Misinformation as Immigration Control
It is wrong to force refugees to return to the countries they fled from. It is similarly wrong, many argue, to force migrants back to countries with life-threatening conditions. I argue that it is additionally wrong to help such refugees and migrants voluntarily return whilst failing to inform them of the risks. Drawing on existing data, and original data from East Africa, I describe distinct types of cases where such a wrong arises. In ‘Misinformation Cases’ officials tell refugees that it is safe to return, when it is not, and refugees return who would have otherwise stayed. In ‘Omission Cases’ officials do not provide any information on countries of origin, and this omission causes refugees to repatriate. In ‘Relevancy Cases’ refugees are misinformed or uninformed, but would have returned even if better informed. In all of these cases, at least some state officials are blameworthy for their failure to inform refugees, and are engaging in a form of wrongful immigration control.
KeywordsImmigration Refugees Informed consent Deportation Repatriation
Every year, hundreds of thousands of refugees repatriate to their countries of origin (UNHCR 2012). Many return with little information on these countries, having lived abroad most of their lives. They instead rely on information from government officials, UN agencies, and non-governmental organizations (NGOs). At times, these officials, agencies, and organizations provide inaccurate information on the risks of repatriating.
Consider the following case. Between 2007 and 2009, roughly 1200 South Sudanese refugees in Cairo crossed the Sinai desert and into Israeli territory, where they were given temporary asylum (Gerver forthcoming). Soon after, a worker for an NGO travelled from Israel to South Sudan on a fact-finding mission, meeting with ministers in the South Sudanese parliament, and taking photographs of clinics, markets, and solid buildings throughout the country. She then flew back to Israel and showed these photographs to refugees in a community centre, informing them that her NGO, Operation Blessing International (OBI), was working with the Israeli government to charter flights to South Sudan. She and government officials explained that Juba, secondary towns, and rural villages had housing, security, schools, universal healthcare, and income-generating opportunities.1
Hundreds of refugees returned to South Sudan, but found none of these amenities. Few had reliable shelter, medical care, regular meals, or clean water. Most drank from contaminated rural wells, and were dependent on strangers or the unreliable charity of distant relatives. Some started small businesses, but they mostly failed. An unknown number died or were killed in ethnic-based violence, and the majority were displaced within two years (Gerver forthcoming).
Current discussions on immigration ethics rarely discuss such cases, focusing instead on forced returns. It is wrong, many note, to deport an individual to a country where they will be persecuted (Miller 2005; Gibney 2004; Betts 2010) or lack necessities to survive (Betts 2010; Gibney 2004), and wrong to compel such individuals to return by denying them basic necessities (1951 Convention; Chimni 2004; Barnett 2001; Bradley 2013; Long 2013; Fouda 2007). But many refugees, asylum seekers, and survival migrants are not forced to return in these senses. They return due to misinformation, as seen in repatriations to Uganda, Iraq, and Afghanistan over the last two decades (Strand 2011; Carr 2014; Walsh et al. 1999; Van England-Nouri 2008). Sometimes, the misinformation is about the right to stay: immigration officials tell refugees they cannot legally stay when they can (Gerver 2014). Other times, government officials and NGOs say conditions are safe in a country of origin, when they are not (Strand 2011; Carr 2014). With over half a million refugees repatriating annually, it is unclear who must provide accurate information, if anyone.
We could claim that lying is wrong, just as it is wrong to physically force refugees to return. But let us assume that officials and NGOs are not lying, because they themselves are misinformed. Unlike physically forcing an individual onto a flight, one can give misinformation without being aware that one is doing so. And unlike deportations, refugees and migrants can avoid the consequences of misinformation if they find information themselves. The question is not merely if individuals were misinformed, but whether agents helping them return have a duty to inform them.
This article defends the following claim: if it is wrong to deport an individual, due to risks she will face, then governments and organizations have moral duties to inform this individual of the risks of voluntarily repatriating. I shall defend this claim, and argue that it holds in four distinct types of cases.
In the following section I will describe what I call ‘Misinformation Cases’. When officials or NGOs provide misinformation, they are culpable for the resultant misinformed returns. I will then describe ‘Omission Cases’. When officials or NGOs merely omit information, they are similarly culpable, though their actions may be less wrong than in Misinformation Cases. In ‘Relevancy Cases’ information seems irrelevant to refugees’ decisions to return, because refugees would have returned even if better informed. In some such cases NGOs and officials still act wrongly when failing to inform. Finally, there are ‘Blameworthiness Cases’, which cut across the above three cases. Officials and NGO staff members may be unaware that they are misinforming, and so lack the intent, motives, or foreseeability to know the consequences of their actions. In some such cases they are still culpable for their actions.
Before presenting the above four claims, it is necessary to clarify certain terms and assumptions.
I shall call agents helping with return ‘repatriation facilitators’. They include NGOs working with the government, as well as civil servants working with immigration authorities. They pay for travel back to countries of origin, at times arranging travel documentation, sometimes offering a modest stipend during the initial months of return.
My focus is on the moral culpability of such facilitators. I assume that culpability requires three conditions, broadly conceived. First, it must be true that agents failed to fulfil various duties, which I shall specify in the next section. Second, there must be what Miller calls ‘outcome responsibility’ (Miller 2007, p. 96), where the failure to fulfil duties leads to an objectionable outcome, such as a reduction in security or autonomy.2 Third, agents must have wrongful intent or motives or, at least, the ability to foresee the negative consequences of their actions. The next three sections focus on establishing whether the first two conditions have been met: whether facilitators failed to fulfil duties associated with information disclosure, and whether this led to uninformed returns or a reduction in autonomy. Only the last section considers the possibility that there was no blameworthiness, and so no culpability.
Throughout this article I shall focus on normative, rather than legal culpability. In particular, I will demonstrate that the duties I describe are consistent with three competing normative theories: deontology, consequentialism, and virtue ethics. Deontological theories hold that we often do not have duties to assist others if the costs are considerably high (Barry and Overland 2012). Whether we do depends on criteria other than the consequences of our failure to assist. Consequentialist theories, in contrast, hold that we can have very costly obligations to help if the benefits are substantial.3 Finally, some virtue ethicists hold that individuals ought to be judged by their motivations, rather than only their actions or the consequences of their actions (Slote 2001). All three theories, I argue, ought to support the notion that misinforming refugees, either intentionally or not, is a form of wrongful immigration control.
Though I focus on misinformation, deportation is still relevant: I wish to explore whether, if deporting an individual is wrong, the government and NGOs must inform this individual of the risks of returning. There is, of course, great disagreement as to who states should not deport, and so by extension there may be disagreement as to who states must inform. For simplicity, I will largely assume that if a country has the capacity to accept individuals, it is wrong to deport them to any life-threatening conditions. I call all such individuals ‘refugees'. Though I make this assumption, one need not accept this assumption to accept my theory regarding misinformation: my theory can be plugged into a range of assumptions. If you believe an individual should only be protected from deportation if they will face persecution in their countries of origin—the common legal standard—my theory is only relevant for cases where individuals return to persecution because of misinformation. If you believe an individual should be protected from deportation if they will face any life-threatening conditions (Gibney 2004; Hidalgo 2015), my theory is relevant for all cases where individuals return to any life-threatening conditions because of misinformation. My focus is on the duties to find information when deportation is wrong, and not when deportation is wrong.
Throughout the article, I shall draw on several sources, such as government reports on returns to Iraq, Sierra Leone and Afghanistan. However, my primary source will be research on misinformation in Israel, drawing upon 126 interviews I conducted with former refugees in South Sudan, Uganda, and Ethiopia. This case is especially useful because the NGO facilitating return from Israel worked to avoid misinformation, travelling regularly to South Sudan. It still seemed to misinform refugees. As such, it was especially unclear whether it acted wrongly, given the efforts invested. Resolving this question requires us to consider not only whether misinformation occurred, but who has a duty to find information, given the costs involved.
Misinformation Cases occur when repatriation facilitators, due to unreliable sources, falsely tell refugees there is food and physical security in their countries of origin, and refugees return home who would have otherwise stayed. Such was the case in the 1990s when the German government relied on official Bosnian sources for information, telling refugees they would receive housing and employment assistance upon return, none of which materialised (Walsh et al. 1999). In 2005 UNHCR, based on existing reports, told Afghan refugees in Iran that it was safe to return, and refugees faced violent attacks on the border during repatriation (van England-Nouri 2008, pp. 157–159). Three years later, the International Organization of Migration (IOM) in Norway told refugees returning to Iraq that there were income-generating activities, but most faced severe food insecurity and homelessness after returning (Strand 2011). In 2011 the European Return Platform for Unaccompanied Minors (ERPUM) set up programmes for the repatriation of Afghan youth in Sweden, Norway, the Netherlands, and the United Kingdom. ERPUM case workers promised children access to safe living facilitates (Lemberg-Pedersen 2015, pp. 15–17), despite the widespread insecurity in Afghanistan.4
We might believe that facilitators in all of the above cases acted wrongly, as they ought to have invested more resources in finding accurate information. But perhaps facilitators must merely relay information they are told, rather than seek information themselves.
When the OBI director in Israel began organizing returns to South Sudan, she tried to avoid the mistakes of past repatriations. In addition to travelling to South Sudan, she hired an organization called the Hebrew Immigrant Aid Society (HIAS) to independently determine if refugees’ consent was informed. HIAS would interview each refugee and, if it felt that a refugee knew little about their rights in Israel, and little about South Sudan, it would tell this to OBI, who would then reject the refugee’s application for return assistance.
Unfortunately, HIAS appeared to know little about South Sudan, and so largely failed to determine if individuals were uninformed. The HIAS training manual for staff only has a short page on the history of the country, and very little information about the risks of returning. For example, the manual states, ‘Many applicants might not be aware of the entire situation in Sudan. Instead, they might only know about the circumstances in their village. This is OK.’ Villages often lacked basic services and employment, and so information on urban centres was essential.
There were different types of misinformation given before return. Refugees were told there was universal free healthcare in South Sudan, when this was not the case.5 After return, three children of subjects in my sample died from illnesses, and of the total who returned, twenty-two children died from illnesses within the first year. In total, sixty-eight of the 126 respondents recall being told that South Sudan was a safe country. Almost a third were told this by government officials, and nine were told this by OBI or a UN official. In reality, internal ethnic-based fighting was continuous as of 2012, when most returned (Ferrie 2012a, b; Al Jazeera 2011; MSF 2014; Small Arms Survey 2011). Of the 110 returnees whose conditions I could confirm by 2014, thirty-two adults and seven children had fled to UN IDP camps, thirty-seven had fled the country, and three adults and two children were killed.
Given the risks of return and widespread misinformation, what are the minimal obligations of repatriation facilitators?
There are a number of principles that can establish these duties. The Good Samaritan Principle holds that, if we can save a life at very little cost to ourselves, or at moderate costs to ourselves, we ought to do so (Barry and Overland 2012; Gibney 2004; Athanassoulis 2000, pp. 219–220). In all of the cases cited, facilitators could have easily conducted at least some research themselves, reading existing reports on malnutrition and healthcare (Green 2012; Brown 2006; Cometto et al. 2010), and warning refugees of the risks, preventing unsafe returns. Even OBI, who travelled to South Sudan, relied on information from ministers in the South Sudanese government, rather than independent sources.
Unfortunately, some information is not easily available, and is very costly to obtain. The World Bank and the International Labour Organization offered no employment statistics on South Sudan as of 2014 (World Bank 2014), and Médecins Sans Frontières cannot provide statistics on the location of health clinics in the country. There are also no recorded mortality rates from ethnic-based violence. Gathering information would often involve considerable costs on NGOs and government agencies.
We might claim that facilitators should gather information because our duties towards others extend beyond Good Samaritan duties. We ought to help others in great need not only when it involves low or moderate costs, but also when it involves high costs only we are able to bear. If an agent has greater access to information, then they ought to find this information and disclose it, if this will significantly help others make informed or safer decisions (Miller and Wertheimer 2009). NGOs and government agencies may have greater abilities to travel regularly to countries of origin, interviewing past returnees, and hiring local research assistants if necessary.
Though greater ability to find information may seem like a good reason to demand more from agents, this conclusion seems inconsistent with intuitions about cases outside the sphere of immigration. Imagine I am a car mechanic selling you a car. I could run a test that you are not able to run, ensuring that the breaks are functioning, and ensuring that you do not find yourself in a car accident. It would seem unfair to claim I have a duty to run this test, while another car owner, who is not a mechanic, would have no such duty when selling you their car. While greater ability may be a reason to disclose what one knows, it does not follow that greater ability to obtain knowledge creates a duty to obtain this knowledge.6
There is another consideration, other than ability. We often have duties to keep our promises. If an NGO promises to only help with informed returns, it has duties to find information, to ensure that return really is informed. This argument, though relevant for many NGOs, is less relevant for government officials, who rarely promise to ensure informed returns. In such cases, the value of promise-keeping seems irrelevant.
We might still suppose that governments should avoid misinforming for another reason: misinforming causes harms, by causing refugees to make decisions without full information, undermining their ability to make autonomous decisions and encouraging them to return to unsafe countries.
This argument is sufficient to claim that governments ought to find accurate information when finding information involves moderate costs. It is not clear that this establishes a duty to find information when this involves especially high costs. For, we do not wrong every person we cause harm to, if avoiding harm is very difficult. Imagine I purchase an airline ticket to Jonglei in South Sudan and ask the pilot if conditions are safe in the port of arrival. She says they are, because she has read this in the Sudanese Tribune. If the Tribune was mistaken, the pilot’s false information may have causally contributed to my harmful decision to purchase the ticket, but it is not clear she has wronged me, given that it would be costly for her to seek a more accurate source of information. To establish wrongdoing, it is not enough to establish that misinformation causes harm, but whether we have a duty to find information to prevent harm.
There is a final consideration which, I believe, creates duties for governments to find information, even when doing so is costly and they never promise to ensure informed returns.
We often have duties to know information that are derived from other duties, unrelated to informed consent. Drivers have duties to look in their rear-view mirror, to know if there is someone behind them. This is derived from their duties to avoid running others over (Smith 2014a, b). Similarly, drivers may have duties to bring their car to be inspected for faulty breaks in an annual inspection, to avoid running others over due to faulty breaks. Sometimes, the information we have a duty to know also happens to be information we must disclose. If I have my car inspected and find the breaks are faulty, and I want to sell you my car, I should tell you about the faulty breaks. I know about the breaks because I have a duty to, in order to ensure safe driving, and once I know this information, I have a duty to disclose it in a subsequent sale. It is not that I have a duty to know about the breaks in order to tell you; I have a duty to know and, by chance, this information is the sort I need to disclose because I know it. If I am negligent, and do not find out about the faulty breaks before selling you my car, is seems a poor excuse to say, ‘But I did not know about the breaks’. For, I had a duty to find out about the breaks to be a safe driver.
We may apply similar reasoning to return migration. States have various duties which create derivative duties to know. States have a ‘Responsibility to Protect’ others from great harm, as outlined in the 2005 UN World Summit (UN 2005), and obligations to avoid deporting individuals to countries where they might experience persecution or inhumane treatment (Osman v. United Kingdom, cited by Gammeltoft-Hansen 2011, p. 196). These duties create duties to know information about persecution and inhumane conditions. If states fail to find information that is necessary to fulfil these general duties, their ignorance may be a poor excuse for their failure to disclose risks to refugees voluntarily repatriating. Even if it is questionable whether states ought to invest costly resources in ensuring informed repatriation, it seems clearer that they ought to invest costly resources in preventing mass atrocities and establishing who is a refugee. And these duties create derivative duties to know about certain countries, creating duties to disclose information about these countries.
To clarify this point: I am not arguing that we must disclose all information we have a duty to know. If my neighbour wants to know about the faulty breaks in my car, despite expressing no interest in buying my car, I have not wronged the neighbour when I misinform her. For, were I to know about the breaks, I would have no duty to tell my neighbour about them, even if it would make her happier. Rather, the argument is that, when we have a duty to disclose information we know to ensure informed consent, it is not an excuse to say, ‘I did not know’ if we ought to have known, due to our other duties.
This reasoning would imply limits to the demands we place on facilitators. No facilitator should be required to risk their lives by travelling to an insecure area to find out if refugees’ lives will be at risk from return. For, they would have no such duty to risk their lives to fulfil their other duties as state officials. But if information is not extremely dangerous to obtain, officials should try to obtain it, if they have other duties that require this information.
While Misinformation Cases involve facilitators providing false information, Omission Cases involve facilitators providing no information at all, true or false. It is not clear if facilitators have acted unethically, given that there was no explicit misinformation.
Consider the case of Iraqi nationals who returned from Norway with the help of IOM, funded by the Norwegian government. A subsequent IOM report found that those who returned were never told of the risks of return, such as the likely inability to secure a reliable income (Strand 2011). Similarly, Iraqi refugees in Denmark recalled not being told information on the security situation in Iraq (Bak Riiskjaer and Nielsson 2008, p. 7) and refugees returning to Sierra Leone from the UK were never informed of the risks of homelessness, common after return (Carr 2014). Refugees returning to Sri Lanka were never warned of security risks, with many arrested, detained, and some tortured and killed by police after repatriation (Carr 2014). In Israel, OBI intended to disclose all risks, but information omissions were common. Refugees were never informed that they could not re-enter Israel once they left,7 as it was assumed they knew this. In general, HIAS assumed refugees had information from family members in South Sudan, or from their own memories, and so did not seek out and disclose information on health, food insecurity, or the death toll in Unity State, Jonglei, and other areas of concern.
Given that information may be lacking amongst some, do facilitators have a duty to provide information?
A central reason we might suppose facilitators have no duty to disclose information, or a lesser duty, is that there is a general distinction between doing and allowing, or ‘positive’ and ‘negative’ acts. An example of a positive act would be injecting arsenic into a victim, and a negative act merely letting a stranger die of arsenic poisoning, rather than calling an ambulance. It seems the former deems a person more culpable than the latter. In a similar manner, perhaps giving false information deems an agent more culpable than merely failing to inform.
There are three reasons to believe that such a positive/negative distinction is normatively relevant. The first reason is that positive acts tend to indicate stronger causal relationships to upshots compared to negative acts (Callahan 1989).8 If someone injects arsenic into a victim, then her actions are necessary and sufficient for the resulting death. If an individual merely fails to save a victim of poisoning committed by another, the failure to call an ambulance is not sufficient for the death: the perpetrator’s injecting of arsenic is also necessary. If a refugee has accurate information, and is then misinformed, she may suddenly believe a falsehood she did not before, and so the misinformation is necessary and sufficient for the outcome of an uninformed return. In contrast, if a refugee already holds false information from other inaccurate sources, and a facilitator never provides accurate information, the facilitator’s omitting information is not sufficient for the refugees’ false belief: the refugees’ other inaccurate sources are also necessary for the false belief. Because most normative theories view causing harm as one criteria (though not the only) for determining culpability,9 then most theories ought to generally hold positive acts as worse than negative acts.
It is not clear, however, that misinforming is causally related to upshots more than failing to provide information. One reason an individual believes misinformation is that she is not provided with alternative information to correct this misinformation. As such, the resultant false belief is the result of both the false information provided, in addition to information omissions from other sources. As such, the misinformation is not sufficient for the false beliefs. Just as information omissions only lead to false beliefs when there is misinformation from another source, misinformation only leads to a false belief when there are information omissions from another source. Omitting information, as such, can be causally responsible in a similar manner to misinforming. Moreover, in many cases omitting information can be interpreted as a communicative act, similar to misinformation. When refugees told HIAS that they were returning to South Sudan to access high-quality education, and HIAS responded with silence, refugees may have interpreted this silence as a confirmation that high-quality education was available, reinforcing their false beliefs.
Even if omitting information causally contributes to false beliefs as much as misinforming, there is a second reason omitting information may be less wrong than misinforming. Some facilitators can legitimately expect refugees to have information, given that refugees are returning to their own home countries, and so can legitimately fail to disclose information. This is the assumption we often hold in other relationships where informed consent is necessary (Manson and O’neill 2007, pp. 68–96). If a surgeon tells a patient about an incision, it is reasonable to expect patients to know that the incision will be made with a scalpel, and so omitting this information is acceptable.
The problem with this approach is that it is difficult to know what would be reasonable for refugees to know. Every refugee is returning to a different socioeconomic circumstance, village, and family network, with different language skills, resources, and expectations. All surgery is fairly similar in some ways, so our knowledge of it is as well, and our expectations of others’ knowledge is as well. The same cannot be said about refugees and their countries of origin.
There is a final way that positive acts may be distinct from negative acts. Positive acts are generally costless to avoid. It is costless to avoid misinforming, as this merely involves keeping one’s mouth closed. It is more costly to give information, as this involves finding information. When an act is costly, then many deontologists would conclude that refraining from the act is permissible. Virtue ethicists may similarly conclude that a virtuous individual can at times fail to help others, given the costs involved.
Even if finding information is more costly that merely avoiding misinformation, finding information can still involve low costs, and so still be obligatory. When HIAS and government officials failed to tell refugees about widespread ethnic-based killings in Unity State and Jonglei, they could have easily changed their actions by searching the internet for ‘death toll in Unity State’ and ‘death toll in Jonglei’, relaying this information without great effort. Similarly, when Swedish and Norwegian case workers failed to inform children about the risks of returning to Afghanistan, they could have easily told them that the government of Afghanistan had warned of its inability to ensure protection (Schuster and Majidi 2013).
Even when negative acts are significantly costly to avoid, they may still involve culpability. We can be culpable for negative acts that involve costs that we are expected to bear, given our unique position, or our other duties, as described in the previous section. If governments and NGOs ought to have information because of their other general duties, or because of promises they made, then we can expect them to bear the costs of finding and disclosing this information. If they do not, they may be acting wrongly, even if slightly less wrongly compared to actively misinforming.
Until now, the examples I have raised concerned facilitators who failed to inform refugees, and this led refugees to repatriate who would have otherwise stayed. There are instances where facilitators fail to inform refugees but refugees would have accepted repatriation even if fully informed. In such cases, the misinformation turns out to be irrelevant, and so it is not clear if an NGO or official committed a wrong.
Consider the case of Stephen, a father of three who approached OBI in 2012. The organization told him it was safe in South Sudan, but he knew this was not the case, having lived in South Sudan relatively recently. He wished to return despite the risks, and so accepted OBI’s free ticket home, boarding a flight for Juba with his wife and children in 2012. Slightly over a year later, the South Sudanese Civil War broke out, and he fled with his family to an Internally Displaced Persons (IDP). Today, he feels OBI’s poor knowledge had no impact on him. He knew the risks, and returned regardless. He does not regret his choice and so, perhaps by chance, OBI did no wrong, or a lesser wrong.
Consider, also, the case of Yasmin. Unlike Stephen, she had no accurate information when she returned, and upon reaching her home village of Aweil she was surprised to find no reliable clean water, no free education, and no safety for her children. She says she would have returned even if given more information. She runs a restaurant today, and is happy to be close to her family.10
In such Relevancy Cases it is not clear if facilitators are culpable for failing to provide information. Though Stephen and Yasmin were never informed of risks, this had no seeming impact on their lives or choices.
We might claim that NGOs and immigration officials did not know that Stephen and Yasmin would have returned regardless, and so their actions were still problematic. Alternatively, we might believe in moral luck. OBI failed to inform Stephen about the risks, but Stephen happened to know about the risks. By chance his consent was informed, and his life unaltered, and so OBI did no wrong, or a lesser wrong.
Even if we believe in moral luck, there is reason to believe Yasmin did not give her valid consent, and really did make her choice based on false information. Yasmin may say today that she would have returned, but this may be because she cannot turn back time, and so may as well view her decision as optimal. It may be that, had she been told information in Israel, she would not have returned. In the case of Stephen, we know he would have returned even if OBI had given him more information, because he had this information before returning, and still returned.
Even if we take Yasmin’s claim at face value—and accept that she really would have returned even if better informed—there is an additional consideration for cases like that of Yasmin.
One of the values of informed consent is that it reinforces the autonomy of the consenter, giving her control over her decisions by allowing her to deliberate over what she feels is best. If an agent lacks control, because she is forced to accept a given option, then her autonomy is undermined, even if she would have chosen the same option regardless. If a doctor forces a patient to accept surgery, he wrongs the patient even if she would have consented to surgery if asked (Groll 2012). The same may be said about misinformation. If we fail to disclose relevant information about a service, we wrong the recipient of the service by denying them information necessary to decide for themselves what option is best. This is true even if the recipient would have accepted the same service had we disclosed the information. In this sense, though Yasmin’s preferences were fulfilled, her autonomy was still undermined at the time she returned, because she could not make a fully informed decision.
We might argue that, even if Yasmin was not in control of her decision, she at least gave her ‘hypothetical consent’. This is the consent an agent would have given had she been fully informed (Waldron 1987, p. 139). Such consent is not true consent, and so Yasmin was wronged, but perhaps she was not quite so wronged. If she would have consented regardless, the act of misinformation did not set back her welfare or interests, and so the consequences were less harmful. The misinformation also did not significantly undermine her autonomy, because her preferences were, in a sense, autonomous: they were the preferences she would have developed had she been fully informed.
Even if hypothetical consent has some value in reducing wrongdoing, there is reason to believe that Yasmin did not even give her hypothetical consent. In discussing hypothetical consent, we must consider not only how one would act if given more information, but how one would act if they knew they were given false information. Imagine a patient is told there are no risks to surgery, and so consents. The surgery goes well, and she wakes up, goes home, and browses Google Scholar, finding that the surgery was, in fact, quite risky. The patient may feel legitimately wronged. She may think, ‘Had I known about the misinformation I was receiving, I would not have consented to surgery with this particular doctor. Instead, I would have gone to another doctor for this surgery’. In determining hypothetical consent, the relevant counterfactual is not only what we would do had we full information; it is what we would do had we known that we were given false information.
Even if refugees would have returned had they been given full information from a given NGO, this does not mean they would have returned via this particular NGO, had they known this NGO was misinforming them. This reasoning is reflected in the actions of one refugee, Bok, who returned in 2011. Bok knew he was being misinformed by OBI and, out of principle, rejected OBI’s assistance, paying for his own flight and arranging his own travel documentation.11 When considering what information is relevant, it is not enough to ask, ‘Would the refugee have consented if they were informed?’ We must also ask, ‘Would the refugee have consented, had they known, at the time of consent, that they were being misinformed?’
In Stephen’s case, we know that he was prepared to accept OBI’s services despite the misinformation he received. In Yasmin’s case, we do not know if she would have been prepared to accept such services had she known she was being misinformed. She may today say she would have returned via OBI even if she knew she was misinformed at the time, but we cannot know what she would have truly done at the time. We must take her memories at face value for this consideration, as well. And the more we rely on memories, the less we can be certain that information really was irrelevant for the choice made and the consent obtained.
In general, we cannot travel to a counterfactual world and see how refugees would act. Except in rare cases like that of Stephen, it is difficult to establish what information was irrelevant. To be safe, NGOs and government officials should change their policies to ensure information is available to all refugees. Just as medical professionals set general standards for informing patients, facilitators should set general standards for informing refugees, telling all about the risks prior to returning.
When NGOs and officials speak with refugees, they rarely know that they are misinforming or omitting information. Because they are not aware that they are misinforming or omitting information, perhaps they are not blameworthy for their actions.
In normative theory there are extensive disagreements as to the conditions for blameworthiness, but three theories are prominent.
Some deontological and virtue ethicists argue that one can be blameworthy for lacking morally important desires or motivations in one’s actions (Smith 2011; Slote 2001). If I fail to call an ambulance to save a person in need right near me, I am blameworthy for her death either because I lack the important desire to save her or because I lack the general motives in life to care about others in need.12 We might conclude that officials are similarly blameworthy if they fail to inform refugees, because they lack a desire to help refugees at that moment, or lack motivation in life to care about refugees. This would be true even if officials are not aware that they are misinforming and not aware that they lack important motives or desires.
Some consequentialists and deontologists argue that, to be blameworthy, one needn’t hold morally objectionable desires or motives. It must only be that one foresees, or ought to foresee, the probable consequences of one’s actions (Bennett 1998; Jackson 1997). NGOs and officials are blameworthy in this sense, if a reasonable person could foresee that failing to find information would increase the probability of an uninformed repatriation, or an unsafe return that would otherwise not take place. This would be true even if NGOs and officials are unaware that they are misinforming refugees.
Finally, a range of deontological theories argue that, to be blameworthy, one must intend one actions, rather than hold certain motives or foresee certain consequences. For one to intend one’s actions, two conditions must be met. First, one must be in control of one’s actions (Tognazzini and Coates 2016). To be in control, one must be aware of what one is doing. If one is not aware of what information is true, as in the cases described, then one is not aware one is misinforming or omitting information. Second, to wrongly intend an action one must have a particular aim (Smith 2011, pp. 124–125). If one wrongly intends to omit information, then one omits information to bring about some objectionable aim, such as the aim of uninformed repatriation. It is not clear that NGOs or officials have any aim in mind, let alone an objectionable one.
There is one reason that NGOs and officials can intentionally misinform even if they are not aware they are misinforming at the time of their actions. Misinformation can be a type of ‘tracing case’, where one’s unintentional act is traced back to an earlier intentional act, as when a doctor unintentionally fails to warn a patient of risks because she earlier intentionally failed to read the latest medical journals (Smith 2014a, b, p. 3). If at an earlier time facilitators intentionally failed to find information, and this led them to unintentionally give false information, then they are culpable. When UNHCR told Afghan refugees in Iran it was safe to return, it may not have known it was misinforming, but it knew it was failing to find information on previous returnees, as part of its general policy of not conducting post-return monitoring (Morris and Salomons 2013). Similarly, when the German government told Bosnian refugees that there would be sufficient services upon return, the government did not know about conditions in Bosnia, but perhaps it intentionally avoided finding information, with the aim of encouraging more to return.
We might argue, though, that not all intent is wrongful intent, and so not all tracing cases equal. In Israel, the director of HIAS did not conduct research on South Sudan because he thought refugees already had information.13 His intentions seemed pure: they were never to encourage misinformed returns, even if he intentionally did not find information.
The Israeli government also established its own repatriation scheme, helping several thousand refugees return to Sudan and Eritrea by 2015. Like the HIAS director, the official heading the scheme never researched the risks of repatriation. He chose to not find information because he wanted to avoid being ‘patronizing’,14 arguing that it would be disrespectful to tell refugees about their own countries. For this reason, he never learned how many had been killed after return,15 or the likelihood that others would be killed. If this official and the HIAS director can be taken at their word, they did not intend the outcome of refugees being misinformed, even if they intentionally did not find information.
Despite this qualification, there are reasons to believe that the HIAS director and the government official were culpable. The HIAS director, though believing refugees had their own information, never validated this belief. He did not merely neglect to find information on South Sudan; he intentionally failed to find out if his belief about refugees’ knowledge was correct. And we do not know why the HIAS director intentionally failed to find out whether his belief about refugees’ knowledge was correct. It may be that his motivations were to encourage return, or at least ensure good working relations with OBI, which wanted to encourage return. If so, then his intentions were problematic.
The government official’s reasons for his actions were not based on any false belief about refugees’ knowledge. He wanted to avoid being patronizing. Many refugees really may feel patronized if told about their own country. As such, the official's intentions were not quite based on false beliefs.
Though it is true that the official’s reasons for his actions were to avoid being patronizing, reasons can be derived from other reasons. The official perhaps chose to avoid being patronizing because this would cause more to be misinformed, and so more would return. If so, then his ultimate intention was not to avoid being patronizing, but to encourage return.
The above analysis assumes we can know the intentions of other agents. Clearly, we cannot reach inside their minds and learn about their aims and reasons. Nonetheless, we can still find evidence of intent, if not decisive certainty. Repatriation facilitators are often in a position where they have an interest in more refugees repatriating to meet their annual targets. From this, we can conclude that facilitators should not have annual targets. Such targets give facilitators reasons to intentionally fail to find information, leading to uninformed repatriation.
When the director of OBI showed photographs from South Sudan to refugees in Israel, she could have distributed information on the lack of healthcare, security, and universal education. Finding some of this information would be costless, and Good Samaritan duties require such costless acts.
When finding information is costly to obtain, and when there are few reports on countries of origin, facilitators may still have duties to disclose accurate information. NGOs have duties to find information derived from their promises to ensure informed returns. States make no such promises, but have duties to find information derived from their general duties to help prevent atrocities abroad, and offer protection to refugees. Such duties create a duty to know about conditions abroad, which in turn creates a duty to disclose information to those repatriating.
Even when facilitators merely fail to disclose information, rather than actively misinforming, they may still be culpable. Acts of omission, even if less wrong than active misinformation, are still wrong if causally contributing to refugees’ decisions to return. Failing to inform refugees can be viewed as a positive communicative act that implies security, encouraging refugees to return who would have otherwise stayed.
Some refugees claim they would have returned even if warned of the risks. In reality, however, it is difficult to know if refugees really would have returned had they been better informed. When refugees today say they would have returned, these feelings about the past may be the result of their inability to change the present. Given our inability to be certain that misinformation had no impact on refugees’ decisions, we ought to still hold NGOs and officials accountable for misinformation.
Many repatriation facilitators may claim that, though they provided false information, they did not intend to give false information, and were not driven by objectionable motives, and so are not blameworthy. To establish the validity of this claim, it is not enough to establish if misinformation was intentional, or the result of objectionable motives, but whether earlier failures to find information were intentional, or the result of objectionable motives. If facilitators chose to avoid finding information in order to encourage return, they are blameworthy for their earlier motives and intent, even if they unwittingly gave misinformation.
Regardless of whether facilitators are blameworthy, they still act wrongly when failing to inform refugees of the risks. Such failures mean many refugees will take the irreversible decision to return, regret doing so, and find themselves again displaced, without asylum or basic necessities. Just as preventing deportations is essential in some cases, so is preventing misinformation, to ensure more ethical immigration control.
Interview with Bol, Juba, 21 December, 2013; interview with Niko, Juba, 14 December 2013; interview with Tareza, Juba, 25 December 2013.
I assume that an agent can be responsible for an objectionable outcome without being culpable, but that one cannot be culpable without being responsible for an objectionable outcome.
I shall assume a largely utilitarian form of consequentialism (see Singer 2015), but nothing I argue is dependent on this formulation of consequentialism.
In this case, the government of Afghanistan warned of insecurity, but the case workers helping children repatriate may have still been misinformed themselves due to inaccurate reports from their superiors.
Interview with OBI, Jerusalem, 6 October 2010.
Even if such demands would ensure optimal consequences, they would create unequal burdens and may, as such, be unfair.
Interview with S, an OBI and HIAS staff member, Tel Aviv, 28 April 2012.
Such a reason would also be supported by deontologists and virtue ethicists who care about avoiding negative consequences.
Even theories which view virtuous motives as primary still determine the rightness of motives partly based on the extent of causal harm. If, for example, a virtuous person is one who cares about others, it seems that caring involves, at least some of the time, caring about whether one’s actions will case harm (see Slote 2001).
Interview with Yasmin, Aweil, 30 March 2012.
Discussion with Bok, Juba, 1 January 2013.
Strictly speaking, virtue ethicists may not couch this in terms of ‘blameworthiness’ but simply in terms of non-virtuous motives. There is much debate over what constitutes a non-virtuous motive. Regardless, if seems clear that one lacks a virtuous motive if one fails to call an ambulance for an individual in urgent need (see Slote 2001).
Interview with director of HIAS-Israel, Jerusalem, 11 December 2012.
Interview with Voluntary Return official, Tel Aviv, 7 August 2013.
- Al Jazeera. 2011. ‘Hundreds dead’ in South Sudan cattle raids. 22 August 2011, downloaded 10 January 2014 from http://www.aljazeera.com/news/africa/2011/08/201182220946583842.html.
- Barnett, Michael. 2001. UNHCR and the ethics of repatriation. Forced Migration Review 10: 31–34.Google Scholar
- Betts, Alexander. 2010. Survival migration: A new protection framework. Global Governance 16(3): 361–382.Google Scholar
- Brown, Tim. 2006. South Sudan education emergency. Forced Migration Review (supplement): 20–21. Available at http://www.fmreview.org/sites/fmr/files/FMRdownloads/en/FMRpdfs/EducationSupplement/13.pdf.
- ECtHR–Hirsi Jamaa and Others v Italy [GC], Application No. 27765/09.Google Scholar
- Fouda, Lauren. 2007. Compulsory voluntary repatriation: Why temporary protection for Sudanese asylum-seekers in Cairo amounts to refoulement. Georgetown Journal on Poverty Law 10(3): 511–538.Google Scholar
- Geneva Convention for the Protection of Refugees. http://www.unhcr.org/protect/PROTECTION/3b66c2aa10.pdf.
- Gerver, Mollie. Refugee repatriation and the problem of consent. British Journal of Political Science (forthcoming).Google Scholar
- Goren, Yuval (Hebrew) Aid organizations: More than 22 refugees expelled to South Sudan killed this year. Maariv 5/6/2013.Google Scholar
- Howard-Snyder, Frances. 2011. Doing versus allowing harm. In The stanford encyclopedia of philosophy, ed. Edward N. Zalta (Winter 2011 Edition) http://plato.stanford.edu/archives/win2011/entries/doing-allowing/.
- Jackson, Frank. 1997. What Effects? In Reading parfit, ed. Jonathan Dancy. Oxford: Blackwell Wiley.Google Scholar
- Ferrie, Jared. 2012a. More Than 200 Die in South Sudan Tribal Feud, Official Says. CNN. 12 March 2012 http://edition.cnn.com/2012/03/12/world/africa/south-sudan-violence/.
- Ferrie, Jared. 2012b. More Than 200 Die in South Sudan Tribal Feud, Official Says. CNN. 12 March 2012 http://edition.cnn.com/2012/03/12/world/africa/south-sudan-violence/.
- Lemberg-Pedersen, Martin. 2015. The rise and fall of the ERPUM pilot: Tracing the European policy drive to deport unaccompanied minors. Refugee Studies Centre working paper no. 108.Google Scholar
- Miller, David. 2005. Immigration: The case for its limits. In Contemporary debates in applied ethics, ed. A. Cohen, and C. Wellman, 193–204. Malden, MA: Blackwell Publishing.Google Scholar
- Morris, Helen and Michael Salomons. 2013. Difficult Decisions: A review of UNHCR’s engagement with Assisted Voluntary Return programs. UNHCR Policy Development and Evaluation Services.Google Scholar
- MSF. 2014. South Sudan: Violence against healthcare. 1 July http://www.msf.fr/actualite/publications/south-sudan-conflict-violence-against-healthcare.
- Osman v. United Kingdom, European Court of Human Rights, Application N. 23452/94, 28 October 1998, para. 1.Google Scholar
- Riiskjaer, Maria Helene Bak and T. Nielsson. 2008. Circular repatriation: The unsuccessful return and reintegration with refugee status in Denmark. Research Paper 165, New Issues in Refugee Research.Google Scholar
- Singer, Peter. 2015. The most good you can do. London: Yale University Press.Google Scholar
- Small Arms Survey. 2011. Fighting for spoils: Armed insurgencies in Greater Upper Nile. November 2011. http://www.smallarmssurveysudan.org/fileadmin/docs/issue-briefs/HSBA-IB-18-Armed-insurgencies-Greater-Upper-Nile.pdf.
- Smith, Holly. 2014a. Negligence. In The international encyclopaedia of ethics, ed. Hugh LaFollette. Oxford: Wiley-Blackwell.Google Scholar
- Strand, Arne. 2011. Review of two societies: Review of the Information, Return and Reintegration of Iraqi Nationals to Iraq (IRRINI) program. Chr. Michelson Institute. http://www.cmi.no/publications/publication/?4155=between-two-societies-review-of-the-information.
- Tognazzini, Neal and D. Justin Coates, 2016. Blame. The Stanford encyclopedia of philosophy, ed. Edward N. Zalta.Google Scholar
- UNHCR. 2012. UNHCR. 2012. Displacement: The New 21st Century Problem. Available from http://www.unhcr.org.uk/fileadmin/user_upload/pdf/UNHCR_Global_Trends_2012.pdf, accessed October 2014.
- United Nations. 2005. The Responsibility to Protect. Outcome Document of the 2005 United Nations World Summit, A/RES/60/1, para. 138-140, http://www.un.org/en/preventgenocide/adviser/responsibility.shtml.
- Van England-Nouri, Anisseh. 2008. Repatriation of Afghan and Iraqi refugees from Iran. International Journal on Multicultural Societies 10(2): 144–169.Google Scholar
- Walsh, Martha, Richard Black, and Khalid Koser. 1999. Repatriation from the European Union to Bosnia Herzegovino: The role of information. In The end of the refugee cycle? Refugee repatriation and reconstruction, ed. Richard Black, and Khalid Koser, 110–125. New York, NY: Berghahn Books.Google Scholar
- World Bank. 2014. Unemployment, total (% of total labour force) (modelled ILO estimate). downloaded on 15 June 2015 from http://data.worldbank.org/indicator/SL.UEM.TOTL.ZS.