The well-known bioethicist Peter Singer mooted the concept of the “expanding circle” in 1981 with his book The Expanding Circle: Ethics, Evolution, and Moral Progress (the link also has an embedded Bloggingheads conversation between Singer and Robert Wright discussing the book and the idea). Although the term “expanding circle” hasn’t exactly become a household term (it doesn’t even have a Wikipedia article yet) the idea does seem to have gained a lot of currency and is expressed in different forms. Possibly, many people have rediscovered the idea without even being aware of Singer’s original formulation. In this blog post, I’m going to address the idea as it appears in many common discussions, and not necessarily Singer’s original formulation. (To be honest, I haven’t read Singer’s book, so I couldn’t critique Singer’s original formulation accurately anyway).

Book cover of The Expanding Circle: Ethics, Evolution, and Moral Progress, Princeton University Press

The “expanding circle” idea relates three things:

1. The changes in people’s circle of moral concern over time.
2. A notion of larger versus smaller circles of moral concern, i.e., a way of meaningfully saying that a certain circle of moral concern is larger than another.
3. The normative ethics of whether larger or smaller circles of moral concern are better.

The strong version of the “expanding circle” idea makes three claims:

• Relating (1) and (2) (empirical): People’s circle of moral concern has expanded over time, and this trend is likely to continue
• Relating (2) and (3) (normative): Expanded circles of moral concern are generally better.
• Relating (1) and (3) (normative): In general, people’s morality improves over time.

To some extent, if we believe in two of these three claims, the third automatically follows.

It’s tempting for optimistic open borders advocates to make a case for open borders based on the expanding circle. The obvious idea is that open borders would be a lot more appealing to people whose circle of moral concern includes foreigners at nearly the same strength as natives. Restrictionists typically argue their case using a combination of citizenism and territorialism, which correspond to a narrower circle of moral concern (put differently, it puts a substantially higher weight on the interests and rights of one subset of humanity compared to another). Optimistic open borders advocates may therefore make three related claims:

• Relating (1) and (2): Open borders advocates would like to argue that people’s circle of moral concern will grow to include non-citizens and those who live far away.
• Relating (2) and (3): Open borders advocates would like to argue that, generally speaking, an expanded circle of moral concern, that does not weigh the interests and rights of citizens/native residents too much higher than the interests and rights of non-citizens/non-residents, is objectively better.
• Relating (1) and (3): Open borders advocates would like to argue that the future will be more favorable to open borders than the present is.

The three claims are linked (and any two imply the third, to an extent). My co-blogger John Lee’s first blog post borrowed its title from Martin Luther King Jr.: “The arc of the moral universe is long, but it bends towards justice” and is a fine example of the expanding circle argument.

A pessimistic open borders advocate may still view (2) and (3) as related, but be pessimistic about whether the trajectory of moral progress will move the world closer to open borders. An empirically agnostic open borders advocates may also view (2) and (3) as related but be agnostic about what the future has in store.

More on the circles and their applicability to open borders

A distinction can be made between two kinds of circles of moral concern: people (or more generally, sentient beings) to whom you feel an obligation of non-aggression (i.e., beings that are perceived to have moral rights) and people to whom you feel positive obligations to directly help. This distinction is very important for libertarians, who generally believe that most obligations to strangers are in the form of an obligation of non-aggression rather than an affirmative action. Thus, libertarians generally view the killing versus letting die (act/omission distinction) as morally significant. As Robin Hanson has pointed out, libertarians may have narrower circles of people to whom they feel positive obligations, but larger circles of people to whom they feel an obligation of non-aggression.

There is an important intermediate case: who are the people for whom you feel you have a positive obligation to expend resources to stop others from violating their rights? With the language of “killing versus letting die” this comes in the “letting be killed” category — a category that seems to be intermediate (I’d like to thank Andy Hallman for raising this question). On a certain view, immigration restrictions involve the use of aggression and coercion, but to argue that individuals are morally obligated to advocate for open borders would require making an argument that they view foreigners as part of the sphere of moral concern that deserves a positive expenditure of resources towards a non-violation of rights.

Another circle that might be of interest is the circle of people, or sentient beings, towards whom we feel a requirement to be honest and contractually fair. Generally speaking, it is considered okay to use deceit or trickery against animals, even if actually torturing those same animals was considered wrong. However, in general, the use of deceit or tricks against humans is frowned upon, with possible exceptions being made for infants, people who are violent and unstable, and people with some special and severe cognitive problems.

All these circles are relevant in different ways to different framings of the case for open borders. The libertarian case focuses on the circle of creatures to whom we are obliged to not aggress — a generally larger circle — and then argues that restrictions on the right to migrate constitute acts of coercion (through the threat of aggression and violence) and are hence immoral. Arguments of a utilitarian or egalitarian nature, on the other hand, require more in terms of one’s circle of moral concern — but at the same time fit better with conventional frameworks that conceded the state’s authority to restrict migration.

Is the expanding circle idea true?

My own take on the matter is both normatively and empirically agnostic.

I am normatively agnostic (i.e., I am agnostic about the relation of (2) and (3)): I don’t think that expanding circles are always better. I think that expanding circles might collide with problems of information — my local knowledge of the people near me is better than that of people far away, so I can do better by concentrating on caring more for their welfare. There are also issues of scaling — the complexity of dealing with a very large circle of people to whom one has positive obligations is tremendous. On the other hand, I do believe that the circle of people, or sentient beings, towards whom one can adopt a stance of non-aggression and non-coercion, and whose natural rights one can acknowledge, can be expanded quite a bit. It’s scalable — it doesn’t require much additional effort to practice non-aggression against a billion people compared to a million people. And non-aggression doesn’t require local knowledge. Incidentally, although I believe that to some extent non-aggression also applies to non-human animals, this does not make a case for “open borders for animals” — I discuss this towards the end of a post on moral egalitarianism and blank slatism.

I am also empirically agnostic (i.e., agnostic about the relation of (1) and (2), and also of (1) and (3)). On the relation between (1) and (3), I do think there is a general tendency towards moral progress, for the reasons Peter Singer outlines: people reason about morality, and they learn through experience with different systems what works and what does not. However, the secular trend is also accompanied by noise — in some cases heavily overwhelmed by noise — and specific predictions are very difficult to make. I also think that claims about an expanding circle can be met with strong (though not completely convincing) refutations. For instance, Gwern has written a lengthy article titled The Narrowing Circle where he argues that in many respects, humans’ circle of moral concern has narrowed, and a more accurate characterization would be a circle that is constantly shifting shape. Cyclical trends in attitudes to abortion are one example. Considerably reduced respect for the wishes of the dead is an example of a narrowing circle offered by Gwern.

My take on the applicability of the expanding circle idea to open borders

Normatively, it is not clear to me that the expanding circle idea offers a compelling framework to make the case for open borders.

I do think it is useful to draw specific analogies between ways that we consider morality to have improved in the past, i.e., moral practices that our ancestors took for granted as part of day-to-day life, but which people today generally consider immoral — and immigration restrictions. The immigration restrictions as Jim Crow analogy is (arguably) one such illustration, in so far as people are convinced of the parallels. This parallel could be closely examined and critiqued, and to the extent that it stands up to scrutiny, it could make the case for open borders.

However, I do not think that a crude argument that an expanded circle of moral concern would imply more support for open borders is in and of itself powerful, because it does not give one enough specifics to really get a sense of the underlying moral intuitions.

Moving the normative question to the background, one can ask the empirical question of whether people’s moral circle will expand over time in terms of valuing foreigners, and whether such an expansion will lead to more open borders. Even if we agree that there is a general trend for the circle of moral concern to expand, it is not at all clear which directions future expansion will be in. There are too many contenders, often pulling in very different directions. I do expect a slight move to open borders, but I doubt that it will become the next cause du jour — even though I hope it will (another open borders advocate, Bryan Caplan, offers offers a better/same/worse breakdown of 45/35/20, and my co-blogger Nathan puts a probability of about 20% on a serious move towards open borders over the next fifty years).

Here are some of the other contenders for where the next big expansion of moral concern will happen:

• Animal rights, veganism, and an end to cruel practices in factory farming, due to the expansion of the sphere of moral concern to non-human animals.
• Abortion being viewed as a much worse crime than it is today, due to the expansion of the sphere of moral concern to fetuses.
• Transhumanism of various sorts, which expands our moral concern to intelligences beyond conscious human intelligence.
• Incarceration being viewed as a much worse crime than it is today, particularly for victimless crimes, but even for other crimes, as torts substitute for jail time.
• A worldwide welfare state as political support is achieved for taxing the global rich and redistributing wealth to the global poor through an expansion of moral concern to people in other countries.
• A substantial expansion of the circle of moral concern to include future generations, leading to much lower discount rates in public policy and personal decisions.

The list is partially culled from a discussion initiated by Kwame Appiah and with participation from Ross Douthat, Will Wikinson, and Tyler Cowen). Will Wilkinson, who is an enthusiastic supporter of open borders, has an insightful take on the matter:

Yet I fear that when predicting the future of contrition we will tend to mount our personal hobby-horses and congratulate ourselves for getting on the right side of history before the right side of history was cool. In this spirit, I would like to congratulate myself for recognising that the global system of nation-states, borders, visas, and their attendant limits on the human rights to free movement and association amounts to a worldwide system of apartheid and is responsible for tremendous avoidable suffering. Though I feel quite sure that this is indeed an unconscionable injustice and a source of immense harm, I am far from certain that history will come to see things my way. My suspicion is that most of us would be quite surprised by the things our grandchildren will condemn us for, and that the more our predictions amount to praise for our current, farseeing moral enlightenment, the more sceptical we ought to be.