The Constituent Assembly of India is usually remembered for what it delivered: a constitution adopted on 26 November 1949, running to 395 articles and eight schedules, for a republic of hundreds of millions of people who had never voted in a general election. That delivery was real, and no serious account of the founding can begin by minimising it. Yet the delivery is the least interesting thing about the Assembly. Read the Constituent Assembly Debates closely, volume by volume, and a different institution comes into view: a body of lawyers, landlords, former prisoners of the Raj, princely nominees and party organisers who repeatedly reached the edge of a decision, recognised that the decision would break the coalition in the room, and chose a form of words that postponed the reckoning. The claim defended in this article is the one the record sustains under pressure. The Constituent Assembly’s most consequential choices were the questions it chose not to answer.

Three questions were left open in that deliberate way, and they organise everything that follows. The first asks who counts as fully Indian, and the Assembly worked on it through the religious settlement: the debates on draft articles that became Articles 25 to 28 of the Constitution, the refusal to write a single principle of secularism into the text, and the postponement of the Hindu Code Bill that Sardar Vallabhbhai Patel and Jawaharlal Nehru were already arguing about in their correspondence of October 1948. The second asks who governs whom, and the Assembly worked on it through the federal settlement: the reorganisation of the whole federal scheme after Partition in August 1947, the decision of 4 November 1948 to call India a Union of States rather than a federation, and the choice to permit outright asymmetry in arrangements such as Article 370 for Jammu and Kashmir and the Sixth Schedule for the tribal areas of the north-east. The third asks which past the present is continuous with, and the Assembly worked on it through the rights settlement: a charter of enforceable Fundamental Rights placed beside a set of Directive Principles that no court could enforce, with no stated rule for which of the two should prevail when they collided. Each settlement held the founding coalition together. Each settlement exported its hardest question into the future, addressed to legislatures, courts and movements that the founders would never sit in.

This way of reading the founding has a scholarly pedigree and a scholarly opponent, and the opposition deserves to be stated at full strength. Granville Austin, in The Indian Constitution: Cornerstone of a Nation, published in 1966, read the Assembly’s work as a coherent achievement: national unity, democracy and social revolution woven, in his famous phrase, into a seamless web by a body that decided by consensus and meant what it built. Madhav Khosla, in India’s Founding Moment, published in 2020, read the same record as a pedagogical project undertaken in full awareness of its own fragility: a constitution written to teach democratic habits to a society that did not yet possess them, whose silences and deferrals were part of the teaching problem rather than accidents of drafting. This article adjudicates between Austin and Khosla on the primary evidence of the debates, the February 1948 Draft Constitution and its marginal record, and B. R. Ambedkar’s closing speech of 25 November 1949. The verdict reached is Khosla’s, with one correction in Austin’s favour that the evidence requires. Where Austin describes the founders’ intentions, he is frequently right about the room. Where the argument turns on what the text actually settled, the deferral reading wins, because the deferred questions can be named, dated and traced.

The Achievement Story and What It Leaves Out

The achievement story begins with arithmetic that no other founding generation has matched. The Assembly that first met on 9 December 1946 in the Constitution Hall in New Delhi had been designed under the Cabinet Mission Plan of 16 May 1946 for a united India of 389 members, of whom 296 were to come from British India and 93 from the princely states. Partition cut that design down. After August 1947 the Assembly that continued the work was the constitution-making body of a smaller and wounded country, and it still completed the task in eleven sessions spread across two years, eleven months and eighteen days, adopting the finished text on 26 November 1949 and bringing it into force on 26 January 1950. The document it produced contained a justiciable bill of rights, a parliamentary executive, an independent judiciary crowned by a Supreme Court, universal adult franchise at a stroke for an electorate that dwarfed every electorate then voting anywhere on earth, and a federal division of powers across three legislative lists. Measured against the stated aims of the Objectives Resolution that Jawaharlal Nehru moved on 13 December 1946 and that the Assembly adopted on 22 January 1947, the delivery rate was extraordinary.

Granville Austin gave this story its canonical form. Austin worked from the debates and from interviews with surviving participants, and his 1966 book argued that the Assembly’s genius lay in holding three commitments together without sacrificing any of them: the unity and integrity of the nation, the institutions and spirit of democracy, and a social revolution to be pursued by constitutional means. The strands formed, in his metaphor, a seamless web, and the Assembly’s preferred method, consensus rather than bare majority voting, was the loom. There is strong evidence for the consensus claim. The Congress Assembly Party meetings, whose deliberations Austin reconstructed, absorbed disagreement before it reached the floor; formal divisions in the chamber were rare; and Ambedkar, presenting the Draft Constitution on 4 November 1948, could plausibly describe a text that the Drafting Committee had carried through more than two years of committee scrutiny. Any reading of the founding that cannot explain this record of completion is not worth holding, and the deferral reading does not try to explain it away.

What the achievement story leaves out is visible only when the questions are sorted by a different test: not whether the Assembly produced text, but whether the text closed the question the text addressed. On 6 December 1948, debating the draft article that became Article 25, the Assembly guaranteed freedom of conscience and the right freely to profess, practise and propagate religion, and in the same sitting heard Lokanath Misra warn that the propagation clause handed a charter to conversion drives, and K. M. Munshi answer that the clause restated a freedom the minorities had been promised. The article passed. The underlying question, what the Indian state is in relation to religion, was answered with a list of permissions and restrictions rather than a principle, and the word secular did not appear in the constitutional text the Assembly adopted. On 19 November 1948, Ambedkar defended the Directive Principles against members who wanted them enforceable, arguing that a government which ignored them would answer to the electorate rather than to a court. The Principles passed. The underlying question, whether the rights of the individual or the directives addressed to the state should govern when the two conflict, was answered by assigning each to a different forum and declining to rank them. On 17 October 1949, N. Gopalaswami Ayyangar moved the draft article that became Article 370, conferring a special position on Jammu and Kashmir, and the Assembly accepted an arrangement for one state that it accepted for no other. The federal question was answered, state by state, in different currencies. A text can be complete and a settlement can be open at the same time. The founding text is complete. The founding settlement, on the three questions that matter most, was left open on purpose.

What did the Assembly finish on 26 November 1949?

On 26 November 1949 the Assembly adopted the Constitution as a finished legal text of 395 articles and eight schedules, covering citizenship, fundamental rights, directive principles, the Union and state governments, the judiciary, elections, emergency powers and amendment. Members signed the document on 24 January 1950, and its main provisions commenced on 26 January 1950, closing the drafting task completely.

The distinction between finishing a text and settling its questions explains a feature of the debates that the achievement story passes over in silence: the founders’ own repeated announcements that they were deferring something. Ambedkar told the Assembly on 25 November 1949, in the most quoted passage of the closing debates, that on 26 January 1950 the country would enter a life of contradictions, holding equality in politics and inequality in social and economic life, and that the contradiction would have to be resolved by those who came after or the political structure would be blown up by those who suffered from it. That is not the language of a draftsman who believes the hard questions have been answered. It is the language of a draftsman inventorying what has been handed on. The same inventory appears in miniature across the record: in the decision to let the Hindu Code Bill wait for a parliament elected by adult franchise, in the decision to describe India as a Union in Article 1 while building a federation in the Seventh Schedule, in the decision to place the uniform civil code among the unenforceable directives in Article 44 rather than among the enforceable rights. The deferrals were not failures of nerve, and this article does not present them as such. They were the price of a text that Congress conservatives, socialists, minority representatives, princely nominees and the Drafting Committee’s chairman could all sign. A founding that had forced each question to a conclusion in 1949 might have produced no constitution at all, or a constitution for a smaller coalition that would have unravelled faster than the compromises did.

How the Assembly Was Built: Membership, Mandate and Method

The Assembly’s method cannot be separated from its composition, and its composition explains both its strengths and its characteristic evasions. Under the Cabinet Mission Plan announced on 16 May 1946 by the mission of Lord Pethick-Lawrence, Stafford Cripps and A. V. Alexander, members were allocated to provinces roughly in proportion to population, one member for about every million inhabitants, divided among general, Muslim and Sikh categories. They were chosen indirectly, by the members of the provincial legislative assemblies voting by proportional representation, and the princely states were to send their representatives by methods their rulers and the Assembly’s States Negotiating Committee would settle. The Indian National Congress swept the general seats in the elections of July and August 1946; the Muslim League took most of the Muslim seats and then, after the collapse of the plan’s grouping scheme and the call for Direct Action Day on 16 August 1946, boycotted the Assembly when it convened. The body that met on 9 December 1946 therefore spoke for a political nation that was already coming apart, and everyone in the Constitution Hall knew it. Sachchidananda Sinha presided as interim president on the French convention model, and on 11 December 1946 the Assembly elected Rajendra Prasad its permanent president.

Partition changed the institution’s character midway through its work, in ways recorded in the debates of August 1947 in Volume V of the Constituent Assembly Debates. Members from the territories that became Pakistan withdrew; the representatives of Bengal and Punjab had to be reconstituted for the divided provinces; the princely states, released from paramountcy by the lapse provisions of the Indian Independence Act of July 1947, acceded on the three subjects of defence, external affairs and communications and sent their nominees into the chamber. The Assembly also sat in a second capacity as the Dominion Legislature, passing ordinary laws under the adapted Government of India Act of 1935, and the double burden shaped its calendar and its habits. Constitution-making proceeded by committee before it proceeded by chamber: a Union Constitution Committee under Nehru, a Provincial Constitution Committee under Patel, an Advisory Committee on Fundamental Rights, Minorities and Tribal Areas under Patel, whose sub-committees did much of the substantive drafting through 1947. The constitutional adviser B. N. Rau prepared an initial draft by October 1947 after consulting constitutional practice abroad, including a meeting with Justice Felix Frankfurter in Washington, whose advice against an American-style due process clause left a mark on the final Article 21. The Drafting Committee, appointed on 29 August 1947 with Ambedkar as chairman and six other members, converted the committee reports and Rau’s draft into the Draft Constitution published in February 1948, a document whose text, together with the committee’s recorded alternatives and the comments received from the public, forms the paper spine of the founding.

That paper spine remains consultable, and its material form matters to the argument here. The Draft Constitution of February 1948 was published for public comment, and the Lok Sabha digital archive preserves the drafting record, including the alternatives the Drafting Committee considered and set aside, which is where several of the deferrals are visible in their earliest written form. Readers working through that record on a ordinary computer meet it as scanned volumes and word-processed transcripts rather than as a single bound book, and an office file viewer for working with spreadsheet, document and presentation files is the practical means by which the February 1948 text, the committee reports and the debate transcripts can be opened side by side and compared clause by clause. The comparison repays the effort. The February 1948 draft already contained the rights-and-directives architecture in its final shape, already named India a Union in its first article, and already carried the religious freedoms as a list without a governing principle, which shows that the deferrals were structural features of the design by the winter of 1948 and not improvisations forced on the founders by the debates of 1949.

The Assembly’s working procedure in its final phase also deserves record, because the achievement story relies on it and the deferral reading must absorb it honestly. Ambedkar presented the Draft Constitution to the Assembly on 4 November 1948 in a speech that answered the main lines of criticism in advance, defending the strong centre, the parliamentary form and the amendment procedure against the Gandhian and federalist objections he knew were coming. The Assembly held a general discussion, then took the text clause by clause in a second reading that ran from 15 November 1948 to 17 October 1949, considering thousands of amendments, most of them withdrawn, negatived or accepted in modified form through the Congress Assembly Party’s internal discipline, and completed a third reading between 14 and 26 November 1949. The method was genuinely deliberative in form and tightly managed in substance, and both halves of that sentence matter. A chamber that was managed by a dominant party could force conclusions when the party’s leadership wanted conclusions, as the votes on adult franchise and on the abolition of untouchability under Article 17 demonstrate. The same chamber left the religious principle, the federal symmetry question and the rights hierarchy open. Deferral, in a body capable of decision, is itself a decision, and it is on that footing that the three tensions are examined in turn.

The Drafting Committee and the February 1948 Text

The seven men appointed to the Drafting Committee on 29 August 1947 were B. R. Ambedkar as chairman, Alladi Krishnaswami Ayyar, K. M. Munshi, Muhammad Saadullah, B. L. Mitter, N. Gopalaswami Ayyangar and D. P. Khaitan; Mitter resigned on grounds of health and was replaced by N. Madhava Rau, and Khaitan died in 1948 and was replaced by T. T. Krishnamachari. The committee worked from the draft that B. N. Rau had prepared by October 1947 out of the committee reports of the preceding summer, and it sat, by the Assembly’s own accounting, for 141 days before presenting its text. The Draft Constitution it published in February 1948 contained 315 articles and eight schedules, and it was thrown open to public comment on a scale no earlier Indian constitution-making exercise had attempted: suggestions and amendments arrived from provincial governments, bar associations, political organisations and private citizens, and the committee’s recorded responses to that commentary, preserved with the draft in the Lok Sabha archive, show it accepting some formulations, rejecting others with reasons, and on the questions that mattered most to this article, restating the formula it had inherited rather than resolving the argument beneath it.

That pattern in the paper record deserves emphasis, because it converts the deferral thesis from an inference into a document. On the religious clauses, the February 1948 draft already carried freedom of conscience and propagation in language close to the final Article 25, already preserved state power to regulate religious practice, and already omitted any definition of the state’s religious character. On the union, the draft already opened with India as a Union of States, months before Ambedkar defended the phrase on 4 November 1948. On the rights architecture, the draft already divided the fundamental rights from the directive principles by the line of enforceability that Article 37 would later state. When the Assembly took the draft clause by clause from 15 November 1948, it amended freely at the margins: the debates of the second reading produced, by the standard count, some 7,635 amendments tabled, of which about 2,473 were actually moved and disposed of. The architecture survived the amending process untouched at exactly the three points where this article locates the deferrals. A chamber that changed thousands of words and left the three silences standing was not overlooking the silences. It was ratifying them, session after session, with the alternatives moved, debated and negatived on the floor in full view.

The personnel of the committee reinforced the pattern. Ambedkar’s authority as chairman was technical and procedural rather than majoritarian: he was not a Congressman, he led a party of his own that the Congress had fought electorally, and his presence at the head of the drafting table was itself one of Patel’s and Nehru’s coalition decisions. The committee’s other members carried the Congress’s internal range, from Munshi’s cultural conservatism to Ayyangar’s administrative liberalism, and the committee’s method, visible in its recorded alternatives, was to carry disagreements upward into the text as paired provisions rather than to settle them downward into a single doctrine. Rights beside principles, Union beside lists, freedom beside regulation: the February 1948 draft is a museum of such pairings, and each pairing marks a meeting at which the committee, like the Assembly after it, declined to choose. Later scholarship that treats the Drafting Committee as a drafting machine executing a settled philosophy mistakes the machine’s product. The philosophy was the pairing.

The Franchise Leap: The Decision the Assembly Took Whole

Against the deferrals stands the one great decision the Assembly took whole and at once, and it belongs in the record of this article because the contrast disciplines the argument. Universal adult franchise was not a colonial inheritance or a committee accident. The Government of India Act of 1935 had enfranchised roughly a sixth of the adult population through property, tax and educational qualifications. The fundamental rights sub-committee of 1947 recommended adult suffrage, the Union Constitution Committee accepted it, and when the relevant articles, the ancestors of Articles 325 and 326, were debated on 15 and 16 June 1949, amendments to qualify the franchise by literacy or property were moved and rejected. Alladi Krishnaswami Ayyar, defending the unqualified franchise on the floor, treated the vote as the constitution’s act of faith in the common citizen, and the Assembly adopted a single electoral roll with no separate electorates and no communal reservation of the franchise, for a country in which, by the census figures the members knew, literacy stood well under a fifth of the population.

The administrative audacity of the decision is easily missed because it succeeded. The Election Commission under Sukumar Sen, appointed Chief Election Commissioner in March 1950, prepared a roll of some 173 million voters for the elections of 1951 and 1952, devised a system of party and candidate symbols to serve voters who could not read a ballot, and conducted the largest election yet held anywhere. Ambedkar’s part in the franchise settlement is sometimes folded into his general chairmanship, yet its logic is his in a precise sense: the vote was the one instrument of equality the state could distribute immediately and identically, and his closing speech’s warning of 25 November 1949 drew its force from the contrast between that distributed equality and the social and economic equality that had been left to the unenforceable principles. The founders could leap on the franchise because the leap required a decision and machinery, not a settlement among the coalition’s warring convictions about religion, land or language. Where a question could be answered by an act of will, the Assembly answered it. Where a question required the coalition to agree about what India was, it deferred. The franchise is the control case that proves the deferrals were choices rather than habits, and it is also the decision that made the deferrals workable, since the questions handed onward would be argued by legislatures and movements resting on the one foundation the founders had laid without flinching.

The First Tension: Religion and the Refused Definition of Secularism

The first tension asks who counts as fully Indian, and in the Assembly it arrived dressed as a drafting problem: how to write the state’s relation to religion for a country whose largest single political trauma, the Partition of August 1947, had just been executed in religion’s name. The raw material was assembled early. The Advisory Committee under Sardar Patel, through its sub-committee on fundamental rights chaired by J. B. Kripalani and its sub-committee on minorities chaired by H. C. Mookherjee, produced interim reports that the Assembly debated on 27 and 28 August 1947, in the sittings recorded in Volume V of the Constituent Assembly Debates, a fortnight after independence and in the middle of the Punjab massacres. Those debates fixed the outer boundaries of the settlement: safeguards for minority culture and language, freedom of religion for every person, and no state religion. Within those boundaries, almost everything that would have amounted to a principle was negotiated away, clause by clause, over the following two years.

Consider what the Assembly actually wrote in the articles numbered 25 to 28. Article 25 guarantees freedom of conscience and the right freely to profess, practise and propagate religion, subject to public order, morality and health, and expressly preserves the state’s power to regulate secular activity associated with religious practice and to throw open Hindu religious institutions of a public character to all classes of Hindus. Article 26 guarantees religious denominations the right to manage their own affairs in matters of religion. Article 27 bars taxation whose proceeds are specifically appropriated for any particular religion. Article 28 regulates religious instruction in institutions maintained or aided by the state. When the draft article behind Article 25 was debated on 6 December 1948, the argument in the chamber ran along two tracks that never met. Lokanath Misra spoke for members who feared that propagation licensed organised conversion and objected to placing all religions in a single undifferentiated clause. K. M. Munshi, a member of the Drafting Committee, defended the clause as the settlement’s necessary generosity, pointing to the explanations that counted the wearing and carrying of kirpans as part of the Sikh religion. Tajamul Husain moved toward the opposite pole, proposing that religion be confined to the private sphere altogether. The Assembly kept the clause substantially as drafted. What it did not do, at that sitting or any other, was adopt a sentence stating what the arrangement meant. There was no Indian equivalent of a separation wall, no French-style declaration of laicism, and no definition of the secular character that later generations would attribute to the founding. The word secular entered the Preamble only with the Forty-Second Amendment of 1976, a full generation later, and its absence from the 1949 text was not an oversight that the debates conceal. Members who wanted stronger formulations said so, on the record, and lost.

The Patel correspondence shows the same refusal being practised at the level of high politics rather than drafting. In the exchange of 12 October 1948 between Sardar Patel and Jawaharlal Nehru, preserved in the sixth volume of Patel’s correspondence covering 1947 to 1948 and rarely used in general accounts of the founding, the two men who between them controlled the Congress organisation and the government worked over the communal temperature of the country and the handling of the Hindu Code question with a caution that the floor debates only hint at. Patel, who chaired the Advisory Committee and carried responsibility for the states and for internal order as Minister for Home and States, wrote as the settlement’s manager; Nehru wrote as its public face. Neither treated the religious question as soluble by declaration. Their instinct, repeated across the correspondence of those months, was to settle each concrete dispute, the entry of Harijans into temples, the protection of mosques, the rehabilitation of abducted women, and to leave the general theory unwritten. Historians who read the founding through its proclamations miss this layer entirely. The Assembly’s leaders were not concealing a secular doctrine they lacked the courage to state. They were deliberately declining to acquire one, because a stated doctrine would have created winners and losers inside the Congress coalition before the constitution itself was safe.

The second half of the religious settlement concerned representation, and here the Assembly did decide, late and in changed circumstances, against its own earlier draft. The Advisory Committee’s 1947 scheme retained joint electorates with reserved seats for religious minorities in legislatures, a continuation of the arrangements the Congress had accepted in the 1946 elections. In May 1949, with Partition accomplished and the Muslim League’s Indian remnant diminished, the Assembly took up the Advisory Committee’s further report and abolished reservation of seats for religious minorities altogether, while retaining it for the Scheduled Castes and Scheduled Tribes. Sardar Patel, moving the change, presented it as the minorities’ own maturing preference and as a step toward a polity in which, in the spirit of his speech, the majority and minority distinction would cease to structure representation. The decision was defended in the chamber as generosity and criticised outside it, then and later, as a majority withdrawing a safeguard once the safeguard’s beneficiaries could no longer bargain. Both characterisations belong to named holders and both should be carried at full strength; the factual core is not disputed. A safeguard that the 1947 Assembly had treated as part of the settlement was removed by the 1949 Assembly by ordinary resolution of the same body, which demonstrates how much of the religious settlement rested on political convenience rather than on entrenched principle. Caste-based reservation survived because its beneficiaries and its Congress champions sat inside the room in force. Religious reservation did not, because its beneficiaries, after Partition, largely did not.

Did the Constituent Assembly make India a secular state?

The Assembly guaranteed religious freedom and barred a state religion without writing the word secular into the 1949 text or adopting any single doctrine of state and religion. It left the relationship to be defined through Articles 25 to 28, personal law and later legislation, so India received a religious settlement rather than a secular definition.

The pattern across the religious clauses is consistent enough to deserve a name, and the name this article uses is settlement by enumeration. Wherever a principle would have forced a general answer, the Assembly substituted a list: a list of freedoms in Article 25, a list of denominational rights in Article 26, a list of venues in Article 28, a list of communities for whom reservation survived in the articles on representation. Lists can be amended item by item and litigated clause by clause, which is precisely their attraction for a coalition that disagrees about the whole while agreeing about the parts. The cost of enumeration is that the parts point in different directions when a hard case arrives: the freedom to propagate against the power to regulate religious practice, denominational autonomy against temple entry reform, minority educational rights against a common citizenship. The founders handed those collisions to the courts without instructions. When the Supreme Court spent the following decades classifying practices as essential or inessential to a religion, it was doing, in judicial dress, the definitional work the Assembly had declined to do in 1948 and 1949. The deferral reading does not require the founders to have foreseen the essential-practices doctrine. It requires only that the definitional work was left undone in the text, and on that point the record is plain.

The August 1947 Minorities Debates: The Settlement’s First Draft

The religious settlement’s first draft was debated before the drafting of the constitution properly began, and Volume V of the Constituent Assembly Debates preserves it in its rawest form. On 27 August 1947 the Assembly took up the interim report of the Advisory Committee on fundamental rights and minorities, a fortnight after independence day and while the Punjab was burning along the new border. The report recommended joint electorates with seats reserved for minorities in proportion to population, rejecting separate electorates, the device under which Muslims, Sikhs and others had voted in segregated rolls since 1909. Govind Ballabh Pant, the Premier of the United Provinces, spoke against separate electorates in the debate’s most quoted sentence, calling the device suicidal and a poison that would destroy the minorities it claimed to protect, and the Assembly adopted joint electorates with reservation as the settlement’s frame. The same sittings carried the cultural and educational safeguards that became Articles 29 and 30, the freedom of religion clauses in their committee form, and the abolition of untouchability in draft, all of it argued by members who were receiving, during the very days of debate, the reports of the massacres that gave the arguments their stakes.

Two features of the August 1947 sittings repay attention for the argument here. The first is the identity of the settlement’s carriers. The Advisory Committee was chaired by Patel, its minorities sub-committee by H. C. Mookherjee, a Christian member from Bengal who later became Governor of West Bengal, and its fundamental rights sub-committee by J. B. Kripalani; the Congress leadership had arranged for the safeguards of minorities to be drafted under a chairman from the majority community’s most organisationally powerful figure and reported through spokesmen drawn from the minorities themselves. The arrangement was coalition management of a high order, and it explains the settlement’s durability at the level of form: the clauses arrived in the chamber pre-negotiated among the interests they bound. The second feature is the sittings’ treatment of the one proposal that would have converted the settlement into a principle. Amendments and speeches pressing for a declaration that the state would hold no religion of its own in any stronger sense, or that religion be treated as a purely private matter, were present in the debate’s atmosphere through members such as Tajamul Husain, and they went the way of all such proposals in the founding record: absorbed into the enumerated clauses, which the Assembly adopted one by one, while the general proposition was neither adopted nor squarely rejected. By the time the finished articles were debated in December 1948, the architecture had been fixed for sixteen months. The chamber was refining a settlement whose defining refusal had already happened, in August 1947, in the shadow of Partition, by a body that understood itself to be drafting under fire and chose enumeration as the form its prudence took.

Citizenship at the Commencement: Belonging Decided for One Day Only

The citizenship clauses show the first tension being managed by the same technique in its most technical form. Articles 5 to 8, debated principally on 10 and 11 August 1949, defined who was a citizen at the commencement of the Constitution: persons domiciled and born in the territory, persons with a parent born in the territory, persons resident for five years, and the special and anguished categories created by Partition migration, those who had moved between India and Pakistan under permits and deadlines that the debates set and reset with evident reluctance. Govind Ballabh Pant and Alladi Krishnaswami Ayyar carried much of the argument, members pressed the claims of migrants in both directions, and the Assembly wrote the clauses with a lawyer’s care for the single date of commencement. Then, in Article 11, it handed the entire subject to Parliament, providing that nothing in the preceding articles should derogate from Parliament’s power to legislate on the acquisition and termination of citizenship. The founding definition of belonging was thus expressly provisional: a snapshot for 26 January 1950, attached to a standing authorisation for the legislature to redefine the membership of the political community by ordinary law.

The consequences arrived on schedule. Parliament enacted the Citizenship Act of 1955 to occupy the field Article 11 had cleared, and amended it repeatedly thereafter, each amendment reopening, in statutory form, the question the Assembly had answered only for commencement day. Jayal’s layered citizenship framework, discussed in the treatment of the third tension in this article, finds one of its clearest illustrations here, because the citizenship clauses divide belonging into a constitutional layer, settled and justiciable for the founding moment, and a statutory layer, revisable by majority, and the boundary between the layers is itself a political decision renewed at every amendment. Members in August 1949 who pressed for a more generous or a more restrictive definition were not defeated on the merits so much as outlasted by the design: whatever they lost in the articles they could contest again in the Citizenship Bill. The Assembly had discovered, in citizenship as in personal law, that the surest way to carry a contested definition through a divided chamber was to date-limit it and delegate its future. Belonging, the most explosive of the three tensions’ subjects in the year of Partition, was the subject on which the founders were most candid that their settlement was a commencement and not a conclusion.

The Hindu Code Deferral: Personal Law Postponed to a Future Parliament

No single episode exposes the mechanics of deferral as cleanly as the Hindu Code Bill, because the episode has dates, named antagonists and an unfinished manuscript. A committee under B. N. Rau had been appointed in 1941 to examine the reform of Hindu law, and reported in 1947 in favour of codification. Ambedkar, as Law Minister in Nehru’s cabinet, took charge of the codification project, and a draft Hindu Code was placed before the Constituent Assembly sitting in its legislative capacity in April 1948. The code proposed to remake the law governing some four-fifths of the population: marriage made monogamous and dissoluble, daughters given shares in ancestral property, adoption and succession regularised, caste disabilities in marriage removed. It struck directly at the first tension, because it proposed that the state define the civil consequences of being Hindu while leaving Muslim, Christian and Parsi personal law untouched, and it struck at the Congress coalition, because orthodox opinion inside and outside the party regarded the code as an assault on the joint family and on scriptural authority.

The Patel and Nehru correspondence of 12 October 1948 catches the leadership mid-retreat. The exchange belongs to the same autumn in which the Assembly was debating the Draft Constitution clause by clause, and it shows the Hindu Code being weighed not as jurisprudence but as political exposure: how much reform could the organisation carry, in a country still absorbing Partition, with a general election approaching in which the code’s opponents could mobilise against it. Patel’s caution and Nehru’s commitment pulled in opposite directions within a shared decision to proceed slowly, and slowly, in the event, meant not at all within the life of the Assembly. The bill was debated in the provisional parliament in 1951, truncated and stalled; Ambedkar resigned from the cabinet on 27 September 1951, and in his resignation statement of 10 October 1951 he named the abandonment of the Hindu Code among his reasons, recording that the bill had been allowed to die after years of labour. The code’s substance eventually became law as four separate Acts of the elected Parliament: the Hindu Marriage Act of 1955 and the Hindu Succession Act, the Hindu Minority and Guardianship Act and the Hindu Adoptions and Maintenance Act of 1956. What the Assembly deferred, the Parliament of 1955 and 1956, answerable to an electorate voting under the new constitution, enacted in pieces.

Three conclusions follow for the argument of this article. The first is that the deferral was conscious at every level: the Law Minister who drafted the code, the Prime Minister who praised it and postponed it, and the Home Minister who doubted its timing all understood that a question was being put off, and said so in documents that survive. The second is that the deferral had a rationale the achievement story cannot supply. Ambedkar himself, in the Assembly’s debates on the uniform civil code in November 1948, had accepted that Article 44 would state an aspiration rather than a command, remarking that a future parliament might bring a code into operation gradually and initially on a voluntary basis; the founders’ theory, such as it was, held that civil law reform required a democratic mandate the nominated and indirectly elected Assembly did not possess. The third conclusion is the one the deferral reading is built on: the postponed questions did not dissolve during the postponement. The uniform civil code remained in Article 44 as an unenforced instruction; the reform of one community’s law and the non-reform of the others created the asymmetry that every later controversy over personal law inherited; and the terms in which those controversies are conducted, majority reform against minority autonomy, state uniformity against denominational freedom, were all fixed in the exchanges of 1948 to 1951. The Assembly did not fail to settle the personal law question because the question was too hard to notice. It noticed the question in committee in 1941, in draft in 1948, in correspondence in October 1948 and in a ministerial resignation in 1951, and it chose, repeatedly and for stated reasons, to hand the question onward.

The Language Settlement: A Formula, a Deadline and a Promise

Language was the first tension’s fourth theatre, and the Assembly came closer there to forcing a decision than anywhere else, which makes the eventual shape of the settlement the more instructive. The question had two layers: the language of the Union’s official work, and the standing of the major regional languages in their own states. On the second layer the Assembly decided early and by deferral of a different kind, accepting the principle of linguistic provinces in debate while postponing the map; the Dar Commission of 1948 advised against reorganisation on language lines at that juncture, and the question was handed to the following decade, which answered it through the Andhra movement, the death of Potti Sriramulu in December 1952 after a fast for a Telugu state, and the States Reorganisation Act of 1956. On the first layer the Assembly fought its most emotional late battle. The compromise known as the Munshi-Ayyangar formula, negotiated through the Congress leadership and moved in September 1949, was debated from 12 to 14 September 1949 in sittings that members described afterward as among the most heated of the Assembly’s life. Seth Govind Das and Purushottam Das Tandon pressed Hindi in the Devanagari script as the national language in substance, members from the south and from Bengal, including T. T. Krishnamachari with a warning against what he called linguistic imperialism, resisted any imposition, and Maulana Abul Kalam Azad and Nehru worked the compromise: Hindi in Devanagari as the official language of the Union under Article 343, English to continue for fifteen years, the Eighth Schedule listing the recognised languages, and no language named as the national language at all.

The formula’s structure is the deferral method in diplomatic dress. It decided the script and the official language, the two matters that could be decided by vote, and it postponed the operative question, whether English would actually yield to Hindi in the Union’s work, behind a fifteen-year deadline expiring on 26 January 1965 and a prime ministerial assurance. Nehru’s assurance, given in the Lok Sabha in 1959 and repeated thereafter, was that English would continue as an associate language for as long as the non-Hindi-speaking states required it, and the Official Languages Act of 1963 was written to carry that assurance into law. When the deadline arrived in January 1965, the anti-Hindi agitations in Madras State, with student self-immolations and deaths in police firing, announced that the deferred question had matured, and the settlement was renegotiated into the Official Languages (Amendment) Act of 1967 and the three-language formula of the National Policy on Education of 1968. The founders who voted in September 1949 cannot be charged with naivety about the deadline they set: the Munshi-Ayyangar formula was understood in the chamber as an armistice, and armistices are deferrals with dates attached. What the language settlement adds to the argument is the demonstration that even the Assembly’s hardest-fought apparent decision dissolved, on inspection, into the same architecture as the other three, an enumeration for the present, a principle withheld, and a date given to the future at which the argument would resume with the parties’ positions intact.

The Second Tension: A Union That Is Not a Federation

The second tension asks who governs whom, and the Assembly’s handling of it was transformed by catastrophe in the middle of the drafting process. The federal scheme the founders first imagined was the scheme of the Cabinet Mission Plan, under which the centre would hold only defence, foreign affairs and communications, with the provinces grouped into sections and the residuary powers resting with the units. The Union Constitution Committee under Nehru reported on 6 July 1947 in favour of a considerably stronger centre than the plan allowed, yet still within a framework negotiated among near-sovereign units. Partition destroyed that framework in a matter of weeks. The sittings of August 1947 recorded in Volume V of the Constituent Assembly Debates show the federal scheme being reorganised under emergency conditions: the provinces of the continuing India were no longer bargaining partners whose consent had to be purchased, the princely states were acceding under the pressure of the lapse of paramountcy, and the refugee columns moving in both directions across Punjab made the case for central power in the most concrete possible form. By the time the Draft Constitution appeared in February 1948, the federation it described was a centralised one, with a Union List of ninety-seven entries, emergency provisions capable of converting the federation into a unitary state, a single judiciary, a single citizenship and All-India Services recruited by the centre and serving in the states.

The naming of that federation produced the debate in which the deferral technique appears in its purest form. Ambedkar presented the Draft Constitution on 4 November 1948 and addressed the point head-on: the Drafting Committee, he said, proposed that India be described as a Union of States although its constitution was federal in structure, because the Indian federation was not the result of an agreement among the states to join a federation, and no state consequently possessed a right to secede; the federation was a Union because it was indestructible. When the first article was taken up on 15 November 1948, K. T. Shah moved to substitute a formulation naming India a secular, federal and socialist union, and the amendment was defeated on each of its three adjectives. The debate resumed in September 1949 over the name Bharat, with members pressing competing claims of antiquity and usage before the Assembly settled on the double formula of the final text: India, that is Bharat, shall be a Union of States. Every participant in these exchanges understood the stakes. To write federation was to concede the compact theory and its corollary, a possible right of exit. To write Union was to claim permanence while preserving a federal distribution of powers underneath the claim. What the Assembly declined to supply was any account of how the permanence of the Union and the autonomy of its units were to be balanced in the disputes that would certainly come: over the reorganisation of state boundaries under Article 3, where Parliament may alter a state’s territory after consulting a legislature whose views it need not accept; over the governor’s office; over the use of Article 356 to displace a state government. The federal settlement stated the Union’s permanence as a premise and left the units’ security as a practice to be negotiated, which is why Indian federalism has been renegotiated politically in every decade since, through the States Reorganisation of 1956, the Sarkaria Commission appointed in 1983 and reporting in 1988, and the Supreme Court’s discipline of Article 356 in the Bommai judgment of 1994, rather than settled jurisprudentially at the founding.

The centralisation itself was not hidden, and its defenders in the chamber argued it from the evidence of 1947 rather than from doctrine. Patel, whose States Department under V. P. Menon had executed the accession of more than five hundred princely states between July 1947 and 1949, spoke for an administrative generation that had watched a subcontinent nearly fragment and had no appetite for a centre that must plead with its units. Ambedkar argued in the November 1948 debates that a constitution could be federal in normal conditions and unitary in emergency, and that this convertibility was a strength the American and Australian models lacked. The critics deserve their strongest form as well. Members from the provinces, and federalists such as K. Santhanam in his writings of the period, warned that a centre controlling finance, services and emergency power would reduce the states to municipalities with legislatures, and that the warning was being ignored in the name of a unity emergency that would pass. On the record, the centralisers won the text and the federalists won a share of the argument’s future, since nearly every subsequent commission on centre and state relations has restated the Santhanam objection in official language. What neither side obtained in 1949 was a boundary. The Assembly distributed powers meticulously in three lists and then equipped the centre with residuary power, emergency override and fiscal dominance, leaving the real federal question, how much self-government the units would enjoy when the centre was held by their political opponents, to be discovered by experience. The discovery process, running from the dismissal of the Kerala ministry in 1959 to the coalition decades when regional parties held the balance in New Delhi, is the second tension’s renegotiation history, and none of it was accidental. It was the working out of a settlement that named the Union indestructible and left destructibility, in every other respect, negotiable.

Why does the Constitution call India a Union of States instead of a federation?

Ambedkar explained on 4 November 1948 that the word Union was deliberate: the federation was not created by agreement among states, so no state could secede from it. India would work as a federation in its distribution of powers while claiming permanence in its name, an indestructible Union composed of states that Parliament could reorganise.

The strongest evidence that the naming choice was a deferral rather than a solution lies in the company it keeps inside the text. A constitution confident about its federal theory would not need Article 3, which lets Parliament form new states and alter areas, boundaries and names with only the formality of ascertaining the affected legislature’s views. It would not need Article 200 and the governor’s power to reserve state bills for the President. It would not need Article 356 at all, or it would fence that power with the precision the Assembly instead lavished on the Seventh Schedule’s lists. The Assembly’s drafting energy went into enumerating the centre’s instruments and the states’ jurisdictions in parallel, without a stated rule for the political conflict between them, in exactly the manner of the religious settlement’s enumerations. Austin read this accumulation as the seamless web’s unity strand doing necessary work, and for the conditions of 1947 to 1949 the reading is persuasive. Khosla’s framework explains the longer record better, because a web implies finished weaving, and the federal clauses have been re-woven by amendment, commission and judgment more often than any other part of the constitutional fabric. The founders built the loom and supplied the thread. The pattern has been under negotiation ever since, which is what a deferral looks like when it succeeds.

Asymmetry as a Design: Article 370 and the Sixth Schedule

The clearest proof that the Assembly knew its federal settlement was provisional lies in the exceptions it wrote into the settlement’s core. On 17 October 1949, in one of the final clause-by-clause sittings, N. Gopalaswami Ayyangar, who had conducted the negotiations with Sheikh Abdullah’s government, moved draft Article 306A, the provision that entered the Constitution as Article 370. It confined Parliament’s power over Jammu and Kashmir largely to the subjects in the Instrument of Accession signed by Maharaja Hari Singh in October 1947, required the state government’s concurrence or consultation for extension of other provisions, and contemplated the state’s own constituent assembly as the authority that would define the relationship’s future. Members objected on the floor, Maulana Hasrat Mohani among them asking why discrimination in favour of one state was being entrenched, and Ayyangar answered from the circumstances of the accession, the war in Kashmir and the reference of the dispute to the United Nations. The article was adopted and then placed, pointedly, in the part of the Constitution headed Temporary, Transitional and Special Provisions. The Assembly thereby wrote into its federal settlement a state whose position was avowedly unsettled, pending processes outside the Assembly’s control. Whatever later generations made of Article 370, the founding intention recorded in October 1949 was not permanence. It was a bridge, built because the far bank had not been reached.

The Sixth Schedule tells the same story from the opposite end of the Union’s geography. The Advisory Committee’s sub-committee for the tribal areas of Assam, chaired by Gopinath Bordoloi, reported in 1947 for autonomous district councils with legislative and judicial powers over land, forests, inheritance, marriage and social custom in the hill districts, and the Assembly debated and adopted the schedule in September 1949 against objections that it created a state within a state and cut the tribal areas off from the national mainstream. Bordoloi defended the arrangements as the price of bringing the hill peoples into the Union voluntarily, and members who knew the frontier, including J. J. M. Nichols-Roy, himself a Khasi representative, supported the autonomy as recognition rather than concession. The founding text that emerged therefore contained at least three distinct federal logics operating at once: the standard logic of the Seventh Schedule for the ordinary states, the accession logic of Article 370 for Jammu and Kashmir, and the autonomy logic of the Sixth Schedule for the tribal districts, with the Fifth Schedule adding a fourth, protective regime for scheduled areas elsewhere. A constitution that distributes belonging in four currencies has not answered the question of who governs whom. It has answered it four times, differently, and left the exchange rates between the answers to politics.

This is the ground on which the adjudication between Austin and Khosla on the federal question must be conducted, and it is ground Khosla holds. Austin’s seamless web treats asymmetry as the web’s elasticity, the founders’ wise refusal to force unlike units into a single mould, and the description fits the intentions of Ayyangar and Bordoloi as the debates record them. The deferral reading adds the observation Austin’s metaphor suppresses: elastic arrangements require later decisions about how far they stretch, and the Assembly supplied no decision procedure for the day the stretching stopped. Article 370’s own text assigned its future to institutions, the state constituent assembly and the President acting on its recommendation, whose sequence no founder could schedule. The Sixth Schedule’s councils were given powers whose adequacy each succeeding generation in the hills would test, in movements that produced the states of Nagaland in 1963, Meghalaya in 1972 and Mizoram in 1987 out of the old Assam. Asymmetry was the founding’s most honest device, an admission recorded in the text that the Union had been assembled from unlike materials at unlike speeds. Honesty is not settlement. The renegotiations the device invited have consumed more political energy than any clause the Assembly actually finished, and they belong in the ledger of the second tension, credited to founders who chose a bridge over a wall and left the bridge’s far end for their successors to build.

The Third Tension: Rights That Command and Principles That Instruct

The third tension asks which past the present is continuous with, and the Assembly encoded it as a question about the future: whether the republic would be continuous with the liberal strand of the freedom movement, which spoke in the language of individual rights against the state, or with its social strand, which spoke in the language of caste abolition, land reform and economic reconstruction and required a state strong enough to act on society. The founding text answered by institutionalising both languages and refusing to subordinate either. Part Three of the Constitution carries the Fundamental Rights, enforceable by the Supreme Court under Article 32, which Ambedkar described on 9 December 1948 as the very heart and soul of the document. Part Four carries the Directive Principles of State Policy, borrowed in conception from the Irish Constitution of 1937, declared in Article 37 to be fundamental in the governance of the country and the duty of the state to apply in making laws, and in the same breath placed beyond the reach of any court. A citizen could sue for a right. A citizen could not sue for a principle. The whole architecture of the third tension stands in that asymmetry, and the debates show it was chosen with open eyes.

The drafting history supplies the dates. The Advisory Committee’s sub-committee on fundamental rights had reported in April 1947 with a division between justiciable and non-justiciable rights already drawn, and the Objectives Resolution of January 1947 had promised both justice in its social, economic and political forms and equality of status and opportunity, giving the committee two mandates that no single chapter could hold. When the Assembly debated the draft articles on the Directive Principles in November 1948, the pressure to close the gap came from the chamber itself. K. T. Shah, the economist member from Bihar, moved amendments to give the Principles teeth and to commit the state expressly to a socialist order; other members sought enforceability for the right to work and to education. Ambedkar’s reply of 19 November 1948 is the canonical statement of the founders’ position and must be quoted in substance to be judged fairly. The Directive Principles, he argued, were like the Instruments of Instructions issued under the Government of India Act of 1935: directions to those who would govern, binding in the forum of politics because a government that disregarded them would have to answer to the electorate at election time, and deliberately not binding in a court, because no generation should fasten its economic programme on the future through litigation. He added the distributive point that whoever captures power under the constitution would not be free to do as he pleased, since the Principles would stand as the standing instruction of the constitution itself. The amendments for enforceability were negatived. Article 37’s formula, fundamental but unenforceable, was adopted as written.

That formula is the third tension in statutory form, and its two halves have been pulling against each other in Indian law ever since. The rights chapters were given the courts; the principles were given the legislatures; and the question of collisions between a property owner’s Article 31 right and a legislature’s directive-driven programme of land reform was left to be fought case by case. Ambedkar’s defence contains a genuine theory, and the deferral reading does it no service by pretending otherwise. The theory assigned each half of the founding promise to the institution fitted for it: courts guard the negative liberties that majorities might invade, electorates discipline governments that neglect the positive programme that majorities demanded. As a division of labour the scheme is intelligible and even elegant. Its defect is jurisdictional. When a land reform statute reached the Supreme Court as a rights violation, as the Bihar legislation did within a year of the republic’s birth, the case arrived in the forum of rights carrying the mandate of the principles, and one forum had to rank the two halves of the promise. The Assembly’s text gave that forum no ranking rule. Ambedkar’s electorate theory works only while legislatures and courts stay in their assigned arenas, and the constitution itself guaranteed they would not, because Article 32 invited the rights-holder into court against precisely the statutes the Principles commanded. The deferral was therefore not a distribution of the question between two forums. It was a duplication of the question into two forums with contradictory instructions, which is the institutional form of leaving a question open.

Niraja Gopal Jayal’s Citizenship and Its Discontents, published in 2013, supplies the analytical frame that connects the rights settlement to the first tension’s question of belonging. Jayal reads Indian citizenship as layered rather than uniform: a universalist layer of equal individual rights laid over a differentiated layer in which the constitution itself distributes recognition by group, through minority cultural rights, caste-based reservation and the schedules for tribal areas, so that the citizen is addressed by the text alternately as an individual and as a member. Her reading explains why the rights-and-directives architecture took the shape it did. A purely individualist bill of rights would have left the differentiated layer invisible and undefended; a purely programmatic charter would have dissolved the individual into the programme. The Assembly wrote both because the country it governed was present in the chamber in both capacities, and Jayal’s layered citizenship is the precise description of the result. The deferral reading absorbs her frame gratefully and extends it one step: each layer was left with its boundary against the other layer unsurveyed. The years since have been spent surveying those boundaries by litigation, and the surveying is the third tension’s renegotiation history.

Why are the Directive Principles not enforceable in court?

Ambedkar told the Assembly on 19 November 1948 that the Principles were directions to future governments, binding through the electorate rather than through litigation. Making them enforceable would have let courts veto the economic programmes of elected legislatures, so Article 37 declares the Principles fundamental in governance while barring any court from compelling their application.

The most instructive single clause of the third tension may be Article 44, the uniform civil code, placed by the Assembly among the unenforceable principles after the debate of 23 November 1948 in which Muslim members, including Muhammad Ismail Sahib and Naziruddin Ahmad, warned against state intrusion into personal law, and Ambedkar and Munshi answered for the aspiration while accepting its postponement. A uniform code was simultaneously a promise of equal citizenship addressed to the first tension and a directive addressed to the third, and the Assembly parked it in the one part of the constitution where a promise binds nobody. No placement in the document states the deferral method more economically. The founders wanted the aspiration on the record, wanted it attributed to the constitution rather than to any party, and wanted no institution authorised to demand its fulfilment on a schedule. Article 44 is the method’s signature, and every later invocation of the article, by courts exhorting legislatures and by governments campaigning on it, has operated inside the space the Assembly cleared in November 1948 by declining to convert an aspiration into a right or a duty with a remedy attached.

The Missing Hierarchy and the Court It Created

The bill for the third tension’s deferral arrived quickly, and its payment schedule shaped the republic’s first three decades. On 9 April 1951 the Supreme Court decided the State of Madras against Champakam Dorairajan, holding that the Fundamental Rights prevailed over the Directive Principles and striking down a Madras government order allocating medical college seats by community, because the order violated the equality articles and Article 37 itself barred the Principles from overriding rights. Nehru’s government answered with the Constitution (First Amendment) Act of June 1951, which amended Article 15 to permit special provisions for socially and educationally backward classes, restricted the free speech guarantee, and created the Ninth Schedule to shield land reform statutes from rights challenges altogether. The sequence matters more than any of its elements. Within sixteen months of the constitution’s commencement, the hierarchy the Assembly had refused to state had been stated by the court in favour of rights, and then revised by the legislature in favour of its programme, each institution supplying the ranking rule the text withheld. The deferral did not avoid the conflict between the constitution’s two promises. It determined that the conflict would be conducted as an institutional contest between court and parliament, amendment against judgment, for as long as the underlying social question remained open.

The contest’s decisive engagement came on 24 April 1973, when a bench of thirteen judges decided Kesavananda Bharati against the State of Kerala by seven votes to six, holding that Parliament’s power to amend the constitution did not extend to damaging its basic structure. The judgment is usually filed under judicial doctrine, and for the argument here it belongs in a different file: it is the third tension’s deferral returning to the forum of rights with institutional force. The Directive Principles had been written as instructions to the state that no court could enforce; the basic structure doctrine, constructed by the court out of the constitution’s silences, made a set of founding commitments enforceable against the amending power itself, including commitments, such as the balance between Parts Three and Four, that the Assembly had refused to rank. When the Supreme Court in Minerva Mills against the Union of India, decided on 31 July 1980, struck down the Forty-Second Amendment’s attempt to give all Directive Principles primacy over Articles 14 and 19, the court stated expressly that the harmony and balance between rights and principles was itself part of the basic structure, and that to destroy the balance was to destroy the constitution’s identity. A hierarchy had at last been supplied, and it was supplied in the form of an equilibrium doctrine whose terms the judiciary alone would administer. The line runs in one direction only. Each stage, Champakam in 1951, the First Amendment the same year, the property amendments and the Golaknath judgment of 1967, Kesavananda in 1973, Minerva Mills in 1980, and the nine-judge bench’s restatement in I. R. Coelho against the State of Tamil Nadu in January 2007, consists of an institution answering the question the founders left open and entrenching its answer against the next institution that tried. None of these actors was improvising a new question. All of them were working the same seam between the constitution’s two promises, and the seam is located exactly where the Assembly left it in November 1948.

Austin’s reading has a ready reply, and it should be given its due before the adjudication closes on this tension. The reply is that a constitution which generates its own corrective doctrines across thirty years has demonstrated the coherence Austin attributed to it: the seamless web held, strain passing from strand to strand without rupture, and Kesavananda itself can be read as the web’s self-repair, the judiciary defending the founders’ balance against a parliament that had, during the Emergency of 1975 to 1977, attempted to tear it. The reply is strong on the level of outcome. It fails on the level of attribution. The balance the court defended in 1980 was a balance the court had defined in 1973, against a parliament claiming the Assembly’s authority for the opposite ranking, in a controversy the Assembly’s text was cited on both sides of without deciding. A settlement that requires a seven-to-six verdict of thirteen judges, twenty-three years after the founding, to determine what it settled, is better described as an inheritance under litigation than as a web woven whole in 1949. Khosla’s founders, who knew they were writing for citizens and institutions that would have to learn the constitution by contesting it, predict every stage of this history. Austin’s founders, who wove the strands into mutual dependence, predict the destination only by calling the whole journey part of the cloth.

What the Constituent Assembly of India Decided, and What It Deferred: Austin Against Khosla

Granville Austin’s The Indian Constitution: Cornerstone of a Nation appeared in 1966, close enough to the founding for its author to interview participants and to treat the Assembly’s self-understanding as evidence of the first rank. The book’s central perception, that the constitution joined national unity, democracy and social revolution in a single fabric whose strands strengthened one another, was not imposed on the record from outside. It can be documented from the record at point after point. The Objectives Resolution of 13 December 1946 and its adoption on 22 January 1947 joined the three commitments in a single preamble of intent before any drafting began. The Assembly’s preference for consensus over division, which Austin documented from the Congress Assembly Party’s working, made the weaving a social fact about the room and not merely a metaphor about the text. His account of the franchise decision, of the abolition of untouchability in Article 17, and of the Drafting Committee’s labours under Ambedkar remains the standard narrative against which later scholarship defines itself, and the deferral reading should be understood as a correction within Austin’s tradition rather than a repudiation of it. Where Austin is strongest, he is very strong: the founders did intend the strands to hold together, and for the founding generation the intention was itself an achievement with few parallels in the century of constitution-making that followed the Second World War.

The correction begins with a distinction Austin’s metaphor cannot express, between coherence of intention and closure of questions. A web can be seamlessly woven around holes. The religious clauses cohere as an intention, to hold a multi-religious polity together without a state church, while leaving the state’s relation to religion undefined. The federal clauses cohere as an intention, to hold a subcontinent together without secession, while leaving the units’ security against the centre unguaranteed. The rights chapters cohere as an intention, to promise liberty and transformation together, while leaving their collisions unranked. Austin, writing in 1966, stood close enough to the founding to take the intentions as the story, and his book’s enormous authority froze that choice into the textbook tradition: the coaching summaries and encyclopaedia entries that repeat the seamless web rarely mention that the Assembly’s record contains Ambedkar’s own contradiction speech, or that the Hindu Code died of postponement in the Assembly’s lifetime, or that Article 370 was filed under temporary provisions by the men who drafted it. Coherence became the whole account because the most influential historian of the founding chose the loom over the holes. The primary evidence supports the intentions and the holes together, and any adjudication honest to the evidence must hold both, which is precisely what the metaphor forbids and the deferral reading permits.

Madhav Khosla’s India’s Founding Moment, published in 2020, starts from the opposite end of the same record. Khosla reads the constitution as a pedagogical project: a text written in a borrowed language and an unfamiliar institutional grammar for a population most of whom encountered the state through it for the first time, intended to constitute a democratic people rather than merely to regulate a government. On his reading, codification itself was the founders’ gamble, the decision to gather the rules of political life into a single authoritative document from which citizens could learn what the state was and what they could demand of it. The pedagogical frame explains features of the record that the coherence frame treats as decoration: the extraordinary length and detail of the text, which reads at many points like a manual for administrators and citizens rather than a frame of government; the founders’ tolerance of ambiguity where a teaching text profits from open texture; and the Assembly’s persistent concern with the educative effect of its own proceedings, from Nehru’s Objectives Resolution speech of 13 December 1946, framing the task as one the future would judge, to Ambedkar’s warnings that constitutional morality was not a natural sentiment in India and would have to be cultivated. A constitution understood as a school does not fail when its lessons are contested later. Contestation is the curriculum arriving on schedule.

Adjudicating between the two readings on the evidence assembled in the preceding sections produces a split verdict, and the split is the article’s considered position. On the religious settlement, Khosla prevails: the refusal of a defining principle, the enumerated freedoms, the abandoned Hindu Code and the withdrawn minority reservation are the acts of founders managing an unteachable disagreement by deferring its theory, and Austin’s web describes their hope rather than their text. On the federal settlement, Khosla prevails with a concession recorded: the Union’s permanence was genuinely settled, as Austin insisted, and the secession question has never seriously reopened; but the settlement’s interior, asymmetry, boundary revision, fiscal dominance and emergency override, was left as open as the religious question, and the concession does not reach it. On the rights settlement, the verdict is Khosla’s outright, because the court-parliament contest from Champakam in 1951 to Minerva Mills in 1980 is unintelligible as the working of a finished balance and fully intelligible as the deferred ranking arriving in instalments. Austin’s residual point survives the adjudication and should be carried forward at full weight: the deferrals held. A constitution whose open questions produced a basic structure doctrine rather than a civil war, whose federal bridge at Article 370 stood for the whole period the founders’ generation governed, and whose personal law question was fought through elections and amendment rather than through communal collapse, achieved by postponement much of what other foundings failed to achieve by decision. The deferral reading credits the founders with that achievement and names its mechanism correctly. The mechanism was not the seamless web. It was the disciplined refusal to force the three questions that would have torn the web before it left the loom.

First-Order Decisions and Second-Order Deferrals: The Strongest Counter

The strongest counter to everything argued so far comes from the founders’ own mouths, and it is associated with the most formidable of them. Ambedkar did not speak like a man postponing the constitution’s meaning. Presenting the Draft Constitution on 4 November 1948, he answered the charge that the text borrowed too heavily from the Government of India Act of 1935 by accepting the fact and denying the shame, defended the strong centre as a deliberate choice for a country of India’s divisions, and rejected the Gandhian village republic as a model in terms that allowed no ambiguity about the decision taken. On the franchise, the Assembly chose universal adult suffrage against the gradualism that every colonial franchise committee had recommended, and the choice was neither hedged nor phased: the electoral roll prepared under Sukumar Sen for the elections of 1951 and 1952 registered some 173 million voters on the authority of a decision the Assembly took in full knowledge that most of the electorate could not read the ballot. On untouchability, Article 17 abolished the practice and forbade its exercise in any form, a sentence with no enumeration, no exception and no sunset. A reading of the founding that sees deferral everywhere dissolves against this record, and the counter must be answered on its merits rather than absorbed by adjectives.

The answer is a distinction the record itself supplies, between first-order choices and second-order choices. A first-order choice settles who holds power, under what institutions, with what formal rights: the parliamentary form, the federal structure’s existence, the franchise, the abolition of untouchability, the independence of the judiciary, the amendment procedure. On the first order the Assembly was as decisive as its defenders claim, and Ambedkar’s speeches of November 1948 are the record of that decisiveness accurately described. A second-order choice settles what those institutions mean when their commitments collide: what the state is in relation to religion when propagation meets regulation, what a state is owed by the Union when the Union is held by its opponents, what a citizen’s right is worth against the state’s directive programme when a statute embodies the second and violates the first. On the second order the Assembly’s record is one of enumerated freedoms without a principle, permanence without a boundary, and two promises without a ranking. Ambedkar’s clarity and the Assembly’s deferral are both real, and they operate on different floors of the same building. His closing speech of 25 November 1949 performs the distinction in a single address: the political structure, he told the members, was settled and good of its kind, and he then spent the speech’s most famous passages on the social and economic contradictions the structure had not touched, warning that political equality beside social and economic inequality would put the political structure itself in peril of being blown up by those who suffered from the contradiction. The chairman of the Drafting Committee closed the founding by separating what had been decided from what had been handed on. This article’s reading follows his separation, and claims no more originality for it than that.

One further piece of evidence disciplines the distinction, and it is the fate of the founders’ own attempt to revisit a deferral within the founding period. The Hindu Code’s death in the provisional parliament in September and October 1951, two years after the Assembly adopted a constitution whose Article 44 aspired to a uniform civil code and whose equality articles had been justified partly by the coming reform of personal law, shows a second-order question defeating the very coalition that had deferred it, even with Nehru’s premiership and Ambedkar’s drafting skill committed to the attempt. Patel’s death in December 1950 had removed the settlement’s manager; the elections of 1951 and 1952 were approaching; the code was dismembered into the four Acts of 1955 and 1956 by a parliament with a fresh mandate. The sequence vindicates the founders’ judgment that the Assembly could not have forced the question in 1948 without breaking itself, and simultaneously vindicates the deferral reading’s account of what the Assembly did. It chose the survival of the text over the settlement of the question, correctly by its own lights, and the question outlived the choice by generations, as deferred questions do.

The Three Tensions Table: What Was Settled and What Was Handed On

The argument of this article can be compressed into a single reference object, and the compression is offered under a name by which it can be cited: the Three Tensions Table of the founding settlement. Each row names one of the Assembly’s three governing tensions, records the position the founders actually adopted between 1946 and 1949, and traces the renegotiation that the adopted position set in motion across the decades that followed, through named decisions and dated turning points rather than through generalities. The table is the article’s findable artifact, and it is deliberately severe in its middle column, since the discipline of stating the founding position in one cell is what separates a settlement from a sentiment.

Tension	The founding position, 1946 to 1949	The renegotiation trajectory
Who counts as fully Indian: the religious settlement	Freedom of conscience and propagation guaranteed in Articles 25 to 28 with no definition of secularism and no state religion; personal law left unreformed; reservation for religious minorities withdrawn in May 1949 while caste reservation was retained	Hindu Code enacted as four Acts in 1955 and 1956; the Shah Bano judgment of April 1985 and the Muslim Women (Protection of Rights on Divorce) Act of 1986 reopened the personal law asymmetry; the Shayara Bano judgment of August 2017 set aside instant triple talaq while the uniform civil code remained an unenforced directive
Who governs whom: the federal settlement	India named a Union of States in Article 1 on 4 and 15 November 1948, indestructible and not founded on compact; three legislative lists with central residuary and emergency power; asymmetry entrenched through Article 370 in October 1949 and the Sixth Schedule in September 1949	States Reorganisation Act of 1956 redrew the internal map on linguistic lines; the dismissal of the Kerala ministry in 1959 tested Article 356; the Sarkaria Commission of 1983 to 1988 restated the states’ case; the Bommai judgment of March 1994 subjected Article 356 to judicial review
Which past the present continues with: the rights settlement	Fundamental Rights made enforceable under Articles 13 and 32 while Directive Principles were declared fundamental yet non-justiciable in Article 37, with no ranking between Parts Three and Four and the uniform civil code parked in Article 44 in November 1948	Champakam Dorairajan in April 1951 ranked rights above principles and the First Amendment of June 1951 answered for the legislature; Kesavananda Bharati in April 1973 created the basic structure doctrine; Minerva Mills in July 1980 made the balance between Parts Three and Four itself unamendable

Read down the middle column and the founding appears as the achievement story describes it: every cell contains real text, adopted after real debate, by a body that knew the alternatives. Read down the third column and the deferral reading makes its case without further argument, because every trajectory consists of the same question returning in a new institutional costume, and none of the returns was caused by the founders’ carelessness. The returns were caused by the questions’ depth. A religious settlement without a principle generates boundary disputes for as long as the communities governed by it disagree about the principle. A federal settlement without a boundary generates centre and state contests for as long as the units and the centre are held by rivals. A rights settlement without a ranking generates court and parliament contests for as long as the social programme and the property order conflict. The table’s use for readers of Indian politics is diagnostic: when a controversy presents itself as unprecedented, locate its row. The controversy’s ancestry is in the middle cell, and its likely forums, court, parliament and street in alternation, are in the third.

The Renegotiation Agenda at Work

Three episodes, one from each row of the table, show the agenda operating as the founders’ deferrals predicted it would, through institutions arguing over an inheritance rather than over fresh ground. In April 1985 the Supreme Court decided Mohd. Ahmed Khan against Shah Bano Begum, holding that a divorced Muslim woman could claim maintenance under the general criminal procedure code notwithstanding personal law, and the judgment’s reasoning reached for Quranic authority to support the result. The political system’s response ran through the first tension’s channels with documentary precision: minority organisations characterised the judgment as interference in personal law, the characterisation being theirs and stated here as attribution; the government of Rajiv Gandhi, commanding the largest majority the Lok Sabha had seated, enacted the Muslim Women (Protection of Rights on Divorce) Act in 1986 to displace the judgment’s effect; and the personal law asymmetry the Assembly had preserved in 1949, reform for one community’s law and autonomy for the others, became the explicit subject of national legislation thirty-six years after the founding. Nothing in the episode was a departure from the founding settlement. The episode was the founding settlement, still in force, still unsettled, working itself out through the court and the legislature in the sequence the deferral had arranged.

The federal row’s episode is the career of Article 356. The Assembly had adopted the emergency provisions after debates in August 1949 in which members including H. V. Kamath and K. T. Shah warned that the power to displace a state government would be abused, and Ambedkar answered with the hope, recorded in the debates of 4 August 1949, that the articles would remain a dead letter and never be called into operation. The power was used against the elected communist ministry of Kerala in July 1959, and used scores of times thereafter by governments of every complexion, until the Supreme Court’s nine-judge bench in S. R. Bommai against the Union of India, decided on 11 March 1994, subjected proclamations to judicial review, required floor tests of majority, and converted a political discretion into a reviewable act. The distance between Ambedkar’s dead letter hope of August 1949 and the Bommai discipline of March 1994 measures the second tension’s deferral exactly: the founders supplied the power and a hope, the intervening decades supplied the abuse, and the court eventually supplied the boundary the Assembly had declined to draw. Federalists who cite Bommai as the constitution working as intended and critics who cite the Kerala dismissal as the constitution working as written are describing the same settlement from its two ends, and the deferral reading holds both descriptions together by assigning each to its date.

The rights row’s third episode belongs to the constitution’s most recent founding argument. On 24 August 2017 a nine-judge bench of the Supreme Court decided Justice K. S. Puttaswamy (Retd.) against the Union of India, holding unanimously that privacy is a fundamental right intrinsic to life and personal liberty under Article 21. The judgment reached back across the whole third tension: it overruled the Emergency-era habeas corpus decision in ADM Jabalpur of 1976, in which the court had surrendered the rights forum entirely; it grounded privacy in the structure of Part Three read with the aspirations of Part Four, the balance Minerva Mills had made unamendable; and it treated the founders’ silences, the absence of a privacy clause in a text written before the surveillance state, as space for the constitution’s principles to work rather than as a prohibition on their working. A reader formed by the achievement story sees in Puttaswamy the seamless web holding under strain. A reader formed by the deferral reading sees the more precise truth: the founders left Article 21’s phrase, procedure established by law, deliberately narrower than the American due process the Assembly had been advised by B. N. Rau’s consultations to avoid, and the courts spent seventy years widening that phrase by interpretation, from A. K. Gopalan in 1950 through Maneka Gandhi in 1978 to Puttaswamy in 2017, because the widening was the work the founding had assigned to the future. The third tension’s renegotiation is the longest of the three because the question it carries, what the republic owes the individual against the state and society, is the one on which the freedom movement’s two languages disagreed most deeply, and the Assembly, having written both languages into the text, left their translation to its heirs.

Why the Deferrals Held: The Coalition Logic of Postponement

The deferrals survived the drafting process because the Assembly’s composition rewarded postponement and punished definition, and the arithmetic can be stated. In the elections of July and August 1946 the Indian National Congress won 208 of the 296 seats allotted to British India, and after the Muslim League’s withdrawal and the Partition reconstitution, Congress members and their allies commanded an overwhelming share of the reduced chamber, with the princely states’ nominees generally aligned with the Congress-led States Ministry on the questions that reached a vote. A body so composed did not lack the votes to define secularism, bound the centre or rank the rights chapters. It lacked a reason to spend its unity on definitions when its dominant party contained, inside a single organisation, the men and women who would have been on opposite sides of each definition: socialists around Nehru and conservatives around Patel, Gandhians and modernisers, Hindi advocates and their southern opponents, all wearing the same party label and all needed for the constitution’s ratification in the country. Rajni Kothari would later name the arrangement the Congress system, a party of consensus containing its own opposition, and the Assembly was that system in its founding session. Definition would have externalised the opposition, giving the party’s defeated wing a reason to campaign against the constitution itself. Deferral internalised every disagreement, converting potential wreckers into stakeholders of a text each faction could read as its own promissory note.

The founding’s calendar conspired with the arithmetic. Mahatma Gandhi was assassinated on 30 January 1948, three weeks before the Draft Constitution was published, removing the one figure whose moral authority might have forced questions of religion and village order to a conclusion the party managers preferred to avoid, and the drafting proceeded under the discipline of mourning and national emergency. Patel died on 15 December 1950, little more than a year after the adoption, and the provisional parliament that buried the Hindu Code in the autumn of 1951 was already a body whose settlement manager was gone and whose prime minister faced the first general election. The deferrals, in other words, were made by a coalition that expected to remain in power and to conduct the renegotiations itself, from office, at moments of its own choosing. That expectation was reasonable in 1949 and false by 1967, when the Congress lost its monopoly in the states, and it explains both the deferrals’ generosity and their risk. The founders handed the three questions to their own party’s future governments. The questions were inherited instead by coalition governments, regional parties, social movements and courts, actors the founders had not seated at the drafting table, and the renegotiations acquired the unpredictability that has marked them since. The deferral system’s success, measured by the republic’s survival with its argument intact, was achieved by institutions improvising on an inheritance designed for other hands, which is the final reason the founding rewards study as an allocation of questions across time rather than as a settlement carved in a single session.

The Evidence Base: Debates, Drafts, Letters and the Historians

An adjudication is only as good as its evidence, and the founding is unusually well documented for a revolution conducted in committee rooms, which is why the argument of this article has been built from four classes of material whose dates and custodians can be stated. The first class is the Constituent Assembly Debates themselves, the verbatim official report of the proceedings from 9 December 1946 to 24 January 1950, published by the Lok Sabha Secretariat and running to twelve volumes, of which Volume Five carries the August 1947 reorganisation and minorities sittings, Volume Seven the great clause-by-clause debates of late 1948 in which Articles 1, 25 and the Directive Principles were fought over, and Volume Eleven the closing debates of November 1949 including Ambedkar’s speech of 25 November. The debates are a self-interested record, kept by the victors of each vote, and the historian’s caution applies: a speech in the Assembly is an act of persuasion addressed to a chamber, not a confession addressed to posterity. The caution cuts against both readings impartially, however, and the deferral reading relies less on speeches than on outcomes, the amendments negatived, the clauses unadopted, the bills unpassed, which the record cannot flatter because it merely registers them.

The second class is the drafting paper: B. N. Rau’s preparatory draft of October 1947, the committee reports of that year, and the Draft Constitution of February 1948 with the Drafting Committee’s alternatives and its digest of the public comments the draft attracted. B. Shiva Rao, who had served in the Assembly’s secretariat and as a member, edited the great documentary collection The Framing of India’s Constitution: Select Documents, published in five volumes between 1966 and 1968, with his analytical companion volume, A Study, and the collection preserves the memoranda, the committee drafts and the correspondence through which the paired provisions of the final text were assembled. The drafting paper matters to the adjudication because it dates the deferrals. A silence in the final text might be an oversight; a silence that appears in Rau’s draft of October 1947, survives the committee stage, is printed in February 1948, is debated in November 1948 with the alternative expressly moved, and is adopted in 1949, is a policy. The religious definition, the federal boundary and the rights ranking each pass that five-stage test, which is why this article has treated the three silences as the founding’s most deliberate products.

The third class is the private paper of the principals, of which the Patel correspondence is the outstanding example for the founding’s managerial layer. Sardar Patel’s Correspondence, 1945 to 1950, edited by Durga Das and published in ten volumes in the early 1970s, carries in its sixth volume the letters of 1947 and 1948 between Patel and Nehru on the communal situation, the states and the Hindu Code, including the exchange of 12 October 1948 on which the treatment of the personal law deferral in this article rests. Private letters correct the debates in a specific way: where the chamber record shows what the coalition could say in public, the correspondence shows what its managers believed in private about timing, risk and the limits of the possible, and on the Hindu Code the two records diverge in the instructive direction, public commitment running ahead of private intention by a distance that explains the bill’s death in 1951 better than any account of parliamentary arithmetic. The correspondence of other principals, and the memoirs of participants such as K. M. Munshi, supply checks on the same episodes from other desks in the same rooms.

The fourth class is the historiography itself, whose sequence is part of the evidence for how the founding has been remembered. Austin’s Cornerstone of a Nation of 1966 founded the field from within the participants’ generation and fixed the coherence reading as the default. The documentary turn of Shiva Rao’s volumes made the drafting paper available to scholars who had not been in the room. A later generation, working after the Emergency of 1975 to 1977 and the basic structure litigation had stress-tested the founding settlement, asked harder questions about power and exclusion in the Assembly, and the most productive recent work, Jayal’s Citizenship and Its Discontents of 2013 on the layered character of belonging and Khosla’s India’s Founding Moment of 2020 on the pedagogical wager, has returned to the primary record with frameworks the first generation did not possess. The adjudication offered in this article is positioned inside that sequence rather than above it: it accepts Austin’s evidence about intentions, accepts Khosla’s framework of a founding that expected to be continued by its readers, and adds the documentary observation that the continuation was necessary because the text, at its three most contested points, was built to be continued. Readers who wish to test the verdict have the means at hand, since every class of material named here is in print or in the public archive, and the debates repay reading in the order the Assembly conducted them, as an argument that its participants knew would be resumed.

The Amendment Door: Article 368 and the Expectation of Revision

One mechanism converted the Assembly’s deferrals from postponements into a system, and it deserves separate treatment because later generations have consistently misread it as a mere procedure. The amending power, debated in its draft form in September 1949 and settled as Article 368, was designed by the founders in three gears: some provisions amendable by simple majority as ordinary legislation, most of the text by special majority of Parliament, and the federal core only with ratification by half the states. Nehru, speaking in the Assembly on 8 November 1948 during the general discussion of the draft, had laid down the philosophy in advance: while the founders wanted the constitution to be as solid and permanent as a structure could be, there was no permanence in constitutions, and a rigid text would stop the growth of a living nation. Members who feared easy amendment, and members who feared difficult amendment, both spoke in the September 1949 debates, and the Assembly adopted the middle instrument with the same equanimity with which it had adopted the deferrals the instrument would service.

The connection between the amendment door and the three tensions is structural, and it closes the article’s evidentiary case. A founding that settles its questions entrenches them against amendment, as the American founders entrenched their settlement behind Article V’s near-impossibility. A founding that defers its questions must instead equip its successors with a workable revision procedure, and the Indian founders, whatever their rhetoric of permanence, built the more workable door of any major twentieth-century constitution. The traffic through the door confirms the design’s purpose. The First Amendment of June 1951 walked through it within eighteen months to revise the rights settlement after Champakam. The Seventh Amendment of 1956 walked through it to execute the linguistic reorganisation the founders had postponed. The Forty-Second Amendment of 1976 attempted to walk through it with the entire settlement in its arms, and the basic structure doctrine that Kesavananda had announced in April 1973 was the judiciary’s answer, a judicially built second door, locked against the first door’s abuse. Even the Preamble’s description of the state was revised through the door in 1976, when the words socialist and secular entered it, twenty-seven years after the Assembly had debated religious definition and declined to write one. Read as a whole, the amendment history is the deferral system’s operating log: the questions the founders left open have been re-opened, on average, more often than once a year since 1950, always through or against the procedure the founders supplied for the purpose. The founders did not merely fail to finish their work. They built the workshop in which the work would continue, staffed it with their successors, and left the three largest commissions on the board.

The Questions Outlasted the Answers

The founding generation’s record, read whole, supports a verdict more demanding than either celebration or debunking. Between 9 December 1946 and 26 November 1949, a body that had lost a third of its country to Partition, that governed through a refugee emergency while it drafted, and that contained within itself nearly every organised interest of a subcontinent, produced a complete constitution for the largest electorate the world had seated. Within that achievement, on the three questions that structure Indian political life, the same body practised a consistent and conscious deferral. It enumerated religious freedom without defining the secular, because a definition would have split the coalition that carried the text. It named the Union indestructible without bounding the centre, because a boundary would have reopened the bargaining that Partition had closed by force. It promised rights and transformation in adjacent chapters without ranking them, because a ranking would have forced the freedom movement’s two languages into a contest one of them would have lost in the chamber and avenged outside it. The namable claim with which this article began can therefore be restated with the evidence behind it: the Constituent Assembly’s most consequential choices were the questions it chose not to answer, and the choices were consequential precisely because the questions were the permanent ones, who counts as fully Indian, who governs whom, and which past the present is continuous with.

The deferral reading also carries a discipline for reading the politics that followed, and the discipline is worth stating as plainly as the claim. When a controversy over personal law reaches the legislature, when a governor’s report reaches the President, when a statute embodying the Directive Principles reaches a courtroom as a rights violation, the controversy is not a malfunction of the founding design and not a betrayal of it. It is the design operating in the only mode the founders left available for such questions, renegotiation among institutions and movements, each armed with a genuine fragment of the founding text. Austin taught readers to see the founding’s unity. Khosla taught readers to see its pedagogy. The record of the debates, the February 1948 draft and the correspondence of October 1948 teaches a third sight that contains the other two: a founding is an allocation of questions across time, and the quality of a founding is measured by whether the questions it defers are the ones its successors can afford to keep arguing about. By that measure the Assembly of 1946 to 1949 chose extraordinarily well, since the republic has argued about its three bequeathed questions for generations without the argument once dissolving the state the argument is about. The tensions were named at the founding, carried in the text, and handed on intact. They remain the agenda because the founders, knowing exactly what they were doing, wrote them into the republic’s permanent business.

Frequently Asked Questions

Q: What was the Constituent Assembly of India?

The Constituent Assembly of India was the body elected in 1946 to frame a constitution for independent India. It first met on 9 December 1946 in the Constitution Hall in New Delhi, with Sachchidananda Sinha as interim president and Rajendra Prasad elected permanent president on 11 December 1946. Composed of representatives chosen by the provincial assemblies and nominees of the princely states, it drafted the Constitution over eleven sessions, adopted the text on 26 November 1949, and doubled as the country’s legislature during the transition from British rule to the republic that commenced on 26 January 1950.

Q: When did the Constituent Assembly meet and how long did it take to draft the Constitution?

The Assembly met for the first time on 9 December 1946 and completed its constitution-making work on 26 November 1949, when the Constitution was adopted. The full span was two years, eleven months and eighteen days, within which the members sat for 165 days across eleven sessions. Much of the drafting was done in committees during 1947, the Drafting Committee under B. R. Ambedkar produced the Draft Constitution published in February 1948, and the clause-by-clause reading of that draft occupied the Assembly from 15 November 1948 to 17 October 1949, followed by a third reading in November 1949.

Q: Who were the key members of the Constituent Assembly?

Rajendra Prasad presided over the Assembly. Jawaharlal Nehru moved the Objectives Resolution of 13 December 1946 and chaired the Union Constitution Committee. B. R. Ambedkar chaired the Drafting Committee appointed on 29 August 1947 and piloted the draft through the chamber. Sardar Vallabhbhai Patel chaired the Advisory Committee on fundamental rights and minorities and managed the integration of the princely states. Around them worked K. M. Munshi, Alladi Krishnaswami Ayyar, N. Gopalaswami Ayyangar, T. T. Krishnamachari and K. Santhanam on the drafting side, with B. N. Rau as constitutional adviser, and members such as K. T. Shah and H. V. Kamath pressing amendments from the floor.

Q: Why is the Constituent Assembly significant for Indian democracy?

The Assembly’s significance rests on the decisions it took at the first order of politics. It adopted universal adult franchise for a population of hundreds of millions at a time when most had never voted, rejecting the property and education qualifications of the colonial franchise. It abolished untouchability in Article 17, wrote an enforceable charter of fundamental rights, created an independent judiciary and a parliamentary executive answerable to an elected legislature, and did all of this through a deliberative procedure whose debates remain on the public record. Indian democracy’s institutional skeleton, and the habit of arguing about politics in constitutional language, both date from its sittings.

Q: Did the Constituent Assembly resolve all major constitutional questions?

It did not, and the record shows the non-resolution was deliberate on the largest questions. The Assembly guaranteed religious freedom in Articles 25 to 28 without defining the secular character of the state, and left personal law unreformed after the Hindu Code Bill stalled. It declared India a Union of States without fixing the boundary between central power and state autonomy, and wrote asymmetry into the settlement through Article 370 and the Sixth Schedule. It placed enforceable Fundamental Rights beside unenforceable Directive Principles without ranking them. Those three deferrals, on religion, federalism and rights, became the recurring agenda of the republic’s courts and legislatures.

Q: What did Ambedkar say in his closing speech to the Constituent Assembly?

Speaking on 25 November 1949, the day before the Constitution was adopted, Ambedkar warned that on 26 January 1950 the country would enter a life of contradictions: equality in politics, with one person one vote, beside inequality in social and economic life. If the contradiction were not removed, he argued, those who suffered from it would blow up the structure of political democracy the Assembly had laboured to build. He also cautioned that a good constitution worked by bad hands would prove bad, that hero worship in politics was a road to degradation, and that India must hold fast to constitutional methods of change.

Q: How does the Constituent Assembly’s work shape contemporary Indian politics?

The Assembly shapes the politics that followed it through both its decisions and its deferrals. Its decisions supplied the operating system: parliamentary government, adult franchise, judicial review and a federal division of powers. Its deferrals supplied the agenda. Disputes over personal law and the uniform civil code conduct the argument the founders postponed in 1948. Contests between the centre and the states over governors, boundaries and Article 356 replay the federal question the founders left unbounded. Litigation balancing fundamental rights against the Directive Principles, from Champakam Dorairajan in 1951 to the basic structure cases, continues the ranking exercise the founders declined to perform in the text.

Q: What is the Granville Austin reading of the Indian Constitution?

Granville Austin, in The Indian Constitution: Cornerstone of a Nation, published in 1966, read the founding as a coherent achievement in which the Assembly wove national unity, democracy and social revolution into what he called a seamless web, each strand strengthening the others and none pursued at the others’ expense. Working from the debates and from interviews with participants, Austin emphasised the Assembly’s method of consensus over majority voting and treated the constitution as the cornerstone on which the new nation rested. His reading became the canonical textbook account, against which later scholarship, including the deferral reading, defines its corrections.

Q: How were the members of the Constituent Assembly selected?

Members were not elected directly by the public. Under the Cabinet Mission Plan of 16 May 1946, seats were allotted to the provinces roughly at one member per million of population, divided among general, Muslim and Sikh categories, and the provincial legislative assemblies elected their representatives by proportional representation through the single transferable vote in July and August 1946. The princely states, allotted 93 of the original 389 seats, sent representatives chosen through negotiation with their rulers rather than by election. After Partition, members from the territories forming Pakistan withdrew, and the Assembly continued with the representation of the continuing Indian provinces and acceding states.

Q: What was the Cabinet Mission Plan, and how did it bring the Constituent Assembly into being?

The Cabinet Mission was a delegation of three British cabinet ministers, Lord Pethick-Lawrence, Stafford Cripps and A. V. Alexander, sent to India in March 1946 to negotiate the transfer of power. Its plan of 16 May 1946 proposed a Union of India holding defence, foreign affairs and communications, with provinces free to form groups and with residuary powers, and it set out the machinery for a constitution-making body of 389 members elected by the provincial assemblies. Congress accepted the plan’s constitution-making scheme while disputing its grouping provisions, the Muslim League first accepted and then withdrew, and the Assembly convened on 9 December 1946 under the plan’s framework.

Q: What did the Objectives Resolution commit the future republic to?

Jawaharlal Nehru moved the Objectives Resolution on 13 December 1946, and the Assembly adopted it on 22 January 1947. It declared the Assembly’s resolve to proclaim India an independent sovereign republic and to draw up a constitution guaranteeing justice in its social, economic and political forms, equality of status and opportunity, freedom of thought, expression, belief, faith, worship, vocation and association, and safeguards for minorities, backward and tribal areas and depressed classes. It promised that the republic would maintain the integrity of its territory and contribute to world peace. The Resolution supplied the template from which the Preamble was later drawn.

Q: Why did the Constitution come into force on 26 January 1950?

The Assembly adopted the Constitution on 26 November 1949, and a small group of provisions, covering citizenship, elections, the provisional parliament and transitional matters, took effect that day. For the main commencement the Assembly chose 26 January 1950 because of the date’s place in the freedom movement’s calendar. The Indian National Congress, at its Lahore session of December 1929 under Jawaharlal Nehru’s presidency, had declared complete independence, Purna Swaraj, as its goal, and 26 January 1930 was observed across the country as Independence Day and commemorated every year thereafter. Beginning the republic on the twentieth anniversary joined the legal founding to the movement’s older pledge.

Q: Why did the founders adopt universal adult franchise in one step?

The Assembly rejected the gradualism of the colonial franchise, under which property, tax and education qualifications had limited the vote to a fraction of adults, holding that sovereignty claimed in the name of the people could not be rationed by the state the people were replacing. Congress commitments ran back to the Nehru Report of 1928 and the Karachi Resolution of 1931, both of which had demanded adult suffrage. Members including Alladi Krishnaswami Ayyar defended the leap in the chamber, arguing that democracy could not be taught by withholding its instrument. The decision, carried into Article 326, required an electoral roll for some 173 million voters at the first general election.

Q: Why did the Constituent Assembly choose parliamentary government over a presidential system?

Ambedkar explained the choice when presenting the Draft Constitution on 4 November 1948. A presidential system on the American pattern, he argued, offers stability at the price of responsibility, since the executive is fixed for its term, while the parliamentary system offers daily accountability of the government to the legislature through questions, motions and debates. The Drafting Committee, he said, preferred more responsibility to more stability. Familiarity also counted: Indian legislators had worked responsible government in the provinces under the Government of India Act of 1935. Amendments moved by K. T. Shah and others for a presidential executive were debated and defeated.

Q: How much of the Constitution was carried over from the Government of India Act of 1935?

A substantial portion of the administrative and federal machinery was adapted from the Government of India Act of 1935, the longest statute the British Parliament had passed for India; the borrowed provisions are often counted at about 250, covering the federal scheme, the judiciary’s structure, the governors’ offices, emergency provisions, public service commissions and much administrative detail. Ambedkar met the criticism directly on 4 November 1948, saying there was nothing to be ashamed of in borrowing, that nobody held patent rights in the fundamental ideas of a constitution, and that variations were made to remove defects and suit the country’s needs.

Q: How many women sat in the Constituent Assembly, and what did they change?

Fifteen women sat in the Assembly, among them Sarojini Naidu, Vijaya Lakshmi Pandit, Sucheta Kripalani, Rajkumari Amrit Kaur, Hansa Mehta, Renuka Ray, Durgabai Deshmukh and Dakshayani Velayudhan, the only Dalit woman member. Their interventions left marks on the text and the record. Hansa Mehta and Rajkumari Amrit Kaur pressed the case for a uniform civil code against the caution of the majority. Dakshayani Velayudhan spoke against separate electorates and objected to the use of the term Harijan for the Scheduled Castes. Begum Aizaz Rasul argued against reserving seats for religious minorities. Their presence also stood as a first-order fact about the franchise settlement the Assembly was writing.

Q: Did the Constituent Assembly also serve as India’s legislature?

Yes. From its sitting of 17 November 1947 the Assembly met in two capacities, as the constitution-making body under Rajendra Prasad’s presidency and as the Dominion Legislature, which took the name Constituent Assembly (Legislative) and elected G. V. Mavalankar as its Speaker. In the legislative capacity it passed ordinary laws and budgets for the Dominion under the adapted Government of India Act of 1935, debated the Hindu Code Bill, and handled the legislation of the transition, including the measures required by Partition and the integration of the princely states. The double duty explains the Assembly’s long calendar and much of its working method.

Q: What does the original Preamble show about the republic the founders intended?

The Preamble as adopted in 1949 opens with the claim of popular sovereignty, We the People of India, and constitutes India as a sovereign democratic republic securing justice, liberty, equality and fraternity, with fraternity specified as assuring the dignity of the individual and the unity of the nation. It was moved and settled late in the drafting, in October 1949, so that its terms would conform to the constitution it prefaced, and it descends directly from the Objectives Resolution of January 1947. The words socialist and secular and the phrase integrity of the nation were added only by the Forty-Second Amendment of 1976, so the original text states the founders’ own selection of commitments.

Q: What is Madhav Khosla’s argument in India’s Founding Moment?

Madhav Khosla, in India’s Founding Moment, published in 2020, reads the Constitution as a pedagogical project. The founders, on his account, were creating democratic citizens in a society without democratic habits, and they chose codification, a single long and detailed text in a common legal language, as the instrument by which ordinary people could learn what the state was, what it owed them and how its power could be questioned. The constitution’s length, its mixture of principle and administrative detail, and its tolerance of open texture are, for Khosla, features of a teaching document. The founding’s significance lies in the wager that a people could be educated into democracy by a text.

Q: What became of the Constituent Assembly after the Constitution took effect?

The Assembly did not vanish on 26 January 1950. Under the Constitution’s transitional provisions it continued as the Provisional Parliament of the republic, legislating for the country until the first House of the People was constituted after the general elections of 1951 and 1952. In that capacity it enacted the Representation of the People Acts of 1950 and 1951 that built the electoral machinery, passed the First Amendment of June 1951, and debated the Hindu Code Bill before that measure stalled and Ambedkar resigned from the cabinet. Its members who sought election entered the new Parliament through the ballot, closing the passage from a nominated founding to an elected legislature.