Centralist thinking has long rested on a seductive but mistaken premise: that national unity requires uniformity. In this view, a nation must share one language, one culture, and one administrative template, imposed by a dominant Centre, in order to remain cohesive. Diversity is treated as a weakness to be erased through homogenisation into a single national mould.
Federalism rests on a fundamentally different understanding of unity — one that accommodates diversity rather than erasing it. This understanding is captured through a number of metaphors used in federal theory.
The “melting pot” metaphor, often associated with assimilationist nationalism, assumes that diversity must dissolve into a single uniform identity. Federalism rejects this premise. Instead, the “salad bowl” metaphor emphasises combination without dissolution: each ingredient retains its distinct character while contributing to a coherent whole.
A similar idea exists through the metaphor of the “warp and weft”. The Union provides the warp — the structural threads that hold the fabric together — while the States supply the weft, adding colour, variation, and texture. Unity is thus woven through diversity rather than cast in a single mould.
A fourth metaphor, developed by Morton Grodzins in 1960, describes federalism as a “marble cake”, where federal, state, and local functions are intermingled rather than neatly layered.
Together, these metaphors convey a common lesson: unity does not require homogeneity. Imposed uniformity produces a poor salad, a dull fabric, or a tasteless cake.
Comparative experience reinforces this conclusion. Nations that have accommodated diversity through constitutional design — Canada, Switzerland, Belgium, South Africa, and Singapore — have sustained stable political systems despite deep linguistic, cultural, or ethnic plurality. By contrast, nations that pursued unity through enforced homogeneity, notably the Soviet Union and Pakistan, found cohesion elusive and short-lived. Unity pursued through uniformity frequently fractures; unity built on accommodation proves far more durable.
Classical “people’s revolutions” such as the French Revolution (1789) and the Russian Revolution (1917) were not merely regime changes but total revolutions. A unified “people” rose against a centralised nation-state to refound society through decisive rupture. Political conflict became binary — the Third Estate against the monarchy, the proletariat against the bourgeoisie — allowing mobilisation along a single dominant fault line. As Hannah Arendt observed in On Revolution (1963), such revolutions sought to “found freedom on a new beginning” by destroying the inherited social and institutional order.
India presents a striking contrast. Despite colonial domination, mass poverty, deep inequality, and recurrent unrest, the country has never experienced a comparable Jacobin or Bolshevik moment — no episode in which a single, unified people rose to violently remake the internal political and social order. The uprising of 1857, the freedom movement led by Mahatma Gandhi, and Jayaprakash Narayan’s mobilisation in the 1970s came closest, but participation was uneven and far from universal.
Since Independence, India has witnessed numerous insurgencies, Naxalite movements, ideological mobilisations, and mass protests. Yet these have remained fragmented and localised rather than a simultaneous pan-Indian upheaval — a pattern best explained by the sociological theory of cross-cutting cleavages.
The theory of cross-cutting cleavages, articulated by Seymour Martin Lipset and Stein Rokkan in Party Systems and Voter Alignments (1967), argues that the configuration of social divisions — more than the intensity of grievances — determines whether political conflict becomes moderate, polarised, or revolutionary.
Where cleavages reinforce one another — class aligning with religion, language, region, and ideology — society polarises into two antagonistic camps. Politics becomes binary and revolutionary mobilisation becomes possible, as in the French Revolution and the Russian Revolution. Where cleavages cross-cut, however, identities fragment rather than aggregate. Individuals belong to multiple overlapping groups, none of which dominates across the polity. Alliances shift across issues, coalitions remain fluid, and grievances fail to converge into a single national rupture.
India is a textbook example. Class cuts across caste; caste across religion; religion across language; and language across region. A Kannada-speaking Hindu shares religion with a Bengali-speaking Hindu but not language; a farmer in Punjab shares occupation with a farmer in Telangana but not caste. Political allies on one axis routinely become opponents on another.
This structure produces three stabilising effects. First, coordination failure: revolutionary mobilisation requires common symbols and enemies, but in heterogeneous societies, a slogan that mobilises one group often alienates another. Second, regionalisation of grievances: discontent erupts intensely but locally — language in Tamil Nadu, agriculture in Punjab, tribal rights in central India — rarely becoming a nationwide conflagration. Third, shifting coalitions: overlapping identities encourage negotiation, accommodation, and compromise rather than total political destruction.
The sociological reality that liberty is safer when society is divided into a multiplicity of interests, factions, and groups was anticipated by James Madison in Federalist No. 10:
“The influence of factious leaders may kindle a flame within their particular states, but will be unable to spread a general conflagration through other states.”
India exemplifies Madison’s insight on a civilisational scale. Its extraordinary diversity of language, caste, religion, and region ensures that no single faction can plausibly claim to represent “the people” as a whole. What may appear, from a centralist perspective, as disorder or fragmentation is in fact a structural safeguard against both revolutionary rupture and majoritarian domination.
For this reason, attempts to homogenise the country through projects such as “One Nation, One Language” or “One Nation, One System” are fundamentally misconceived. India’s plural identities cannot be artificially or coercively erased — whether through compulsory linguistic uniformity, mandated inter-caste or inter-religious marriages, forced conversions, or similar interventions. Even if such homogenisation were possible, it would likely prove counterproductive and dangerous to the country’s long-term unity. India remains united precisely because it is diverse — and because its federal structure has, however imperfectly, learned to live with that diversity rather than deny it.
To be concluded
The author is retired IAS officer of Tamil Nadu cadre, former Vice-Chancellor of Indian Maritime University, Chennai, and Member, High-Level Committee on Union-State Relations constituted by the Government of Tamil Nadu