Among the most urgent measures envisaged in the European Agenda on Migration (the EU Commission plan to face the so-called “refugee crisis” in Europe) one finds a “scheme to relocate 40,000 people from Italy, Greece and other Member States” (a number later increased to 160,000) and a “scheme to resettle over 20,000 people from outside the EU”. Whereas some countries resolutely refused the plan, the Portuguese government supported the initiative and agreed to host more than 4,500 people. As a consequence, an issue that had gone almost unnoticed in the country (Portugal registered only 477 asylum requests in 2014), took center stage and became a matter of public debate. Opponents remarked that country-nationals impoverished by the economic crisis should be prioritized, while a number of local organizations and NGOs responded with an extraordinary public mobilization to receive the expected newcomers. While some refugee advocates were certainly driven by the interest to gain visibility and state commissions, the majority was motivated by considerations of solidarity and commitment that cannot simply be reduced to self-interest and profit. Although the images of death and despair coming from Syria, Turkey, Greece and Eastern Europe played a significant role, the wish to be “useful” and alleviate human suffering constituted a reason of collective ethical engagement. As time passed, however, it became clear that refugees, and particularly the awaited Syrians, were not arriving. Greek and Italian authorities were occasionally blamed for their supposed lack of organization, while some began to think that the Syrians might not want to come to Portugal. Based on this example and on interviews carried out with policymakers and community workers, I discuss how institutional representations of refugees hinders the possibility to consider their autonomy, and in particular their autonomy to move. I use this case to reflect upon the frequent clash between regimes of inclusion and expectations of migrants and refugees, but also on the imaginary geographies of power that underpin their itineraries. Drawing also on my research in other areas (Morocco, Tunisia and Italy), I argue that the notion of refuge as it is currently defined and enforced turns out to be a new tool of immobilization and subordinate inclusion. In this sense, the forms of “radical mobility” enacted by migrants in different ways today constitute a challenge to the common notions of refuge and require new ways of thinking reception and hospitality.