Does International Refugee Law offer adequate protection for forced migrants? Should it protect them

There is an assertion that ‘refugee and forced migration studies could be treated as separate but allied' (Hathaway). Nevertheless, I will argue that forced migrants and refugees should have the same standards of protection because they are both categories of people who are in desperate need of haven. International Refugee Law (IRL) contains many principles that offer protection to forced migrants; however, there are difficulties with the laws. This is partly because of the interpretations of the definitions and the fact that the scope does not cover enough deserving people. Nevertheless, the main problem is the way in which States implement the laws. It will be shown that much more needs to be done by international actors before the system is completely successful.
Who are forced migrants? Why should we protect them?
The term ‘forced migrant’ applies to, but is not limited to, refugees. It also includes ‘internally displaced people, … people displaced by natural or environmental disasters, chemical or nuclear disasters, famine, … development projects’, or conflicts (Mailman School of Public Health). 51 million people are displaced today (UNHCR), so this is a huge problem. Hathaway is right that ‘refugees should not be privileged over [forced migrants]’, yet IRL is focused on the protection of refugees.

Laws
The 1951 Refugee Convention is the cornerstone of refugee law. The most important principle is in Article 1A, as amended by the 1967 Protocol, which states that a refugee is someone 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 or, owing to such fear, is unwilling to avail himself of the protection of that country’.
This is supposedly a ‘very clear definition’ (David J. Whittaker), but every single word of this has been interpreted by at least one country around the world, with varying results (Machiel Salomons). There are also too many requirements that must be fulfilled in order to obtain this protection.
The requirement of direct persecution is especially problematic. The Adan case held that the claimants could be refused protection as refugees, even though they were escaping the dangers of civil war, because they were not directly persecuted. The rationale was that ‘all sections of society in northern Somalia are equally at risk.’ This is harsh, but it is following the legal requirement. Therefore, this is not a good principle in practice because it doesn't take account of the reality on the ground; you wouldn't hang around and wait to be personally persecuted, would you? In recognition that this is too onerous, there have been two regional conventions (OAU Convention and Cartegina Declaration), but this problem needs to be recognised and rectified universally to ensure justice for all.
Moreover, the ‘reasonable fear’ requirement of Article 1A is troublesome. ‘Reasonable’ is a notorious concept in law anyway, but there are further disagreements over whether there should be an objective or subjective standard for this criterion. It is ‘in one sense hypothetical’, yet the requirement that there be real reasons for a fear is afforded great significance in States’ assessments (Whittaker). Unfortunately, the information States use may not be reliable, so how can we trust their interpretations?
It is not only the contents of the 1951 Convention that causes problems, but also what is excluded. Forced migrants are often excluded from refugee status in situations where there is not one identifiable persecutor. Article 1F 1951 Convention is especially controversial because it excludes people from protection who would otherwise fit the definition of a refugee if there are ‘serious reasons for considering that’ they have committed certain crimes or other prohibited acts (ELENA). The possible rationale is that they don't deserve protection, but this is a throwback to victors’ justice. There may be good reasons why the person committed those acts; they are fleeing persecution after all. So, the principle of double peine may not be upheld in these cases. Clearly, these exclusions do not help to achieve adequate protection for refugees.
Complementary protection covers those who ‘cannot be returned to their home country, because there is a real risk that they would suffer certain types of harm that would engage [the State’s] international non-refoulement … obligations (ELENA). Although there are numerous other complementary laws in place, the principle of non-refoulement is perhaps the most important example here because migrant status is irrelevant under this principle. Although it is codified, it is probably also customary international law, and it may even override the provision in the Convention itself. This means that no State can send migrants, even non-refugees, back to their home country if their life or freedom would be threatened there because they possess one of the protected grounds. This is significant as it goes some way to solving some of the problems of exclusion under the 1951 Convention, especially those under Article 1F. Take, for example, the Abu Quatata case. Abu Quatada had been granted refugee status in the UK, but was then convicted in absentia in Jordan of various offences that fall under Article 1F 1951 Convention. The UK tried to deport him, but, after considering non-refoulement, the ECtHR held ‘that the applicant’s deportation to Jordan would be in violation of Article 6 [ECHR]…’ This is a victory for IRL. He isn’t likeable, but he is still a forced migrant with real concerns about torture in his home country, therefore he should be adequately protected.
The lack of protection for internally displaced persons (IDPs) in the 1951 Convention is also problematic. The only substantial difference between refugees and IDPs is that refugees have crossed border; IDPs have not. This distinction is in place because States are very jealous about their sovereignty; the UN is equally reluctant to be involved in State’s internal affairs. IDPs still retain their rights under international human rights law, IHL and complementary protection (ICRC), but if they have been forced to flee their homes it is unlikely that the State they remain in will protect them. Despite it not technically being in their mandate, the UNHCR try to help IDPs; however, they do not have enough time or money to guarantee adequate protection. The safeguards would be more useful if the protection for refugees was extended to IDPs.
There are many more problems, but these are the most important ones. They show that the legal differentiation between refugees, asylum seekers, IDPs and other migrants is not always fair. It is probably not always understood either. For example, the term ‘asylum seeker’ seems to have the most prominence in everyday discourse. Unfortunately, this common use ‘is vague, ambiguous, [and] often censorious…’ (Whittaker). The ‘fearful scenario of millions on the move, in turmoil and danger, deprived of a normal life’ seems to be neglected (Whittaker). It is clear then that the definition of refugees under the 1951 Convention should be understood, updated and universalised before the situation for forced migrants can be improved.
Implementation
Implementation has been discussed briefly with regard to the above problems, but further examples will help to highlight the issue. IRL will not be fully efficient unless it is effectively implemented. Many States are signed up to the relevant treaties; however, they often do not respect them as much as they should.
The biggest areas of non-compliance are probably the ones relating to non-refoulement and situations where States do not accept migrants when they should. Although there is no right to asylum under in international law, States still must adhere to general principles of IRL and international human rights law.
Rich, Western countries probably have the resources to protect forced migrants. Nonetheless, often the only way to stay in these States is to use the asylum channels, yet the conditions of these channels are regrettable. Australia is a good example of this. Although it is signed up to the 1951 Convention and other relevant treaties, it uses loopholes in order to sidestep its obligations, including rejecting migrants at the border and sending them to other territories. Rejection at the border is not covered by the principle of non-refoulement. The problem with this is that many of the States these people are sent to are not signed up to the 1951 Convention, nor do they comply with many fundamental IRL, IHL or human rights norms. Consequently, these States may then send the migrants back to the States they were fleeing. Australia even tries to convince forced migrants, who have real fear for their safety, to return home themselves.This flagrantly violated IRL obligations. Unfortunately, instances like this are not uncommon in other States.
Conclusion
Clearly there are many problems with IRL, but none that cannot be solved. Only a few States have been analysed here, yet it is clear that the international community, do not even implement IRL properly. Rich, Western states have the means to support more forced migrants than they do. Yet they do not adequately share this task with other States that may be unable to protect forced migrants sufficiently. This shold not be seen as a burden; humanitarian concerns and respect for human dignity should be afforded more weight in practice. The international community should improve its protection of forced migrants out of respect, dignity and humanity, no matter who they are or where they come from.