—Md. Sakib Sarkar, Law Graduate, Eastern University, Bangladesh, and Associate Editor, NILS Bangladesh; and M. Jonayed Al Ratul Rasel, LLM Candidate, Eastern University, Bangladesh


Air pollution kills nearly seven million people each year. Given a toll that size, one might expect the world’s constitutions to provide a guaranteed right that those individuals can invoke and courts can enforce when the air becomes unsafe to breathe. Most do not. Five jurisdictions, including Bangladesh, India, Kenya, Pakistan, and South Africa, show what follows from that choice, and what it takes to undo it. We suggest that the lack of such a right is a flaw in the way constitutions are designed, one common to many constitutions drafted during decolonization. Rather than granting citizens the right to a healthy environment, these constitutions generally imposed a duty on the state to protect the environment and placed that duty beyond the courts’ reach. That design choice explains why the right to clean air, where it exists at all, usually arrives through another right instead of standing on its own. In this post, we argue that a constitutional right to clean air only becomes real once people have a real way to invoke it, and naming the right without preserving that access produces the appearance of protection rather than protection itself.
The Architecture of Deferral
The design takes the form of directive principles of state policy. The model began with Ireland’s Constitution of 1937, took root in India’s Constitution in 1950, and spread across the postcolonial world, reaching Bangladesh’s Constitution of 1972. Comparative legal scholarship has paid surprisingly little attention to this category. David Boyd’s global survey of environmental constitutions treats the directive-principle model as a stage that many states have outgrown by converting duties into rights. Tarunabh Khaitan reads them more sympathetically, as a device that let constitutional framers secure the consent of political groups who might otherwise have rejected the constitutional bargain. Either way, directive principles bind the state in principle, but no individual can enforce them in court.
There are three reasons that may lead framers to shield a provision from judicial enforcement. The provision may depend on resources only the political branches control, it may turn on technical judgements that courts are poorly equipped to make, or it may remain too contested to be settled through adjudication. In all of these cases, a directive principle can still shape how a court interprets a separate right or justify deference to the executive. It cannot ground a claim on its own, and the state keeps discretion over whether, and how quickly, to act. The line between a principle a court may consult and a right a citizen can enforce determines who controls the pace of environmental protection.
Four Paths from the Same Starting Point
Four jurisdictions built on this template show different paths. Indian Supreme Court began reading a right to a healthy environment into Article 21’s guarantee of life as early as the 1980s, through M.C. Mehta’s long campaign of environmental litigation. Yet, India’s constitutional text has never caught up with its own case law. The environment remains, formally, a directive principle under Article 48A. The reach and impact of the Indian case law therefore still depends on which case reaches a bench and on how that bench interprets the right to life.
Kenya took the same interpretive path but reached a different outcome. In Peter K. Waweru v Republic (2006), Nairobi’s High Court read the pre-2010 Constitution’s right to life as including a clean environment, drawing on the public trust doctrine. Kenya’s 2010 Constitution then made this interpretive move unnecessary, with Article 42 guaranteeing a clean and healthy environment directly. Moreover, Kenyan courts have since applied it without inferring it from another right. This case suggests a simple lesson that once a constitution names the right explicitly, courts no longer need to improvise.
Pakistan’s experience complicates that lesson. Its courts used the same right-to-life technique in Shehla Zia v WAPDA (1994), three decades before the Twenty-Sixth Amendment added Article 9A in 2024, which guarantees an environment free from pollution as a standalone right. But that same amendment curtailed the Supreme Court’s suo motu jurisdiction and left the new right without implementing legislation or a clear procedure for citizens to invoke it. In other words, it is not self-executing, but rather a right written into a constitution that still needs a procedural door that people can walk through.
South Africa shows what happens when text and procedure arrive together. Its 1996 Constitution skipped the interpretive stage altogether. Section 24 guarantees an enforceable right to an environment that does not harm health or well-being. In the 2022 Deadly Air case, the Pretoria High Court held that chronic pollution on the Mpumalanga Highveld breached Section 24 directly. The court treated the guarantee as immediately enforceable and ordered the government to enforce its own air quality plan; the Supreme Court of Appeal upheld in 2025. The remedy followed from the text that named the right, not from an inference built on a different one.
Bangladesh has taken none of these paths, and sits closest to India. Article 18A, added in 2011, commits the state to protect the environment, but it sits within the constitution’s Fundamental Principles of State Policy, which Article 8 places beyond judicial enforcement. Absent a standalone right, the Appellate Division held in the 1997 Mohiuddin Farooque case that the right to life includes the enjoyment of pollution-free water and air. Protection still depends on which case reaches a bench, and according to Section 7(4) and 6(3) of the Environment Court Act 2010, a citizen can sue only after first petitioning a Department of Environment inspector and waiting sixty days for a response.
The General Case
Across all five jurisdictions, the pattern shows that naming environmental protection in a constitution is necessary but not sufficient. What separates an enforceable right from an aspirational one is not the eloquence of the text but whether an individual can easily access a court and obtain a remedy. An enforceable right to clean air requires both naming the right and preserving the machinery to enforce it. A reform that achieves the first without the second is not a partial win, but rather a new kind of gap.
Kenya converted an interpretive right into an explicit, enforceable one and changed outcomes. South Africa never needed the interpretive stage because it built the right and the remedy in the same instrument. India and Bangladesh remain dependent on courts willing to read “life” expansively, and in Bangladesh’s case, on which government agency is willing to bring the case at all. Pakistan shows that even the act of naming the right can be undercut by the same amendment that grants it, if the reform quietly narrows the court’s jurisdiction to hear it.
The lesson that follows applies to every one of these five constitutions. In jurisdictions where the constitution still treats clean air as a directive principle, these examples force the state to confront a clear choice. Environmental protection must either remain a mere promise that politicians can indefinitely ignore, or become an actionable right that individuals can take to court for a real remedy. The evidence from this comparison supports the latter option, but only in so far as the right and the standing rules necessary to enforce it are developed together. The present reform process in Bangladesh is one instance in which this choice is being made in real time; it will not be the final one.
Suggested citation: Md. Sakib Sarkar & M. Jonayed Al Ratul Rasel, The Clean Air Battles: Five Constitutions and One Unresolved Right, Int’l J. Const. L. Blog, Oct. 6, 2026, at: http://www.iconnectblog.com/the-clean-air-battles-five-constitutions-and-one-unresolved-right/