Federalism

Constitutional Provisions

Political science derives its theory of federalism largely from the United States, which established the world's first federal government, so the standard account of the concept leans on American constitutional features. Etymologically, federalism traces back to the Latin term Foedus, which means a contract, and for this reason federalism is often termed government by contract. The American experience began when thirteen colonies decided to unite, giving the US what is called the 'coming together model'. Theory suggests that a number of impulses typically drive the creation of a federation:

• Security - member units come to realise that they cannot defend themselves individually.
• Economic - a federation creates a vast common market with smooth movement of goods, capital, services and people.
• Dicey's view that federalism is a political contrivance adopted by communities seeking unity without uniformity.

Because a federation rests on a contractual arrangement, its defining features include:

• Constitution - an agreed set of written rules that serves as the contract document.
• Independent Judiciary - which acts as an umpire settling conflicts that arise between the contracting parties.
• Division of powers - a clear allocation of subjects to each level of government; in India this takes the form of the union, state and concurrent lists.

Although the US provides the earliest model, it would be a mistake to treat it as the universal ideal, since several other federal models exist. Granville Austin regarded every federation as sui generis, shaped by its own distinctive circumstances and needs. The US represents dual federalism, while Canada pioneered the cooperative variant, and the Indian arrangement stands closer to Canada's approach.

Any serious discussion of federalism must begin by separating federation from confederation and the unitary system.

A federation is a political arrangement in which authority is constitutionally shared between a national government and regional governments such as states, with each tier enjoying distinct spheres of sovereignty and governing independently within those spheres. The United States and Germany are commonly cited examples.

A confederation, by contrast, leaves most power with regional governments while the central authority remains weak. Member states assign limited responsibilities, usually confined to matters such as defence or foreign policy, to the centre, creating a loose association in which central authority functions only with the consent of the members. The member states keep their sovereignty and may withdraw at will. The European Union and ASEAN illustrate this form.

In a unitary system, a single central government holds the bulk of governing power and delegates authority downwards to regional or local bodies as it chooses. Britain, France and Japan are standard examples.

The Indian arrangement blends both federal and unitary characteristics.

The US Supreme Court has described a federation as an indestructible union, while a confederation remains destructible.

Units in a federation permanently surrender external sovereignty, which means they can neither secede from the union nor conduct an independent foreign policy. Article 253 of the Indian Constitution illustrates this, since the normal division of powers between the centre and states yields to India's international treaties and agreements.

The former USSR functioned as a confederation whose units enjoyed the right of secession and the freedom to pursue foreign policy, which is why India maintained diplomatic ties with Central Asian republics such as Tajikistan even while they remained part of the Soviet Union.

In short, a confederation may be described as a loose federation.

A basic contrast between the American and Indian federations lies in symmetry: the US is a symmetrical federation while India is asymmetrical. Symmetry here refers to the status of constituent units. Every American state sends two senators to the Senate irrespective of population, whereas Indian states are not equally represented in the Rajya Sabha.

India also grants special arrangements to particular states, such as Art 371(A) for Nagaland, and applies the Fifth Schedule to some states and the Sixth Schedule to others.

Alfred Stepan described India's asymmetrical arrangement as demos-enabling and America's symmetrical model as demos-constraining, with demos meaning the people. The Indian design, in which Rajya Sabha representation follows population, is considered more democratic because it ensures that citizens are not disadvantaged by living in a large or small state. In his view, developing countries should look to India's model rather than America's.

Article 3 is sometimes cited as a non-federal feature of the Indian Constitution, but this reading is mistaken.

Centre-State Relations

India's federal design attempts to balance power between the union and the states, fostering cooperation while accommodating regional diversity, yet the same division of authority repeatedly breeds disputes and friction. Centre-state conflicts in India stem from disparities in the powers allotted to the two tiers across political, administrative and fiscal fields, and such tensions can hobble governance and slow the socio-economic progress of states.

In the legislative domain, the cards are stacked in favour of the centre. The 42nd Amendment enlarged union powers still further while trimming the subjects available on the state list.

Even outside emergency situations, the union can legislate on state subjects.

Under Article 249, the Rajya Sabha may pass a resolution authorising Parliament to make laws on an entry in the state list.

Article 252 allows Parliament to legislate on a state-list subject with the consent of at least two states. Whereas Article 249 operates as a temporary arrangement, Article 252 effects a permanent transfer of legislative power.

Article 253 likewise favours the union, empowering it to pass laws implementing international agreements.

Beyond these parliamentary powers, Article 201 gives the governor the discretion to reserve a state bill for the President's consideration.

Governors have exercised this discretion with gross arbitrariness. A bill that sails through in one state is reserved in another even when the two are similar. The Maharashtra Control of Organised Crime Act (MCOCA), enacted in 1999 to fight organised crime and terrorism, had a counterpart in the Gujarat Control of Organised Crime Act (GUJCOCA), yet the Gujarat legislation languished without presidential assent for years and only received it in 2019 after a change of government at the centre.

No time limit binds the President on bills reserved for assent, and some bills have remained pending for decades. Such arbitrary conduct by the union executive undercuts parliamentary democracy within the states.

Thus, in the legislative arena the states find themselves distinctly weak.

Though the Constitution describes India as an executive federation, it simultaneously hands the union enormous bargaining strength. Several issues in the administrative sphere are particularly contentious:

• Article 365 requires a state to comply with executive directions of the union, and failure to do so may be treated as proof that the state government cannot be run in accordance with the Constitution, furnishing a ground for President's rule.
• Article 356, which permits the imposition of President's rule, has been the most controversial provision of all. This uniquely Indian arrangement was borrowed from the colonial Government of India Act, 1935.

Article 356 is cited as a leading cause of secessionist sentiment in states such as Jammu and Kashmir and Punjab.

Constituent Assembly members raised objections to it, but Ambedkar reassured them that the provision existed only as a precaution and promised it would remain a 'dead letter'.

Practice has belied that promise, for the provision has been invoked around 115 times in fewer than seventy years of the republic. Its heavy use continued through the 1980s, after which misuse declined, thanks chiefly to coalition governments at the centre and judicial activism, with the credit going above all to the judiciary.

A classic instance of arbitrary invocation came in 1977, when the freshly elected Janata Party dissolved nine state governments run by the Congress on the logic that, having lost the Lok Sabha election, the Congress could be presumed to have lost the confidence of voters in the states too. Indira Gandhi repeated the same act in 1980, dissolving nine state governments on identical grounds.

In S R Bommai (1994), the Supreme Court declared federalism part of the basic structure and laid down guidelines regulating recourse to Article 356.

The Constitution bars courts from inquiring into the advice tendered by the Council of Ministers on the basis of which the President invokes Article 356. In Bommai, however, the Supreme Court held that the judiciary may demand the 'material evidence' that would lead any rational person to take such a decision, and that this evidence must take the form of a 'speaking document', self-evident on its face. The Court further insisted that Article 356 should be a last resort, to be used only when absolutely necessary.

The Court also ruled out automatic dissolution of the state assembly until Parliament approves the proclamation; until then the assembly remains suspended, and even after parliamentary approval the matter is open to judicial review. If the Court finds mala fide intent, it may restore the suspended assembly, including the government.

Arunachal Pradesh in 2016 supplied the first instance in which the Supreme Court actually restored a dissolved assembly.

The all-India services are a colonial legacy and a distinctive Indian feature, yet they weaken parliamentary democracy.

Officers of these services are appointed by the union and can be removed only by the union; a Chief Minister may suspend them but no more. Because ultimate disciplinary authority rests with the centre, these officers function as 'agents of the union', and there are cases where a suspended official has been reinstated by the union to the very same post and department.

This arrangement not only erodes the Chief Minister's position but also politicises the civil services. The Supreme Court directed the centre to set up a civil services board, a step that has never been taken; such a board could decide promotions, transfers and disciplinary matters.

Law and order is a state subject, yet the 42nd Amendment authorised the union to deploy its forces to aid the civilian authorities of a state. Article 355 supplied the constitutional basis, declaring it the Union's duty to protect each state against internal disturbance.

Concerns raised by the state governments

• Suo motu deployment, though the Supreme Court has held that such deployment does not violate federalism.
• The powers, privileges and immunities of these forces are determined exclusively by the union, and the states object that they should be consulted.

The American system is labelled coming together and is said to be centrifugal, with the force pulling towards the periphery, while India is described as holding together and centripetal, with the force pulling towards the centre.

In theory a federal system calls for a presidential form of government, for a parliamentary system rests on party discipline and parties tend to distort federalism, whereas a presidential system does not depend on party structures.

It has become an open secret that India's centre-state disputes are really party-to-party disputes. The system works smoothly when the same party rules both the centre and a state, but once the combinations diverge, friction appears and cooperative federalism degenerates into bargaining.

M P Singh identified two axes of the Indian political system: the parliamentary axis and the federal axis.

The parliamentary axis signifies a strong centre, while the federal axis signifies strong states. When a single party commands an absolute majority at the centre, the parliamentary axis dominates, as under Nehru, Indira Gandhi and the Modi government of the last ten years. When coalition politics, which embodies regionalisation, takes over, the federal axis becomes prominent.

Federalism in India has thus evolved continuously with the changing complexion of the party system.

Cooperative vs Competitive Federalism

K C Wheare set the Indian model against the American and judged it quasi-federal, maintaining that India is federal in form but unitary in spirit, and describing the Constitution as carrying subsidiary federal features alongside prominent unitary ones.

Wheare's method is legal-constitutional in character, confining itself to legal and institutional structures. Such a static approach leaves out the dynamic dimension of federalism.

M P Singh counters that federalism is a dynamic concept which must be studied in its socio-cultural context rather than through statutes and institutions alone, and he points out that both Indian and American federalism have changed over time.

Even the American federation has not remained what it was. Roosevelt, who built the welfare state in the US, spoke of 'new federalism', by which he meant an unmistakable centralising trend.

In India, the position of state governments has strengthened since 1989 owing to the new economic policy and coalition politics. Indeed, state bargaining power grew so markedly that scholars such as Balveer Arora concluded that India was moving from quasi-federal to quasi-confederal.

Since 2014, however, the ruling party has enjoyed an absolute majority and something close to one-party dominance. Suhas Palshikar describes this as the emergence of a 'BJP system', with clear centralising tendencies.

The fractured mandate of 2024 has reopened the question of what federal dynamics the coming years will produce.

Wheare's assessment is also ethnocentric, for in branding India quasi-federal he took the US Constitution as the benchmark. Because the Indian federation was not modelled on the American one, he concluded it was not truly federal.

Granville Austin argues that every federation is sui generis, the product of its own peculiar circumstances, and that a country threatened by secessionist tendencies, whether India or Canada, is bound to adopt a model with a strong centre. Austin considers the Indian model an example of 'cooperative federalism'.

M P Singh notes that cooperative federalism entered the vocabulary of federal theory only in the 1930s, when the American welfare state arrived, so that by the time the Indian Constitution was drafted the idea was already well established.

The US Constitution exemplifies dual federalism, while India embodies cooperative federalism. Under dual federalism the two levels of government are wholly separate entities, like two watertight compartments.

Cooperative federalism is built on mutual dependence rather than separation. The two governments are not watertight compartments but interlocking ones, each made dependent on the other. The states depend on the union financially, while the union depends on the states to implement its laws and policies, since union laws are executed by state machinery. India is therefore often described as an example of 'executive federalism', unlike the legislative federation of the US, and the significance of states emerges in the context of administration.

Yet M P Singh cautions that in most situations cooperative federalism remains an ideal, while practice turns it into bargaining federalism.

Cooperation implies trust between the two levels of government, expressed through consultation and mutual confidence. Prime Minister Modi captured this spirit in his phrase 'team India'. Within cooperative federalism, the centre is expected to act as a friend, philosopher and guide.

Bargaining federalism, by contrast, finds the centre behaving like a bullying elder brother or a patriarch. Because the centre's bargaining power is greater, bargaining federalism reveals the disadvantaged position of the states, though the regionalisation of the party system has enlarged the states' bargaining strength.

Modi's vision is of 'competitive, cooperative federalism', a neo-liberal idea built on 'minimum government, maximum governance'. It implies rolling back the state: as developmental functions are withdrawn from state governments, the private sector fills the vacuum, and so state policy must be crafted to attract private investment.

Balveer Arora doubts that competitive federalism suits the Indian context, since conditions differ widely across states; it would spawn unfair competition and unbalanced development.

Political observers note that the union has discarded the formula-based approach of the Planning Commission and enlarged its own discretionary power to strike bargains. M P Singh charges the present government with having introduced a spoils system.

Regional Aspirations

Federalism never operates in a vacuum, and its working is shaped by a host of environmental factors. Alongside the party system, the economic policy and political economy of the state condition federal dynamics: a neo-liberal developmental model strengthens the states, whereas a welfare-oriented model strengthens the centre.

Granville Austin reminded us that every federation is sui generis and that the US cannot be treated as the ideal type. The Indian model, a cooperative federation, was logically dictated by India's historical circumstances. Ambedkar was explicit that India is an indestructible union of destructible states, in other words a holding together model.

Louise Tillin, in her book Remapping India, observes that India's internal territorial map is still not settled. Indian boundaries have historically been colonial rather than natural, and Gandhi himself accepted and campaigned for the reorganisation of India on linguistic lines. Administrative efficiency also requires some homogeneity, which made it natural for the Constitution to provide a flexible process for forming new states. Without Article 3, had the redrawing of boundaries been frozen into rigidity, India would scarcely have come through its 'dangerous decades'.

Demands for new states have been persistent, but scholarly opinion is sharply divided.

Ramchandra Guha and Bibek Debroy, along with political leaders such as Advani and Mayawati, favour the creation of smaller states.

Their central argument rests on the logic of good governance: smaller states are easier to administer, and efficient administration fosters development and inclusive growth. In their view there should be a rationalisation between the strength of the administration and the size of the population.

Debroy argues that India should have at least fifty states. The US, with a quarter of India's population, has fifty states, whereas India has only twenty-nine. Uttar Pradesh alone, if counted as a country, would rank fifth in the world by population.

M P Singh and Sudha Pai, for their part, oppose the creation of smaller states:

• There is no necessary relation between size and good governance. Tamil Nadu, though large, is well governed, while small states like Jharkhand and Chhattisgarh can hardly be cited as models.
• Nor is there any relation between size and development. Punjab's growth story is over, and Chhattisgarh's mining-based prosperity rests on an unsustainable approach, while Maharashtra, despite its size, remains among the more developed states.
• New states mean new ministries and new infrastructure for government institutions, entailing unnecessary expenditure.
• More states mean more inter-state disputes.
• Smaller states tend to be politically unstable, and defection becomes easier, as the example of Goa shows.
• The economic viability of many demanded states is doubtful; they would depend on union grants much like union territories.
• The impact of carving out a new state on the mother state must also be weighed.
• Certain demands carry strategic risks, such as Gorkhaland, which would lend weight to Nepal's allegations that India seized its territories during the colonial period.
• With the 73rd Amendment, the rationale for smaller states collapses, since good governance and development call for strengthening Panchayati Raj institutions rather than multiplying states.
• E-governance and information technology can bring government to the doorstep without territorial reorganisation.
• None of this means every demand should be rejected: genuine popular aspirations may be conceded, but purely political motivations should not drive the process.

In Remapping India, Tillin offers a comprehensive account of state formation in India, showing that new states have been created in distinct phases, each marked by a different logic.

• The first phase, spanning the 1950s and 1960s, reorganised the Indian peninsula, with language as the dominant criterion.
• The second phase, in the 1960s and 1970s, covered western and north-eastern India, where ethnicity supplied the basis in the north-east.
• The third phase, in the twenty-first century, reorganised the Hindi heartland and produced Telangana, driven by the logic of 'good governance' and development.

Beyond these narratives, academics have offered theories to explain the creation of smaller states.

1. The sociological theory of Ramchandra Guha and Yogendra Yadav sees demands for separate statehood as assertions by marginalised sections seeking a share in power and development.

2. Atul Kohli's political economy approach holds that economic policy dictated state creation. Chhattisgarh, for instance, was carved out of the central government's neo-liberal approach to mining; once formed, the new state had no option but to open up its mines and minerals.

Tillin herself, however, concludes that political considerations have dominated. The timing and feasibility of state formation have been dictated chiefly by the interests of the ruling party at the centre. The Congress repeatedly postponed Telangana and made its announcement only on the eve of elections; the BJP created Uttarakhand to consolidate its own position; and Lalu Prasad Yadav agreed to the bifurcation of Bihar because it strengthened the RJD.

Christophe Jaffrelot likewise treats state formation, including linguistic reorganisation, as an essentially political phenomenon rooted primarily in caste. His analysis makes three points:

• Nehru opposed linguistic states because he feared for the future of the Congress system, though he maintained that multilingual states showcased India's unity in diversity.
• Even after the States Reorganisation Commission accepted the case for Telangana, Nehru rejected it, offering a 'gentleman's promise' instead, the real reason being that Telangana was dominated by communists.
• Linguistic reorganisation served the ambitions of particular castes to consolidate political power, producing dominant castes in different states who could then form their own governments.

Inter-state Disputes

Commentators predict that the wars of the future will be 'water wars', for water is fast becoming the new oil. South Asia, with its agrarian economy, depends heavily on water and ranks among the most water-stressed regions in the world, in fact possessing the lowest per capita availability anywhere.

India is involved in water conflicts with almost every one of its neighbours, while inside the country most of the major rivers run within a single state. Water disputes are special, however, for the Constitution treats them differently under Article 262, which keeps the Supreme Court's jurisdiction out of this one category of disputes.

Water finds its place in the state list, specifically Entry 17.

Entry 56 of the Union List addresses the regulation and development of rivers that flow through more than one state, together with their valleys.

The Constitution therefore makes the union answerable for inter-state rivers. Yet because water is so deeply politicised, the union has preferred to deal with inter-state rivers and valleys as if they belonged to the states. Nirvikar Singh and others blame the union government, which has never exercised to the full the powers the Constitution gives it, and this abdication has made inter-state water conflicts intensely political.

A number of constitutional articles and institutions govern the settlement of inter-state river disputes.

Article 131 vests the Supreme Court with original jurisdiction over disputes between states or between a state and the central government. It is through this provision that the Court can step in as umpire in inter-state conflicts, river disputes included, and states may bring their grievances directly before it.

Article 136 allows the Supreme Court to hear special leave petitions against any judgment or order pronounced by any court or tribunal in the country. In river disputes, it gives aggrieved states a further avenue of appeal, ensuring that tribunal decisions receive the closest judicial scrutiny.

Article 262 empowers Parliament to provide by law for the settlement of disputes concerning the waters of inter-state rivers or river valleys. Relying on this provision, Parliament enacted the Inter-State Water Disputes Act, 1956, under which tribunals are constituted and their awards are made binding. The same article bars the Supreme Court and all other courts from entertaining such disputes, a deliberate design in favour of specialised adjudication.

Article 263 enables the President to establish an Inter-State Council for investigating and discussing matters in which the union and the states, or the states among themselves, share a common interest, and river disputes fall within this remit. The Council is meant to be a forum for dialogue and consensus, easing negotiations and proposing amicable solutions.

Zonal Councils, statutory bodies instituted by the States Reorganisation Act, 1956, exist to foster cooperation among neighbouring states. They offer states within each zone a forum to take up matters such as river disputes and work them out through discussion and mutual consent, with the centre offering guidance.

• From the perspective of the Constituent Assembly, the preferred route was the Inter-State Council under Article 263, because such councils can be created with ease by a presidential resolution. Unfortunately this facility was never used, and given that water disputes resist purely legal solutions, the platform should have been activated.
• Zonal Councils emerged as an extra-constitutional arrangement under the States Reorganisation Act, 1956, designed partly to cushion the fallout of reorganisation, especially over water. Home Minister G B Pant famously said that 'rivers know no linguistic boundaries', but no dispute was ever sent to these councils.
• Article 131, the Supreme Court's original jurisdiction, was the principal option until 1956, when the Inter-State Water Disputes Act removed water disputes from the Court's jurisdiction. Even so, the Supreme Court continues to act as arbiter through special leave petitions under Article 136.
• States routinely run to the Supreme Court against tribunal awards, rendering the entire tribunal exercise futile and squandering public money.
• The Supreme Court also overextends itself at the cost of its regular work, given the enormous pendency of appeals.
• A special leave petition is, by design, exceptional and meant to be used rarely, but it has become so routine that it is hardly special any more; more than 30,000 special leave petitions are pending.

Pratap Bhanu Mehta calls the Indian judiciary a 'self-perpetuating institution', one that never misses an opportunity to burnish its image and enlarge its powers.

Water disputes were the sole category from which the Supreme Court's jurisdiction was excluded, and for that reason the Court should have exercised restraint.

The American experience is instructive here: the US Supreme Court has original jurisdiction over such disputes but refuses to hear appeals, which in effect forces the states to sort out their differences through negotiation.

Fali S Nariman recommends restoring the pre-1956 position, abolishing the tribunal system and returning to original jurisdiction.

Under the Constitution, Parliament is free to oust the Supreme Court's jurisdiction in inter-state river disputes and to prescribe its own statutory machinery for their settlement. This is exactly what it did when the states were reorganised in 1956, passing:

• The River Boards Act, 1956
• The Inter-State Water Disputes Act, 1956

The scheme was deliberately proactive, seeking to stop disputes before they could arise. Under the River Boards Act, boards were to be set up to guide the planned, combined development of inter-state rivers and their valleys, with the expectation that co-riparian states would cultivate a habit of cooperation through joint planning, execution and monitoring. No such board has ever been created, however.

The Inter-State Water Disputes Act remains full of holes, some of which the 2002 amendment attempted to close. A further amendment bill was passed by the 17th Lok Sabha, but when that House was dissolved in 2024, the bill lapsed in accordance with clause (5) of Article 107 of the Constitution.

The Kaveri dispute is the classic case for understanding the Act's weaknesses.

• The Act enumerates certain categories of disputes as water disputes, essentially those concerning the use of water. Disputes over other matters, such as the height of a dam as in the Mullaperiyar case, can go directly to the Supreme Court under Article 131.
• When a dispute arises, the interested states approach the central government, which first attempts mediation; only if that fails does it constitute a tribunal.

A tribunal consists of at least one Supreme Court judge and two other judges drawn from the Supreme Court or the High Courts. Indian tribunals, it must be noted, do not differ from courts in practice. The Punchhi Commission recommended interdisciplinary tribunals drawing members from other fields, but that suggestion was never adopted.

1. Originally there was no fixed time frame, and tribunals took extraordinarily long. The Kaveri tribunal delivered its award only after seventeen years, in 2007, and even then only after Tamil Nadu moved the Supreme Court. The 2002 amendment made it mandatory for a tribunal to give its award within three years, extendable by two more in exceptional circumstances, and this is one reason the Kaveri award finally appeared in 2007.

Non-implementation has proved the biggest challenge of all. The Act provides that a tribunal award stands on par with a decree or order of the Supreme Court, and that the union government is responsible for implementing it. In practice, neither the union nor the state governments implement it.

Even after the Kaveri tribunal's award, Tamil Nadu had to approach the Supreme Court because the award remained unimplemented; the Court then directed the Manmohan Singh government to constitute an authority, which was set up under the leadership of the Prime Minister. Still nothing moved on the ground.

The Supreme Court intervened, in an act of judicial activism. Up to that point the Court had refrained from altering tribunal awards, entertaining special leave petitions only over procedural irregularities. In February 2018, however, the Court altered the award itself for the first time.

Highlights of the Judgement

• Tamil Nadu's share was cut and Karnataka's raised, so that the drinking water demands of the global metropolis of Bengaluru could be met.
• It directed that the availability of groundwater be taken into account.
• It declared inter-state rivers to be national assets.
• The Court held that drinking water needs should take priority.
• It rejected the doctrine of historical rights.
• It ordered implementation of the decision within six weeks.

The union government dragged its feet, wary of the Karnataka elections, and only after renewed orders from the Court did it notify the Kaveri Management Authority. This body operates at two levels: a management board above and a regulatory committee below, which keeps an eye on the situation on the ground.

The episode demonstrates that even after Supreme Court intervention, state governments may not implement an award, and the union government may lack the political will to see it through.

The 1924 agreement between Madras and Mysore (later Karnataka) was due to expire in 1974, so Tamil Nadu approached the union government as early as 1970 seeking a tribunal. It took the union twenty years, until 1990, to constitute one, and then only after Tamil Nadu moved the Supreme Court.

Because the tribunal was constituted after such a long delay, Tamil Nadu pressed for an interim award. The tribunal refused, and only after Tamil Nadu again approached the Supreme Court did it produce one.

Karnataka maintained that it was under no obligation to implement the interim award, forcing Tamil Nadu back to the Supreme Court, which directed Karnataka to comply.

Karnataka next argued that it need not implement the award until it was published in the Gazette. Tamil Nadu returned to court, and only then did the union government publish the award.

There followed a bandh called by Karnataka's Chief Minister, a fast undertaken by Tamil Nadu's Chief Minister Jayalalithaa, localised violence, and even suicides by farmers.

The tribunal gave its final award in 2007; challenged in the Supreme Court, it received its final verdict only in 2018.

The Indus Water Treaty of 1960 with Pakistan hit Punjab, an agrarian state, hard.

When Haryana was carved out in 1966, it demanded a share of the Sutlej's waters, which Punjab refused on the ground that the Sutlej does not flow through Haryana.

Water became one of the grievances fuelling the Khalistan movement. The union government intervened, proposing to link the Sutlej and the Yamuna through a canal and bringing two more states, Delhi and Rajasthan, into the project.

The Rajiv-Longowal Accord of 1985 led to the constitution of a tribunal, which reduced Punjab's share. So controversial was the award that it has still not been published in the Gazette.

In 2004, the Punjab assembly took the unprecedented step of terminating all its water agreements with other states.

The union government referred the matter to the Supreme Court under its advisory jurisdiction to test constitutionality, and in 2016 the Court concluded that the act was unconstitutional.

With elections due in 2017, the Punjab government took another unprecedented step: it de-notified the land acquired in Punjab for the link canal and began returning it. Haryana, which had completed its side of the link and even issued an advance cheque to Punjab, approached the Supreme Court, but the Punjab government had already returned the cheque.