What Bangladesh can learn from past referendums worldwide
As the idea of a referendum reappears in political discussion, the question is not simply whether it can be done, but whether it can be trusted
On 17 October 2025, Chief Adviser Muhammad Yunus and leaders from 24 parties signed the July National Charter at the South Plaza of the Jatiya Sangsad Bhaban.
The July charter is an 84-point reform plan born from the student uprising that ended Hasina's 16-year tenure. While cameras captured handshakes and national anthems, tension flared outside as the National Citizens' Party (NCP) boycotted over missing legal safeguards and protesters clashed with police.
Weeks later, unity began to fray over a referendum on 48 constitutional changes, including proportional representation and new democratic safeguards. BNP wants the vote on the national elections day while Jamaat-e-Islami and other Islamist parties demand it beforehand; NCP is fine with either.
The National Consensus Commission (NCC) ignored dissent, enforcing a single yes-no ballot and requiring the next parliament to pass reforms within 270 days, or have them automatically adopted, a move critics call unconstitutional.
Keep updated, follow The Business Standard's Google news channel
The interim government has now given parties seven days to agree or face unilateral action.
As the idea of a referendum resurfaces in Bangladesh's political discourse, it has stirred both curiosity and suspicion — seen by some as a way to renew democratic participation and by others as a tactic to centralise power. Globally, referendums have served to empower citizens but also to control them, their impact depending on who wields them and how.
Switzerland is the prime example, with over 600 federal referendums since 1848 on issues from constitutional reform to taxation. Though turnout averages 35-45%, this reflects normalisation, not apathy, in a stable, high-trust democracy.
For a referendum to have meaning, it needs an independent election commission, a free press, and a political environment where all sides can campaign without fear. Without those conditions, even a high turnout or a decisive result may not convince the public that the process was genuine.
However, in South Asia, referendums have rarely empowered the people, instead legitimising rulers. Pakistan held three - in 1960, 1984, and 2002 - all under military rule. Ayub Khan's 1960 vote, through the Basic Democracies system, gave him 96% approval, reported The New York Times, despite being indirect and tightly controlled.
Zia-ul-Haq's 1984 Islamisation referendum claimed 98% support with 62% turnout, while Musharraf's 2002 vote secured 97%. All were suspiciously one-sided, reflecting not unity but a lack of freedom.
Bangladesh experienced a similar event in 1977 under Ziaur Rahman, who organised the country's first national referendum to approve his policies and programmes. The official results were astonishing — showing around 98% support with an 88% turnout.
Many saw it as an effort to legitimise Zia's authority after years of political instability. The referendum gave the appearance of public endorsement but offered no real choice.
Later, Hussain Muhammad Ershad held another referendum in 1985 under martial law, claiming 94% support with 72% turnout - results that were also widely disputed. Since then, Bangladesh has held only one more referendum, in 1991 with around 35% turnout and around 83% voting in favour of the bill, which restored the parliamentary system through a democratic process.
The contrast between Switzerland's modest 45% turnout and Bangladesh's or Pakistan's official 90% shows a paradox of democracy. High turnout does not always mean high legitimacy. In established democracies, people can afford not to vote because they trust institutions. In weaker systems, high turnouts often reflect control, intimidation, or political theatre.
Even in advanced democracies, referendums often divide rather than unite. The UK's 2016 Brexit vote (52%-48%) deeply split the nation despite high turnout. Scotland's 2014 independence referendum saw 84% participation, showing that identity-driven votes boost engagement but also division. Similarly, Turkey's 2017 referendum, narrowly expanding Erdoğan's powers, blurred the line between democracy and control.
Most of the referendums took place in Europe where institutions are strong while only a fraction occurred in Asia and Africa, often under authoritarian or transitional regimes. In most cases, the outcomes reflected not just the people's will but the political climate of the time.
That context matters deeply for Bangladesh today. As the idea of a referendum reappears in political discussion, the question is not simply whether it can be done, but whether it can be trusted.
For a referendum to have meaning, it needs an independent election commission, a free press, and a political environment where all sides can campaign without fear. Without those conditions, even a high turnout or a decisive result may not convince the public that the process was genuine.
Nowhere is this tension clearer than in the standoff over the July Charter, as major parties clash over its implementation. BNP, calling itself the "biggest stakeholder," wants the referendum held alongside the national elections next year to cut costs and prevent it from overshadowing the polls. They accused the NCC of betraying it by ignoring "notes of dissent" on key reforms like proportional representation that, it claims, benefit smaller rivals.
In contrast, Jamaat-e-Islami and other Islamist parties demand a November referendum to secure reforms such as an upper house before elections, warning that delays threaten the July uprising's spirit and accusing BNP of stirring discord to maintain dominance.
NCP backs holding the referendum before the polls, but only if the Chief Adviser issues an executive implementation order nullifying dissenting notes and prioritising the Charter's "uprising supremacy." While dismissing the BNP's dispute as a political drama that obstructs justice and stability, NCP itself has yet to sign the charter.
This three-way conflict over when to hold the referendum, whether to include dissenting opinions, and how to make it legally binding has exposed how fragile the agreement really is.
The interim government has now given a seven-day deadline to settle the disputes, as protests, court challenges, and economic pressure grow. What started as a promise of unity after the uprising has turned into a risky test for Bangladesh's democratic future.
Bangladesh's democratic institutions continue to struggle with credibility as voter confidence wanes after consecutive one-sided elections, and civic participation remains largely symbolic. In this fragile context, launching a referendum without broader political reform risks repeating past patterns of limited inclusion disguised as public consultation.
