—Arafat Hosen Khan, Visiting Senior Fellow, LSE Law School
INTRODUCTION: A MOMENT SQUANDERED
Constitutional moments, in Bruce Ackerman’s foundational account, are those rare episodes in a polity’s life when an exceptional degree of popular mobilisation, institutional urgency, and political will converges to permit transformative constitutional change that would be impossible in ordinary politics. The July 2024 mass uprising in Bangladesh — in which hundreds lost their lives, Sheikh Hasina’s long rule and brought down the Awami League (AL) government, and a Nobel Laureate took office leading an interim administration — appeared, by any measure, to constitute precisely such a moment. The interim government of Professor Muhammad Yunus seized this opening by establishing a Constitutional Reform Commission (CRC), which submitted its proposals in January 2025, and later a National Consensus Commission (NCC) intended to anchor those proposals in a broad political settlement. A referendum held in February 2026 was presented as the democratic culmination of this process. Yet as Bangladesh now transitions into a political environment shaped predominantly by the Bangladesh Nationalist Party (BNP), the question that confronts constitutional scholars is stark: did the post-July reform process genuinely capitalise on its constitutional moment — or did it squander it through structural exclusion?
This two-part commentary argues that Bangladesh’s reform process failed to honour the promise of the constitutional moment it inherited. In Part I, I examine the design of the National Consensus Commission and the procedural defects that rendered it, in the words of senior Supreme Court lawyer Sara Hossain, speaking at a SHUJAN roundtable on 21 April 2026, incapable of constituting a democratic process in any recognisable sense. I situate this critique within comparative constitutional theory, drawing on the scholarship of Amal Sethi on equitable elite bargaining, and the landmark comparative work of Elkins, Ginsburg, and Melton on constitutional endurance. In Part II, I examine how the transition to BNP-dominated political reality has rendered the constitutional moment not merely incomplete but potentially foreclosed — and what, if anything, can be salvaged.
WHAT A CONSTITUTIONAL MOMENT DEMANDS: THE THEORETICAL STAKES
Before engaging Bangladesh’s specific trajectory, it is worth articulating precisely what constitutional moment theory requires of its actors. Ackerman’s framework, developed across his We the People trilogy, posits that higher law-making — the kind that alters constitutional fundamentals — requires a different quality of political engagement than ordinary legislation. It demands breadth of participation, depth of deliberation, and a process by which constitutional decisions can be meaningfully attributed to “the People” as a whole rather than to a faction, a movement, or a victorious party.
The comparative constitution-making literature converges on a related but distinct insight: that the inclusiveness of the process, not merely its formal procedures, determines both the immediate legitimacy and the long-term durability of the resulting constitutional text. Elkins, Ginsburg, and Melton have shown empirically that constitutions built through broad stakeholder compromise — South Africa’s 1996 post-apartheid constitution being the paradigmatic case — display markedly greater endurance than those imposed by dominant factions or crafted through technocratic commissions with limited political representation. Amal Sethi’s theory of equitable elite bargaining extends this analysis by insisting that constitutional legitimacy in deeply divided societies depends not merely on formal participation, but on structurally equitable bargaining among the major political elites whose future behaviour the constitution must regulate. A constitutional settlement that excludes major political forces does not merely suffer from a legitimacy deficit in the abstract — it systematically undermines its own capacity to bind the very actors whose compliance is most essential.
Bangladesh’s constitutional history offers a painful confirmation of these theoretical propositions. The 1972 Constitution, though born of genuine revolutionary legitimacy, concentrated power so dramatically in the Awami League’s hands that it catalysed the cycles of military intervention and constitutional rupture that have defined the country’s subsequent history. The 13th Amendment (1996), introducing the non-partisan caretaker government system, succeeded precisely because it was the product of cross-party pressure and mass mobilisation; its abolition through the 15th Amendment (2011), accomplished without genuine multiparty agreement, generated the electoral credibility crisis that ultimately fuelled the 2024 uprising. History, in other words, has already issued its verdict on exclusive constitutionalism in Bangladesh. The post-July process was supposed to break this cycle. The evidence suggests, instead, that it reproduced it.
THE NATIONAL CONSENSUS COMMISSION: SELECTIVE PROCESS, MANUFACTURED LEGITIMACY
The National Consensus Commission was established to transform the CRC’s technocratic recommendations into a politically grounded charter capable of commanding cross-party support. In theory, this was sound design: the CRC’s proposals — including semi-parliamentary restructuring, the creation of a National Constitutional Council as a meta-guarantor institution, term limits, and electoral reform — were substantively ambitious and in many respects well-calibrated to Bangladesh’s structural pathologies. As I have argued in earlier writing for this blog, the post-July constitutional process was animated by a genuine desire for structural transformation. The NCC was designed to be the mechanism through which that desire was translated into durable political agreement.
What the NCC delivered instead was, in Sara Hossain’s precise formulation, something that cannot be called a democratic process in any way. Her critique, delivered at the SHUJAN roundtable of 21 April 2026, was threefold. First, the process was selective: a small circle of participants made decisions without meaningful input from outside that circle. Second, there was no female member on the NCC — a remarkable omission in a society where women constitute half the population and where the July uprising itself was substantially shaped by female students. Third, and most fundamentally, the public had no clear understanding of what they were voting for in the referendum. Hossain posed the challenge directly to those assembled: how many could articulate the thirty constitutional proposals that formed the referendum’s third question? Her implicit answer — that virtually none could — raises the most basic question of democratic legitimacy: a vote whose object is unknown cannot meaningfully be attributed to the popular will.
This critique is not merely procedural. It strikes at the constitutional core of the NCC’s legitimacy. As Velasco-Rivera and Colón-Ríos have argued in their analysis of permanent constituent power, the exercise of constituent authority — the authority to remake or fundamentally alter a constitutional order — requires a form of popular engagement that is qualitatively different from ordinary electoral participation. The ballot is not self-legitimating; it legitimates only when those casting votes understand what they are deciding and have had a genuine opportunity to shape the options before them. The NCC process, on both counts, fell short.
The selective character of the NCC process was not accidental. It was legally entrenched. On 11 May 2025, the interim government issued the Anti-Terrorism (Amendment) Ordinance, 2025- a gazette notification banned all activities of the Awami League and its affiliated organisations. This was not a temporary administrative measure: it invoked the anti-terrorism framework of the Anti-Terrorism Act, 2009 — a law designed to combat political violence — to prohibit the participation, organising, and public expression of Bangladesh’s largest opposition party and its entire organisational ecosystem. The NCC process that followed was therefore not merely one from which the AL was absent: it was one from which the AL was legally excluded on pain of criminal sanction. To speak of “selective” participation in those circumstances is, if anything, an understatement. It was structurally enforced constitutional silence.
The comparative record underlines the stakes. Chile’s constitutional Convention of 2021–2022, despite being a far more elaborate and formally inclusive process, nonetheless failed partly because the deliberative architecture created internal dysfunction and public disconnection — the resulting text was rejected by a substantial majority in a ratification referendum. Tom Ginsburg and Isabel Álvarez have observed that the Convention’s procedural failures — including internal fragmentation and the absence of a coherent deliberative framework — proved fatal regardless of the substantive quality of individual provisions. Bangladesh’s NCC never achieved even the level of procedural elaboration that Chile’s Convention, despite its failures, possessed. The procedural architecture was, from the outset, compromised by the exclusion of the political forces whose buy-in was most necessary.
THE EXCLUSION OF THE AWAMI LEAGUE AND JATIYA PARTY: A CONSTITUTIONAL CATEGORY ERROR
The most consequential structural flaw in the NCC process was the exclusion of the Bangladesh Awami League (AL) and the Bangladesh Jatiya Party (BJP), along with other significant political formations. This exclusion was, to a substantial degree, understandable as a matter of immediate post-uprising politics: the AL’s government had overseen mass violence against protesters, and the legitimacy of the interim administration rested in part on its differentiation from AL governance. Yet understandable does not mean constitutionally sound. There is a categorical difference between transitional justice — holding the AL accountable for specific acts — and constitutional exclusion — structurally denying the AL’s constituency any voice in determining the constitutional framework under which they will be governed.
The Anti-Terrorism (Amendment) Ordinance, 2025, and the gazette ban that followed it the next day, operationalised this category error in the most consequential way available to an executive: by statute. The anti-terrorism framework was applied not to specific acts of violence — for which ordinary criminal law and transitional justice mechanisms exist — but to the entire organisational existence of a political party and all its affiliates. A party that represents, by conservative estimates, well over a third of the Bangladeshi electorate was not merely excluded from the deliberative table: it was declared, by executive order backed by anti-terrorism law, legally incapable of participating in political life. The constitutional implications are severe. As the African Court of Human and Peoples’ Rights recognised in Noudehouenou v Republic of Benin (Application No. 003/2020), the suppression of a political party’s right to participate in public affairs — even where that party has been associated with governance failures — raises fundamental questions of both democratic legitimacy and human rights frameworks. The blanket organisational ban, without judicial determination of specific criminal liability, transformed constitutional exclusion into legal erasure.
This distinction matters for several reasons. First, as I have noted in earlier analysis of Bangladesh’s constitutional crossroads, the AL, even in its post-2024 weakened state, retains the support of a substantial portion of the electorate — conservative estimates suggest well over a third of the voting population has historically aligned with AL or its affiliated formations. A constitutional settlement that does not account for this constituency’s preferences is not merely politically fragile; it is constitutionally incomplete. Adem Kassie Abebe’s analysis of constitutional amendment legitimacy in Africa provides instructive comparative context: he demonstrates that constitutional processes designed by post-authoritarian governments to lock out the preceding regime’s supporters tend to generate the very instability they seek to prevent, because the excluded constituency retains sufficient capacity — electoral, institutional, and organisational — to challenge or circumvent constitutional settlements made without them. In Bangladesh, this dynamic is additionally complicated by the anti-terrorism ban: a party whose organisational activities are criminalised cannot exercise its latent constituent authority through ordinary political channels. The result is not that the excluded constituency’s constituent power disappears — as Velasco-Rivera and Colón-Ríos’s framework predicts, it persists — but that its expression is driven underground, into informal networks, electoral successors, or, in the worst case, extra-constitutional mobilisation. Banning a party does not resolve the constitutional problem it represents; it defers and intensifies it.
Second, the exclusion of the Jatiya Party presents a distinct but related problem. The Jatiya Party, whatever its complex genealogy and its association with the Ershad-era authoritarian period, has maintained a consistent electoral presence in specific regions and social constituencies. Its exclusion from the NCC reproduced the same logic of selective legitimation that has historically characterised Bangladesh’s constitutional politics: the idea that constitutionalism is the prerogative of the currently powerful rather than a framework binding on all. As De Visser and Son Bui have documented in their analysis of Asian constitution-making, post-authoritarian transitions that fail to include parties associated with the prior regime — even discredited ones — routinely generate constitutional texts that are subsequently challenged or dismantled by those parties upon their return to power.
Third, and perhaps most theoretically significant, the exclusion of major political formations transforms the constitutional moment from an exercise in popular constituent power into an exercise in factional governance. Hahm and Kim’s comparative analysis of constitutional founding in postwar Japan and South Korea illuminates the pathology: when a constitutional founding is accomplished by one group “making ‘we the people'” without genuine participation by significant segments of the population, the resulting constitution is formally a popular charter but substantively a partisan instrument. Bangladesh’s interim government, in excluding the AL and the Jatiya Party from the NCC, did not merely commit a political error. It committed a constitutional category error — mistaking the exclusion of a discredited government for the exclusion of an indispensable political constituency.
THE REFERENDUM AS SUBSTITUTE: PLEBISCITE OR REFOUNDING?
The February 2026 referendum was presented as the popular democratic validation that would transform the NCC’s selective deliberations into genuine constitutional authority. As I have argued with Sangita F. Gazi in The Diplomat, there is a fundamental distinction between a plebiscite — a majoritarian instrument that registers popular sentiment on a pre-determined question — and a constitutional refounding — a participatory process through which the people genuinely constitute themselves as the author of their basic law. The February referendum fell unmistakably into the former category.
This matters because the constitutional legitimacy of the outcomes depends on which category the referendum inhabits. The Thai Constitutional Court’s ruling in Re Constitution Amendment Bill (Ruling 4/2564, 2021), and the Kenyan Supreme Court’s analysis in Attorney General v David Ndii (Petition No 12 of 2021), both establish the comparative principle that referendums do not confer unlimited constituent authority on the outcomes they validate. Constitutional amendments — even those approved by popular vote — can be struck down or delegitimised if they exceed the bounds of legitimate revision, or if the process generating them failed to satisfy the requirements of genuine popular participation. Haiti’s Article 284-2 similarly entrenches the non-amendability of certain constitutional provisions even in the face of majoritarian preferences.
Bangladesh’s Article 256(6) of the Thai Constitution (2017) equivalent — the question of whether a referendum approving changes through an inadequate process creates binding constituent authority — has not been directly adjudicated. But the theoretical answer is clear: a referendum whose participants did not understand the questions, held without the participation of major political forces in the preceding deliberative process, cannot constitute the kind of transformative popular act that converts ordinary political authority into higher law-making. Sara Hossain’s observation that the public could not articulate what they were voting for is not merely a practical concern about public education. It is a constitutional diagnosis: the February 2026 referendum did not capture a genuine constitutional moment; it registered a political preference under conditions of informational inadequacy and structural exclusion.
The AL’s legal prohibition from participation in any public political activity also means that the February 2026 referendum — whatever its other deficiencies — was conducted in the absence of organised opposition from a party representing a substantial share of the electorate, whose campaigning, public communication, and organisational activity were all criminalised under the anti-terrorism ordinance. This is not a referendum in any constitutional sense that democratic theory can recognise: it is a plebiscite held in conditions of enforced political silence for a major political force.
CONCLUSION: THE PROMISE NOT KEPT
The Yunus-led interim government came to power on the back of extraordinary popular mobilisation and a genuine mandate for constitutional transformation. It had access to one of the most significant constitutional moments in Bangladesh’s post-independence history. The Constitutional Reform Commission produced substantively serious proposals; the July National Charter offered a legitimate basis for cross-party dialogue; the Supreme Court’s November 2025 judgment restoring the caretaker system — whatever its prospective application problems — suggested a judiciary willing to align itself with democratic reforms.
Yet the promise was not kept. The National Consensus Commission was, as Sara Hossain correctly observed, a selective rather than a democratic process. The exclusion of the AL and the Jatiya Party from meaningful participation transformed what could have been a constitutional moment into a factional settlement. The referendum amplified rather than corrected these deficiencies. The result is a constitutional reform package that is formally enacted but substantively precarious — resting on a foundation of selective legitimacy that the excluded parties have every incentive, and considerable capacity, to challenge.
Part II will examine what this failure means under the political reality now emerging in Bangladesh, and whether the constitutional moment can yet be recovered, or whether it has been lost entirely.
Suggested citation: Arafat Hosen Khan, Bangladesh’s Broken Constitutional Moment Part I: The Architecture of Failure – The National Consensus Commission and the Problem of Selective Constitutionalism, Int’l J. Const. L. Blog, May 8, 2026, at: http://www.iconnectblog.com/bangladeshs-broken-constitutional-moment-part-i-the-architecture-of-failure-the-national-consensus-commission-and-the-problem-of-selective-constitutionalism/