The Citizenship Amendment Act alters the definition of ‘illegal immigrant’ for the purposes of the Citizenship Act 1955. It provides that ‘persons belonging to minority communities, namely, Hindus, Sikhs, Buddhists, Jains, Parsis and Christians from Afghanistan, Bangladesh and Pakistan’ who have received specified exemptions from the Central Government ‘shall not be treated as illegal migrants for the purposes of this Act’ – and are hence not exempted from applying for citizenship of India.
Most readers of this post will know that the Citizenship Amendment Act has been hugely controversial. Nivedhitha K has persuasively argued that the amendment makes impermissible classifications based on religion and country of origin (see also Shivam Sanghania’s response). Mihika Poddar has similarly argued that the amendments ‘seek to read religious divisions explicitly rejected by the framers of the Constitution into citizenship law’, that they exclude without justification multiple communities similarly subjected to religious persecution in India’s neighbours, that the revised naturalisation process’s preference for non-Muslims is ‘suggestive of a broader sectarian agenda’ and that the amendment hence violates India’s obligations under international law. The amendment has prompted mass protests in India and internationally.
This post does not re-state the broader arguments for and against the constitutionality of the Citizenship Amendment Act. It is much more narrowly focused. Public debate around the Citizenship Amendment Act has often turned on its apparent purpose of protecting vulnerable minorities in India’s neighbouring countries (albeit with severe criticism of how it purports to do so). If, as defenders of the amendment’s constitutionality argue, the Citizenship Amendment Act is meant to provide a ‘measure of protection for [a] recognizable and internationally accepted target group’, then who actually falls within this ‘target group’? Who is a ‘Christian’ or a ‘Hindu’ or a ‘Buddhist’ for the purposes of the Citizenship Amendment Act? What are the boundaries of the protected categories, and what does this mean for its constitutionality and its operation in practice?
These are not new debates in refugee law. Indeed, the Citizenship Amendment Act’s failure to address these questions speak of how far removed it is from conventional international models for the protection of persecuted people. The Refugees Convention (as amended by the Refugees Protocol) requires a refugee, relevantly, to possess a well-founded fear of persecution ‘for reasons of race, religion, nationality, membership of a particular social group or political opinion’. Unlike the ‘minority communities’ recognised by the Citizenship Amendment Act, these five categories under the Convention are not fixed. ‘Religion’ includes, for example, religion as belief, religion as identity and religion as a ‘way of life’; critically, persecution for reasons of religion can include persecution on the basis of perceived religious identity as much as actual religious identity, or the way in which one chooses to express one’s religion (for example, through evangelism or practicing one’s faith in community with fellow believers) as much as the actual content of one’s religious beliefs. One can, for example, be a Muslim and identity as a Muslim but nonetheless be persecuted ‘for reasons of religion’ because of beliefs or practices which are individual or heterodox.
Viewed in this light, the Citizenship Amendment Act raises enormous questions. Even though identity as a ‘Hindu’ or a ‘Christian’ may be capable of determination in other respects in Indian law (for example, in determining the applicability of personal laws), there is nothing in the Citizenship Amendment Act to suggest that how those communities are defined under other statutes should be ‘imported’ into the new amendments or given a consistent meaning. Is a person who is perceived as a Christian, and fears harm on that account, a ‘Christian’ for the purposes of the Citizenship Amendment Act regardless of the content of their beliefs or how they personally identify? That interpretation would be consistent with the supposed aim of the Citizenship Amendment Act to protect religious minorities targeted for persecution. But it sits uneasily with the definition’s use of the term ‘persons belonging to minority communities’ – can one ‘belong’ to a community with which one does not identify? Is it membership of the community or beliefs, actual or perceived, which is the protected characteristic? Is a person a ‘Christian’ because they believe themselves to be a Christian; because their persecutors believe them to be a Christian; because their persecutors believe them to observe Christian beliefs or rituals (regardless of whether they are a ‘Christian’ in any deeper sense); because the Christian community in their country of origin believes them to be a Christian; because the Christian community in India would regard them as a Christian; or because they would be regarded as a Christian for the purposes of determining which set of personal laws would apply to them, in India or in their country of origin? Are converts to the protected faiths eligible for protection under the Citizenship Amendment Act – and, if so, who determines what that conversion must entail and whether that conversion is ‘genuine’?
These are not just theoretical questions. In refugee law, people often claim to fear persecution because they have converted to new faiths or because they have adopted new religious practices. These are some of the most difficult cases to prove and to determine. As I have written elsewhere, decision-makers often assess these claims in troubling ways – deciding that asylum seekers are not ‘true’ converts because they cannot explain religious doctrines in a detailed or persuasive way (even though detailed understanding of doctrine is a poor proxy for sincerity of religious belief) or because they maintain heterodox practices from their prior life (even though genuine believers substantially vary in how they understand and practice their faiths). When I was younger, I tried to explain this in terms of Said’s Orientalism:
[R]eligious faith must not be viewed completely in a vacuum, without regard to influences arising from any other cultural or individual factor, and claimed religious identities should not be judged purely by virtue of how they match a supposedly-‘objective’ conception of the faith. Such modes of thought are both erroneous and (where applied to asylum seekers from the imagined ‘non-West’) reflect an ‘Orientalist’ mode of thought – embodying the notion that, as Said put it with regard to Islam, ‘[t]here are still such things as an Islamic society, an Arab mind, an Oriental psyche’ (Said 1978, 301), with the Other thereby ‘classified’ for the purposes of scrutiny and assessment (Said 1978, 41). The extraordinary complexity and diversity of Islam as lived and practiced around the world is mirrored by that of other religions, complicating any attempt to judge the truthfulness of a claimed conversion by reference to believers’ knowledge of doctrine. Where held by decision-makers, the belief that one can ‘divide human reality… into clearly different cultures, histories, traditions, societies, even races’ (Said 1978, 45), discrete from one another and capable of infallible definition even in the messy circumstances of individual lives, amounts in practice to the enforcement of this belief upon applicants (through adverse consequences for those asylum seekers who do not meet decision-makers’ assumptions).
Now that I’m older, I can admit that’s a pretty thin reading of Said. But the point I was trying to make is this: refugee law is not and should not be based on ‘essentialised’ notions of personal or community identity. If it is to provide meaningful protection against persecution on the basis of religion, it needs to reject false classifications of particular communities or faiths as somehow clearly divisible from one another and as containing shared, intrinsic traits, with all those within the community possessing those traits and all those outside the community being somehow clearly differentiated (or belonging to their own essentialised communities, discrete and sharply divisible from one another). Decision-makers in refugee law all too frequently fall into these traps by, for example, acting upon a false and ahistorical idea of what a ‘Christian’ is and rejecting claimants who do not fit that model.
The Citizenship Amendment Act invites decision-makers to fall into that same error. On one hand, it creates a binary between people who belong to particular religious communities (and who will therefore receive protection) from those who do not – suggesting, again, that such a binary is capable of identification in the messy circumstances of individual lives and that it is membership of a community, rather than individual belief and practice, that defines a person’s religious identity. But it exacerbates this problem because it does not say how these ‘religious communities’ are to be defined; it is left open to individual decision-makers to proceed on the basis of their own preconceived notions of who falls within the amendments’ protection and who does not.
Returning to the question of the Citizenship Amendment Act’s constitutionality: the Act must provide intelligible differentia for the ‘reasonable classifications’ which it creates. As noted above, others have argued that the Act impermissibly discriminates on the basis of religion and nationality. But one can take that critique further. By failing to define who is, in fact, a person belonging to a minority community for the purposes of the Act, and hence who is protected and who is not, it is reasonably arguable that the Act confers uncanalised discretion upon individual decision-makers (that is, those officials of the Government of India charged with determining whether a person is an ‘illegal immigrant’ or not, and who will enforce the Act in practice) without clear criteria or guidance; the terms ‘Christian’ or ‘Hindu’, for example, are impermissibly broad without an indication of how they are to be understood in practice and who falls within their ambit. The law’s vagueness and uncertainty provide further scope for challenging its constitutionality.