A refugee, by legal definition, is someone who is fleeing a well-founded fear of persecution for reasons of race, religion, nationality, political opinion, or membership in a social group. But the category has always exceeded its legal core. The classic image of the refugee, also understood often as an individual in exile, used to make room for the heroic: the freedom fighter, the revolutionary. Political persecution was seen as the exemplary form of persecution. A refugee was an active, not passive, figure — even when this was romanticized. But over the past fifty years, the category of the refugee has shifted in meaning. If asylum was only ever for exceptional cases, as Hannah Arendt acknowledged in 1951, it nevertheless allowed for an exceptionally active person.

Innocence is now the key qualifier for someone who claims to be a refugee. Paradoxically, as part of this moralized regime, innocence is also claimed by those who grant asylum. The qualification that refugees must be seen as innocent can change their fate at different points. Those requesting asylum may not even be allowed to make a claim, if they are not considered innocent. They may be deported before this happens, since discretionary power over who can claim asylum is increasingly given to border control officials, especially in the United States. And if people reach the territory, this implicit qualification plays a role in the evaluation of their claims.

But what is innocence? Commonly understood, to be innocent is to be pure, without guile, without knowledge, and without intent. Indeed, innocence is defined as “freedom from sin, guilt or moral wrong in general,” “freedom from specific guilt,” and “freedom from cunning or artifice.” This negative freedom, of “freedom from,” is so free that it is seemingly free of content: it purports to be a state of moral and epistemic purity. Such purity structures moral categories, filling out binary notions of deserving and undeserving, ultimately giving shape to what philosopher Ian Hacking would call different “human kinds” — different orders of being.

Since the long summer of migration in 2015, the media and public officials in both Europe and the United States have insisted on the distinction between “refugees” and “illegal economic migrants.” Asylum in this case is primarily a moral, not legal, distinction that purports to separate the innocent from the guilty, but it can have legal impact. “Real” refugees are seen as innocent, fleeing real well-founded fears of persecution. As anthropologist Liisa Malkki has observed, refugees, especially African refugees, are figured as “a ‘sea’ or ‘blur of humanity,’” as “a spectacle of a ‘raw,’ ‘bare humanity.’” They are understood as passive and in need of saving. Innocence is about vulnerability and naiveté. Children and puppies come to mind first. Economic migrants, by contrast, are counterposed against refugees and portrayed as wily, trying to lie their way into welfare and other benefits, and to undermine national security and national values. They are the innocents’ binary other: the guilty.

Yet who are these innocents, these exemplary refugees? Who are the people who have not lived, known, desired? Perhaps unsurprisingly, those associated with innocence tend to be at humanity’s edges. They mark its border, in the sense that they are not corrupt (as is normative humanity), yet nor are they fully human, in the Enlightenment sense of having reason, will, or autonomy. Innocence produces worthiness only insofar as it is also a space free from desire, will or agency — the lack leaves one incapable of being a thinking, engaged, active, informed subject. Indeed, in valorizing this kind of purity, innocence produces unknowing dupes. These are not the people who confront their circumstances and try to change them. These are not the people who work to escape poverty, who react to the global system of racialized inequality, and who, in trying to escape it, also expose it. Neither are these the people who live ordinary lives of joy and hardship. In their want of knowledge, these “innocent refugees” cannot take care of themselves. They certainly cannot challenge a system.

In this sense, by valorizing refugees as innocent, vulnerable, naïve, and lacking responsibility for their circumstances, the category enables the criminalization of all those who want to change their circumstances, who see better lives and refuse to accept inequality or its accompanying forms of violence. It criminalizes poverty, but it leaves wealth untouched. There is no possible place of innocence for the racialized poor. Consider this in contrast to the newest category of legitimate refugee: the LGBT asylum seeker. While it is surely a good thing that people fleeing homophobic violence and persecution can apply for asylum, this does not include everyone persecuted for their gender or sexual orientation. It covers the LGBT person, cleaned up and rendered innocent. To be gay in this sense is to be born this way, to have a biological condition; no choice or desire is involved, since to be innocent is to be without knowledge or desire. Paradoxically, this “biological” condition is only recognized in certain scripted forms: only masculine women or feminized men qualify. Perhaps more crucially, recognizing the persecution of legitimized LGBT asylum seekers does not undermine the innocence of those receiving them.

Innocence produces hierarchies among people on the move, enabling some to be prioritized over others, but it also enables a distinction between those who help and those whom they help. Innocence always carries with it the desire to protect and take responsibility for those whom — in their want of knowledge — cannot take care of themselves. Guilelessness evokes the need for care: innocents cannot take responsibility for themselves. But this means that it props up a feeling of control in those who care for the innocent. It assures them not only of their power but also of their knowledge, insofar as the innocent person is oblivious. It creates a class of saviors. As a space of purity, innocence itself appears outside history, and as such, it allows those who work as saviors to ignore the political and historical circumstances that created these victims.

This not only allows saviors to feel powerful or knowledgeable but also enables them to simultaneously capture innocence — to purify or absolve themselves. In other words, innocence also creates a savior class or subject, and they too make claims to innocence. If refugees are understood as innocent, not responsible for their circumstances, and one is only intervening to help them, how can this not be considered innocent too? Here I want to draw attention to another meaning of innocence: while the Latin etymology of innocence focuses on harm (in- + nocens, “not harmful”), I want to draw attention to a different etymology of in- + nocere, “not to know,” which I think is particularly significant. What does it mean not simply to be empty of knowledge, but specifically to not know? This can be a willful ignorance.

James Baldwin considered this kind of willful ignorance or unknowing in terms of racial (white) innocence: the American refusal to deal with deeply entrenched forms of racial injustice by holding onto ideologies of equality that undergird the American Dream rather than facing the actual historical evidence. Baldwin was describing the way Americans want to maintain the fantasy of a race-blind present and future. Their ignorance allows the posture of innocence, which can be a politically useful deflection for those in power. It allows white people to benefit from the system without acknowledging the way the system operates or their central role in its reproduction.

In a “Letter to my Nephew,” Baldwin writes, “I know what the world has done to my brother, and how narrowly he has survived it and I know, which is much worse, and this is the crime of which I accuse my country and my countrymen and for which neither I nor time nor history will ever forgive them, that they have destroyed and are destroying hundreds of thousands of lives and do not know it and do not want to know it…[I]t is not permissible that the authors of devastation should also be innocent. It is the innocence which constitutes the crime.” Baldwin condemns the innocence claimed by those in power to prop themselves up, while at the same time purifying or absolving themselves of the harm they inflict in the process. That is, a claim to innocence writes out responsibility.

After all, in its Judeo-Christian interpretations — which continue to resonate today — the innocent were bound by divine authority; they could not act on their own. Only after the Fall, when Eve eats from the tree of knowledge, does human life acquire a will and responsibility of its own. How can people who did not know, who did not intend to do harm, to discriminate, be held accountable? How can those helping the most passive, the most vulnerable, the most pathetic, not also be innocent? How can they be held responsible for the distinctions this furthers, the types of the hierarchy it valorizes? This builds on colonial logics, where innocence was also invoked as a technique of rule; colonial violence was rationalized in the name of tutelage, or betterment.

The refugee system has always allowed nation-states that accept refugees to feel superior, to claim a higher status of civilization, against the less enlightened regimes who unjustly persecute people. It has been an instrument of foreign policy. The U.S. government could welcome those being persecuted by the Soviet regime, for instance. But now, asylum not only allows a claim to superiority but also produces purity. When innocence is put front and center — when innocence is what determines legitimacy and deservedness — it absolves everyone of responsibility for the situation of the refugee. It not only allows but also enables global capitalism and its effects to go unconfronted and unchecked. It is one of capitalism’s safety valves. It allows all of us to not know how our lives and privileges are the cause of poverty elsewhere. It permits us to absolve ourselves of responsibility for people increasingly needing, wanting, or dying to move. In this sense, we must question whether the category of refugee does more harm than good, and whether we need both new moral and legal vocabularies.

Miriam Ticktin is Associate Professor of Anthropology at the New School for Social Research and author of Casualties of Care: Immigration and the Politics of Humanitarianism in France (2011). She is currently writing a book on innocence as a political concept.