Ethics and the Contemporary World

David Edmonds

• Publisher: Routledge
• Year: 2019
• Print ISBN: 9781138092013
• eBook ISBN: 9781351608015
• Topic: Philosophy
• Subtopic: Ethics & Moral Philosophy

Arguments about ethics often centre on traditional questions of, for instance, euthanasia and abortion. Whilst these questions are still in the foreground, recent years have seen an explosion of new moral problems. Moral and political clashes are now as likely to be about sexuality and gender and the status of refugees, immigrants and borders, or the ethics of social media, safe spaces, disability and robo-ethics.

How should we approach these debates? What are the issues at stake? What are the most persuasive arguments? Edited by best-selling philosophy author David Edmonds, Ethics and the Contemporary World assembles a star-studded line-up of philosophers to explore twenty-five of the most important ethical problems confronting us today. They engage with moral problems in race and gender, the environment, war and international relations, global poverty, ethics and social media, democracy, rights and moral status, and science and technology.

Whether you want to learn more about the ethics of poverty, food, extremism, or artificial intelligence and enhancement, this book will help you understand the issues, sharpen your perspective and, hopefully, make up your own mind.

Part One
Race and gender

One
Profiling and discrimination

David Edmonds

Here is a story about queue-rage, injustice, and profiling.

Each week, I spend a day or two in the British Library. For as long as I can remember, there has been some sort of security in operation at the main doors. It used to operate like this. The security staff would allow the vast majority of people to walk through unmolested. But occasionally they would stop someone and ask to check their bag. This very rarely happened to me (I have a strikingly innocent face) and, when it did, it was only a minor inconvenience, delaying me by just a few minutes. Still, it would annoy me, and meant that I was pipped to my favourite library seat. The Library has now changed its policy – with an interesting psychological consequence, to which I will return later.

In London, as in many parts of the world, we have had to become accustomed to living with the threat of terrorism. The British Library security guards are presumably on the lookout for would-be terrorists, although the Library is not the most obvious of targets.

Readers, I expect, will find it easy to sympathise with my irritation at having someone rummage through my stuff, and slowing me down. We’ve all experienced this sort of situation – at public buildings or events, at airports. But now shift the perspective. Imagine not that somebody is searching you, but that you are doing the searching. Imagine that you are one of those guards. Whom would you choose to stop, and why?

Suppose the next two people entering the building were an octogenarian woman and a young male of Middle Eastern appearance. Would it make more sense for you to inspect the young man than the old woman?

Would that be rational? Would it be fair? Would it be racist?

*

The example of targeting specific groups of people for additional security checks is just one instance of profiling. Profiling – attempting to extrapolate information about a person based on a trait or characteristic they possess – is a deeply contentious topic. It is also ubiquitous. We could hardly make our way through the world without making generalisations. Profiling occurs at many levels. When you meet someone new, you make assumptions about them based on how they look or sound. Much of this takes place at a subconscious level. You are processing information, their accent, their clothes, their race, their smile. You use this information to make judgements about them. When an insurance company asks you questions, where you live, what you do, your age, your gender, they are using this information to assess risk.

In our Brave New World, in which decisions are increasingly subcontracted to artificial intelligence, profiling is only going to become more common. Algorithms run by companies will dictate what kind of advertisements you see, what loans you’ll be able to access, what utility tariffs you have to pay.

Sometimes you may benefit from profiling and other times you may lose out. If I put down ‘journalist’ on a car insurance form, I will be asked to pay a higher premium than if I enter ‘academic’. Journalists appear to be more prone to accidents than academics. The box I tick won’t affect the safety of my driving or the likelihood of my being in a collision. But it does impact how I’m profiled. Profiling relies on there being groups, or categories of people.

Let me talk a little about the logic of groups.

The Logic of Groups

In the loose way in which I use the term, each of us can be categorised into an infinite number of groups. Male or female? American or British or Nigerian or Nepalese? Christian or Buddhist or Jewish or Muslim? They need not be exclusive: I can be American and British.

The school soccer team is a sort of group. You may be a member of the group of people above six foot, or the group of people below six foot. You may be a conservative, a liberal or a socialist. These are all types of groups. In the standard case, members of a group share at least one characteristic in common – e.g. males have a Y-chromosome.

Normally we talk about particular groups only if there’s a purpose in doing so. It would be weird to categorise together people above six foot unless there was some reason why this height was pertinent. One could imagine circumstances in which this might make sense. Perhaps a particular disease was prevalent among the tall – or perhaps a clothing company categorised people by height after deciding to specialise in catering to the needs of the tall.

Most groups will have fuzzy borders. It may not be entirely clear whether a person does belong into a particular category. For example, most of us are a tiny bit taller in the morning than the evening – consequently, some people will be above six foot only at certain times of the day. There are some categories, such as sex, which were once regarded as essentially binary: either you were male, or you were female. These days, there’s an acknowledgement that it’s not as straight-forward as that. Some people, for example, have both male and female physical characteristics. A few people have the Y chromosome but female genitalia.

There are a few more points to make about groups. First, the individuals within some groups will have voluntarily chosen to belong (the soccer club) but in other groups there has been no choice (a person born biologically male did not choose his sex). Second, in my expansive understanding of ‘groups’, it is not necessary to know that you are a member of a group to be a member of it. If you have never measured yourself, you may not be certain whether you belong to the group of humans who are taller than six foot. Third, groups differ in their structure. Some groups are constituted in a formal sense – a soccer club might have written rules for membership – whilst others don’t have anything like this (there is no written constitution governing the male group).

Finally, and importantly, groups can be more or less broad and can be subdivided. For example, as well as the group of men, there are subsections of this group, such as the group of men aged between 25 and 40. This narrowing process can continue indefinitely – for example, the group of men aged between 25 and 40, who live in London, are married, and in a job earning less than £35,000 a year etc.

As I have mentioned, at some level generality and categorisation is inevitable: to identify someone as a human is already to categorise them. But why is it useful to categorise humans into sub-groups? Well, one obvious reason is that we can use membership of groups to make predictions. That is to say, if I know that you are a member of a particular group, I can use this information to assess the likelihood that you will have another characteristic or that you will behave in various ways. That could save me a huge amount of time and a huge investment of resources. If a medical test for a disease is expensive, it would be costly to give it to everyone, when instead it could be targeted at those who are most prone to have the disease.

In general, knowing about your membership of some groups will be more useful than knowing about your membership of others. Knowing that you belong to the group of people whose first name has only one syllable conveys little of use to me. Knowing your sex is much more useful. Sex is correlated with all sorts of other things. On average, women live longer than men. Women are safer drivers than men. Women still earn less than men.

Would it ever be wrong to make use of these statistical links?

The Profiler

We should begin on a cautionary note. Often people believe that there is a correlation between a proxy such as race or sex and something else, when there really isn’t. There might be a deeply ingrained belief that individuals in a particular group are less friendly or more violent or more corrupt than others. And this might be completely untrue. There might be no such statistical relationship. Sometimes there might be an innocent reason for why people believe something that is untrue. But the explanation for the belief in an unfounded correlation can usually be summed up in one word – bigotry. There is a caricature that Scots are mean. As far as I know, there is no evidence at all to back that claim up. The belief that Scots are mean is pure prejudice.¹ Many of our assumptions about individuals based on their membership of groups will be of this form.

But suppose there is a statistical relationship? What then? Even so, there are reasons for the would-be profiler to be wary. Why? Well, the relationship might be very weak indeed. It might be the case that there’s a link between 49.9% of women and some other characteristic, X, and 50.1% of men and the same characteristic. For most purposes, that’s hardly a wide enough statistical gap to justify treating men and women differently.

What’s more, even if there were a more statistically significant correlation, we would still need to worry about false negatives and false positives. If we were using a proxy like religion to determine whom to search in a security operation, then a false positive would involve stopping someone who was innocent, and a false negative would be failing to stop somebody of a different religion who posed a danger. Suppose that almost all terrorists in the past have emerged from one religious group X. Still, these terrorists would represent only a miniscule percentage of total X. So if all members of X are searched by the security forces, the vast majority of people targeted in this way would be innocent. And if a few members of other religious groups were would-be terrorists, ignoring this possibility could prove extremely costly.

Numbers, and statistical relationships, are not fixed forever. Correlations are not carved in stone. Society evolves. So another problem with profiling is that the very practice of profiling might itself alter the statistics about the world and in ways we should regret. There are at least two routes this might take.

First, human beings are adaptable. If a terrorist organisation is determined to blow up a plane, but realises that everyone with a certain profile – a particular age, race, sex, nationality – is closely questioned at airports, they will shift their tactics and try to recruit bomb carriers from another demographic group. Using profiling to combat terrorists would then be self-defeating. Because institutions tend to be sluggish about reforming their methods and policies, the statistics on which they base their policies may be out of date.

Second, and conversely, profiling might entrench a statistical difference between two groups and so cement divisions in society. Imagine that one racial group was disproportionately responsible for street crime. If, consequently, the police chose to target members of this racial group, there is a strong possibility that individuals would feel harassed and oppressed. They would resent being suspected. They might feel under a constant threat of being stopped. They might become hostile to the police, unwilling to cooperate in the battle against crime and increasingly alienated from the wider community. Narrowing in on one racial group might contribute to driving members of that group into crime, which in turn might encourage profiling – a vicious circle.

The Profiled

All of these are reasons why, from the point of view of the profiler, we should be cautious about profiling. What about the perspective of those being profiled?

The benefits and the costs of profiling are not distributed among the population evenly. This generates a further asymmetry. Those who benefit from profiling are likely to take their privileged status for granted. If you belong to, let’s say, the wealthy majority community, you probably won’t even notice how favourable assumptions are constantly being made about you.

By contrast, those who bear the brunt of profiling are likely to be acutely conscious of it. Their justified belief that they have been searched, or might be searched, because of, say, their ethnicity, will make that search a humiliating and alienating experience. Profiling often works to the disadvantage of already disadvantaged groups. This is not always the case – medical profiling, allowing doctors to identify our medical risks, may benefit everyone and benefit those most at risk most of all. But it is because profiling often harms the disadvantaged that it is usually divisive and can be destabilising for society as a whole.

The Future

The practice of profiling appears to be riddled with drawbacks and risks. Is there any way in which it can be salvaged?

Let us begin by noting that not all forms of profiling cause the same damage.

One reason why those stopped by the security forces because of – they suspect – their ethnicity, feel so hurt and resentful is because ethnicity is powerfully linked to identity. So is religion, gender, nationality, perhaps even class. By identity, I mean the sense of attachment, or belonging, or loyalty one feels to particular groups. Not all members of, say, an ethnic minority, will feel a strong sense of identity with it, but many will. Often, people will marry within their ethnic group and their values, their social and their work lives will be fundamentally shaped by their background, as perhaps will their choice of neighbourhood in which to live.

For this reason, knowing somebody’s ethnicity can be usefully predictive. If you discover someone was born on a Monday, that tells you almost nothing useful. People born on Mondays share nothing beyond this trivial fact in common. By contrast, ethnicity will be correlated with many important aspects of life – including income, educational attainment, health, the likelihood of being a victim of a crime, the likelihood of living in a particular part of town, and so on.

The groups to which people feel the strongest sense of attachment or belonging are usually the groups that are also most predictive about their lives. This is not always the case. For example, Manchester United supporters might feel very strongly about their group, yet their support of Manchester United might not be very predictive. However, on the whole, identity and predictability are linked. People born on a Monday don’t feel a sense of affinity to others born on a Monday. And it is no coincidence that knowing that a person is born on a Monday is a useless piece of information for a policy maker or insurance broker.

Now, there may be some groups to which people belong without having a strong sense of identity which, nonetheless, are usefully predictive about their lives. Take height. We know that tall men on average earn a bit more than short men. Suppose, as a short man, you discovered that a bank was using your shortness as a factor in calculating the probability of your being able to repay a mortgage. You might be upset. But, I contend, not as upset as you would be if you thought they were profiling you by race or ethnicity. Why? Because, for most of us (even us short men), height is not an identity marker. We do not feel a strong bond with people of the same height.

And this is where the future of profiling holds opportunities and dangers.

The data revolution – the huge explosion of data tracking our lives – combined with the artificial intelligence revolution, means that profiling is going to become easier and more ubiquitous, and will increasingly be done by machine. The optimistic scenario is this. Sophisticated profiling algorithms, which do not simply rely on crude correlations involving emotive identity groups such as race, will be both more effective and may possibly arouse less resentment. Time will tell; but if our insurance comes to be based on 50 characteristics (rather than just a few blunt ones), we may take less umbrage when it occurs. And profiling will increasingly offer real benefits in...

Table of contents

• Cover
• Half Title
• Title
• Copyright
• Contents
• Notes on contributors
• Preface and acknowledgements
• Part One Race and gender
• Part Two The environment
• Part Three War and international relations
• Part Four Global poverty
• Part Five Ethics and social media
• Part Six Democracy
• Part Seven Rights and moral status
• Part Eight Science and technology
• Index