retaining some factory farming could maximize the aggregate good. Likewise, maintaining a small underclass of slaves could yield the best balance of pleasure over pain for society as a whole.
Furthermore, the hedonistic calculus demands knowledge we cannot properly claim to know. How are we to know what sonar feels like to a bat? How are we to know water feels like to a plant? How are we to know how these subjective experiences, many alien to our own, feel, let alone compare? Though we may apply general rules to guess the nature of these experiences, we are still at a profound loss as to what exactly they are and how they can possibly be reduced to just a few numbers.
This uncertainty about the feelings of other beings and entities has resulted in hilarious fringe initiatives like Brian Tomasik's essay Is there suffering in fundamental physics? and an organisation called People For The Ethical Treatment of Reinforcement Algorithms. Some might say that PETA itself is an older example, although strictly speaking PETA is not utilitarian - it ignores the suffering of wild animals and only considers suffering caused by human action, not human inaction.
Even if we restrict our attention to human suffering, and try to rely on supposedly good evidence like randomised controlled trials, as effective altruists often do, the problem is that human societies are very complex and constantly changing. This was demonstrated when Giving What We Can, a self-proclaimed effective altruist charity, responded to a 2015 Cochrane Collaboration casting doubt on certain types of deworming initiatives, and in passing quoted a criticism that "public health nutrition has changed, worm burden has declined and this probably accounts for the lack of effect on biomedical outcomes in contemporary studies". Despite the initial evidence being flawed by programming errors, statistical errors and it being of uncertain relevance to today's world in any case, Giving What We Can continues to recommend the same deworming initiatives anyway.
- Ayn Rand and Objectivism - "You can make those others happy when and if those others mean something to you selfishly."
- Hedonic Calculus
- John Stuart Mill
- Roko's basilisk
- Peter Singer
- Average utilitarianism can be divided into positive-average-utilitarianism and negative-average-utilitarianism. The hedonic/preference distinction makes 8 combinations.
- Osborne, Richard. Philosophy For Beginners
- Less Wrong: Torture vs Dust Specks
- ‘Utilitarian’ judgments in sacrificial moral dilemmas do not reflect impartial concern for the greater good. Cognition, 2014
- Is deworming really effective? A response to recent deworming studies
- Ben Goldacre (July 22, 2015). "Scientists Are Hoarding Data And It’s Ruining Medical Research".