The Citizenship Amendment Act 2019 debate is conducted, on most days and in most places, about a statute that neither side quite describes. Supporters defend a measure of relief for persecuted minorities, stated at a level of generality the text does not use. Opponents attack a measure of exclusion aimed at one community, stated at a level of generality the text does not use either. Between those two descriptions sits a short amending Act of five sections, the Citizenship (Amendment) Act, 2019, Act 47 of 2019, which received the assent of the President on 12 December 2019 and was brought into force on 10 January 2020. This article sets out what that Act provides, element by element, and what it does not provide. It then states the case made in its support and the case made against it, each at full strength and in the terms its own proponents use, and it records the procedural position of the constitutional challenge that will decide between the legal parts of those cases. It adopts no position of its own, and a reader who finishes it should be able to say which side the writer holds only by guessing.

The discipline matters because citizenship is the gateway status in the Indian legal order. It determines who may vote, who may hold specified constitutional offices, who may move and reside without permission, and who stands inside the community to which the state owes its fullest duties. A change to the rules of admission is therefore never a small thing, even when the change is textually narrow. The argument of this article is the argument of the series of which it forms part: that a statute is frequently narrower than the debate conducted about it, that the narrowness is not a trick or an evasion, and that stating the text precisely is the precondition for engaging with the debate honestly rather than a way of avoiding the debate. The 2019 amendment is the instance in which that discipline is hardest to hold and most necessary, because the implications each side draws from the text are large, sincerely held, and incapable of being settled by the text alone.

Three separations organise everything that follows. The first is between the text and its implications: the amendment does a small number of defined things, and nearly everything contested about it concerns what those things signify, portend or presuppose. The second is between the class the amendment names and the population each side talks about: the statutory class is closed, dated and conditioned in ways that public description on both sides routinely drops. The third is between the legal question and the political question: the legal question, whether the classification the amendment makes is one the Constitution permits Parliament to make, is pending before the Supreme Court and will be answered there; the political question, whether the classification is wise or just, is not before any court and will not be answered by one. An account that keeps those separations visible can be fair to both sides without being empty, and that is the standard this article sets itself.

What the Citizenship Amendment Act 2019 provides, stated exactly

The Citizenship (Amendment) Act, 2019 amends the Citizenship Act, 1955, the principal statute governing the acquisition and termination of Indian citizenship, whose architecture is treated in the article on the Citizenship Act 1955 explained. The amending Act makes four substantive changes and one consequential change. Each is stated here in the order in which it appears in the statute book, and each is stated without the vocabulary of either side of the debate, because the vocabulary is itself one of the things in dispute.

The first change is to the definition of illegal migrant in clause (b) of sub-section (1) of Section 2 of the principal Act. The definition, inserted by the amendment of 2003, treats as an illegal migrant a foreigner who entered India without valid travel documents, or who entered with such documents and remained beyond the permitted period. To that definition the 2019 Act adds a proviso. Under the proviso, any person belonging to the Hindu, Sikh, Buddhist, Jain, Parsi or Christian community from Afghanistan, Bangladesh or Pakistan, who entered India on or before 31 December 2014, and who has been exempted by the Central Government under clause (c) of sub-section (2) of Section 3 of the Passport (Entry into India) Act, 1920 or from the application of the provisions of the Foreigners Act, 1946 or any rule or order made under that Act, shall not be treated as an illegal migrant for the purposes of the Citizenship Act. Every element of that sentence is a condition, and the later sections of this article take the elements one at a time, because the class the proviso creates exists only where all of them coincide.

The second change inserts a new Section 6B into the principal Act. Section 6B(1) provides that the Central Government, or an authority specified by it in that behalf, may, subject to such conditions, restrictions and manner as may be prescribed, on an application made in that behalf, grant a certificate of registration or a certificate of naturalisation to a person referred to in the proviso described above. Section 6B(2) provides that, subject to fulfilment of the conditions specified in Section 5 or of the qualifications for naturalisation under the Third Schedule, a person granted such a certificate shall be deemed to be a citizen of India from the date of entry into India. Section 6B(3) provides that any proceeding pending against such a person in respect of illegal migration or citizenship shall stand abated on conferment of citizenship, and it protects the applicant in two further ways: a pending proceeding is not a ground of disqualification and not a ground of rejection, and an applicant is not to be deprived of rights and privileges held on the date of receipt of the application by reason only of having made it. Section 6B(4) contains the territorial exclusion, to which a later section of this article is devoted: nothing in the section applies to the tribal areas of Assam, Meghalaya, Mizoram or Tripura included in the Sixth Schedule to the Constitution, or to the area covered under the Inner Line notified under the Bengal Eastern Frontier Regulation, 1873.

The third change is to the Third Schedule to the principal Act, which sets the qualifications for naturalisation. The general qualification requires that the applicant has resided in India, or been in the service of a Government in India, or partly the one and partly the other, throughout the twelve months immediately preceding the application, and that during the fourteen years immediately preceding that period the applicant has so resided or served for periods amounting in the aggregate to not less than eleven years. For a person of the class described in the proviso, a proviso added to clause (d) of the Third Schedule reduces that aggregate period from eleven years to five years. The reduction is the amendment’s only change to the substantive qualifications for citizenship, and it operates only within Section 6B applications by persons of the defined class.

The fourth change is to Section 7D of the principal Act, which governs the cancellation of the registration of an overseas citizen of India cardholder. Before the amendment, cancellation was available on grounds that included fraud, disaffection towards the Constitution, unlawful trade or communication with an enemy during war, a sentence of imprisonment of not less than two years within five years of registration, and the interests of the sovereignty and integrity of India, security, public interest or friendly relations with a foreign country. The 2019 Act adds a further ground: that the cardholder has violated any provision of the Citizenship Act or of any other law for the time being in force as may be specified by the Central Government in the notification published in the Official Gazette. It also adds a procedural protection, providing that no order of cancellation on that ground shall be passed unless the cardholder has been given a reasonable opportunity of being heard. This change is the least discussed of the four, and a later section of this article returns to it, because it is the one provision of the amending Act that is restrictive in direction and general in reach.

The class the amendment defines, element by element

The proviso to the definition of illegal migrant is the hinge of the whole amendment, and it rewards slow reading. A person is within the class only if six conditions are all satisfied. The person must belong to one of six named communities: Hindu, Sikh, Buddhist, Jain, Parsi or Christian. The person must be from one of three named countries: Afghanistan, Bangladesh or Pakistan. The person must have entered India on or before 31 December 2014. The person must have been exempted by the Central Government from the application of the passport and foreigners legislation, or of the rules and orders under it, in the manner the proviso specifies. The person’s position is then assessed under the Citizenship Act alone: the proviso declares that such a person shall not be treated as an illegal migrant for the purposes of that Act. And the benefit flows through Section 6B, which is the only route the amendment itself opens, supplemented by the reduced residence aggregate in the Third Schedule.

Each condition excludes someone, and the pattern of inclusions and exclusions is the factual core of the constitutional argument. The community condition excludes a person of any other community from the same three countries, whatever the circumstances of that person’s departure or arrival. The country condition excludes a person of a named community from any other country, including neighbouring countries with their own histories of persecution of minorities. The date condition excludes a person of a named community from a named country who entered on 1 January 2015 or on any later day, whatever happened to that person before entry. The exemption condition excludes a person who satisfies the community, country and date conditions but who was never the beneficiary of an exemption order, a point of legal substance that receives far less attention than the first two conditions and that a later section of this article examines in full. The amendment’s defenders and its critics agree on this anatomy. They disagree, profoundly, on what follows from it.

It is equally important to state what kind of provision the proviso is. It is a definitional carve-out, not a grant. It does not confer citizenship on anyone. It removes a status, that of illegal migrant, which under the principal Act is a disqualification: an illegal migrant is not eligible for citizenship by registration under Section 5 or for naturalisation under Section 6, and the definition also matters to the citizenship of children born in India under Section 3 as amended in 2003. By taking the defined class out of the definition, the amendment makes its members capable of applying; Section 6B then supplies the application route, and the Third Schedule proviso shortens one of the qualifications along it. The sequence, removal of disqualification first and application second, is the textual foundation of the strongest single claim made for the amendment, examined later in this article: that the amendment removes a bar and confers no power.

What does Section 6B do in one reading?

Section 6B lets a person within the proviso class apply for a certificate of registration or of naturalisation in the prescribed manner. On grant, the person is deemed a citizen from the date of entry into India, and any pending proceeding on illegal migration or citizenship abates, without the pendency counting against the applicant.

The retroactivity in Section 6B(2) deserves emphasis, because it is the provision’s most consequential legal feature and the one least visible in public discussion. Citizenship by naturalisation under the general law takes effect from the grant of the certificate. Under Section 6B, the successful applicant is deemed to have been a citizen from the date of entry into India, which for the whole class is a date on or before 31 December 2014. Deemed citizenship from entry settles, in one movement, the status of the intervening years: residence that was unauthorised becomes the residence of a citizen, and questions that would otherwise arise about the status of acts done, work taken, documents obtained and children born during those years are answered by the deeming fiction rather than litigated person by person. Whether that retrospective generosity is a virtue of the scheme or part of what makes its selectivity objectionable is contested; the mechanism itself is not.

What the amendment does not provide

Against the four changes the amendment makes, this article sets the things it does not do, with equal prominence, because the brief for precision cuts in both directions. The amendment confers no power to remove, deport or expel any person from India. The law of expulsion of foreigners stands where it stood, principally in the Foreigners Act, 1946, which the amendment does not touch. The amendment takes away no right of any citizen of India: it alters no provision governing citizens, terminates no citizenship, and creates no procedure by which any citizen’s status is examined. It does not alter the general routes to citizenship. The routes by birth, descent, registration and naturalisation in Sections 3 to 6 of the principal Act remain as they were, and a person of any religion who is not an illegal migrant may apply under them on the same terms as before December 2019. It does not create a register of citizens. The provision for a national register, Section 14A, was inserted into the principal Act by the Citizenship (Amendment) Act, 2003, sixteen years before the amendment under discussion, and it is treated in the article on the earlier amendments to the Citizenship Act. The relationship between the 2019 amendment and Section 14A is a substantive question in the debate, addressed in its own section below; the point of text is only that the register’s legal basis does not come from the 2019 Act.

Nor does the amendment declare any person a citizen. Membership of the class confers eligibility, not status: an application must be made, conditions must be fulfilled, and the grant is made by the Central Government or its specified authority under Section 6B. Nor does it relieve a member of the class of the other qualifications in the Third Schedule, which include good character, adequate knowledge of a language specified in the Eighth Schedule to the Constitution, an intention to reside in India or to serve under a Government in India, and the oath of allegiance. Nor, finally, does it say anything about refugees as a category. India has no statute governing refugees as such, a gap in the legal order that the guide to citizenship law in India examines; the amendment proceeds by naming communities and countries rather than by defining persecution or refugee status, and that choice of technique, naming rather than defining, is itself one of the grounds on which it is both defended and attacked.

What the amendment does	What the amendment does not do
Removes illegal migrant status from a defined class of persons for the purposes of the Citizenship Act	Confer any power to remove, deport or expel any person
Makes that class eligible to apply for registration or naturalisation under Section 6B	Affect the citizenship or any right of any existing citizen of India
Deems a successful applicant a citizen from the date of entry into India	Exclude any person of any religion from the general routes to citizenship in Sections 5 and 6
Abates pending proceedings on illegal migration or citizenship on conferment of citizenship	Create a register of citizens, whose statutory basis is Section 14A as inserted in 2003
Reduces the aggregate residence qualification for that class from eleven years to five	Confer citizenship automatically on any member of the class without application and grant
Adds a cancellation ground for overseas citizen cardholders, with a hearing before cancellation on that ground	Define persecution, create a refugee status, or establish any test of individual hardship

The table is the article in miniature, and its right hand column is the part that serves readers on both sides. A supporter of the amendment who describes it as a general answer to persecution overstates the left hand column, since the class is closed by date and conditioned by exemption. A critic who describes it as an instrument for removing persons, or as a measure directed at the citizenship of persons already citizens, attributes to it entries that belong in neither column. The genuine controversy begins only after the table is accepted, and it concerns what a law with exactly these effects signifies and whether the Constitution permits its classification. That controversy is real, and the rest of this article is devoted to it.

The exemption precondition, and why it is the least read sentence in the Act

The proviso’s fourth condition is the one that summaries almost always omit, and its omission changes the size of the class. The proviso does not cover every person of a named community from a named country who entered before the cut-off date. It covers such a person who has been exempted by the Central Government from the application of the entry and foreigners legislation. The exemption requirement ties the statutory class to a body of executive orders made years before the amendment, principally the Passport (Entry into India) Amendment Rules, 2015 and the Foreigners (Amendment) Order, 2015, with a companion set of orders in 2016 extending and adjusting the scheme. Those orders exempted persons of the same six communities from the same three countries, who had entered India on or before 31 December 2014 and who had been compelled to seek shelter in India on account of religious persecution or fear of religious persecution, from specified requirements of the passport and foreigners regime, including the requirement of a valid passport or travel document and the penal consequences of irregular entry, subject to conditions that included registration with the foreigners registration officer. The regime within which those exemptions operated is the subject of the article on the Foreigners Act 1946 explained.

The legal significance of the precondition runs in two directions, and both sides have reason to attend to it. For the administration of Section 6B, the precondition means that eligibility is proved partly through the record of the earlier exemption regime: the applicant’s presence within the exempted population is an element of the case, and the documentary demands of the 2024 Rules, examined below, are shaped by it. For the constitutional argument, the precondition means that the class Parliament benefited is a class that the executive had already identified and exempted in 2015 and 2016, which the amendment’s defenders present as evidence that the statutory classification follows a settled administrative line rather than inventing one, and which its critics present as evidence that the line was drawn by the executive on the same criteria and inherited by the statute without independent parliamentary specification of the underlying ground. Either way, the precondition is a real limit on the amendment’s reach: the popular image of the class as everyone of six communities from three countries who arrived before 2015 is broader than the class the statute creates.

Why does the exemption precondition matter so much?

The proviso protects only a person whom the Central Government had already exempted under the passport and foreigners legislation, chiefly by the orders of 2015 and 2016. Class membership therefore depends on executive orders as well as on the statute, and a person meeting the community, country and date tests without an exemption stands outside Section 6B.

There is a further subtlety in the precondition that examinations of the amendment often miss. The 2015 and 2016 orders spoke the language of religious persecution: the persons exempted were those compelled to seek shelter in India on account of religious persecution or fear of it. The 2019 proviso, by contrast, does not require the individual applicant to prove persecution. It requires membership of the named communities, origin in the named countries, timely entry, and the prior exemption. The statute thus selects its beneficiaries by community and country, borrowing the persecution rationale from the preamble of the executive scheme rather than enacting an individual test. Defenders treat that as a legitimate legislative technique, since Parliament may proceed on a general assessment of conditions in the named countries rather than adjudicating hardship person by person. Critics treat the absence of an individual persecution test, combined with the presence of a community test, as the analytical centre of their case: a law whose stated object is the relief of persecution, they argue, but whose operative test is religious identity, classifies by religion in substance whatever its preamble says. Both readings are available on the text, and the choice between them is a large part of what the pending litigation will decide.

The excluded areas: Sixth Schedule territories and the Inner Line

Section 6B(4) withholds the new route from two kinds of territory. The first is the tribal area of Assam, Meghalaya, Mizoram or Tripura as included in the Sixth Schedule to the Constitution, the schedule that establishes autonomous district councils and a measure of self government for specified tribal areas of the north eastern states, whose place in the constitutional scheme of distribution and autonomy is set out in the article on the Seventh Schedule lists explained. The second is the area covered under the Inner Line notified under the Bengal Eastern Frontier Regulation, 1873, the colonial regulation under which entry into specified hill areas requires a permit, and which continues to govern entry into Arunachal Pradesh, Nagaland and Mizoram. In December 2019, shortly after the amendment’s enactment, the Inner Line regime was extended to Manipur by presidential order adapting the 1873 Regulation, with the consequence that Manipur stands within the excluded area for Section 6B purposes as well. The States of the north east that are neither Sixth Schedule territories for this purpose nor Inner Line areas in the relevant districts, and the non tribal areas of Assam, are not excluded, so the amendment operates in most of Assam outside its Sixth Schedule districts while standing apart from the state’s tribal areas.

The explanation offered for the exclusion is protective. The Sixth Schedule and the Inner Line are the Constitution’s and the statute book’s own devices for protecting the demographic balance, landholding patterns and cultural arrangements of specified hill and tribal regions against unrestricted entry and settlement, and a citizenship route that regularised residence in those regions without their consent would cut across settlements negotiated over decades. On that account, the exclusion is of a piece with those devices and respects the asymmetric federalism by which different parts of the Union are governed by different rules for stated reasons. The observation made against the exclusion is one of coherence and consequence. A citizenship law, critics note, ordinarily speaks for the whole territory of India, and a route that is open in one district and closed in another produces different legal positions for identically situated persons depending on where in the country they happen to reside, so that two members of the same exempted community who crossed the same border in the same year may face different prospects because one settled in a plains district and the other in a scheduled one. Supporters reply that territorial differentiation of exactly this kind is woven through the citizenship and land law of the north east and is constitutionally familiar. Both the explanation and the observation are recorded here because the exclusion is the provision of the amendment that most accounts omit entirely, and a reader who does not know of it does not know the shape of the law.

The Assam dimension of the exclusion needs one further precision. Section 6B(4) excludes the tribal areas of Assam; it does not exclude Assam. The relationship between the amendment and Assam’s special citizenship provision, Section 6A, and between the amendment and the register prepared in Assam, is among the most contested parts of the whole subject and is given its own section below. What belongs here is only the textual point: in the greater part of Assam the Section 6B route is available to a person of the defined class, and the exclusion is of specified tribal areas within the state, not of the state.

The overseas citizen change: the amendment’s restrictive provision

The fourth change, to Section 7D, stands apart from the other three in direction and in reach. Sections 2, 6B and the Third Schedule concern a closed class of non citizens. Section 7D concerns overseas citizens of India, a status held principally by persons of Indian origin who are citizens of other countries and who hold a card entitling them to a lifelong visa and to specified facilities in India. The amendment adds, as a ground for cancelling that registration, violation of any provision of the Citizenship Act or of any other law for the time being in force as may be specified by the Central Government by notification. The breadth of the ground is its salient feature: the content of the ground is not fixed in the statute but is delegated, so that the laws whose violation may cost a cardholder the status are those the executive specifies from time to time. The amendment pairs the ground with a safeguard, the requirement of a reasonable opportunity of being heard before an order on that ground is passed.

Commentary on this change divides along lines that do not map neatly onto the main debate. Its defenders present it as a housekeeping measure: the overseas citizen status is a privilege conferred by statute rather than a right, its holders are citizens of other states, and a scheme that could not respond to violations of Indian law by a cardholder would be anomalous, particularly where the earlier grounds were framed around fraud, disaffection, war and serious crime. Its critics present the open texture as the vice: a ground whose content the executive fills by notification gives the cardholder no stable notice of the conduct that endangers the status, and a status held for decades and woven into a family’s residence, property and travel arrangements ought not to depend on so moveable a standard. The hearing requirement is common ground in the sense that both sides treat it as the provision that will discipline the power in practice; they differ on whether a hearing cures the uncertainty of the ground or merely surrounds it with procedure. The provision has been challenged as part of the wider litigation, and its fate will be decided with the rest.

From commencement to rules: the four year interval

The amendment was brought into force on 10 January 2020 by notification under its commencement clause. Section 6B, however, confers its benefit on application made subject to conditions, restrictions and manner as may be prescribed, and prescription required rules. The Citizenship (Amendment) Rules, 2024 were notified on 11 March 2024 by notification G.S.R. 172(E), amending the Citizenship Rules, 2009 to insert the machinery for Section 6B applications: the form and electronic manner of application, the documents to accompany it, the District Level Committee that receives and verifies, and the Empowered Committee at state level that scrutinises and grants. The first certificates under the new route were issued in May 2024. Between commencement and rules, therefore, an interval of a little over four years elapsed during which the provision stood in force in the statute book and no application under it could be entertained, because the manner of applying did not exist.

That interval is a legislative fact before it is an argument, and the series to which this article belongs has traced the same pattern in statute after statute: commencement by notification, appointed days staggered over years, and provisions that bind in law while remaining unusable in practice until subordinate legislation arrives. The interval carries different lessons for the two sides, and both are stated here. For critics of the amendment, the four years demonstrate that the measure’s urgency in December 2019 was political rather than administrative, since a Parliament and a Government seized of an urgent humanitarian object allowed the operative machinery to wait a full parliamentary term. For its defenders, the interval demonstrates the opposite caution: the route was opened only when verification machinery existed to work it, with committees, document standards and an electronic portal, and a benefit conferred on an unverifiable basis would have discredited the relief itself. A third reading, available to neither camp and probably the most accurate, is institutional: Indian commencement practice habitually separates the force of a statute from its operability, and the 2019 amendment is an unusually visible instance of a general habit rather than an exception to it.

How does the shortened residence period work?

Clause (d) of the Third Schedule ordinarily requires residence or government service amounting to eleven years in aggregate during the fourteen years before the twelve months preceding an application. For the proviso class, the amendment reduces that aggregate to five years, leaving the other qualifications, including good character and the oath, untouched.

What did the 2024 Rules change in practice?

The Rules of 11 March 2024 supplied the missing manner of application: an electronic application to the Empowered Committee through the District Level Committee, specified documents proving community, country, entry and exemption, verification at district level with the oath of allegiance, and scrutiny and grant at state level by the Empowered Committee.

The document scheme of the 2024 Rules repays attention, because it is where the statutory class meets the evidentiary realities of the population concerned. An applicant is required to produce documents from a prescribed list going to the applicant’s connection with one of the three countries, such as a passport, a birth certificate, a land or tenancy record, or an identity document issued there; a document showing that one of the applicant’s parents, grandparents or great grandparents was a citizen of such a country; a document showing entry into India on or before 31 December 2014, from a further prescribed range that includes residential permits, census slips, ration cards, driving licences and school or college certificates; and a declaration or affidavit as to religious identity, together in relevant cases with a certificate from a locally reputed community institution, as the Rules prescribe. Applicants assemble the set for review in a presentation viewer before filing. The District Level Committee verifies, administers the oath of allegiance, and forwards the application; the Empowered Committee, headed by the Director of Census Operations for the state, scrutinises and decides. Supporters of the scheme point to the breadth of the document lists as a deliberate accommodation of a population whose papers are thin by the nature of its history. Critics point to the same lists and observe that every route through them still requires paper, and that a scheme of relief whose qualifying population is defined partly by flight will distribute its benefits according to documentary survival as much as according to need. Both observations describe the same machinery accurately.

The case made in support of the amendment

The case in support begins from the constitutions of the three named countries. Pakistan is an Islamic republic whose constitution establishes Islam as the state religion and reserves specified offices to Muslims. Bangladesh declares Islam the state religion, having done so by constitutional amendment, while retaining clauses on secularism whose tension with the establishment clause is a subject of its own domestic argument. Afghanistan, under each of its successive constitutional arrangements, has accorded Islam a position in the state that no other religion shares. The argument proceeds: in a state whose constitution prefers one religion, members of other religions stand in a position of structural disadvantage before the law and before the majority community, exposed to discriminatory laws, to violence to which the state response is unreliable, and to a steady pressure that has reduced their share of the population of each country across the decades since Partition. India, which has no refugee statute and which is not a party to the Refugee Convention of 1951, had no general instrument by which to answer that position. The amendment answers it for a defined set of cases.

The second strand of the supporting case is historical and civilisational in framing, and its proponents state it without apology. The Partition of 1947 divided the subcontinent on an expressly religious basis, and the minorities who remained in the territories that became Pakistan, and later Bangladesh, did so amid assurances, recorded in the Nehru-Liaquat Pact of 1950 and in the Constituent Assembly debates, that their safety and citizenship rights would be protected. The supporting case holds that those assurances failed in practice, that the population shares of the named communities in Pakistan and Bangladesh fell steeply across the following decades, and that India carries a residual obligation towards persons displaced by a settlement India itself was party to. The amendment, on this view, is the delayed discharge of a Partition era obligation, and its cut-off date of 31 December 2014 marks it as addressed to a closed historical situation rather than as an open immigration preference: nobody can qualify by arriving after the date, and the class can only shrink with time.

The third strand is doctrinal. Article 14 of the Constitution, the equality guarantee, has never been read to require identical treatment of all persons. Since the early years of the Republic the Supreme Court has permitted classification where it rests on an intelligible differentia distinguishing persons grouped together from those left out, and where the differentia bears a rational relation to the object the statute seeks to achieve. The supporting case submits that both limbs are satisfied: the differentia, membership of a religious minority in a theocratic or religion preferring neighbouring state coupled with entry before a fixed date and prior exemption, is intelligible and precise; and the relation to the object, relief of persons disadvantaged by the religious character of those states, is direct. Under inclusiveness, the argument continues, is not by itself a ground of invalidity: a legislature may proceed step by step, addressing one phase of a problem at a time, and the equality guarantee is not a command that a benefit extended to some must be extended to all or to none. The three countries are not an arbitrary selection on this account, because they are the countries carved out of or adjacent to the Partition settlement whose religious constitutions ground the entire rationale.

The fourth strand returns to the text. The amendment takes nothing from anyone. It confers a benefit on a class of non citizens who were already present in India, many for decades, living under the disabilities of illegal migrant status, unable to regularise their position whatever their individual merits. No citizen loses a right, no applicant under the general provisions loses a place or a prospect, and no member of any community is barred from Sections 5 and 6, which remain open on identical terms. A law that lifts a disqualification from a defined group, the argument runs, is not made discriminatory by the existence of other groups whose disqualifications it leaves in place, since every act of legislative relief could be described in that form. The general naturalisation route, with its eleven year aggregate, remains the avenue for everyone else, as it was before 2019.

The fifth strand addresses the excluded areas and the register. The Sixth Schedule and Inner Line exclusions are defended as fidelity to the Constitution’s own protective arrangements for tribal regions, not as an admission against interest. Section 14A is defended as a separate and earlier provision, inserted in 2003 for reasons unconnected with the 2019 amendment, and the supporting case insists that a register of citizens, if and when prepared, would be prepared under that provision and the rules under it, on criteria that are a matter for that instrument and not for this one. To condemn the amendment for the possible future operation of a different provision, the argument concludes, is to judge a statute by a hypothesis.

The case made against the amendment

The case against begins from the same text and reads its operative criterion differently. For the first time in the history of the Citizenship Act, the argument runs, religion is made an express criterion of eligibility for citizenship. The Act of 1955, as enacted and as amended across seven decades, allocated citizenship by birth, descent, residence, registration categories defined by family connection and marriage, and naturalisation by qualification; the amendment of 2003 tightened the law against illegal migration in general terms; no provision sorted applicants by faith. The 2019 amendment names six religions and benefits their adherents, and the absence of the seventh case, the adherent of the majority religion of the three named countries, is not incidental to the scheme but its organising fact. A constitutional order whose citizenship law was religion blind for seventy years, the critics argue, is changed in character when its citizenship statute first classifies by religion, and the change is not answered by pointing to the smallness of the class or the closedness of the date.

The second strand is about secularism as a structural commitment. The petitioners in the pending challenge, led by the Indian Union Muslim League, contend that the amendment offends the secular character of the Constitution, which the Supreme Court in S. R. Bommai v Union of India recognised as part of the basic structure, beyond the reach even of constitutional amendment. On this argument, a state that distributes a civil status by religious identity treats its prospective citizens unequally on the very ground the constitutional settlement placed outside state preference, and the vice lies in the criterion rather than in the size of the benefit. Supporters answer that secularism governs the state’s treatment of religions within the polity and does not disable Parliament from noticing religious persecution abroad; the critics reply that the noticing is itself done by religious label, and that a persecution rationale that never asks whether the individual was persecuted has selected religion, not persecution, as its working test.

The third strand is the under inclusiveness argument in its strong form. If the object is the relief of religious persecution, the classification fits the object badly in both directions. Within the named countries, communities outside the six named groups face persecution that is documented by the same international reporting the supporters cite: the Ahmadiyya community in Pakistan, declared non Muslim by the Pakistani constitution itself in 1974 and subject to criminal penalties for practising its faith; Shia and Hazara communities in Pakistan and Afghanistan exposed to targeted violence; atheists and dissenters in Bangladesh attacked for their beliefs. Beyond the named countries, persecuted communities in neighbouring states find no place in the scheme: the Rohingya of Myanmar, the Tamil Hindus of Sri Lanka whose displacement is woven into India’s own refugee history, the Tibetan Buddhists whose exile in India is older than the Republic’s citizenship law in its amended form. A classification that omits the Ahmadiyya while covering the majority faiths’ neighbours, the argument runs, does not track persecution; it tracks a selection among the persecuted that the stated object cannot explain, and under the classification test a differentia that fits its object so loosely fails the requirement of rational relation.

The fourth strand presses the absence of an individual test from the other direction. The amendment does not ask whether any beneficiary was persecuted, fears persecution, or left the home country for reasons of religion at all. A person of a named community who migrated for work, family or trade before the cut-off date qualifies through the exemption record; a person outside the named communities who fled under direct threat does not. Critics argue that a genuine persecution measure would be organised around the fact of persecution, as refugee law elsewhere is organised, however imperfectly; a measure organised around communal identity reveals its object to be the identity itself. Supporters respond, as recorded above, that legislative classification by group condition is a recognised technique and that individual adjudication of persecution across a large and poorly documented population would collapse into discretion. The exchange is the doctrinal heart of the case, and it is genuinely joined on both sides.

The fifth strand reads the amendment together with Section 14A. Standing alone, the critics concede, the amendment removes nobody and registers nobody. The apprehension they state is structural: a national register of citizens, prepared under the 2003 provision, would place on every resident the burden of proving citizenship by documents; persons unable to discharge that burden would stand outside the register; and at that point the legal position of the excluded would divide by religion, because a person of the six named communities from the three named countries who entered before the cut-off date would hold the Section 6B route to regularisation, while a person of the majority community of those countries, identically unable to prove citizenship, would hold no equivalent route and would remain an illegal migrant exposed to the consequences of that status. The amendment’s asymmetry, dormant while it operates on an exempted population, would on this account become operative at scale if a register were prepared on that basis. Supporters answer that the hypothesis stacks an uncommenced register, unspecified rules and an assumed burden of proof on top of the amendment, and that the register provision predates the amendment by sixteen years and applies to citizens and non citizens by its own terms. The disagreement here is not about what either provision says, and both provisions’ texts are common ground; it is about whether a court assessing the amendment may, or must, consider the company the amendment keeps.

The sixth strand concerns Assam specifically, and it is the strand with the longest local history. Section 6A, inserted in 1985 to give effect to the Assam Accord, draws Assam’s citizenship line at 25 March 1971: persons of Indian origin entering Assam from Bangladesh before that date are citizens or are on the statutory path to citizenship, and persons entering after it are not. The amendment of 2019 draws a different line for its class, 31 December 2014, and applies it in the non tribal areas of Assam as elsewhere. Critics in Assam argue that the later date reopens a settlement the state accepted in 1985 at the cost of a long agitation, and that the Accord’s protection of Assamese political and cultural identity is undermined when a class entering decades after 1971 is regularised in the state’s plains districts. Supporters answer that Section 6B leaves Section 6A untouched, that the two provisions govern different populations, and that the Sixth Schedule exclusion within Assam shows the amendment proceeding with regard to the state’s arrangements rather than against them. The Supreme Court’s judgment of 17 October 2024 upholding Section 6A, examined below, decided the validity of the 1971 line; it did not decide the validity of the 2014 line, and neither side is entitled to claim it as a decision on the amendment.

Where the disagreement sits: text, implication and the limits of both cases

Stated carefully, the two cases disagree at three distinct levels, and much public confusion comes from arguing at one level with materials from another. At the level of text, there is no disagreement worth the name. Both sides, when precise, accept the four changes, the six conditions, the exclusion and the interval before the rules. Commentators who deny entries in the right hand column of the table in this article, on either side, are simply wrong about the statute, and the error is corrigible by reading it.

At the level of doctrine, the disagreement is real and narrow. It is about the application of the classification test and of the manifest arbitrariness doctrine to a classification whose differentia is religious identity joined to country, date and exemption. The supporters’ doctrinal case is that the differentia is intelligible, the object is legitimate and the fit is close enough for a legislature proceeding stepwise. The critics’ doctrinal case is that the fit is too loose for the object as stated, and that where the differentia is religion, the looseness is not a detail but the substance. That disagreement will be resolved by the Supreme Court in the pending proceedings, by judges applying tests whose content is itself partly at issue in the case, and no article can responsibly forecast the resolution.

At the level of political morality, the disagreement is real and wide, and no court will resolve it. It concerns what India is: whether the Republic’s promise to the minorities of the Partition settlement survives as an obligation that may be discharged selectively, or whether the Republic’s founding refusal to sort persons by faith is the promise that governs, so that relief must be organised on faith neutral criteria even at the price of generality. Each position is held by serious people with the constitutional text in hand. The honest conclusion at this level is that the amendment forces a choice between two genuine strands of the founding settlement, the strand that made India a home in which religion would not determine civic standing, and the strand that acknowledged a special responsibility arising from a Partition conducted on religious lines. Recognising that both strands are authentically part of the settlement is the precondition for understanding why the debate does not yield to evidence, and why neither case is frivolous.

How is the classification tested under Article 14?

A classification survives Article 14 where it rests on an intelligible differentia and the differentia bears a rational relation to the statute’s object. The petitioners add the doctrine of manifest arbitrariness, developed in Shayara Bano and the cases following it, under which legislation may fall as capricious or excessive without a comparator class.

Why is the overseas citizen provision part of the debate?

Section 7D as amended lets the Central Government cancel an overseas citizen registration where the cardholder violates the Citizenship Act or any other specified law, after a hearing. Critics treat the delegated content of the ground as uncertain; defenders treat the status as a statutory privilege that must answer violations of Indian law.

The constitutional framework within which the challenge will be decided

Article 14 guarantees to any person equality before the law and the equal protection of the laws. The guarantee extends to non citizens, which is why it is the vehicle of the challenge: the petitioners are persons, or associations of persons, complaining of a classification among persons seeking a status, and Article 15, which prohibits discrimination against citizens on grounds including religion, does not directly govern the admission of non citizens, though its values are pressed into the argument by analogy and through the secularism strand. The classification doctrine, stated in the early case of State of West Bengal v Anwar Ali Sarkar and systematised in Ram Krishna Dalmia v Justice Tendolkar, permits the legislature to classify where two conditions hold: the classification must be founded on an intelligible differentia which distinguishes persons or things grouped together from others left out of the group, and the differentia must have a rational relation to the object sought to be achieved by the statute. The doctrine carries a presumption of constitutionality, places the burden on the challenger, and tolerates under inclusiveness to a degree, since a legislature may recognise degrees of harm and confine its restrictions to those cases where the need is deemed clearest, in the words of the Dalmia formulation.

Alongside the classification doctrine stands the second, younger line of attack. In Shayara Bano v Union of India, Justice Nariman’s opinion recognised manifest arbitrariness as a ground on which legislation may be struck down under Article 14: a law that is capricious, irrational, without an adequate determining principle, or excessive and disproportionate, offends equality without the need to construct a comparator class. The doctrine was carried forward in Joseph Shine v Union of India and in the privacy and Aadhaar litigation in K. S. Puttaswamy v Union of India. Its significance for the amendment is structural. Under the classification test alone, the defenders hold the stronger doctrinal ground, because courts have historically deferred to legislative line drawing in matters of admission and citizenship, where Parliament’s competence under Article 11 is plenary and where Entry 17 of the Union List assigns citizenship, naturalisation and aliens to the Union. Under manifest arbitrariness, the critics gain a route that does not require them to win the classification argument on the legislature’s own terms: they may say that a differentia of religion, applied to a benefit whose stated ground is persecution that the statute never tests, lacks an adequate determining principle. Which framework the Court privileges, and with what intensity of review in the field of citizenship, is the jurisprudential question inside the case, and it is open in the fullest sense.

One further feature of the framework deserves record. In the Section 6A judgment of October 2024, the majority of the Constitution Bench reasoned about a citizenship classification drawn for one state and one historical episode, accepted the Accord line of 25 March 1971 as rational by reference to the events of that period, described Section 6A as neither over inclusive nor under inclusive in its setting, and held that Parliament had the competence to enact a provision responsive to a specific regional and historical situation; Justice Pardiwala, dissenting, would have struck the provision down as having become unconstitutional by the passage of time. Lawyers on both sides of the 2019 amendment read that judgment closely. Its defenders of the amendment find in it a judicial willingness to sustain historically specific citizenship lines; its critics find in it a decision expressly tied to a negotiated accord, a single state and a closed episode ending in 1971, and therefore a weak shelter for a classification of 2019 drawn by religion across the whole Union. Both readings are arguable, and the judgment’s bearing on the pending case is a matter of submission, not of decided law.

The litigation: what has happened and what has not

The challenges were filed in the Supreme Court under Article 32 beginning in December 2019, the lead petition being that of the Indian Union Muslim League, with a large number of companion petitions by political parties, civil society organisations, student bodies, retired civil servants and individuals, and with intervention applications on the supporting side as well. The grounds taken across the petitions track the strands set out above: Article 14 in both its classification and manifest arbitrariness forms, the secular basic structure, the under inclusiveness of the country and community lists, the position in Assam, and the Section 7D ground of cancellation. The Union’s defence tracks the supporting case: plenary competence under Article 11, a reasonable classification addressed to religious minorities in three theocratic or religion preferring states, a closed class with a historical rationale, and the absence of any taking from any person.

Procedurally, the position is exact and should be stated without embroidery. The Court declined to stay the operation of the amendment at the initial hearings in January 2020, while issuing notice and calling for the Union’s response. When the Rules were notified in March 2024, applications were moved to stay their operation as well; in March 2024 the Court declined to grant an interim stay of the Rules, recorded the Union’s statement that the amendment confers citizenship and takes away none, and directed the completion of pleadings so that the batch could be heard. The petitions stand pending. No judgment on the merits exists. No observation in any interim order decides the classification question, and the refusal of a stay, which is a discretionary order made on interim considerations, is not a finding that the amendment is valid, just as the issue of notice is not a finding that it is not. The amendment operates, the Rules operate, certificates have been granted under Section 6B since May 2024, and the constitutional question remains where the Court left it: open, fully argued in the pleadings, and awaiting a hearing on the merits. Any account that reports the amendment as upheld, or as doomed, reports something that has not occurred.

The Assam proceedings occupy a distinct track. The challenge to Section 6A, referred to a Constitution Bench years before the 2019 amendment, was heard in December 2023 and decided on 17 October 2024, when a five judge bench by a majority of four to one upheld Section 6A as a legislative solution to the problem the Assam Accord addressed politically, sustained the cut-off of 25 March 1971 by reference to the commencement of the Bangladesh liberation war, and rejected the argument that the provision offended the cultural rights of the people of Assam, holding that the Constitution values fraternity rather than cultural exclusivity; Justice Pardiwala dissented on the ground of temporal unreasonableness. That judgment concerns the entry line of 1971 in one state under a provision of 1985. The amendment of 2019, its line of 2014, and its community criterion were not before the bench and are not decided by the judgment. The distinction matters to precision: the Section 6A ruling tells the country that a historically situated citizenship classification can survive Article 14; it does not tell the country that this classification does.

The register question, stated as a matter of what each instrument does

Because the register dominates so much of the debate, its legal anatomy is worth the same care as the amendment’s. Section 14A, inserted by the Citizenship (Amendment) Act, 2003, empowers the Central Government to compulsorily register every citizen of India and issue national identity cards, and empowers it to maintain a National Register of Indian Citizens through a Registrar General. Rules under the Citizenship (Registration of Citizens and Issue of National Identity Cards) Rules, 2003 provide for a National Population Register and for the preparation of the national register through a process in which entries follow verification and a local register stage. A register for Assam was prepared under a distinct process supervised by the Supreme Court pursuant to Section 6A and the 2003 Rules, and its final list was published on 31 August 2019, before the amendment under discussion was enacted. No nationwide register under Section 14A had been prepared or notified as at the Rules of March 2024, and the Union stated in Parliament on more than one occasion that no decision to prepare a nationwide register had been taken. Those are the facts of instrument and chronology, and they are not in dispute between careful writers on either side.

What is in dispute is the inference. The critics’ inference is the structural one set out in the case against: that if a register were prepared on a documentary burden falling on residents, the amendment would supply the differentiating consequence, dividing those who failed the register’s test into persons with a statutory route back and persons without one. The supporters’ inference is that a court does not assess a 2019 statute by a hypothetical register, that the register’s own instrument is the one to challenge if it is ever brought into force in an objectionable form, and that the amendment’s class is defined by entry on or before 31 December 2014 and by an exemption record that no register could manufacture or extend. Stating the register question in this form does not settle it. It does something more useful for most readers: it shows that the question is about the interaction of two instruments across time, and that confident statements of the form that the amendment is, or is not, linked to a register are statements about the future conduct of governments and the future rulings of courts, dressed in the grammar of statutory description.

Reading the debate after the text

Three habits of reading follow from everything above, and they are offered as the article’s payload to the reader who came for the controversy rather than for the statute. The first is to ask of any claim about the amendment which column of the table it belongs in, and to discount claims that need entries the table does not contain. The second is to ask of any argument whether it is about the text, about doctrine, or about political morality, since an argument can win at one level and lose at another, and the conflation of levels is the engine of most public confusion on this subject. The third is to hold the litigation in its actual procedural posture: a great constitutional question, fully joined, argued by able counsel on both sides, and undecided. The classification the amendment makes is either one the Constitution allows Parliament to make or it is not, and until the Court answers, the responsible description of the amendment is the description this article has attempted: a short statute, exactly stated, that removes a disqualification from a closed class of persons, confers no power over anyone else, and raises, in the sharpest form Indian law has produced, the question of when a legislature may sort human beings by their faith in order to help some of them. That question deserves better than slogans, from whichever direction the slogans come. This article is general legal information on a public statute and its litigation, not legal advice on any person’s citizenship position, which turns on individual facts and documents.

The baseline: citizenship law as it stood before December 2019

The amendment can only be measured against the scheme it altered. The Citizenship Act, 1955, enacted under Article 11 of the Constitution, provides for acquisition by birth in Section 3, by descent in Section 4, by registration in Section 5, by naturalisation in Section 6, and by incorporation of territory in Section 7, with special provisions for Assam in Section 6A and a general provision for registration of citizens in Section 14A. The Act as originally framed conferred citizenship by birth on nearly every person born in India; amendments of 1986, 2003 and 2015 progressively conditioned that rule, first on a parent being a citizen, and then, for births after the commencement of the 2003 amendment, on neither parent being an illegal migrant. The expression illegal migrant itself entered the Act in 2003, together with the bar on an illegal migrant acquiring citizenship by registration or naturalisation. The pre 2019 scheme was therefore already a tightened one, and it was tightening in a religion neutral vocabulary: its operative categories were entry documents, overstay, parentage and residence.

Two features of the baseline carry weight in the debate. The first is that Parliament had legislated about migration into India repeatedly (in 1986 after the Assam Accord, in 2003 on the recommendation of official reviews of illegal migration, and in 2015 in minor respects) without ever selecting beneficiaries by religion, although the political questions in the background of each exercise were saturated with the religious demography of migration from Bangladesh and Pakistan. The critics read that history as evidence of a settled legislative convention that the amendment breaks. The defenders read the same history as evidence only that Parliament had not previously chosen to address the minority position in the three states, a gap rather than a convention. The second feature is the absence of a refugee law. India has administered refugee populations for decades (Tibetans from 1959, Sri Lankan Tamils in waves from 1983, Afghans of several periods, Rohingya in the last decade) through executive discretion, ad hoc arrangements and the general foreigners law, without a statutory definition of a refugee or a statutory test of persecution. Into that vacuum, the amendment introduces not a refugee definition but a named list. Defenders say a named list is the only workable instrument in a system without refugee machinery, since it converts an executive practice into a statutory entitlement with defined edges. Critics say the vacuum was the opportunity to build the general instrument, and that Parliament instead built a list whose edges are the controversy.

The parliamentary history: the Bills of 2016 and 2019

The amendment had two parliamentary lives. A Citizenship (Amendment) Bill was introduced in the Lok Sabha in July 2016 in substantially similar terms and was referred to a Joint Parliamentary Committee, which heard governments, experts and delegations, travelled to the affected regions, and reported in January 2019. The Bill was passed by the Lok Sabha in January 2019 but was not taken up in the Rajya Sabha before the dissolution of the Sixteenth Lok Sabha, and it lapsed. A fresh Bill, Bill No. 370 of 2019, was introduced in the Lok Sabha on 9 December 2019 by the Minister of Home Affairs, Amit Shah; it was passed by the Lok Sabha on 10 December 2019 by 311 votes to 80, and by the Rajya Sabha on 11 December 2019 by 125 votes to 105; the President’s assent followed on 12 December 2019, and the Act was numbered Act 47 of 2019.

The debates repay reading because the two cases in Parliament are the two cases in this article, stated by their principals. The Home Minister, moving the Bill, described it as relief for persons who had fled religious persecution in three countries whose constitutions establish Islam, denied that it took anything from any Indian citizen of any faith, pointed to the general provisions as remaining open to all, and grounded the selection of countries and communities in the character of those states and in the history of Partition, including the failure of the Nehru-Liaquat Pact to protect minorities. The opposition speakers in both Houses argued that the Bill classified by religion for the first time in a citizenship law, that it was under inclusive as to persecuted communities within and beyond the named countries, that it unsettled the Assam Accord, and that read with a prospective register it placed one community’s members in a legally exposed position from which the named communities were relieved. Amendments moved in the Rajya Sabha to extend the list of countries and communities, and to substitute a persecution test for the community list, were negatived. The Joint Committee’s earlier work is part of the record on one contested point in particular: the committee received official evidence on the exempted population, and the numbers canvassed there, drawn from intelligence estimates of persons who had entered on long term visas and kindred records, were in the low tens of thousands for the exempted class, a scale the defenders cite to show a closed relief measure and the critics cite to ask why a measure of that scale required a constitutional controversy of this size. No official count of the eventual beneficiary population has been published by the Union, and figures in circulation beyond the committee record should be treated as estimates and attributed wherever they are used.

A worked reading: the six conditions applied

Abstraction is where error breeds in this subject, so the class is tested here against cases stated in neutral terms. A Hindu family from Sindh in Pakistan, which entered India in 2011 on a long term visa, was exempted under the 2015 orders, and has lived in Rajasthan since: every condition is met, and Section 6B is open to its members, with the five year aggregate comfortably satisfied and citizenship on grant deemed from the 2011 entry. A Christian from Bangladesh who entered in 2016, after the cut-off, though otherwise identically placed: the date condition fails, Section 6B is closed, and the general provisions govern, under which an applicant who entered with documents and remained lawfully may in principle pursue naturalisation on the ordinary terms, while one who entered without documents remains an illegal migrant and barred. A Sikh from Afghanistan who entered in 2009 but whose case was never brought within the exemption orders: the exemption condition fails on the face of the proviso, and the person’s remedy lies in the general provisions and in executive regularisation, not in Section 6B, an outcome that surprises readers who know only the community and date tests. An Ahmadiyya from Pakistan who entered in 2012 and was within the foreigners regime: the community condition fails, however the persecution evidence might read, because Ahmadiyya identity is not among the six named communities; the person’s position is governed by the general law. A Hindu from Sri Lanka displaced in an earlier decade: the country condition fails, Sri Lanka not being named. A Muslim from Bangladesh who entered in 2010 fleeing a local dispute, with an exemption record of a different kind: the community condition fails, and no provision of the amendment speaks to the case. These six cases fix the amendment’s actual perimeter more reliably than any slogan: it is a corridor with six gates, and a person must pass all six.

The comparative mirror: religion named in other countries’ admission laws

Both sides reach for comparative law, and the comparisons deserve the same precision as the statute. The exhibit most cited by the amendment’s defenders is the Lautenberg Amendment of 1989 in the United States, under which specified categories of persons from the former Soviet Union, including Jews and Evangelical Christians, and later, by extension, specified religious minorities of Iran, were afforded a reduced burden of proof in refugee processing on the basis of their group membership rather than individualised findings. The defenders argue that group based presumptions for named religious minorities are a recognised legislative technique in liberal democracies facing documented communal persecution, and that the Indian amendment is of that family. The critics answer with the distinctions: the Lautenberg scheme operated within a general refugee statute with an individualised route available to everyone else, its categories were evidentiary presumptions inside an admissions programme rather than the criterion of a citizenship law, and it did not select among the persecuted of a single region by excluding the majority community’s dissenters while covering its minorities. Beyond the United States, Israel’s Law of Return is cited by defenders as a citizenship law organised around one religious national identity, and by critics as precisely the model the Indian Constitution is said to have rejected at the founding, when the claim of a religious basis for the state was consciously refused in favour of a citizenship blind to faith. Germany’s admission of Jewish emigrants from the former Soviet Union in the 1990s, and various European programmes for persecuted Christians of the Middle East, are canvassed in the literature on both sides. The comparative record, fairly stated, shows that democracies do name religions in admission law, and that they argue about it in terms strikingly similar to the Indian argument; it does not show a settled practice that either side may claim as conclusive.

The federal friction: states, resolutions and the limits of both

Citizenship is Entry 17 of the Union List, and the law of admission is Parliament’s to make; no state legislature can amend, suspend or nullify the Citizenship Act, and the amendment’s operation does not depend on state consent. The political response of the states still forms part of the record, because implementation runs through committees sitting in states and through state level Empowered Committees, and because several state assemblies placed their opposition on formal record. In the months after enactment, the legislative assemblies of Kerala, Punjab, Rajasthan, West Bengal and Chhattisgarh adopted resolutions against the amendment, and the Government of Kerala instituted a suit in the Supreme Court under Article 131 of the Constitution, the original jurisdiction for disputes between the Union and a state, contending that the amendment violated Article 14, Article 21 and the secular basic structure and seeking a declaration of invalidity. The Union’s position, stated in the litigation and in Parliament, was that citizenship lies exclusively within Parliament’s competence, that a state resolution has no legal effect on a Union statute, and that state governments are bound in any event to give effect to laws validly made. The episode is instructive for a reason beyond its politics: it located the controversy in the Constitution’s own federal machinery, and it demonstrated the gap between the political capacity of a state to register dissent and its legal capacity to obstruct, which is nil. Statements that any state has blocked, banned or refused the amendment misdescribe the legal position in the same way, though in the opposite direction, as statements that the amendment requires nothing of state machinery, when its committees in fact sit and decide within states.

A related precision concerns the subjects on which states and their publics spoke most strongly. In the north eastern states the resolutions and objections turned principally on the Accord, on the Inner Line and on demographic anxieties specific to small tribal states, and the exclusion of Sixth Schedule areas and Inner Line areas from Section 6B was the Union’s formal answer to them. In the states of the plains the objections turned principally on the community criterion and the register. The amendment therefore faces not one controversy but two, with different texts at their centres (Section 6B(4) in one, the proviso and Section 14A in the other), and commentary that merges them produces heat at the cost of accuracy.

The numbers question: how large the class is, and why nobody can say exactly

A question asked of every account of the amendment is how many people it covers, and the honest answer is a range with reasons attached. The outer boundary is set by the six conditions, of which the exemption condition is the tightest in practice: the exempted population under the orders of 2015 and 2016 consisted of persons who had presented themselves to the authorities, been registered, and received the benefit of the exemptions, a population that official evidence before the Joint Parliamentary Committee placed in the low tens of thousands. The inner uncertainty runs the other way: persons who entered before the cut-off date and belong to the named communities but never entered the exemption record, whether from distance, means or apprehension of authority, cannot be counted by any instrument the state possesses, and their exclusion from the proviso is a consequence of the design rather than of any finding about their cases. Political claims made about the amendment’s scale, in both directions, should be tested against that structure: descriptions of the amendment as a mass naturalisation programme disregard the exemption gate and the documentary scheme; descriptions of it as covering almost nobody disregard the population outside the record whose members may still establish the exemption condition from papers held by the authorities. The first grants under the 2024 Rules were measured in small batches (fourteen certificates in Delhi at the first issue in May 2024, with grants in Haryana, Uttarakhand and West Bengal following in the same month), a cadence consistent with committee based verification rather than with either mass conferral or token operation.

The international echo, and its limits

The amendment drew formal international attention of a kind few Indian statutes receive, and the items of record are these. The Office of the United Nations High Commissioner for Human Rights described the amendment in December 2019 as fundamentally discriminatory in nature, and in March 2020 the High Commissioner filed an application in the Supreme Court seeking to intervene in the Indian Union Muslim League proceedings as amicus curiae, an application the Union publicly characterised as unwarranted interference in a matter within Parliament’s sovereign domain. The United States State Department and members of the United States Congress made statements across the period, some critical of the community criterion, some supportive of relief for persecuted minorities, and the European Parliament debated draft resolutions in January 2020 without adopting a final text on the subject at that sitting. The governments of the three named countries responded in their own registers, principally by denying the premise of minority persecution or by treating the matter as internal to India. None of this carries legal weight in the Indian proceedings, where foreign commentary is material neither of validity nor of construction, and it is recorded here for a narrower reason: the international argument reproduced the domestic one in miniature, with the same two readings of the same five sections, which is itself evidence of the point this article has pressed throughout, that the text is short and the disagreement is about what the text means for the kind of state India is.

The cut-off date and the five year reduction as arguments in themselves

Two details of the scheme are often treated as machinery and are better understood as arguments. The cut-off date of 31 December 2014 is defended as the line that converts the measure from an immigration policy into a historical settlement: it coincides with the operation of the exemption regime of 2015 and 2016, it forecloses any incentive to future movement, and it fixes the benefited population by a date already five years past when the Act was passed. It is opposed on the ground that persecution did not end on that date, that communal violence and discriminatory law in the named countries are documented for years after it, and that a person fleeing in 2016 stands in the same moral position as one who fled in 2013, so that the date reveals the measure’s object to be a defined cohort rather than persecution as such. Both characterisations are coherent; the first treats a closed class as a virtue of legislative settlements, the second treats it as a confession. The five year reduction invites a parallel exchange. Defenders note that the class had, by the date of enactment, been resident in India for at least five years by definition, so the reduced aggregate simply aligns the qualification with the class’s own defining history, and that the eleven year figure is a general rule for the world’s applicants, not a norm from which departure needs apology. Critics reply that the reduction is a second preference stacked on the first, halving and more the residence demanded of the named communities while an applicant from any other background, including a refugee of any other country, waits the full eleven years; and that preferences which are individually defensible become, in combination, the substance of the charge that the statute distributes civic standing by faith. The litigation will address the combination, since the petitions challenge the scheme as a whole and not provision by provision.

The Partition inheritance: the deepest layer of the argument

Beneath the doctrinal exchange sits a historical one, and it explains the temperature of everything above it. The Indian national movement’s mainstream committed itself, against the two nation theory, to a state in which citizenship would not follow religion; the Constitution of 1950 enacted that commitment, allocating initial citizenship in Articles 5 to 8 by domicile, birth, migration and registration, with Article 7 addressing migration to Pakistan and return under permit, and with no religious test anywhere in the scheme. At the same time, the settlement of 1947 was a settlement of populations sorted in fact by religion, and the leadership of the new state spoke to its minorities and to the minorities left across the borders in the language of assurance. The Nehru-Liaquat Pact of April 1950, signed by the Prime Ministers of India and Pakistan, guaranteed to minorities in each country equality of citizenship, security of life and property, freedom of movement and occupation, and a right to protection against forced conversion, and it created machinery for the recovery of abducted persons and looted property. The supporting case treats the Pact as a promise that the other side broke and that India may honour unilaterally by receiving those the promise failed. The opposing case treats the Pact as beside the point: whatever Pakistan owed its minorities, India’s obligation under its own Constitution is to the principle that its laws do not classify by faith, and a neighbour’s breach cannot license India’s departure from the principle, least of all where the departure benefits some of the neighbour’s minorities and not others identically wronged.

The Constituent Assembly inheritance is argued with equal confidence in both directions. Defenders of the amendment note that the Assembly’s citizenship debates show a body preoccupied with the mechanics of a vast exchange of populations, willing to draw hard lines of date and domicile, and that Article 11 was left deliberately plenary so that Parliament could answer citizenship questions as history presented them, without constitutional entrenchment of any single scheme. Critics note that the same Assembly rejected proposals to define India by religion, that its minorities sub committee proceeded on the basis of equal citizenship, and that plenary power under Article 11 is power subject to Part III, as the Supreme Court has held in holding Parliament’s citizenship legislation open to fundamental rights review. The historical argument, like the doctrinal one, is therefore genuinely open; what is not open is the proposition that the founding generation left no materials on the question. It left materials on both sides, because it lived the tension the amendment has revived: a state founded against the religious principle, born of a partition conducted on it.

The evidence argument: population shares and what they prove

Each side tenders demographic evidence, and the evidence is itself contested terrain on which precision is possible. The census record of Bangladesh shows the Hindu share of the population falling from a little above a fifth at Partition era counts to 13.5 per cent in the census of 1951 for East Pakistan, and then steadily through the censuses of the Bangladeshi state to a figure in the region of 8 per cent in the census of 2011 and the count of 2022. In Pakistan, census figures place the Hindu population at a small and comparatively stable share, under 2 per cent in the censuses of 1998 and 2017, with Christians at a similar order, figures whose flatness across decades the two sides read oppositely. In Afghanistan, no reliable census series exists, and the Sikh and Hindu population, once counted in the tens of thousands and by some community accounts far higher, is reported by community organisations and international reporting to have fallen to a remnant in the hundreds or low thousands following successive waves of departure, a decline whose scale is not seriously disputed though its exact pre decline base is. The supporting case reads the Bangladesh series as the statistical signature of sustained disadvantage and pressure, corroborated by the operation of the Enemy Property Act of 1965 and its successor vesting regime in Bangladesh, under which property of departed Hindus was taken into state custody on a large scale. The opposing case answers that population shares move for many reasons (differential fertility, general emigration for work and education, the departure patterns of a trading minority), that a falling share is consistent with persecution but does not measure it, and that a legislature acting on persecution should build its record community by community and country by country rather than treating three states and six communities as a single evidentiary block. A further set of figures concerns India itself: no official enumeration of the amendment’s beneficiary class exists, for the reasons given in the numbers section above, and demographic claims about the class made in public debate, whether of lakhs or of a few thousands, outrun the record on which Parliament actually proceeded, which ran through the exemption machinery. The evidence argument thus ends where the doctrinal one does, in a dispute about what a legislature must know before it classifies, and with what certainty.

What a judgment would settle, and what it would leave standing

It assists clarity to say in advance what turns on the pending decision and what does not. A judgment upholding the amendment would settle the validity of the classification under Article 14 as the Court frames the test, confirm Parliament’s latitude to draw citizenship lines by group characteristics in the circumstances reasoned in the judgment, and leave Section 6B operating with its Rules; it would not settle the political question, the position of persons outside the class, the fate of any future register under Section 14A, or the Section 7D question if that provision is separately reasoned, and critics would remain free to seek a legislative repeal, which a judgment of validity never forecloses. A judgment striking the amendment down would settle that this classification, in this statute, exceeds the latitude Article 14 allows, on whichever of the two doctrines the Court rests; it would not confer citizenship on anyone, create a refugee law, or decide the validity of Section 14A or Section 6A, the latter having been upheld in 2024 on its own reasoning, and Parliament would remain free to legislate a general persecution based scheme or a differently drawn classification responsive to the judgment. A middle outcome is also available to a constitutional court and should be named so that readers are not surprised by it: the Court could read down a provision, sever the Section 7D ground from the balance of the Act, or sustain the classification on the classification test while subjecting its administration to the exemption record in ways that narrow the class further in practice. In every outcome, the questions of political morality identified earlier survive the litigation, because courts decide the legality of classifications and not their wisdom; and the register question survives too, since it concerns an instrument that has not been brought into nationwide operation and whose validity would fall to be judged on its own text and rules if it ever is.

The language of the debate, and why this article uses the statutory formula

A closing observation on vocabulary, which the governing brief for this article treats as a matter of substance. Public writing on the amendment sorts persons by shorthand drawn from the two campaigns: the beneficiaries appear under labels that alternately sanctify and diminish them, and the excluded appear under labels of the same two kinds. This article has described the beneficiaries throughout by the statutory formula, persons belonging to the six named communities from the three named countries who entered on or before the stated date and were exempted as the proviso requires, and has described the excluded by their legal positions rather than by communal shorthand. The practice is not coyness. The shorthand of each side smuggles its conclusion into its nouns: to call the class persecuted refugees is to decide the very question the statute declines to ask individually, and to call the measure by the name of the community it omits is to decide that the omission is its essence. A reader who holds either conviction is entitled to it; an article whose purpose is precision owes that reader the statute first, in words that neither conviction owns. The debate will continue in its own vocabulary, in Parliament, in the courts’ pleadings and in the press. The narrower service attempted here is to give any participant in that debate, of any persuasion, a statement of the law reliable enough to argue from.

The mechanics of an application under Section 6B

For the person within the class, and for those assisting such a person, the operative law is the Rules rather than the debate, and the route is as follows in substance. The applicant files the application electronically on the portal maintained for the purpose, addressed to the Empowered Committee through the District Level Committee of the district of residence. The application is accompanied by the prescribed documents under the three heads described earlier in this article (connection with one of the three countries, entry on or before 31 December 2014, and identity and community), together with the affidavit of religious identity and, where the Rules call for it, the certificate of a locally reputed community institution attesting the applicant’s belonging to one of the six communities. An electronic acknowledgement is generated. The District Level Committee, chaired by the jurisdictional Senior Superintendent or Superintendent of Posts as designated officer, verifies the documents, arranges the oath of allegiance, and forwards the application with its record. The Empowered Committee for the state, chaired by the Director of Census Operations and including officers of the intelligence, foreigners registration, informatics and postal services, scrutinises the application for completeness and for satisfaction of the conditions of Section 6B, and decides; on grant, the certificate of registration or of naturalisation issues, and citizenship is deemed from the date of entry. An applicant who fails to appear to subscribe the application and take the oath, after reasonable opportunity, has the application forwarded for consideration of refusal. The scheme’s design choices are visible in that route: a postal officer rather than the district magistrate chairs the verifying committee, the intelligence and foreigners machinery sits within the granting committee, and the state government’s political executive is present by invitation rather than by chairmanship, arrangements the Union defended as standardising and insulated, and which critics in the states read as the removal of the process from the accountable district administration. The applicant’s protections in the statute itself remain those noted earlier: pendency of proceedings is no disqualification, and the act of applying forfeits no right or privilege held at the date of application.

Objections of detail: five subsidiary arguments that outlast the headlines

First, the deeming provision and the next generation. Because a successful applicant is a citizen from the date of entry, children born in India to that applicant during the years before grant are born to a person the law deems a citizen at their birth, which answers in the applicant’s favour a question that Section 3 would otherwise put harshly, since a child born after the 2003 changes to a parent who is an illegal migrant does not acquire citizenship by birth. Supporters count this intergenerational settlement among the scheme’s humane effects; critics who accept the deeming logic within the class note that it sharpens the contrast with the child of an identically placed parent outside the class, whose birth status follows the parent’s unrelieved position.

Second, abatement and even handedness. Section 6B(3) abates pending proceedings on illegal migration or citizenship when citizenship is conferred. Defenders treat abatement as the necessary corollary of regularisation, since a state cannot coherently prosecute a person for an entry it has deemed the entry of a citizen. Critics answer that abatement is unobjectionable within the class and objectionable only in the contrast it manufactures: proceedings against persons outside the class, arising from entries in the same years and the same circumstances, continue to their conclusion under the Foreigners Act and the Citizenship Act. The exchange repeats, at the level of criminal process, the structure of the whole debate.

Third, the long term visa population and administrative overlap. A substantial share of the class lived in India for years on long term visas issued under executive policy to the same communities from the same countries, renewable and carrying permissions to work and to hold accounts and documents. For such persons the amendment converts a renewable executive tolerance into a statutory status, and the conversion is defended as the difference between living at discretion and living by right. The same conversion is cited by critics as evidence that the emergency the statute answers was already administratively managed, and that the statute’s marginal effect is therefore smaller than its symbolic charge, a charge each side values oppositely.

Fourth, the specified law ground under Section 7D and the discipline of notification. The ground operates only upon laws the Central Government specifies by notification in the Official Gazette, which means the ground’s content can be inspected, debated and, if a specification is itself excessive, challenged; defenders rely on that publicity as the answer to the uncertainty objection. Critics reply that a status holder cannot organise a life around a list that may grow by executive act, and that the hearing safeguard, valuable as it is, addresses the fairness of the decision rather than the foreseeability of the rule. The provision awaits both specification practice and judicial treatment, and its eventual weight in the scheme is genuinely unknowable on the material available.

Fifth, the relationship with the right to seek asylum in practice. India determines the position of asylum seekers through executive processes in which the United Nations High Commissioner for Refugees conducts status determination for specified nationalities in the capital, while other populations deal directly with the foreigners machinery. The amendment neither creates nor displaces any of that. Its critics observe that a statutory corridor for six communities of three countries, placed beside an executive system for everyone else, hardens the difference between the two tracks; its defenders observe that the executive system remains available to all, that the amendment withdraws nothing from it, and that the existence of one statutory corridor argues for building others, by legislation addressed to further defined cases, rather than for demolishing the one that exists. Here, as everywhere in this subject, the same fact carries both arguments, and the reader who has followed the article this far will recognise the pattern without further assistance.

Why reasonable readers divide: the contested premises collected

The argument resists settlement because it rests on five premises that are each contestable and rarely stated aloud, and collecting them is a service to readers of every persuasion. The first premise concerns the unit of legislative attention. Supporters proceed as if Parliament classifies situations (the position of religious minorities in states constitutionally committed to another religion), while critics proceed as if Parliament has classified persons by faith. A classification can be described either way, and constitutional doctrine has never fully decided which description governs when the two diverge; the case will require the Court to choose, or to explain why the choice does not matter.

The second premise concerns the relevance of the unprovided case. For supporters, the person outside the class is simply not before Parliament in this measure, and the general law treats that person exactly as it did before, neither better nor worse; equality, on this premise, measures the law against the position it found. For critics, a benefit conferred by the state is a public good distributed under Article 14, and the position of the excluded is not the old position but a new and relative one, since civic goods are positional and a corridor built for some redesigns the landscape for all. Constitutional law contains materials for both premises, in the step by step doctrine on one side and in the substantive readings of equality associated with the manifest arbitrariness line on the other.

The third premise concerns time. The amendment’s defenders ask that it be judged as a settlement of a past, the decades of displacement that followed Partition and the exemption regime of 2015 and 2016; its critics ask that it be judged as a precedent for the future, the first entry of a religious criterion into the citizenship code, available to be widened by later Parliaments. Courts are equipped to judge settlements; they are also, in basic structure adjudication, equipped to judge precedents; and the choice of temporal frame may decide the case before the doctrines are reached. The Section 6A judgment of 2024 was, at its core, a decision to judge the 1985 provision as a settlement of its own past, with the dissent insisting on the provision’s present operation; the same methodological fork awaits the bench in the 2019 case.

The fourth premise concerns the register. Every participant in the debate knows that the amendment’s practical significance for persons outside the class depends on whether a nationwide register under Section 14A is ever prepared and on what terms, and no participant can know that. Positions on the amendment therefore encode predictions about the future conduct of the state, and disagreement that presents itself as disagreement about a text is in part disagreement about political expectation. An article can expose that encoding; it cannot remove it, and readers should be suspicious of any account, friendly or hostile, in which the amendment’s meaning appears to depend on nothing outside its five sections.

The fifth premise concerns the judicial role itself. One view holds that admission to citizenship is the paradigm of a political question committed to Parliament, where judicial review is at its most deferential and confined to the outer bounds of rationality; the other holds that a classification by religion touches the identity of the polity, the very subject on which judicial review exists to check majorities. The Constitution Bench that hears the case will have to locate citizenship classification on the spectrum of review intensity, and that location, more than any single doctrinal formula, is what lawyers on both sides are waiting to learn. Until it is known, the most accurate one sentence account of the amendment’s legal position is the least dramatic one: a law in force, operating through rules in force, of contested validity, awaiting judgment.

Consequences inside the statute book: provisions the amendment touches without amending

A statute is a system, and the amendment’s effects ripple into provisions it never names. Section 9 of the principal Act, giving effect to Article 9 of the Constitution, terminates the Indian citizenship of a citizen who voluntarily acquires the citizenship of another country; it does not bear on applicants under Section 6B directly, but the Third Schedule’s qualifications do, since a naturalisation applicant must undertake, if the application is granted, to renounce any other citizenship held and to reside in India or serve under a Government in India. The deeming provision interacts with Section 5 as well: Section 6B(2) sends the applicant to the conditions of Section 5 for registration and to the Third Schedule for naturalisation, so the two existing routes supply the conditions while Section 6B supplies the standing to use them notwithstanding the former illegal migrant status. Section 10, the deprivation provision under which the Central Government may deprive a citizen by registration or naturalisation of citizenship on grounds including fraud, disloyalty and specified sentences, applies to a Section 6B citizen as it applies to any citizen by grant, so the status obtained is full citizenship with its ordinary incidents, including its ordinary liabilities to deprivation proceedings under the Act’s safeguards. The Citizenship Rules’ forms, the oath of allegiance in the Second Schedule, and the certificate forms all operate on the new route without distinction. Two silences in the scheme are worth recording. The amendment makes no provision for the spouse or children of a beneficiary as such; family members qualify, if at all, on their own satisfaction of the six conditions or through the general law, a point of practical hardship raised in the parliamentary debates and answered there by pointing to Section 5’s family based registration categories. And the amendment makes no provision for a beneficiary’s status during the pendency of an application beyond the protections of Section 6B(3) itself, so the applicant’s interim position continues to rest on the exemption orders and on whatever visa or stay arrangements the foreigners regime affords. These silences are not defects to be charged to either side; they are the ordinary incompleteness of an amending Act that builds on an existing code, and a reader assessing the amendment’s reach should count them as part of its actual shape.

The persons at the centre, considered as applicants rather than as symbols

The debate is conducted about the class far more often than it is informed by the class’s circumstances, and a measure of ground level fact disciplines both sides. The exempted population from which Section 6B draws is concentrated in particular geographies: families from Sindh and the border districts of Pakistan settled across Rajasthan, Gujarat, Madhya Pradesh and Delhi; Chakma and Hajong residents of Arunachal Pradesh and Assam, whose position arises from a displacement of the 1960s connected with the Kaptai dam in the Chittagong Hill Tracts and whose claims under the citizenship law were the subject of Supreme Court directions in the National Human Rights Commission proceedings of 1996 and again in 2015, and who are, it should be noted precisely, of Buddhist and Hindu affiliation from a territory that is part of Bangladesh, so that many among them answer the statute’s description; and smaller communities of Afghan Hindus and Sikhs in Delhi and elsewhere, enlarged by later departures that fall outside the cut-off date and therefore outside the Act. The circumstances common to much of this population are long residence, Indian born children, livelihoods in trade and manual work, and a documentary record that is thin in the particular ways the 2024 Rules attempt to accommodate. Whatever view is taken of the amendment’s principle, its administrative success or failure will be decided in committee rooms on files of this character, and assessments of the Act that float free of those files (whether celebrating a historic homecoming or predicting a machinery of mass exclusion) are answering questions the statute does not pose. The Chakma and Hajong case also supplies the debate’s neatest illustration of its own method: their inclusion within the class is welcomed by the amendment’s supporters as proof of the scheme’s fit to documented displacement, and cited by its critics, who note that the Supreme Court had directed the processing of their citizenship claims decades earlier under the unamended law, as proof that the general law, administered with will, was never as closed as the amendment’s necessity argument supposes. Both uses of the same community are available on the record; neither is a misuse of it.

Frequently Asked Questions

Q: What does the Citizenship Amendment Act actually say?

The Citizenship (Amendment) Act, 2019, Act 47 of 2019, makes four changes to the Citizenship Act, 1955. It adds a proviso to the definition of illegal migrant for persons of six named communities from Afghanistan, Bangladesh or Pakistan who entered on or before 31 December 2014 and were exempted under the entry and foreigners legislation. It inserts Section 6B, under which such persons may be granted registration or naturalisation, deemed from the date of entry, with pending proceedings abated. It reduces the residence aggregate for that class from eleven years to five. And it adds a cancellation ground for overseas citizen cardholders, with a hearing safeguard.

Q: Which communities and countries does the Citizenship Amendment Act cover?

The proviso names six communities: Hindu, Sikh, Buddhist, Jain, Parsi and Christian. It names three countries of origin: Afghanistan, Bangladesh and Pakistan. Both lists are closed, and every condition must coincide in the same person. Entry must have occurred on or before 31 December 2014, and the person must have been exempted by the Central Government from the passport and foreigners legislation, chiefly under the orders of 2015 and 2016. A person of a named community from any other country, a person of any other community from a named country, and a person arriving after the cut-off date all stand outside the provision.

Q: Why were the rules under the Citizenship Amendment Act delayed?

Section 6B confers its benefit on application made in the manner prescribed, and prescription required rules that the Citizenship (Amendment) Rules, 2024 supplied only on 11 March 2024, a little over four years after the Act commenced on 10 January 2020. The interval is a legislative fact with two readings. Critics treat it as proof that the urgency of December 2019 was political rather than administrative. Defenders treat it as the time needed to build verification machinery of committees, document lists and an electronic portal. The pattern itself, commencement long before operability, is common in Indian statute practice.

Q: Which areas are excluded from the Citizenship Amendment Act?

Section 6B(4) excludes two kinds of territory. The first is the tribal area of Assam, Meghalaya, Mizoram or Tripura as included in the Sixth Schedule to the Constitution. The second is the area covered under the Inner Line notified under the Bengal Eastern Frontier Regulation, 1873, which governs entry into Arunachal Pradesh, Nagaland and Mizoram, and which was extended to Manipur in December 2019. The explanation offered is protection of the demographic and constitutional arrangements of those regions. The observation made against it is that the citizenship route then differs between districts for identically placed persons. Assam itself is not excluded; only its Sixth Schedule tribal areas are.

Q: On what grounds is the Citizenship Amendment Act defended?

The defence rests on five grounds. The three named countries accord a preferred constitutional position to one religion, so their minorities face structural disadvantage. The amendment discharges a Partition era obligation towards persons the assurances of that period failed to protect. A closed cut-off date addresses a historical situation rather than opening continuing immigration. Article 14 permits reasonable classification, and under inclusiveness alone is not a ground of invalidity, since a legislature may proceed step by step. And the amendment takes nothing from anyone: it lifts a disqualification from a defined class while the general routes remain open to persons of every religion on unchanged terms.

Q: On what grounds is the Citizenship Amendment Act opposed?

The opposition rests on connected grounds. The amendment makes religion an express criterion in a citizenship statute for the first time, which is said to offend the secular character of the Constitution recognised as part of its basic structure. The classification is said to be under inclusive against its stated object, omitting persecuted communities within the named countries, including the Ahmadiyya, and persecuted groups in neighbouring countries not named. No individual test of persecution is applied, so the operative criterion is identity rather than hardship. Read beside the register provision inserted in 2003, the amendment is said to divide the position of persons unable to prove citizenship along religious lines.

Q: How does the Citizenship Amendment Act relate to a register of citizens?

Textually, it does not create one. The provision for a National Register of Indian Citizens is Section 14A of the Citizenship Act, inserted by the amendment of 2003, with rules of the same year providing for its preparation; an Assam register was completed under that framework in August 2019, before the 2019 amendment was enacted. The connection argued by critics is structural rather than textual: if a register placed a documentary burden on all residents, persons of the named class who failed it would hold the Section 6B route back, and others would not. The Union has stated that no decision to prepare a nationwide register had been taken.

Q: What is the status of the challenges to the Citizenship Amendment Act?

The petitions, led by the Indian Union Muslim League and filed under Article 32 from December 2019, stand pending before the Supreme Court. The Court declined to stay the amendment at the initial hearings in January 2020, and in March 2024 it declined to stay the operation of the Citizenship (Amendment) Rules, 2024, while directing that pleadings be completed for a hearing on the merits. No judgment on validity exists, and an interim refusal of stay is not a decision on the merits in either direction. Certificates under Section 6B have been granted since May 2024 while the challenge awaits hearing.

Q: What documents does an applicant under Section 6B have to produce?

The Citizenship (Amendment) Rules, 2024 prescribe documents under several heads. The applicant produces a document connecting the applicant with one of the three countries, from a list that includes a passport, birth certificate, land or tenancy record or identity document issued there, or a document showing that a parent, grandparent or great grandparent was a citizen of such a country. A second document shows entry into India on or before 31 December 2014, from a list including a residential permit, census slip, ration card, driving licence or school certificate. An affidavit of religious identity accompanies the application, with a certificate of a locally reputed community institution where the Rules call for one.

Q: What do the District Level Committee and the Empowered Committee each do?

The two committees divide verification from decision. The application is filed electronically to the Empowered Committee through the District Level Committee of the applicant’s district. The District Level Committee, chaired by the jurisdictional Senior Superintendent or Superintendent of Posts as designated officer, verifies the documents, administers the oath of allegiance, and forwards the application with its record; an applicant who fails to appear for the oath after reasonable opportunity has the file forwarded for consideration of refusal. The Empowered Committee at state level, chaired by the Director of Census Operations and including intelligence, foreigners registration, informatics and postal officers, scrutinises the application against the conditions of Section 6B and grants or refuses the certificate.

Q: From what date is a successful applicant treated as a citizen?

From the date of entry into India. Section 6B(2) provides that a person granted a certificate of registration or naturalisation under the section, having fulfilled the conditions of Section 5 or the qualifications of the Third Schedule as they apply, shall be deemed to be a citizen of India from the date of entry. Since every member of the class entered on or before 31 December 2014, the deeming reaches back at least a decade in most cases. The fiction settles the status of the intervening years, including the position of children born during them, without person by person litigation of each year.

Q: What happens to pending proceedings when citizenship is granted under Section 6B?

They stand abated. Section 6B(3) provides that any proceeding pending against the person in respect of illegal migration or citizenship shall stand abated on conferment of citizenship. Two protections surround the abatement. A pending proceeding is not a disqualification from applying, and the Central Government or its specified authority may not reject an application on the ground that a proceeding is pending if the applicant is otherwise qualified. And an applicant is not to be deprived of rights and privileges held on the date of receipt of the application by reason only of having applied. Defenders treat abatement as the necessary corollary of regularisation; critics note the contrast with proceedings against persons outside the class.

Q: What did the amendment change for overseas citizen cardholders?

It added a ground of cancellation to Section 7D of the Citizenship Act. Alongside the earlier grounds of fraud, disaffection towards the Constitution, enemy dealings in war, specified imprisonment and the interests of sovereignty, security and public interest, the Central Government may cancel an overseas citizen registration where the cardholder has violated any provision of the Citizenship Act or of any other law for the time being in force as may be specified by notification in the Official Gazette. The amendment added a safeguard with the ground: no cancellation order on that ground may be passed unless the cardholder has been given a reasonable opportunity of being heard.

Q: Which executive orders supply the exemption on which the proviso depends?

The proviso incorporates exemptions granted under clause (c) of sub-section (2) of Section 3 of the Passport (Entry into India) Act, 1920 and from the application of the Foreigners Act, 1946 and its rules and orders. The operative instruments are the Passport (Entry into India) Amendment Rules, 2015 and the Foreigners (Amendment) Order, 2015, with companion orders in 2016 extending and adjusting the scheme. Those orders exempted persons of the same six communities from the same three countries, who had entered on or before 31 December 2014 and had been compelled to seek shelter in India on account of religious persecution or fear of it, from specified document requirements and penal consequences, subject to registration conditions.

Q: Why is the cut-off date of 31 December 2014 contested?

The date is defended as the feature that makes the measure a closed historical settlement: it matches the operation of the exemption regime of 2015 and 2016, it forecloses any incentive to later movement, and the benefited class can only shrink with time. It is opposed on the ground that persecution in the named countries did not end on that date, that discriminatory laws and communal violence are documented for the years after it, and that a person who fled in 2016 stands in the same moral position as one who fled in 2013. On the first view the date is a virtue of legislative settlements; on the second it shows the object to be a defined cohort rather than persecution itself.

Q: What is the manifest arbitrariness argument against the amendment?

It is the petitioners’ alternative to the classification test. The classification test asks whether a differentia is intelligible and rationally related to the statute’s object, and legislatures usually survive it. Manifest arbitrariness, recognised in Shayara Bano v Union of India and developed in Joseph Shine and the Puttaswamy litigation, allows legislation to fall under Article 14 where it is capricious, irrational, without an adequate determining principle, or excessive, without constructing a comparator class. The petitioners argue that a differentia of religious identity, attached to a benefit whose stated ground is persecution that the statute never tests in the individual case, lacks an adequate determining principle. The Union answers that the doctrine does not displace classification review in a field of plenary legislative competence.

Q: Does the amendment change the position in Assam under Section 6A?

No. Section 6A, inserted in 1985 to give effect to the Assam Accord, keeps its own line of 25 March 1971 for persons of Indian origin entering Assam from Bangladesh, and the amendment does not alter it. The amendment’s class line of 31 December 2014 operates in Assam outside the Sixth Schedule tribal areas, which Section 6B(4) excludes, and critics in Assam argue that the later date reopens the Accord settlement in the plains districts, while defenders answer that the two provisions govern different populations. On 17 October 2024 a Constitution Bench upheld Section 6A by four to one; that judgment decided the 1971 line, not the amendment of 2019.

Q: Can a person outside the named class still seek citizenship under the general provisions?

Yes, on the terms that existed before December 2019. Sections 5 and 6 of the Citizenship Act, governing registration and naturalisation, were not amended in their general operation, and a person of any religion who is not an illegal migrant may apply under them, subject to the Third Schedule qualifications including the eleven year aggregate of residence. The amendment’s effect on the general routes is nil; what it changes is the position of its own class, whose members were barred as illegal migrants and are barred no longer. A person who remains an illegal migrant under the unamended definition, of whatever religion, remains ineligible for both routes.

Q: How did the Supreme Court approach the requests to stay the amendment and the Rules?

It declined both, without deciding validity. At the initial hearings in January 2020 the Court issued notice on the petitions and called for the Union’s response but granted no stay of the Act, which had commenced on 10 January 2020. When the Citizenship (Amendment) Rules, 2024 were notified on 11 March 2024, stay applications followed; in March 2024 the Court declined an interim stay of the Rules, recorded the Union’s position that the amendment confers citizenship and takes none away, and directed completion of pleadings toward a merits hearing. Interim orders of this kind rest on interim considerations, and neither side may treat the refusals as a ruling on the classification’s validity.

Q: What did the Section 6A judgment decide, and how is it distinct from the challenge to the 2019 amendment?

In the judgment of 17 October 2024 in the Section 6A reference, a five judge Constitution Bench by a majority of four to one upheld Section 6A of the Citizenship Act, the Assam Accord provision of 1985. The majority held that Parliament had the competence to enact it, that the cut-off of 25 March 1971 was rational in the setting of the Bangladesh liberation war, and that the provision was neither over inclusive nor under inclusive for its purpose; Justice Pardiwala dissented on the ground that the provision had become unconstitutional by the passage of time. The case concerned one state, one accord and the line of 1971. The community criterion and the line of 2014 in the 2019 amendment were not before that bench and remain undecided.