Captive Guardians

**Sanjeevan Bhuyan

Nine months separated a landmark Supreme Court ruling on the Election Commission’s independence from Parliament’s quiet, constitutionally compliant reversal of it. That speed was not political boldness. It was constitutional logic and understanding it is the first step toward fixing it.

Nine months. That is all it took. In February 2023, a unanimous five-judge Supreme Court bench delivered what appeared to be a landmark democratic victory. In Anoop Baranwal v. Union of India, the Court declared the Election Commission of India (ECI) a ‘guarantor institution’ of constitutional democracy and directed that its members be appointed through a committee that included the Chief Justice of India. By December 2023, Parliament had passed the Chief Election Commissioner and Other Election Commissioners Act 2023. The Chief Justice was out. A Cabinet Minister nominated by the Prime Minister was in. The executive secured a structural two-to-one majority on the appointment panel. This was not a constitutional crisis. It was the constitution working exactly as designed.

The Problem Is Architectural, Not Political

The instinct is to read the nine-month reversal as political bad faith. That instinct misses the deeper truth. The reversal was constitutionally available because of a foundational design flaw built into India’s accountability institutions at independence itself and understanding that flaw is what any meaningful reform must begin with.

India’s principal fourth branch institutions: the ECI; the Comptroller and Auditor General (CAG, the constitutional body that audits government expenditure); the Central Vigilance Commission (CVC, which supervises anti-corruption investigations); and the Central Information Commission (CIC, which enforces the right to information) were never built to be genuinely independent. They were built to be administratively useful. There is a profound difference between the two.

Agents vs. Trustees: The Distinction That Changes Everything

KEY CONCEPT  AGENT VS. TRUSTEE

An agent acts on behalf of a principal, derives authority from the principal’s grant, and remains subject to the principal’s direction. A trustee holds authority for a beneficiary and owes duties that cannot be overridden by whoever created the trust. In a constitutional context, the beneficiary is the citizenry not the government.

India’s constitution treats its fourth branch institutions primarily as agents. Article 324(1) says electoral superintendence shall be ‘vested in’ an Election Commission. The verb matters: it denotes delegation of a function, not the declaration of an independent institution. Compare Article 124(1), which says ‘there shall be a Supreme Court of India.’ That is a constitutional declaration of existence. The Supreme Court cannot be abolished by Parliament because the constitution declared it into being. The ECI was not declared. It was delegated to.

The CAG has stronger personal protection tenure equivalent to a Supreme Court judge, a mandatory post-retirement employment bar. But these shield the individual officer, not the institution. Parliament can reconstitute the commission, reduce it to one member, or expand its mandate without touching a single job-security provision. Personal independence and institutional independence are not the same thing. India has pursued the former while neglecting the latter. The executive did not abuse the Constitution. It followed it.

What South Africa and Kenya Got Right

Chapter 9 of the South African Constitution does not merely protect the Public Protector or the Auditor-General; it declares them as independent institutions and imposes on Parliament and the executive a duty to ‘assist and protect’ that independence. These are structural declarations, about what the institution is, not protections for whoever currently holds office. When President Zuma defied the Public Protector’s remedial orders, the Constitutional Court in Economic Freedom Fighters v Speaker of the National Assembly [2016] ZACC 11 did not need to construct new authority. It gave effect to the authority the Constitution had already established. With a delegated institution, courts must build that authority case by case. With a declared one, the Constitution has already done it.

Kenya’s 2010 Constitution went further, building civil society participation directly into appointment processes for all independent commissions, so no executive-controlled combination can dominate appointments regardless of who holds power. Colombia’s Fiscalía General de la Nación has its independence constitutionally declared under Article 249 of the 1991 Constitution. Research by Iván Cepeda Castro shows this has produced an institutional culture in which prosecutorial independence is professionally self-reinforcing; prosecutors understand themselves as owing duties to the constitutional order, not to whoever holds executive power. Constitutional declaration changes not just the text. It changes how institutions understand their own authority.

The ECI: When One Person’s Courage Becomes an Institution’s Weakness

TN Seshan, appointed Chief Election Commissioner in 1990, demonstrated that Article 324 read by the Supreme Court in Mohinder Singh Gill v Chief Election Commissioner (1978) and TN Seshan v Union of India (1995) as conferring plenary authority over all matters incidental to free and fair elections could support transformative enforcement. Under Seshan, the ECI operated for a sustained period as though its authority derived from constitutional trusteeship rather than executive delegation.

But his successors exercised the same formally available powers differently. Association for Democratic Reforms’ (ADR) analysis of Model Code enforcement across election cycles shows that outcomes have tracked individual commissioners rather than any stable institutional standard. Milan Vaishnav’s research on money and electoral politics confirms that the structural incentives enabling candidate misconduct have been contained where commissioners were assertive, not addressed structurally. Where outcomes depend on who was appointed rather than what the institution is designed to do, the appointment becomes the operative mechanism of control. There is no need to instruct an appointee when the selection itself does the work. This is subtler than capture, and harder to litigate. It is what you get from a system designed this way.

The legitimacy cost of all this shows up in the data. ADR surveys document declining public confidence in the Commission’s impartiality among opposition supporters and younger voters. The Supreme Court’s electoral bonds judgment (February 2024) put on record the scale of asymmetric political financing that the Commission, despite years of formal authority over the electoral process, had left unaddressed.

The RTI Queue Nobody Discusses

Of the four institutions, the CIC’s decline is the one that costs ordinary people the most directly. The Right to Information Act 2005 did something real: it gave any citizen, without a lawyer or a court fee, a legal right to request information about government decisions that affected them. In its first decade, the CIC built substantial jurisprudence and enforced disclosure obligations against reluctant public authorities.

The Right to Information (Amendment) Act 2019 gutted the structural foundation. It converted fixed statutory commissioner tenures into executive-determined periods, which meant commissioners no longer held office on terms insulated from government preference; commissioners who could not be sure of continued tenure began, naturally enough, reading contested cases differently. PRS Legislative Research documented these consequences at the time.

In several states, a citizen who files an RTI appeal today will wait longer than the full parliamentary term of the legislature that enacted the RTI Act before a commissioner who serves at executive pleasure hears her case. The human cost of the agent model does not arrive as a constitutional event. It builds in RTI queues now running past five years, in disclosure orders that lapse without follow-up. What the Act once delivered and what a first-time applicant today actually receives are not the same thing.

Why Courts Cannot Carry This Weight Alone

The post-Emergency Supreme Court built India’s functional fourth branch through purposive constitutional interpretation. Nick Robinson has shown this judicial expansion was a deliberate institutional strategy rather than simply a response to litigant demand. The results were real electoral disclosure norms from Union of India v Association for Democratic Reforms (2002), CBI supervision from Vineet Narain v Union of India (1998). These are genuine governance contributions.

But a fourth branch whose independence is judicially constructed is not genuinely independent. Gautam Bhatia has documented that the judiciary’s willingness to champion fourth branch independence has varied considerably across institutional contexts, more assertive in some domains than others. The same Court that recognised ECI independence as a constitutional imperative in Anoop Baranwal did not intervene when the 2019 RTI amendments structurally dismantled the CIC’s independence.

Anoop Baranwal itself warrants scrutiny beyond the nine-month reversal it produced. The basic structure doctrine established in Kesavananda Bharati (1973) to mean that certain constitutional essentials cannot be amended even by Parliament, already required treating ECI independence as unamendable. SR Bommai (1994) recognised democracy as a basic structure. Indira Gandhi v Raj Narain (1975) confirmed that free and fair elections are constitutive of that democratic basic structure. If ECI’s independence is necessary for free and fair elections, then the basic structure doctrine already protects it. Parliament’s authority to legislate it away is considerably narrower than the CEC Act 2023 assumes. The Court in Anoop Baranwal saw this implication and chose not to follow it. It invited Parliament to legislate instead of declaring the matter constitutionally closed. Parliament then legislated in the direction it preferred. Authority built through judicial interpretation remains vulnerable to legislative redesign. Authority declared in constitutional text requires governments to confront the basic structure doctrine directly, a contest that has not gone well for any government that has tried.

A Three-Tier Reform Framework

Constitutional reform proposals fail when they treat the political conditions they wish existed as the conditions that exist. The three-tier framework here starts from what is achievable under existing parliamentary arithmetic and works outward toward the constitutional redesign that genuine independence ultimately requires.

TIER ONE  STATUTORY REFORM: AVAILABLE NOW

A constitutional amendment requires a two-thirds parliamentary supermajority in both Houses plus ratification by half the states a threshold current arithmetic does not provide. But ordinary legislation is achievable. A Guardian Institutions Independence  Act should restructure appointment committees so no single actor holds a majority, fix commissioner terms without renewal, bar post-tenure employment in regulated or audited sectors, and require public reasons for major decisions. The Law Commission’s Report No. 255 on Electoral Reforms provides a ready drafting foundation. Centre for Policy Research analysis shows that funding inadequacy is among the strongest correlates of regulatory capture in India. A statutory resourcing formula charging all fourth branch institutions to the Consolidated Fund at a prescribed minimum percentage of Union revenue receipts would address this directly.

TIER TWO  JUDICIAL DOCTRINE: AVAILABLE THROUGH COURTS

A future Constitution Bench should declare that the independence of institutions performing guarantor functions for the democratic basic structure is itself a basic structure element unrevisable through ordinary legislation. This would not create new constitutional authority. It would recognise authority the basic structure doctrine already implies and that Anoop Baranwal identified but declined to claim. Any future legislative reversal would then require confronting the basic structure doctrine directly, a contest no government has successfully won.

TIER THREE  CONSTITUTIONAL AMENDMENT: THE LONG HORIZON

A constitutional amendment should declare the ECI, CAG, CVC, and CIC as independent constitutional institutions rather than functional receptacles; establish a Guarantor Institutions Council for appointments requiring a supermajority no two-actor coalition can satisfy; and impose on Parliament a constitutional duty to ‘facilitate’ institutional independence, replicating South Africa’s Section 181(3). International IDEA’s global analysis shows successful constitutional entrenchment has followed demonstrated statutory performance. Tiers One and Two do the immediate work of strengthening these institutions while the political conditions for Tier Three develop.

The Legitimacy Stakes Are Not Abstract

V-Dem’s 2024 Democracy Report documents a measurable decline in India’s electoral management independence. Transparency International’s Corruption Perceptions Index has stagnated around 38–40 for several consecutive years, despite India having more formal anti-corruption institutions than at any earlier point. Rawls, in Political Liberalism, made the structural point: the conditions of fair democratic participation cannot be left to ordinary majority politics, because the majority will always have reasons to adjust those conditions in its favour. In Between Facts and Norms, Habermas argued that democratic legitimacy depends on the quality of the processes through which decisions are made and justified to the people they affect. An institution that cannot credibly show its independence from the executive cannot provide that quality of process. The formal authority behind those decisions starts to look thin.

Analysis (From Agents to Trustees)

The agent-trustee cleavage was built into the Constitution at independence. For seven decades, informal norms of executive restraint made the flaw manageable. Those norms have frayed, and the underlying design is now plain. The nine months between Anoop Baranwal and the CEC Act 2023 did not reveal a constitutional crisis. It revealed a constitution working precisely as its designers left it.

The ECI and the CAG should hold their authority not because a particular commissioner is personally courageous, a particular bench is willing, or a particular political conjuncture makes audit findings actionable. They should hold it because the Constitution says they hold it in trust for the democratic order that India’s founding committed it to sustaining.

The nine months posed a question India has not yet fully answered: whether it wants institutions that are independent because the Constitution declares them to be, or institutions that are powerful only when courts, political conjunctures, and individual commissioners happen to align in their favour. The first requires constitutional imagination and political will. The second requires only continuing as before. The choice and the cost, borne not by constitutional scholars but by citizens filing RTI appeals into multi-year queues is the real subject of this piece.

Adapted from ‘Captive Guardians: Constitutional Trusteeship, Democratic Legitimacy, and the Fourth Branch in India,’ submitted to the 12th D.D. Basu Essay Writing Competition, Constitutional Law Society, WBNUJS & The Vidhi Centre for Legal Policy.

Sanjeevan Bhuyan is a fifth-year law student at the KIIT School of Law, Bhubaneswar. He has a keen interest in Competition Law and Environmental Law, with a particular focus on the intersection between sustainability, and evolving market structures. He strongly believes that fundamental rights in a democratic country should not remain confined to constitutional text alone, but must be genuinely accessible and meaningful for every individual. His academic interests are rooted in critical legal inquiry and the role of law in shaping a more equitable society.

**This essay was placed first in the DD Basu Essay Competition, organised by Charkha, the constitutional law team, in collaboration with the NUJS Constitutional Law Society.

**Disclaimer: The views expressed in this blog do not necessarily align with the views of the Vidhi Centre for Legal Policy.