Sunday, October 10, 2021

Safe Third Country Agreements and the Availability of Other Rescuers

Former President Trump signed safe third country agreements with Guatemala, Honduras, and El Salvador. On February 6, 2021, President Biden suspended former President Trump’s agreements with these Central American states. Safe third country agreements are bilateral or multilateral agreements between or among states that allow the return of asylum seekers to the first safe third country that the asylum seekers traveled through to have their applications for asylum adjudicated there. Under international law, migrants may have a right to emigrate and a right to seek asylum, but they do not necessarily have a right to choose where they resettle. Safe third country agreements serve certain purposes to ensure the efficient administration of international and domestic asylum law, but they also raise some important ethical and legal issues. 

One question to consider is whether the availability of other “safe” countries discharges or diminishes a particular state’s moral responsibility to adjudicate a claim for asylum. 

Under the 1951 Convention and 1967 Protocol relating to the status of refugees, article 1, a refugee is a person who, “owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality” and is unable for good reason or unwilling to return. We can describe a refugee as a person in need of rescue. Under article 33, a participating state cannot “expel or return (‘refouler’) a refugee in any manner whatsoever to the frontiers of territories where his life or freedom would be threatened on account of his race, religion, nationality, membership of a particular social group or political opinion.” Once a person steps on the shores of a safe country and seeks asylum, that country cannot send him or her away without adjudicating the person’s claim. 

Consider the following hypothetical situation involving three states. State A borders state B and state B borders state C, but state A and C do not share any borders. State A is an outlaw state, using John Rawls’s term, and routinely engages in oppressive practices such as kidnapping and torturing political dissidents. Peter lives in state A and is a vocal opponent of the current government. Peter has reason to believe that officers of state A may knock on his doors at any moment and whisk him off for “questioning.” Under the cover of night, Peter and his family flee state A, secretly cross through state B by foot over the course of several weeks, and then step foot in state C seeking asylum. State B is a weak democratic state that is overrun by criminal elements. Peter believes that he and his family would not be safe in state B and, if caught there, may be returned to state A. There is a third safe country agreement between state B and C that allows state C to return migrants to state B. Under these facts, does state C have a moral responsibility to consider Peter’s claim of asylum?

This question brings to mind Peter Singer’s consideration of an objection to a duty to the global poor based on the availability of others. He says rhetorically, “Should I consider that I am less obliged to pull the drowning child out of the pond if on looking around I see other people, no further away than I am, who have also noticed the child but are doing nothing?” (Famine, Affluence, and Morality, Philosophy and Public Affairs 1(3) (1972): 229-243, 233). He also offers the following argument that an objector may make about giving aid. 

1. If everyone in circumstances like mine gave $10, there would be enough to provide food, shelter, and medical care for the refugees.

2. There is no reason why I should give more than anyone else in the same circumstances as I am.

3. Therefore, I have no obligation to give more than $10.

Premise 2 can be construed as an application of some principle of fair share: a person should be responsible only for doing his or her fair share, no more and no less. You can certainly do more but this would be supererogation and not a matter of duty. Singer points out that while the argument may initially look good, it lacks soundness. It would be sound if the conclusion was also stated as a conditional:

3*. Therefore, if everyone in circumstances like mine gave $10, then I have no obligation to give more than $10. 

Stated in this way, Singer further explains that the argument would have no bearing on the situation because it is not the case that everyone in circumstances like mine will give $10. 

Safe third country agreements can be construed as efforts to say, it’s not my responsibility because there are others who are also available and even better positioned to provide asylum. At this point, one may argue that safe third country agreements are not similar to the “availability of others” situation; it is more accurately a situation involving the “availability of better-positioned others.” 

Theoretically yes, but in practice no. In practice, many safe third country agreements put vulnerable migrants in even more vulnerable positions by leaving them in potentially unsafe situations, even possibly at risk of refoulement, and incentivizing other practices such creating a market for smuggling networks deeper into countries such as state B above. 

Third safe country agreements are designed to create a more efficient system of adjudicating claims of asylum and deter new asylum applications. But, in practice, they have little effect on deterring new asylum applications because of the challenges with enforcement and many of these supposed “safe” third countries lack the capacity to absorb new asylum claims and grant effective protection. Shirking responsibility by forcing countries with poorly equipped asylum systems to adjudicate thousands of asylum applications is the equivalent of providing no protection. Many of these applications will not see the light of day. 

So much for the “availability of better-positioned others.”

We are back at, it’s not my responsibility because of the availability of others. Applying Singer’s critique of such arguments here, we can say that the availability of others does not diminish a person’s responsibility to rescue those in need. If everyone in situations like state C’s provided asylum to a certain number of refugees, then state C would have no obligation to help more refugees. Stated conditionally, we can see that this statement has no bearing on the situation here because it is not the case that other states are opening up their borders and welcoming refugees, at least not nearly enough. According to the United Nations, there are 26.4 million refugees and, in 2020, 34,400 people were resettled in safe countries (this number may be low because of COVID, but, in previous years, the number of refugees resettled globally was usually about double this number). 

This is not to say that safe third country agreements never work or that, at some point, a host country that is admitting hundreds of thousands of refugees cannot complain, “wait a minute, I’m taking on more than my share.” But we are not anywhere near this point yet. This is also not to say that states cannot impose their own orderly system of admitting refugees, vetting them, etc. 

The main point I’m trying to make here is that the availability of others does not diminish one state’s moral responsibility to rescue refugees in dire need of a safe place to resettle. 

Chong Un Choe-Smith
Philosophy Department
Sacramento State


  1. I really enjoyed this Chong, thank you for this post.
    Sadly, I know little about the legal issues related to this topic, but this makes me think of the Bystander Effect, and how this ‘third party agreement’ sounds, in some ways, like putting that effect into policy. This made me think about how the Bystander Effect, therefore, is similar, and how it isn’t.
    So, as we know, with the Bystander Effect, when one sees another in a perilous situation, people may often not help because 1) they are afraid and 2) they think someone else will help. With the famous case of Kitty Genovese, as we know, there were people who actually witnessed an assault take place, didn’t even pick up the phone to call the police, as they were afraid to get involved and in turn maybe assaulted as well, and they thought someone else would call the police. The result, as we know, was the murder of this woman.
    Perhaps we do not have more ‘duty to help’ type of laws because the fact of a situation being perilous for the bystander seems to excuse moral responsibility for some, and of course in especially dangerous situations (such as diving into shark infested waters to save someone just attacked by a shark).
    But in the case of helping refugees seek asylum, states seem to be two steps removed, at least directly, from that perilous situation. Not only are the states not the one experiencing the danger, nor are the states directly seeing the danger – the state is just hearing of the danger not experiencing that particular refugee’s danger, and the state knows in a general sense how unsafe particular parts of the world can be. It is interesting to consider how the perilous situation of the refugee is not a direct threat to the state, yet the state may still decline to help and yet the factor of absolving moral responsibility because of the threat it poses to help, is gone. So, there are clearly other elements that are considered, such as in regard to economics, resources, etcetera. Is it, then, that in regard to those other elements that maybe we can lump together as ‘general resources of the state,’ the concern is running out—but isn’t this a slippery slope fallacy? Meaning, exhausting the general resources of the state is not an inevitable outcome.
    We send vaccines abroad because of the urgent situation we find ourselves in, because we are helping others, and we are helping ourselves because it is to everyone’s benefit. So, perhaps key here is showing how it is to our benefit of taking in refugees, not just for the obvious reason of helping our fellow human beings in dire need, but maybe some need to hear more reasons that benefit the self. For example, we know that being an active bystander can reduce bullying and crime in general. But maybe this is also naive optimism.
    In other words, I guess where I am going with this is that, as you conclude:
    “The main point I’m trying to make here is that the availability of others does not diminish one state’s moral responsibility to rescue refugees in dire need of a safe place to resettle.”
    And I would also like to keep talking about when moral responsibility increases. A state being uninvolved because it is left to the intermediary state, isn’t being uninvolved, since it is a choice not to make a choice in the matter and leave it to others. Being still a choice, there is moral responsibility.

  2. Marnie, thank you for your comment. Your reference to bystander effect brought to mind how in many ways we watch and do nothing or very little as refugees suffer around the world. Applying the bystander effect to states is interesting, but I don’t see why it cannot work for a collective body like a state. The bystander effect may help to explain why people fail to help those in need, but it doesn’t offer a moral justification for not helping.

    When we’re speaking of refugees, there is both a legal duty and moral duty. The legal duty is based on the Convention and Protocol on refugees and, in the US, the Refugee Act of 1980, which includes the same provisions. The problem is that each state is left to its own devices to interpret the provisions, including who qualifies as a refugee, and this is where a moral duty may apply so that states do not interpret and apply the law in ways that are merely self-interested or in pursuant to some inappropriate “political” agenda.

    Whether the state has a moral duty and to what extent will depend on what moral theory we’re applying, but even the nationalists agree that we have some duty to rescue foreigners in need (“burdened societies” using Rawls terminology again). Most moral theories, except egoism, do not require self-interested reasons for a duty of rescue.

    I think what you may be getting at is that, in a political context, democratic states require both external justification and internal justification for acting consistent with its democratic commitments. The external justification may be the duty to rescue burdened societies. The internal justification involves acting consistently with the will of the people, as expressed through democratic processes. In a genuine democracy, the decision ultimately lies with the people and people do not always act morally (enter tribalism, prejudices, selfishness, apathy, excessive fear, etc.).