What every rating is measured against, the reasoning behind each one, and how to read it.
Overview
The heat map translates the analysis in the white paper, A New Digital Frontier: A Roadmap for Rights-Respecting Digital Governance Laws in Bangladesh, into an accessible, high-level visual assessment of the potential human rights impact of six key laws across three dimensions: freedom of expression; civic participation and democratic process; and privacy and data protection.
It is based primarily on the author's review of relevant provisions of each statute in the white paper, considered individually and as part of the legislation as a whole. An in-depth assessment of the six statutes is available in the white paper and should be read alongside the heat map for a fuller understanding of the basis, context, and limitations of each classification.
Several methodological considerations are important to understanding how these classifications were reached and how they should be interpreted.
Considered as a whole in the relevant dimension, the law contains features that create a substantial likelihood of, or are associated with a documented pattern of, serious adverse impact on the protection or exercise of the relevant rights. Relevant factors may include statutory design, the breadth of powers or offenses, inadequate safeguards, enforcement experience, institutional arrangements or a combination of these factors. Actual harm need not already have occurred where the design and likely operation of the law indicate a sufficiently serious risk.
∕Medium risk
Medium Risk of Adverse Impact (“Medium Risk”)
This classification indicates material but qualified or mitigated risks to the relevant rights, including where rights-protective provisions, procedural safeguards, reforms, limited scope or other mitigating factors reduce — but do not eliminate — the potential adverse impact. It may also apply where protective and restrictive features of the law, considered alongside the institutional environment, enforcement experience and foreseeable implementation, produce a mixed overall assessment.
−No risk
No Risk of Adverse Impact (“No Risk”)
The assessment identified no material adverse impact sufficient to warrant a risk classification in the relevant dimension, based on the law's text, scope, likely operation and available evidence at the time of review. This does not mean that there is zero conceivable risk, nor does it preclude a different assessment if implementation, interpretation or circumstances change. Rather, it indicates that, on balance and at the time of assessment, the law was not considered to present a material risk of adverse impact on the protection or exercise of the relevant rights.
+Positive impact
Positive Impact
Considered as a whole in the relevant dimension, the law is assessed, on balance, as more likely than not to materially strengthen the protection or exercise of the relevant rights compared with the legal position otherwise applicable. A positive classification does not endorse every provision, establish that safeguards are sufficient in all respects, or predict effective and rights-respecting implementation; weaknesses may remain and are identified in the full white paper analysis.
Holistic and contextual assessment, not a numerical score
Classifications in the heatmap are based on holistic, contextual assessment, rather than provision-by-provision scores. A statute may contain provisions that strengthen a particular right or interest alongside others that weaken its protection. The classification reflects the author's overall assessment of the likely net effect of the statute when the relevant provisions are read together, rather than a numerical aggregation of individual provisions or the application of a fixed scoring rubric.
Assigning predetermined numerical weights to individual provisions can produce misleading results by failing to account adequately for their scope, severity, practical significance and interaction with other parts of the law. A single broadly framed enforcement power, for example, may have greater practical implications for a right than several narrowly framed protective provisions, while a safeguard that appears strong in isolation may operate differently when read alongside exemptions or executive powers elsewhere in the statute.
Accordingly, a single problematic provision does not necessarily determine the classification of an entire statute; equally, a positive classification does not constitute an endorsement of every provision within it. As such, this exercise necessarily involves reasoned legal and policy judgment.
Central to the assessment is the context extending beyond the text of individual provisions. Where relevant, the classification considers statutory powers, offenses, penalties, procedures, exemptions and safeguards, alongside institutional design, historical patterns of enforcement, implementation capacity, judicial interpretation and the wider constitutional, political, legal and regulatory environment in Bangladesh. This is particularly important because identical or similar statutory language can operate very differently depending on variations in one or more of these factors.
All classifications should therefore be understood as holistic, contextual assessments, current as of May 2026, rather than definitive findings about every present or future application of a law.
Finally, the three dimensions — freedom of expression; civic participation and democratic process; and privacy and data protection — should be understood as analytical lenses rather than self-contained categories. They can and frequently do overlap: restrictions on expression may affect civic and political participation; surveillance may have consequences for privacy while chilling expression or association; and the collection and use of personal data may affect participation in democratic processes.
The heat map separates these dimensions to make the analysis navigable, not because their effects can always be neatly separated in practice. This interrelationship is also one reason why the methodology relies on reasoned legal and policy judgment rather than predetermined numerical weights or a fixed scoring rubric.
Benchmarks
The assessment is grounded principally in Bangladesh's constitutional framework and internationally recognized human rights standards, including, without limitation, the International Covenant on Civil and Political Rights. It is also informed by relevant interpretations and standards developed through the United Nations human rights system, as well as comparative legislation, jurisprudence, regulatory frameworks and policy approaches from other jurisdictions.
Comparative law is used as a reference point, not as a presumption of good practice or a model for automatic adoption. A provision is not treated as rights-respecting merely because a similar approach exists in the European Union, Australia, the United Kingdom or another jurisdiction. The analysis instead asks whether the underlying objective, safeguards and institutional arrangements are appropriate to Bangladesh's constitutional, political, socioeconomic and institutional context, and how the provision is likely to operate in practice.
For example, the white paper considers whether erasure right under the Personal Data Protection Act, 2026, while strengthening individual control over personal information, could also create an alternative pathway for content removal or restrictions on legitimate public-interest expression without adequate substantive and procedural safeguards.
The relevant comparative question is therefore not simply, “Does another jurisdiction have this rule?” as a basis for deciding whether a similar provision should be adopted, but rather, “What would this rule do, for whom, and with what safeguards, in Bangladesh?” A foreign framework is therefore referenced selectively to identify potentially relevant approaches, safeguards and governance models, rather than to advocate their wholesale adoption.
What every rating is measured against.
Every rating is benchmarked against, among others, the International Covenant on Civil and Political Rights (ICCPR), the Constitution of Bangladesh, and comparative laws.
The heat map should not be read as the output of a mathematical model, mechanical checklist or quantitative human rights index. It represents the author's informed qualitative legal and policy assessment, drawing on statutory interpretation, comparative legal analysis, constitutional and international human rights standards, available evidence of implementation and enforcement, and the wider institutional and political context examined in the white paper.
The underlying analysis was also informed by consultations and interviews with legal and constitutional experts, civil society representatives and people with direct experience of the effects of the legislation, enabling consideration of both formal legal rules and their operation in practice.
The evidentiary basis necessarily varies between statutes. Where a law has an established history of implementation, documented enforcement and judicial interpretation inform the assessment. For newer or not-yet-effective legislation, the assessment is necessarily more prospective, placing greater weight on statutory design, foreseeable application and the institutional environment in which the law is expected to operate. The absence of documented harm is therefore not, by itself, treated as evidence of the absence of risk.
The classifications necessarily involve interpretive judgment, and reasonable researchers may reach different conclusions about the weight to be given to particular provisions, safeguards, implementation risks or other factors. They should be read as reasoned analytical judgments, supported by the fuller analysis in the white paper, rather than objective measurements or predictions of every way in which a statute will operate.
They reflect the laws and information available as of May 2026 and may warrant reassessment as legislation is amended, courts interpret its provisions, regulatory frameworks develop, or implementation and enforcement practices evolve.
