Joseph Carens has written a brilliant and stimulating book. I can’t remember the last time I filled a book with so many marginal jottings, either because he had made a striking point that I wanted a reminder of, or because what he said was so thought-provoking, or, often, both.
I agree with the vast majority of Professor Carens’s conclusions. It would make a boring symposium contribution to just list points of agreement, so I’m going to spend a bit of time here on a few points where I don’t agree. Now I’m sure you’ve heard a philosopher give an introduction like that once or twice before, and it can sound rather trite. So I want to start with a couple more positive things.
The fact that the book is so rich, that there are things worth talking about on basically every page, means that it would be a joy to teach. I don’t think there are many philosophy departments around that currently have on the curriculum a course on the ethics of immigration. Here’s some free advice to my fellow philosophers: Add such a course, and have Professor Carens’s book be a central text in it. You’ll get a topic, and a text, that are interesting to people who normally wouldn’t take philosophy classes. You’ll get more topics for fruitful discussion than you can easily handle. And, especially in a university with any kind of diversity, you’ll get the chance for you, and the students, to learn from how the lived experiences of the different members of the class interact with the theoretical issues at hand. I know many universities have been adding, with great success, courses on the ethics of food. A course on the ethics of immigration could have a similar kind of success.
Carens’s book encourages us, among other things, to take two new perspectives on issues about immigration. One is that of the migrants themselves, who have played a surprisingly small role in much academic discussion. A lot of the focus has been on the states constructing the borders, and the legitimacy of their doing so. Carens’s thought experiments often cast the reader as the central protagonist, in many cases the would-be migrant. (E.g., “Imagine you want to move from New York to Los Angeles.”) I think the use of second person examples in philosophy is excessive, and often dangerous. But I think it works here, because part of the point is to see things through the would-be migrant’s eyes. Someone who wants to move, say, across Lake Erie from Ontario to Pennsylvania faces a system of legal impediments, backstopped by the coercive force of the greatest power the world has seen. And that system is not obliged, either legally or according to folk morality, to provide much justification for directing that force at them, should it decide that the impediments will become insurmountable. Indeed,the fact that the person has not provided a positive reason to move that the system finds acceptable is taken to be a sufficient reason to use force against you. This is not what we usually find acceptable in a non-oppressive regime. The state has to justify why it should use force against me; I don’t have to justify why it shouldn’t use force against me. Maybe when I’m an immigrant all of that is right and good. But even so, I think it is right for us to be occasionally shocked by it, and be reminded that it is a state of affairs that stands in need of a very strong justification.
The second change of perspective involves thinking simultaneously about movement within states and between states. Folk morality distinguishes sharply between the two. Compare how you would react if you found out the following facts about three states, A, B and C, that you had previously never heard of.
- A has restrictions on travelling between its two largest cities. To travel between the two requires an internal transit visa.
- B and C sharply restrict movement between each other. To move between one and the other requires official approval well in advance from the country one is headed to. This approval is both costly to obtain, and often denied for what seem like trivial administrative reasons.
I think that most people, especially if the two cases are not presented side-by-side, would think that the facts provide more evidence that A is repressive than that B or C are. Internal movement visas are things that we associate with repressive regimes. But the facts about B and C wouldn’t excite anything like the same reaction. Indeed, they are consistent with B and C satisfying most of the standards that folk morality requires for being a free country.
I haven’t done a careful study to check whether what I am saying about folk morality is true. (Here is a place where careful experimental philosophy may be useful!) But I suspect that it is; that most people living in liberal democracies find the idea of internal movement visas repulsive, but think it is perfectly acceptable to have the kind of sharp restrictions on moving into a country that we see in B and C. And it is worth thinking hard about what we can learn both from the fact that liberal democracies allow free internal movement and that we largely think it is a requirement of freedom that they do so. (Though how much we really think this will be an issue we’ll return to presently.)
The fact that there is free movement within liberal democracies undermines some of the more outlandish predictions of the opponents of open borders. Carens quotes Michael Walzer saying that we need closed borders to preserve “communities of character”. But Vermont has open borders (on three sides at least) and is (or at least contains) as good a community of character as one might aim for in designing policy. By the same token, reflection on currently existing polities without borders should tamper our enthusiasm for the benefits that open borders would bring. There aren’t any closed borders, in the sense Carens is interested in, between Trenton, NJ and Princeton, NJ. There are even trains that run between the two dozens of times a day. And yet the wealth of one doesn’t seem to have done much to benefit the residents of the other. This hasn’t happened by chance, and it is worth worrying that the mechanisms involved would be seen more in a world of open borders.
It is tempting to think of borders between states as much more important than borders within states because of the greater inequalities between rather than within states. Indeed, some of Carens’s qualifications to his conclusions seem to acknowledge this temptation. For instance, he writes: “Our deepest moral principles require a commitment to open borders (with modest qualifications) in a world where inequality between states is much reduced.” (288, emphasis added). But I’m not sure that is a relevant concern given the world we actually live in. Already, inequality within states is almost as significant as inequality between states.1 And if Piketty is right, the trend is that inequality within states is rising, perhaps substantially, while all the evidence suggests that inequality between states is falling.
And this brings us back to the worry that folk morality, distinguishing sharply as it does between inter- and intra-state migration, is incoherent. Something like this plays a key role in one of Carens’s arguments for open borders. I think he’s onto something, but the argument doesn’t prove as much as he takes it to. For a first pass at the argument, return to my story about A, B and C above. And assume now that A is a newly formed loose federation of two states, B and C, each of which retains substantial sovereignty. Does this change of status constitute a new form of repression, or not? Folk morality, taken literally, would seem to suggest the new state is more repressive than its predecessors, but this is hardly plausible. More generally, advocates of free movement within states, and strong borders around states, have tricky challenges in determining what, in the relevant sense, is a state.2
Here’s how Carens tries to leverage this tension in orthodox theory into an argument for open borders.
- “Freedom of movement within a state is widely recognised as a human right.” (238)
- There is a close analogy between freedom of movement within a state, and freedom of movement between states.
- So, freedom of movement between states should be recognised as a human right.
I’m not entirely sure I’ve got the best quote to represent Premise 1 in the intended argument. That’s because I’m not sure whether the analogy is meant to start with a claim about folk morality, or a claim about our current practice, or a claim about institutional commitment, or something else. This unclarity isn’t really a weakness of the argument, though, because it suggests there are a number of distinct arguments by analogy one could pursue in defence of one or other kind of open borders conclusion. And I think Premise 2, properly understood, is true.
My worry is that premise 1 is not true. I’m going to offer a defence of what Carens calls the view that free movement within a state is a ‘membership-specific’ right, rather than a human right. At least, according to folk morality, current practice, and the principles embodied in our existing institutions, it is a membership-specific right. That is, the key intuition is that members of a state should be free to move freely within it, up to resettling in different parts of the state, but that range of freedoms need not extend to non-members. When I say that is the intuition, I don’t mean I agree with it, since I’m an open borders believer. But I do fear it is the intuition most people in the debate have. I fear this because of though experiments like these two. (Both are fictional in their details, though Example Two is I think close to something Australia did recently.)
The US government, in an attempt to boost tourism to Hawai’i, drops the requirement of having an ESTA for people only coming to Hawai’i. So Shirou, wanting a holiday from his home outside Tokyo, gets on a plane with no more documentation than his passport, and flies to Honolulu. While he’s there, he meets someone who tells him some great things about San Francisco. He would like to go, but he needs ESTA authorisation to go. And he can’t apply for that in Hawai’i, nor can he afford to fly back to Tokyo, then on again to San Francisco. So he simply returns home at the end of his holiday.
The Australian government is desperate to get more doctors into rural areas. So it sets up a class of visa for doctors who will work in a rural area for three years. Arya applies for such a visa, is accepted, and moves from Delhi to Broken Hill. While she is there, she is free to visit Melbourne for a weekend, but she isn’t free to move to Melbourne, since there is no practical way she can work in Broken Hill and live in Melbourne. And there is no prospect of her getting a visa that would let her live in Melbourne before the three years are up.
Now if premise 1 of Carens’s argument is right, each of these examples should strike us as grave injustices, either violations of Shirou and Arya’s fundamental human rights, or violations of a good international law. But I don’t think they strike us that way. In both cases they might be instances of bad public policy. Example one may even be a violation of the U.S. Constitution. But I don’t think they strike us as things that are, or should be, violations of human rights treaties.
And that’s a problem because it suggests that fundamentally, we are committed to free internal movement as a membership-specific right, not a general human right. Shirou isn’t a member of the United States, so he isn’t entitled to move freely around it. And Arya doesn’t become a member of Australia until her three years have passed, so she can be restricted in where in Australia she lives. And if that’s right, the argument by analogy doesn’t go through.
It might be objected that Arya, at least, is not really restricted in her movements. She can head to Melbourne for a decent cup of coffee any time she wants. But she can’t move there should she need good coffee every morning, not just on occasional weekends. If freedom of internal movement means the kind of freedom Arya enjoys in her first three years in Broken Hill, the analogy Carens uses doesn’t support open borders. It suggests, at most, that everyone in the world should be free to visit Melbourne on holiday, but the government is allowed to put tight restrictions on which non-Australians can live there. And that isn’t the conclusion open borders advocates should want.
So I don’t think the argument by analogy is dialectically effective. It isn’t one that should make closed borders advocates think that they are tacitly committed to open borders in virtue of their prior commitment to free movement internally.
If that’s right, why do I believe in open borders? I don’t have a well worked out theory here, but here’s the kind of approach I favour. (And I should note that my thinking here has been influenced by several of the other things Carens says.) I think there is a very small class of cases in which a state is entitled to use force against an individual. And in every one of those cases, the state is obliged to explain its use of force in a way that could at least make sense to the target of the force. I don’t think the goals that people hope to achieve by closed borders are sufficient to fall into one of those cases. (That’s in no small part because I think Carens’s objections in Chapter 12 to the attempts to articulate such goals are really compelling.) At the very least, the goals that would justify the use of force have not been clearly articulated. And hence they haven’t been clearly articulated to the targets of the force. So the use of force is unjust. But borders without the use of force to police them just are open borders. I’m not in principle opposed to positive arguments in favour of open borders, of the kind Carens attempts in Chapter 11. But I think there is a simple negative argument in favour of open borders, namely that closed borders require state-sanctioned violence, and state-sanctioned violence requires very strong justification, and that justification has not been provided.
I want to end with a quibble, or perhaps a pet peeve. I think Professor Carens often understates his conclusion. Here is how he sometimes puts it.
The ideal of justice i have identified has two components: a right to freedom of movement across borders and relatively little inequality between states. (278)
But as has been often pointed out, we really shouldn’t care much about the ideal. (Ralph Wedgwood makes this point in a recent blog post.) For one thing, ethicists set themselves too easy, too useless a task if all they do is talk about what heaven will be like. My vision of heaven certainly doesn’t contain guards with guns policing the borders of its different parts, and I suspect yours doesn’t either, but little of practical importance follows from that. What really matters is what would make things better, and what we should aim at doing. But properties of the ideal are not a good guide to this. It certainly doesn’t follow logically from the fact that something is true in the ideal that we should aim to make it true, or be happy that it is actually true. The ideal test-taker makes an even number of mistakes, but little follows from that for the non-ideal. In some circumstances one might think that knowing something is true in the ideal is at least inductive evidence we should aim for it. But that needs to be shown, and it often won’t be true.
This is a quibble because I think Carens’s arguments aren’t really just about the ideal. They are arguments, often strong arguments, that removing barriers to free movement of humans is a good thing, and something we should be aiming for. In the ideal there would be few, or perhaps no, such barriers. But even in the here and now it would be better if there were fewer. That’s the big claim open borders advocates believe, and one we should make clearly.
Here’s a quick bit of data on this. Take the PPP adjusted GDP per capita of the world’s 187 countries, according to the World Bank. Calculate the Gini coefficient for a society that had 187 individuals in it, each assigned an income equal to this GDP per capita of one country. That turns out to be a little under 0.53. Now compare this to some real-world Gini coefficients. Most western democracies have a Gini coefficient that is higher than 0.53 before taxes and transfers, though a fair bit lower afterwards. Taxes and transfers matter, so there is a difference here between intra- and inter- state inequality, but it is less than you might have thought. ↩
A personal anecdote to bring out the real-world complications. I recently caught a ferry from Macau to Hong Kong. At the end I spent an hour standing in an immigration queue to come into Hong Kong. And the paperwork I used wouldn’t have been sufficient to enter mainland China. Was this an example of unjust restrictions on internal movement, or legitimate policing of state borders? ↩