Would you kill one person to save five? Depends if you’re a millennial or not

Unthinkable: Should we be worried about the apparent rise of utilitarianism?

The western world has been gripped by moral panic in recent years.

The spread of populism. The erosion of trust. The coarsening of dialogue. All these are big concerns but we are missing perhaps a greater and potentially much more troubling change in how people deliberate on moral questions, namely the rise of utilitarianism.

As a way of thinking, it is distinguished from the two other main schools of moral philosophy: virtue theory and deontology (a fancy word for rule-based ethics). Utilitarians measure what’s right and wrong – not by whether an act is good for your character, nor by some abstract commandment – but simply by consequences.

It was an Ulster man, the 18th-century Presbyterian Francis Hutcheson, who came up with one of the most famous formulations of the philosophy, saying you should act in a way to produce "the greatest happiness of the greatest numbers". The principle has influenced many social reforms over the past 200 years but has also been twisted to legitimise atrocities in the name of "the greater good".


In recent years, studies have been carried out to measure just how utilitarian we are becoming. The research focuses on a moral conundrum designed in 1967 by the English philosopher Philippa Foot in which people are asked to consider what to do about a runaway tram in different scenarios.

This “trolley problem” has entered the cultural mainstream – it crops up in an episode of the NBC series The Good Place. In its detailed form, it presents three scenarios and asks whether one is better than the next.

In the first, “switch” scenario, the empty boxcar is heading towards five workers on the main track and will kill them all unless you pull a switch diverting it onto a side track where one person will be killed.

In the second, “loop” scenario, you can divert the trolley onto a side track that bends back onto the main track. It means killing one person in the knowledge that the collision will bring the trolley to a halt, preventing further casualties on the main line.

In the third, “footbridge” scenario you can stop the train killing the five workers only by pushing a man off a bridge above the track into the train’s path, bringing it to a halt.

In each situation, one person gets killed to save five but with the loop, and more so, footbridge options the manner of the sacrifice is perceived as more troubling.

In initial surveys, only about 10 per cent of people would agree to push the stranger off the bridge when presented with the choice. Significantly, that 10 per cent scored high on a measure of psychopathic personality traits, such as lack of empathy, glibness and impulsivity.

However, a 2017 trolleyology study – yes, it has now become an entire sub-discipline – examined surveys over several decades and it discovered the endorsement rate for "footbridge" has been slowly creeping up. This "begins with individuals born approximately in the 1960s, and accelerates among birth cohorts after 1990", the US and Brazilian research team writes.

“Recent cohorts (often referred to as millennials) are significantly more likely to support utilitarian sacrifice than their predecessors (especially baby boomers, born before 1970) - a divide which may contribute to patent disagreement between younger and older adults in real-word debates about ethics and policy.”

A major study published earlier this month, analysing the responses of 70,000 participants in 42 countries to the trolley problem, shows just how far things have shifted. Across all countries, the average endorsement rate for the footbridge sacrifice was 51 per cent. It was higher in most western countries (61 per cent in the US; 56 per cent in Ireland). In contrast, most Asian countries were below average, with just 32 per cent of Chinese people finding the footbridge sacrifice morally acceptable.

The report tries to explain what accounts for this cross-cultural difference? Is it average income? Levels of individualism? Religiosity? The authors suggest a different variable at play: relational mobility (RM), or the ease with which people in a given society can develop new relationships.

Pushing someone off a bridge is the sort of thing that would give you a bad reputation in your community. But in a society where you can develop new friends the risk of being ostracised is reduced. Or as the report puts it, “People in low relational mobility societies may be less likely to express and even hold attitudes that send a negative social signal.”

One of the authors Edmond Awad told Unthinkable they focused on this variable for two reasons. "First, the relationship between ethical decision-making and RM was our strongest a priori theoretical prediction", based on previous studies. "Second, RM does indeed have the strongest and most consistent predictive power."

The study is part of a broader research project started in 2015 when Awad co-developed Moral Machine, a website that gathers human decisions on moral dilemmas faced by driverless cars. His research highlights the challenge of developing ethical standards that can be accepted globally – potentially a major obstacle to the introduction of autonomous vehicles.

Will it be possible to ever agree on a universal moral code?

“I remain agnostic regarding this point,” Awad replies. “It is possible that people don’t have to agree on a moral code. After all, we have different traffic rules in different countries, and this seems to be working fine.”

He notes most of the policy reports on autonomous vehicles are being produced in Europe and the US while many experts believe the technology will be tested in other regions first. “So it’d probably be interesting to see whether actual deployment and regulations would end up being too misaligned or whether one would shape the other.”

Whatever about driverless cars operating on utilitarian principles, should we worry that people are increasingly doing so?

Presented with the trolley problem, a virtue theorist would highlight the damage done to your character by pushing someone off a bridge, and might also suggest that had you properly cultivated virtues such as courage and self-discipline you could swing into action, James Bond-like to avert disaster.

Deontologists – who have heavily influenced human rights theory – would say people are always an end in themselves, and pushing someone off a bridge violates human dignity.

But still it’s not easy to explain if, and why, the footbridge option is morally unacceptable.

Something has been lost in recent years – a sense of honour perhaps, or a religious attachment to fate. Something also has been gained – unemotional, technology-driven logic. Whether you find this concerning may depend on your historical perspective, given so many atrocities of recent decades have been conducted in a spirit of the ends justifying the means.


Ask a sage:

The Koran says if you kill a human being, it is like killing all mankind but if you save one you save all mankind. So should I kill one to save five?

Bernard Williams replies: "If utilitarianism is true . . . then it is better that people should not believe in utilitarianism."