The Indian Party System: Evolution and Transformation

The Indian Party System: Evolution and Transformation

Every mass democracy has to answer one question: how do millions of dispersed preferences become a government? India has given several different answers in seventy-five years, and the shifts between them are the substance of this topic. What follows moves from the concept of a party, to the concept of a party system, to the system India built, broke and rebuilt.

What a political party is

Political parties are now a universal feature of politics, in democracies and dictatorships alike, and the field that studies them has its own name — stasiology.

• The defining function of a party is interest aggregation: converting scattered individual demands into a manageable number of policy packages that can actually be voted on.
  • Pressure groups articulate interests; parties aggregate them — the cleanest line between the two, and the pressure-group side of it is developed elsewhere in this unit.
• Two rival definitions of what a party is have organised the literature since the eighteenth century.

	Normative view	Behavioural view
Associated with	Edmund Burke	Behavioural political science; Anthony Downs in the economic tradition
A party is	A body of people sharing a principle, acting jointly for the national interest	A machine for capturing power; a “powerhouse”
Ideology is	Constitutive — the party exists because of it	Instrumental — a resource, discarded when it costs votes
Test of success	Whether the shared principle is advanced	Whether the party wins office
Weakness	Describes almost no real party	Cannot explain why parties differ at all

“A party is a body of men united, for promoting by their joint endeavours the national interest, upon some particular principle in which they are all agreed.” — Edmund Burke

• Otto Kirchheimer’s catch-all party is the concept that closes the gap between the two. Writing on post-war Western Europe, he described parties that shed ideological baggage, strengthened the leadership, widened interest-group access and appealed to the whole electorate rather than to a class.
  • The price is that programmatic difference narrows and voters lose a reason to identify with any party in particular. The Indian relevance is direct: almost every large Indian party is catch-all in Kirchheimer’s sense.
• Katz and Mair’s cartel party extends the sequence further — parties become agents of the state, funded from public resources, colluding to limit competition. India has no public funding, so the form is undeveloped here, but the collusive resistance of all parties to disclosure and internal-democracy regulation is recognisably cartel behaviour.

What parties actually do

• Formation of government — supplying the personnel and the majority that make an executive possible, and keeping it alive between elections.
• Formation of the opposition — the function most easily forgotten. A system without a credible opposition produces accountability without alternatives.
• Political recruitment — parties are the ladder into public office; almost nobody enters an Indian legislature except through one.
• Political education and socialisation — transmitting political information, vocabulary and loyalties, which in a low-literacy electorate parties did before the state or the media could.
• Interest aggregation — the input function above.
• Linkage — connecting the citizen to the state continuously rather than once every five years. This is the function Indian parties perform worst.
• Integration, a sixth function added for developing societies — building a national political community out of regions, languages and castes that had never before shared an electorate.

What a party system is

A party system is not a count of parties. It is a count of parties that matter. This is Giovanni Sartori’s central contribution, and it is the single most useful idea in the topic.

• Simple counting treats a party with two seats and no bargaining weight as equivalent to one that decides who governs. Sartori replaced the count with two relevance tests.
  • Coalition potential — has the party been, or could it plausibly be, needed to form a governing majority? A small party repeatedly inside coalitions is relevant; a larger party permanently excluded may not be.
  • Blackmail potential — even if never in government, does the party’s existence alter the tactics and direction of competition among the others? An anti-system party that pulls competition outward has blackmail potential.
• Applied to India, the tests do real work. At independence there were dozens of parties, yet the system was one-party dominant because almost none had either kind of relevance. Between 1996 and 2004 a dozen parties with under twenty seats each were highly relevant, because each could make or unmake a Prime Minister.

Sartori’s typology

Type	Number of relevant parties	Character of competition	Examples
Predominant party	Several relevant, one wins repeatedly	Genuinely competitive, but alternation does not occur	India to 1967, Japan’s LDP, Sweden’s Social Democrats
Two-party	Two	Centripetal; realistic prospect of single-party alternation	Britain, the United States
Moderate pluralism	Three to five	Centripetal; coalition government the norm	Germany, the Netherlands
Polarised pluralism	More than five, with anti-system parties and bilateral oppositions	Centrifugal — competition pulls away from the centre	Weimar Germany, Italy’s First Republic
Atomised	No party of significance	Residual category	Fragmented post-conflict systems

• Sartori’s insistence that the predominant-party type is competitive matters: it describes a democracy in which one party happens to keep winning, not a system in which it cannot lose.
• India after 1989 sits awkwardly between moderate and polarised pluralism. It had far more than five relevant parties, which suggests polarised pluralism; but competition remained largely centripetal, coalitions formed across ideological distance rather than refusing to, and genuinely anti-system parties were marginal. The mismatch is itself informative: Sartori’s scheme was built on ideological distance along one axis, and India’s cleavages do not lie on one axis.

Weiner and LaPalombara’s classification

Myron Weiner and Joseph LaPalombara cut the field first by whether competition is real at all, and only then by number.

Category	Sub-type	Illustration
Non-competitive	One-party hegemonic — other parties banned or subordinate	China
Competitive	One-party dominant — free elections, one party keeps winning	India to 1967; the African National Congress
	Two-party, ideological	Britain
	Two-party, pragmatic — little ideological space between them	The United States
	Multi-party, extremely polarised — unstable coalitions	Italy; India is often placed here
	Multi-party, consociational or consensus-based	Germany, Switzerland, the Netherlands

• The value of this scheme for India is the first cut. India’s dominant-party phase belonged in the competitive column with South Africa, not in the non-competitive column with China — a point that answers the commonest misreading of the Congress system.

Duverger, and why India is the standing exception

Maurice Duverger proposed the most-cited generalisation in the field, usually stated as Duverger’s law: single-member plurality (first-past-the-post) tends to produce two-party competition, and proportional representation tends to produce multipartism.

• The mechanism has two parts. A mechanical effect — plurality rules systematically under-reward parties whose vote is spread thinly. And a psychological effect — voters, anticipating this, desert hopeless candidates, and donors and elites desert them too.
• India is the standing exception: a first-past-the-post system that has produced one of the most fragmented party systems in the democratic world.

The resolution is that Duverger’s law holds in India where it was always meant to hold — inside the constituency — and simply fails to aggregate upwards.

• Duverger’s law is a claim about districts, not countries. William Riker, restating the law, made this explicit and treated India as the deviant case that clarifies the rule: the law predicts two serious contenders per seat, and India largely delivers that. What it cannot predict is whether the same two contenders appear across all seats.
• Because India’s principal cleavages are regionally concentrated — language, caste-region complexes, regional political memory — a different pair of parties is serious in Tamil Nadu, in Kerala, in West Bengal and in Uttar Pradesh. Summing many local two-party contests with different local participants yields a multi-party legislature.
• Chhibber and Kollman, comparing India with Canada, Britain and the United States, argued that what determines whether local party systems aggregate into a national one is not the electoral rule at all but the centralisation of political and fiscal authority. Where power and money sit at the centre, elites have an incentive to build national parties; where authority devolves to provinces, local parties survive. India’s fiscal centralisation drove aggregation in the Nehru years, and the loosening of that centre after the 1960s drove de-aggregation.
• The corollary: the post-2014 renationalisation of the vote is not a repeal of Duverger’s law but a change in the aggregation incentive — centralised welfare delivery, campaigning and fiscal architecture again reward national labels.

Duverger’s other two contributions

• Mass parties versus cadre parties. Cadre parties are loose caucuses of notables, financed from personal wealth and built for a restricted franchise. Mass parties have formal membership, dues, branches, an ideology and an ancillary network of unions, student wings and cultural bodies. Duverger’s expectation was that mass parties would drive out cadre parties as the franchise widened.
  • Indian application: Congress under Gandhi was converted from a cadre body into a genuine mass party in the 1920s, and its later decline is in part a reversion — a mass party’s membership rolls kept while its branch life died.
• Four internal structures, Duverger’s structural rather than ideological classification.

Structure	Basic unit	Typical party family	Indian illustration
Branch	A territorial branch open to all members	Liberal and social-democratic parties	The formal organisational chart of most Indian parties
Cell	Small, tight units at the workplace	Communist parties	The CPI(M)’s branch-and-cell structure
Caucus	A small closed group of notables	Conservative and cadre parties	The “high command” of most Indian parties in practice
Militia	A paramilitary formation attached to the party	Fascist parties; the Red Guards	Uniformed volunteer wings in the Indian tradition

• The Indian lesson from this table is the gap between column three and column four: parties whose written constitutions describe a branch structure operate in practice as a caucus.

Michels, Lenin and what they predict for Indian parties

• Robert Michels, in Political Parties, set out the iron law of oligarchy. He studied the German Social Democratic Party precisely because it was the most internally democratic mass party in Europe and the one most committed to equality — if oligarchy appeared there, it would appear anywhere.
  • His finding was that it did. Organisation itself generates oligarchy: large bodies need full-time officials, who acquire expertise, information and control of communication, while members are dispersed and grateful. Ideology makes no difference.
  • His prediction for India is unforgiving and largely correct: the high command culture is not an Indian pathology but the ordinary condition of party organisation, aggravated here rather than invented here. That is why it appears in the communist parties too.

“Who says organization, says oligarchy.” — Robert Michels

• Lenin’s vanguard party, set out in What Is To Be Done?, is the deliberate embrace of what Michels described as a defect.
  • Marx never theorised a party — a party creates hierarchy, which sits badly with equality. Lenin’s departure was the claim that workers left to themselves develop only trade-union consciousness, bargaining for concessions inside the bourgeois order; revolutionary consciousness must be brought from outside by professional revolutionaries.
  • Hence a party that is small, secret, disciplined and pyramidal, with cells at the base, distinct from trade unions, and organised on democratic centralism: free discussion until a decision is taken, unconditional implementation afterwards.
  • Its Indian afterlife is the CPI(M). Democratic centralism explains both the party’s discipline and its immobility — it produced the most cadre-based organisation in India and simultaneously suppressed the internal dissent that might have let it adapt, which is a standing item in accounts of the left’s decline.

Should there be political parties at all?

The Indian debate on parties includes a strand, unusual among democracies, that questions whether parties should exist.

• Harold Laski states the conventional position: parties are indispensable to representative democracy, converting an unmanageable diversity of opinion into a choice between governments. Without them the electorate faces a crowd of individuals and no way to hold anyone collectively responsible.
• Gandhi doubted the party form. His last written suggestion, drafted in January 1948, was that the Congress had completed its purpose as a political instrument and should be disbanded and reconstituted as a Lok Sevak Sangh doing constructive social work — a rejection of the idea that competitive electoral parties should be the primary channel of public life.
• M.N. Roy, moving from communism to Radical Humanism, dissolved his own party and argued for party-less democracy — decentralised people’s committees exercising power directly, on the ground that parties substitute the pursuit of power for the pursuit of the public good.
• Jayaprakash Narayan gave the position its fullest Indian statement. In A Plea for Reconstruction of Indian Polity he proposed a communitarian, participatory democracy built upward from village communities — a partyless panchayati raj in which representation would be indirect and consensual rather than competitive.
  • JP’s “total revolution” from 1974 — a simultaneous transformation of social, economic, political, cultural, educational, moral and spiritual life — was the mature form of that argument, and it also names its limitation: the movement he led produced the Janata Party, an ordinary electoral coalition, within three years.
• The Constituent Assembly assumed parties and did not regulate them. The Constitution as adopted nowhere mentioned political parties, although the whole scheme of responsible government presupposed them.
  • Parties entered the constitutional text only in 1985, with the Tenth Schedule inserted by the Fifty-second Amendment — and even then only to punish defection, not to require parties to be democratic, transparent or accountable.
  • The consequence is a regulatory vacuum: the entities that select every candidate, whip every legislator and raise every rupee of campaign money are governed almost entirely by their own constitutions.

How Indian parties have been characterised

• Zoya Hasan sets out six unusual features of Indian parties, which together amount to a claim that they are organisationally shallow.
  • No serious ideological concern, and a general drift toward catch-all formations.
  • Office-oriented rather than policy-oriented — the party exists to win, and programmes follow.
  • A disconnect between parties and people outside election season.
  • Links with corporate groups, extending, notably, even to communist and socialist parties.
  • Unusual nomenclature — names describing a region, a community or a lineage rather than a doctrine, as with the AIADMK.
  • No grand design — no party carries a comprehensive vision of the society it intends to build.
• Paul Brass identifies the paradox at the centre of Indian party organisation: a Western bureaucratic form — written constitutions, elected committees, membership registers — grafted onto indigenous practice of faction, patronage, kinship and personal loyalty. The form is modern; the functioning is not.
• Yogendra Yadav offers the sharpest single formulation: Indian parties are undergoing institutionalisation and de-institutionalisation simultaneously.
  • Reach is widening — parties touch every district, every social group and a larger electorate than ever.
  • Depth and intensity are declining — membership means little, branch life is thin, and the citizen’s relationship with the party has shrunk to the instrumental act of voting.
  • What is lost is the broader agenda of nation-building that parties in the first decades took as their own work.

Indian parties have never been more present in citizens’ lives and never less inhabited by them.

The Congress system

Rajni Kothari’s account of the first two decades is the founding text of the subject, and its power lies in a single move: he stopped describing Congress as the largest party and began describing it as the system itself.

• Congress occupied the centre deliberately, rejecting communalism on one side and communism on the other and leaving both flanks outside a broad middle most voters could accept.
• It contained internally the ideological range that other countries distribute between parties — conservatives, Gandhians, liberals, socialists and a Congress Socialist tradition all had a home in it, along with regional and caste blocs.
• Disputes were settled inside the party, by factional bargaining. Factions competed for tickets, offices and patronage; winners had to compensate losers to keep them in; and the process reproduced, within one organisation, the work that competition between parties does elsewhere.
• Consensus, not alternation, was the mechanism of representation. A group that wanted redress did not vote Congress out; it worked on a Congress faction until Congress absorbed the demand.
• Kothari’s paired terms follow directly. Congress was the party of consensus; the others were parties of pressure — located, in his phrase, at the margins of the system, unable to displace the ruling party but able to influence it by amplifying the demands of factions inside it. Their functional status was closer to that of pressure groups than of oppositions.
• Kothari also called Congress a rainbow coalition, a grand coalition in itself — a programme wide enough to carry very different sections at once.
• Kothari’s own defence of the arrangement rested on that continuity: because Congress was in power without interruption and faced no united or effective threat to its authority, the political process gained, on his reading, incomparable advantages of continuity and unity.
• W.H. Morris-Jones supplied the term that entered general use — the dominant party system, or one-party dominance. His emphasis was complementary: a dominant party coexists with real opposition parties that are not, for the foreseeable future, alternative governments, and the dominant party’s internal life carries the real politics.

It was not a one-party system

This is the point on which the Congress system is most often misdescribed, and Kothari’s own argument depends on it.

	India, 1952–67	A one-party state
Other parties	Legal, funded, contesting, winning seats	Banned or subordinated
Elections	Genuinely competitive — results not known in advance	Ratifying, not choosing
Where competition occurred	Inside Congress, between factions	Nowhere
Function of the opposition	Pressure and criticism — a check on autocracy	Ornamental
Comparator	Japan’s LDP, Sweden’s Social Democrats	China

• China has a one-party system without democratic character; India had a dominant party within a democratic system. Congress won because it kept winning votes, and the parties outside the margin — small, scattered, ideologically distinct — kept the ruling party from becoming autocratic.

Its social basis and why it held

• The social coalition was inherited from the national movement, and it was extraordinarily wide: upper castes, Dalits, Muslims, and in most states the dominant landholding castes — held together by patronage flowing downward and by the prestige of the freedom struggle.
  • This was a coalition of the top and the bottom against the middle, which is why the rise of the intermediate castes later broke it.
• The national movement’s legacy gave Congress a monopoly of legitimacy no competitor could contest; opposing it looked, for a generation, like opposing the nation.
• Gandhi’s transformation of Congress into a mass party — membership at four annas, linguistic provincial committees, a permanent organisation between elections — gave it a reach nobody else approached.
• The centrist agenda made Congress the default choice of anyone with no strong reason to defect.
• Nehru’s internal climate mattered as much as his popularity: freedom of expression inside the party, toleration and accommodation, and demonstrative sensitivity toward minorities. He lost internal arguments and accepted losing them.
• The conviction that Congress was necessary to the nation’s survival was widely held, and for two decades self-fulfilling.

The breakdown from 1967

The Congress system broke not because Congress became unpopular but because the machinery that converted its popularity into consent stopped working.

• The 1967 general election is the hinge. Congress retained power at the Centre with a reduced majority but lost its majority in nine states — non-Congress governments took office in eight, and in the ninth Congress could govern only in coalition.
  • Kothari himself stressed that in most of those states the new Chief Minister was a former Congressman. The Congress system did not lose to an ideological rival; it fragmented into its own factions, which is a very different kind of defeat.

Internal causes

• De-institutionalisation within the party — committees survived on paper while decisions migrated upward.
• Concentration of power in a single leader, which removed the incentive for anyone else to build an independent base.
• The collapse of organisational elections. The Congress has not conducted elections across its organisational tiers since 1972; the presidency has been contested by ballot only in 1997, 2000 and 2022, and the Raipur plenary of 2023 again decided against electing the Working Committee, authorising the president to nominate it instead.
• Loss of contact with grassroots leaders, as state units came to be appointed from Delhi rather than thrown up from below.
• Dependence on loyalists at the centre, and no scope for internal criticism or dissent — which meant that a faction with a grievance had no route but exit.

External causes

• Yogendra Yadav’s argument reverses the usual reading: the decline of Congress tracks the deepening of Indian democracy, not its failure. As new social groups learned to use the vote, they stopped needing an intermediary.
  • The first democratic upsurge brought the OBCs into active participation and took them out of the Congress coalition.
  • The second did the same for Dalits, whose independent political assertion the BSP organised.
  • What is left is instructive: Congress’s residual base became minorities and Scheduled Tribes — the two groups for whom no independent vehicle emerged.
• The Green Revolution enriched the intermediate peasant castes in the north-west and made them politically self-confident; economic power sought political expression, and Congress had no place at the top of its hierarchy to put it.
• Linguistic reorganisation of states gave the dominant caste of each new state a governing prize worth capturing, and a state-level arena in which to capture it.
• Lohia’s non-Congressism supplied the missing strategy. Ram Manohar Lohia’s argument was that the opposition’s ideological differences were less important than the shared objective of removing Congress, and that any anti-Congress combination was legitimate. It made defeat conceivable, which no amount of criticism had done.
  • Its product was the Samyukta Vidhayak Dal governments of 1967 in the Hindi belt — improbable coalitions of socialists, Jana Sangh, Swatantra and Congress defectors. Most fell within months, and the era of defection as the currency of state politics began, running until the Tenth Schedule of 1985.
• The 1969 split and Indira Gandhi’s reconstruction of the party changed its character permanently. What had been a federal, factional organisation — state units with independent bases, bargaining with a centre — became a centralised, leader-focused one in which state units and Chief Ministers were appointed from Delhi. The plebiscitary campaign of 1971 confirmed the method.
  • The decisive irony: Indira Gandhi’s centralisation won elections and destroyed the machine that made the Congress system work, since factional bargaining requires factions with somewhere to stand.
• The Emergency (1975–77) and the Janata interlude (1977–79) completed the demonstration. Janata proved that Congress could be defeated; its collapse within two years proved that its opponents could not yet govern.

Where the Congress stands now

• Suhas Palshikar makes the harder claim: 2014 marked not the decline of Congress but its death as a system-defining party. Its social base and its territorial base have both shrunk — it is a rump in Uttar Pradesh, Bihar, Madhya Pradesh and Rajasthan, states where it was once the natural party of government.
  • The measure is pointed: the 2014 tally was lower than the party’s total immediately after the Emergency, when it had been comprehensively repudiated.
• The 2024 recovery needs to be read carefully. Congress rose to 99 seats from 52, which is a real revival of parliamentary weight and of the opposition’s morale. But its vote share moved only from under 20% to just over 21% — the seat gain came overwhelmingly from alliance arithmetic and better vote transfer, not from a broader appeal.
  • The 2026 state results cut both ways: the party led the winning UDF in Kerala, where it has an organisation, and was a junior partner in the winning coalition in Tamil Nadu, where it has not led for half a century. Where it has neither an organisation nor a partner it continues to poll near zero, as in Delhi in 2025.
• Pratap Bhanu Mehta’s prescription addresses the party’s self-understanding rather than its tactics.
  • Leave the majority–minority complex — a party that defines itself by defending minorities against a majority accepts the majoritarian framing of politics.
  • Learn to make strategic alliances, treating them as an ordinary instrument rather than a humiliation.
  • Become a party of transformation rather than of defence — offering a project, not the preservation of an inheritance.
  • Return to organising social movements instead of restricting itself to electoral calculus, which is the activity Congress was originally built to do and stopped doing.

The phases of the Indian party system

The standard periodisation runs in six stages, each defined by a change in the structure of competition rather than by a change of government.

Stage	Period	Character	What defines it
1	Up to 1967	The Congress system	One-party dominance within a competitive democracy; competition internal to Congress
2	1967–1977	Breakdown of the Congress system	Congress loses nine states; SVD coalitions, defections, the 1969 split, centralisation, the Emergency
3	1977–1979	The two-party moment	Janata versus Congress — the only period in which India looked like a two-party system, and it lasted two years
4	1979–1989	Emergence of multipartism	Congress recovers office but not the system; regional parties consolidate in Andhra Pradesh, Assam, Punjab, Tamil Nadu, West Bengal
5	1989–2014	Multi-party competition and the coalition era	No single-party Lok Sabha majority for twenty-five years; national government requires state parties
6	2014 onwards	A new dominant party	Single-party majorities in 2014 and 2019; renationalised vote share; coalition arithmetic returns in 2024

The cleaner three-systems framing

A second scheme, widely used in the scholarship, compresses the same history into three party systems, each produced by a different social coalition.

• The first party system, 1952–67 — Congress dominance resting on the national movement’s coalition.
• The second party system, 1967 (or 1989) to 2014 — the erosion of that coalition and the rise of state-level challengers, culminating in coalition politics.
• The third party system, from 2014 — a new dominant party with a different social coalition and a different mode of dominance.

	Six-stage scheme	Three-systems scheme
Organising criterion	Government formation and party arithmetic	The underlying social coalition
Treats 1977–79 as	A distinct two-party phase	An episode inside the second system
Treats 1989 as	The start of the coalition era	Either the start of the second system or its deepening
Strength	Descriptive precision; matches the electoral record	Explanatory power; says why the system changed
Weakness	Mistakes short-lived arithmetic for structural change	Compresses genuinely distinct decades

• The two schemes answer different questions. The six-stage scheme asks what the party system looked like; the three-systems scheme asks what produced it.
• The three-systems framing does more analytical work, because it identifies the mechanism: a party system changes when the social coalition sustaining the dominant party breaks up, and not when a government falls. On that test, 1977–79 was an interruption and 1989–99 was a transformation.

The 1989–99 shift: four causes and their interaction

The decade produced an epochal change in the Indian party system, and no single cause explains it. The interaction is the explanation.

Mandal

• Implementation of the Mandal Commission’s recommendation in 1990 — 27% reservation in central government employment for Other Backward Classes — nationalised OBC politics, converting a set of state-level assertions into a single axis of competition across the Hindi belt.
• It gave the Samajwadi Party, the Rashtriya Janata Dal and comparable formations a durable social base that did not depend on any leader’s popularity in a given year.
• Its effect on Congress was structural rather than electoral: it broke the Congress coalition from below, detaching the numerically largest block in north India from a party whose leadership remained upper-caste.

Mandir

• The Ram Janmabhoomi mobilisation and the events of December 1992 produced the counter-mobilisation — an appeal to a Hindu constituency defined across caste, which was precisely the constituency Mandal had divided.
• The electoral result is the steepest ascent in Indian party history: the BJP went from 2 seats in 1984 to 85 in 1989, 120 in 1991, 161 in 1996, and to government in 1996 and 1998.
• Mandal and Mandir were simultaneous, not sequential, and each was partly a response to the other — which is why treating either alone distorts the account.

Market

• Liberalisation from 1991 removed economic policy from the arena of party competition. Every serious contender converged on broadly the same macroeconomic direction, and the differences that remained were about pace and compensation.
• The consequence was decisive: with the economic axis closed, parties competed on identity and on delivery — who represents you, and what did you get.
• It also changed party finance: a liberalised economy generated far more private money than a licence-permit economy could, and campaign costs outran any disclosure regime.

Regionalisation

• State parties became indispensable to national government — the TDP, DMK and AIADMK, Trinamool Congress, Biju Janata Dal, JD(U), Shiv Sena and Akali Dal held the balance in Delhi.
• No party won a Lok Sabha majority between 1989 and 2014 — seven general elections in which every government was a coalition or a minority.
• The state, not the nation, became the effective unit of competition — a general election became an aggregation of separate contests with different participants. Regional parties are treated elsewhere in this unit; their rise is what made the third party system a system rather than an interlude.

How the four interacted

• Mandal fragmented the electorate along caste; Mandir attempted to reunify part of it along religion; the market removed the axis on which the two might have been mediated; and regionalisation gave every fragment an institutional home with a governing prize attached.
• The order matters less than the simultaneity. A caste upsurge without state-level arenas produces pressure groups, not parties; a religious mobilisation in an economy still argued over in class terms competes with a rival axis. It was the coincidence that made the change structural.
• What the decade produced was not merely more parties but a different basis of competition: coalition as the norm, pre-poll alliances with written common programmes from 1998, the Prime Minister’s dependence on allies, a collapse in the use of Article 356, and legislatures that looked more like the society electing them.

The 1989–99 decade did not change who won; it changed what winning required.

After 2014: a fourth party system, or a phase within the third?

This is the live analytical question, and the honest answer is that the evidence points both ways — which is itself the finding.

The case for a new party system

• Nationalisation of the vote share. A single party again polls well over a third of the national vote — 31.3% in 2014, 37.4% in 2019 and 36.6% in 2024 — and does so across regions rather than in a heartland, having entered Odisha, the North-East, and made real inroads in West Bengal.
• Weakening of several regional parties, some absorbed, split or reduced to junior partners.
• A centralised campaign in which national leadership, message and resources substitute for state-level organisation.
• A resource and media advantage no competitor approaches — the declared income alone dwarfs the rest of the field.
• Party identification has realigned. The share of Indians who feel close to any party has stayed near a third for five decades; what changed is which. In 1971 close to 70% of partisans named Congress; by 1996 the BJP and Congress were level at roughly 27% each; by 2019 the BJP had more than twice Congress’s share.
• The dominance is centripetal, which is the analytically important point. A genuinely dominant party draws civil society, local elites and even its opponents into its ambit, so that its ideological centrality matters as much as its electoral superiority. Opponents begin to mirror its issues and tactics — the surest sign that a party has become system-defining.
• Suhas Palshikar calls this a movement towards hegemony: the BJP has become the central pole electorally, expanded its social base from competitive to markedly dominant, and saturated the world of ideas with a twin emphasis on Hindu nationalism and a new developmentalism, at a moment when the Congress legacy of secular nationalism has fallen out of favour.

The case against

• State parties remain strong where they have deep roots: the 2024 Lok Sabha returned SP 37, TMC 29, DMK 22, TDP 16 and JD(U) 12.
• Coalition arithmetic returned in 2024. The BJP won 240 seats, 32 short of a majority, and formed a government on 9 June 2024 depending on the TDP and JD(U) — the first time since 2014 that the leading party needed allies. (It kept Home, Finance, Defence and External Affairs, so the dependence has not yet translated into shared control of the state’s core.)
• National dominance does not extend to the states. The pattern of the last decade has been for the dominant party to shed a large share of its national vote when the same voters vote for a state government — a disjuncture between the two arenas that a truly consolidated dominant-party system would not show.
• The centralisation that produces national dominance weakens state units, because regional leaders with independent bases threaten the centre. The system is therefore strong nationally and brittle locally, by construction.

Reading the 2026 state results

The 2026 assembly results are the newest evidence and they cut both ways, which is exactly why they should be used with care.

• For the fourth-system reading: the BJP took West Bengal with 208 seats, ending fifteen years of TMC government in the last large eastern state that had resisted it, and won Assam for a third consecutive term. That is continued territorial expansion, not a plateau.
• Against it: in Kerala the Congress-led UDF displaced the LDF, and in Tamil Nadu the entirely new Tamilaga Vettri Kazhagam became the single largest party with 108 of 234 seats in its first election, forming a coalition with Congress and outside support from the left, VCK and IUML.
• The Tamil Nadu result is the sharper datum. It ended the DMK–AIADMK duopoly that had governed the state without interruption since 1967 — but by replacing one regional formation with another, not by nationalising the state’s politics. A state party system can be overturned entirely without the national dominant party gaining anything, which counters the claim that regional politics is being absorbed.
• The safe conclusion: the dominant party’s reach is still growing in the east and north-east, and states with autonomous party systems remain autonomous.

How BJP dominance differs from Congress dominance

	Congress dominance, 1952–67	BJP dominance, 2014–
Mode of accommodation	Internal factional bargaining — groups brought inside and given a share	Co-optation from outside — local elites, civil society bodies and opponents drawn into its ambit
Where competition happened	Inside the party	Between parties, with one much larger
Nature of ideological centrality	Consensual — a wide centre most could accept	Programmatic — a defined project pursued deliberately
Relation to opposition	Opposition as parties of pressure at the margins	Opposition mirrors the dominant party’s agenda and tactics
Internal structure	Federal, with powerful state units, until 1969	Centralised from the outset; state leaders deliberately kept weak
Social coalition	Top and bottom — upper castes, Dalits, Muslims, dominant landholders	Cross-caste Hindu consolidation, with upper-caste over-representation in office restored
Source of cohesion	Patronage and the movement’s prestige	Ideology, leadership and organisational discipline

• The substantive difference is the direction of movement. Congress absorbed challengers by opening its own interior to them; the present dominant party absorbs them by extending its exterior over them. The first produced a party internally plural and organisationally soft; the second, one internally disciplined and externally expansive.

Registration and recognition: a distinction routinely confused

Registration and recognition are two different things, done under two different instruments and with two very different levels of difficulty.

	Registration	Recognition
Legal basis	Section 29A, Representation of the People Act 1951	Election Symbols (Reservation and Allotment) Order, 1968
Test	Formal — an application within 30 days of formation, a party constitution, and a declaration of allegiance to the Constitution, to socialism, secularism and democracy, and to the sovereignty, unity and integrity of India	Performance — thresholds of votes and seats actually won
Difficulty	Very easy; almost no party is refused	Hard; most parties never qualify
What it brings	Legal existence; income-tax exemption under Section 13A; a symbol from the free list, allotted afresh each election	A reserved symbol across the relevant territory, free broadcast time on public radio and television, free electoral rolls, a single proposer for nominations, and up to 40 star campaigners against 20

• National recognition follows from any of three routes: recognition as a state party in four or more states; 6% of valid votes in four or more states plus at least four Lok Sabha seats; or 2% of Lok Sabha seats from at least three states. State recognition turns on comparable thresholds within one state.
• There are six recognised national parties: the BJP, the Congress, the BSP, the CPI(M), the AAP and the National People’s Party. The April 2023 review granted national status to the AAP and withdrew it from the Trinamool Congress, the NCP and the CPI; the NPP, recognised in 2019, was the first from the North-East.
• Registered unrecognised parties numbered roughly 2,850 before 2025 and now stand at about 2,050. Most have never contested seriously. The Commission’s delisting drive removed 334 in August 2025 and 474 in September 2025, with proceedings against several hundred more — the suspicion being that dormant registrations exist largely for the tax exemption. Detailed classification belongs elsewhere in this unit.

Intra-party democracy

The absence of internal democracy is the single organisational fact from which most other defects of Indian parties follow.

• Organisational elections are largely notional — uncontested, ratified rather than fought, and held long after they were due. The Congress’s record is set out above; it is not the outlier it is sometimes taken for.
• Candidate selection is centralised almost everywhere. The formal position — a state committee recommending, a central body confirming — inverts into a central body deciding, often days before nominations close.
• The high command culture crosses every ideological line — the Congress, the BJP, the regional parties where it fuses with family control, and the communist parties, where democratic centralism supplies a doctrinal justification for it. Michels’s point stands: ideology makes no difference to the internal distribution of power.
• Dynasty is a consequence, not a separate defect. Where there is no internal contest there is no mechanism for producing a successor, and the only transfer of authority uncontroversial within the organisation is a hereditary one. Family succession fills the vacuum left by the absence of elections, which is why it appears in parties of every ideology and region. Dynasty as a legislator-profile fact belongs elsewhere in this unit.

The Law Commission’s 255th Report and the limits of regulation

• The Law Commission’s 255th Report on Electoral Reforms (2015) is the fullest official attempt at the problem. It recommended a new chapter in the Representation of the People Act regulating parties directly, requiring:
  • a party constitution meeting statutory minimum standards, including a commitment to shun violence for political gain;
  • internal elections at specified intervals;
  • a defined candidate-selection procedure set out in the constitution rather than left to practice;
  • annual audited accounts, and disclosure of all contributions above ₹20,000, with a daily penalty of ₹25,000 for each day of non-compliance;
  • de-registration powers for the Commission where these obligations are ignored.
• Its other recommendations bear on the same problem from other directions: disqualification on the framing of charges for serious offences; restricting a candidate to one constituency by amending Section 33(7); extending the opinion-poll blackout to print media; rejecting compulsory voting and the right to recall; and moving anti-defection decisions from the Speaker to the President or Governor on the Commission’s advice.
• The leverage problem is fundamental. The Commission can recommend; only Parliament can legislate; and Parliament is composed entirely of the parties that would be regulated. No party has an interest in a statute that would open its own candidate selection to challenge, which is why the 2015 recommendations remain unenacted. The Commission has sought de-registration powers for years and still does not have them.

Every party in India agrees that internal democracy is desirable, and no party has an incentive to be the first to have it.

Personality, leader-centred politics and divisionism

• Indian parliamentary campaigns have been presidentialised. An election formally selects 543 members; it is fought and understood as a choice between two or three individuals for Prime Minister. The pattern is not new — the plebiscitary campaigns of 1971 and 1980 were of the same kind — but its intensity, and the technology sustaining it, are.
• The leader has become the party’s principal electoral asset, which changes the internal balance in three ways.
  • Organisation atrophies where a national campaign can substitute for it, because the returns to building a branch fall relative to the returns to proximity to the leader.
  • Internal accountability collapses, since a leader who is the reason the party wins cannot be held to account by people who owe their seats to that fact.
  • Succession becomes a crisis rather than a process — the classic vulnerability of parties built on a personality, and the reason several regional parties have not survived their founders.
• The counter-argument is empirically strong. Indian voters distinguish between arenas. Split-ticket voting — backing one party for the Lok Sabha and another for the assembly, sometimes weeks apart — is well documented, and popular national leaders have repeatedly seen their parties lose state elections in the same states months later. The Indian voter has proved capable of admiring a leader and refusing him a state government, which is a real limit on the personalisation thesis.
• The balanced position: personalities dominate campaigns; parties still dominate outcomes. No individual in India has won an election without a party machine, a symbol and an alliance, and the parties that most depend on a personality are the most fragile.

Divisionism: factionalism and splits as a structural feature

Splits are not accidents in Indian party life; they are the predictable output of organisations with no internal procedure for losing an argument.

• The Congress has split repeatedly — 1969, 1978, the Tiwari Congress of 1995, and 1999 when Sharad Pawar left to found the NCP — each time along the fault line between a leadership and a group that could not be accommodated.
• The Janata Dal is the extreme case: assembled in 1988 out of several formations, it fragmented within a decade into Janata Dal (United), Janata Dal (Secular), the Rashtriya Janata Dal, the Biju Janata Dal, the Samata Party and others — most surviving as state parties.
• The CPI split in 1964, producing the CPI(M), over the assessment of the Congress government and the Sino-Soviet dispute; the CPI(M) in turn shed the Naxalite tendency in 1967–69.
• The Akali Dal has divided along leadership lines through most of its history, successive factions carrying the same name with a qualifier attached.
• The contemporary cases are the Shiv Sena and NCP splits. In February 2023 the Election Commission awarded the Shiv Sena name and the bow-and-arrow symbol to the Eknath Shinde faction, resting on the majority of the legislative wing — 40 of 55 MLAs and 13 of 18 MPs — and declining to apply the party-constitution test because the constitution as amended in 2018 had itself centralised control undemocratically. In February 2024 it awarded the NCP name and the clock symbol to the Ajit Pawar faction, the founder’s group becoming the NCP (Sharadchandra Pawar).
  • These adjudications proceed under paragraph 15 of the Symbols Order, and the tests derive from Sadiq Ali v. Election Commission (1971): the aims and objects test, the party constitution test, and the majority test.
  • The criticism is structural rather than partisan. Because the legislative-wing majority has become the decisive test, the party label follows whoever can assemble the legislators — which rewards exactly the conduct the Tenth Schedule was enacted to deter, and means that a party’s identity is settled by its defectors rather than by its members.

Party finance

Money is the deepest unreformed problem in the Indian party system, because it is the one area where every party’s interest runs the same way.

• The position before 2018. Contributions above ₹20,000 had to be reported under Section 29C of the RPA; below it, unlimited aggregation of undisclosed “small” donations was lawful. Companies faced a cap on political donations and a duty to disclose them. Candidate expenditure was capped; party expenditure was not, and that exclusion is the central weakness of the regime.
• The Electoral Bonds Scheme 2018 let a donor buy a bearer instrument from a designated bank and hand it to a party, the donor’s identity known to the bank and the government but to nobody else. Accompanying amendments removed the profit-linked cap on corporate donations under Section 182 of the Companies Act 2013 and the duty to disclose the recipient.
• In Association for Democratic Reforms v. Union of India, decided on 15 February 2024, a five-judge Constitution Bench — Chief Justice Chandrachud with Khanna, Gavai, Pardiwala and Misra — unanimously struck the scheme down.
  • The holding: anonymous political funding violates the voter’s right to information under Article 19(1)(a). Who funds a party is information about what it is likely to do in office, and the state’s interest in curbing black money did not justify the least transparent of the available means.
  • The Companies Act amendment fell with it as manifestly arbitrary, restoring the position that unlimited, undisclosed corporate donation is not permissible.
  • The State Bank of India was directed to disclose past purchases; 22,217 bonds bought between 1 April 2019 and 15 February 2024 were disclosed, and the Commission published donor and recipient lists in March 2024.
• No replacement scheme has been legislated. Funding has reverted to electoral trusts and direct declared donations. Trust donations rose from roughly ₹1,218 crore in 2023-24 to about ₹3,826 crore in 2024-25, the bulk of it to the largest party — a shift in channel rather than in transparency, since the trust still stands between the donor and the recipient.
• The scale is the point. ADR’s analysis of the six national parties for FY 2024-25 shows declared income of ₹7,960 crore, roughly 85% from donations — the BJP ₹6,769 crore, the Congress ₹918 crore. It was a general-election year, so the total is inflated; the ratio is not.
• Parties remain outside the Right to Information Act, despite a 2013 ruling of the Central Information Commission that the national parties are public authorities — a ruling simply not complied with.

The state-funding debate

Position	Source	The argument
Partial state funding, in kind	Indrajit Gupta Committee, 1998	Funding only for recognised national and state parties, not independents; in kind (facilities, materials, broadcast time) rather than cash; only partial funding is economically feasible
Desirable, but only with a prior ban	Law Commission’s 170th Report, 1999	State funding is worth having only if parties are simultaneously barred from other sources; a regulatory framework must come first
Regulate parties first	National Commission to Review the Working of the Constitution, 2001	Declined to endorse state funding; an appropriate framework for regulating parties must precede it
Transparency and internal democracy first	Law Commission’s 255th Report, 2015	Audited accounts, disclosure and internal democracy, enforced by penalty and de-registration

• The case for state funding: it reduces dependence on undisclosed private money and the obligations that follow it; it creates a level playing field for parties without corporate access; and it makes party finance auditable.
• The case against: it does not displace private money unless private funding is simultaneously capped and enforced, and enforcement capacity is exactly what is in doubt; it entrenches existing parties by tying eligibility to past performance; and, in the cartel-party analysis, it makes parties agents of the state rather than of society.
• The convergence in all four sources is instructive: each says that funding reform is downstream of party regulation, and party regulation is what no Parliament has enacted.

Federal structure, electoral system and social cleavages

The party system India has is the joint product of three variables, and none of them explains it alone.

• The electoral system. First-past-the-post is disproportional — it converts pluralities into majorities and punishes parties whose support is thin and wide. In India it does something unusual: it punishes small national parties and rewards small regional ones, because 4% of the national vote concentrated in one state wins seats while 4% spread evenly wins none. FPTP is therefore an engine of regionalisation in India, not of two-party consolidation.
• The delinking of national and state elections. The first four general elections were simultaneous with assembly elections; premature dissolutions from the late 1960s broke the cycle, and 1971 was the first Lok Sabha election held separately. Separate cycles created a state-level arena with its own timing, issues and leadership — and it was there that regional parties were built.
• Federalism supplies the prize. A state government is a genuine object of power — police, land, education, most of what citizens meet daily. A regional party can therefore aim at governing rather than at protesting, which distinguishes India’s regional parties from ethnic minority parties in unitary states.
• Social cleavages, and why they fragment rather than polarise.Lipset and Rokkan explained Western party systems as the frozen imprint of historical cleavages — centre–periphery, church–state, land–industry, owner–worker.
  • India’s cleavages are numerous and cross-cutting rather than few and reinforcing. Caste does not align with language; language does not align with religion; class cuts across all three.
  • Cross-cutting cleavages produce fragmentation, not polarisation. Because no single line divides the whole society, no two-bloc structure can form on one axis; instead, many parties form on many axes, each strong where its axis is locally salient.
  • The corollary is a warning: the fragmentation is protective. A politics organised on one reinforcing cleavage would be more stable and considerably more dangerous, which is why any project to align all the cleavages onto a single axis is a change in the party system and in something larger than it.

What the party system has and has not delivered

India began with two assets few new democracies had — an established civil service and a well-organised mass party — and the charge is that the record since is nonetheless dismal. It deserves a serious answer in both directions.

• What the system delivered.
  • Peaceful transfers of power, repeatedly, including from a party that had just suspended the Constitution.
  • Integration: regions, languages and castes that had never shared an electorate were brought into one political arena rather than out of it, which is the achievement of the Congress system and the reason India did not fragment.
  • Progressive inclusion: successive upsurges brought OBCs, Dalits and Adivasis into active participation, and legislatures became more representative of the society electing them.
  • Channelling conflict into elections rather than violence, in a society with more mobilisable cleavages than almost any other.
• What it did not.
  • Internal democracy — absent everywhere, with dynasty and centralised selection as its consequences.
  • Programmatic competition — parties are, as Zoya Hasan puts it, office-oriented rather than policy-oriented, and manifestos bind nobody.
  • Linkage between elections — the citizen’s relationship with the party has, in Yogendra Yadav’s phrase, shrunk to the instrumental act of voting.
  • Clean finance — undisclosed money remains the norm, and the one attempt at a statutory scheme was struck down for making the problem worse.
  • Clean candidates — the share of MPs with declared criminal cases has risen across every recent Lok Sabha, a party-organisational failure before it is a legal one; the jurisprudence is treated elsewhere in this unit.
• The causes of the shortfall are organisational, not cultural. Parties were never regulated, entering the constitutional text only in 1985 and only to punish defection; the institution with the standing to regulate them cannot de-register them; and the only body that can legislate is composed of them. These failures are the predictable output of a design that left the system’s most powerful actors outside the law’s reach.

India built an elaborate architecture of accountability for the state and almost none for the organisations that capture it.