Fundamentals

Internet Shutdowns and Digital Rights: What They Are and Why They Matter

By Daniel Sardá · Published on October 08, 2026

7 min read1,354 words

An internet shutdown is not just any loss of connection: it is a deliberate interruption that affects essential infrastructure for communication, work, and collective action.

Someone tries to check an emergency alert, but the page will not load. A shop cannot process a payment. A group of neighbours loses the channel it used to coordinate. The immediate symptom is the same in all three cases—a missing connection—but the cause matters: a technical failure is not the same as an intentionally ordered disconnection.

In discussions of internet shutdowns and digital rights, that distinction is the starting point. An internet shutdown is an intentional interruption of online communications for a population or territory, ordered by a government or carried out on its behalf. It may cover an entire country or a particular area, fixed or mobile networks, and it may take the form of a total cut-off or a slowdown that makes the connection effectively unusable.

Not every outage, then, is a shutdown. Nor do all digital restrictions work in the same way. Naming the problem accurately makes it possible to measure it, assign responsibility on the basis of evidence, and assess whether an intervention respects rights.

Key idea: The visible effect may be “there is no internet,” but only an investigation into the cause can distinguish a failure from a deliberate restriction.

Shutdowns, outages, and blocks are not synonyms

An accidental interruption may result from a power cut, damage to a cable, congestion, or technical errors. Even when its consequences are serious, it does not by itself show that someone intended to prevent communication. Claims about intent or responsibility require further evidence.

A selective block, by contrast, prevents access to particular websites, applications, or services while the rest of the connection remains available. A shutdown in the narrower technical sense cuts or degrades connectivity broadly. Some international institutions use a wider definition that also includes the deliberate disruption of essential communications services. Each analysis should make clear which meaning it is using.

These distinctions do not make selective blocks unimportant. They help identify differences in scope and collateral harm. Blocking one destination and disconnecting an entire region may both affect rights, but they do not affect the same number of people or call for precisely the same technical analysis.

Two questions must also be kept separate. The first is factual: what stopped working, for how long, and for whom? The second is legal and institutional: who made the decision, under what authority, and on what justification? Network data may establish that an anomaly occurred; it will not always answer the second question.

Why connectivity matters for exercising rights

Internet access need not be treated as an absolute right to recognize its importance. It is infrastructure through which people exercise liberties that are already protected. Article 19 of the International Covenant on Civil and Political Rights protects freedom of expression—the freedom to seek, receive, and impart information and ideas—and Article 21 protects peaceful assembly. Today, a substantial share of those activities takes place or is organized online.

When connectivity is cut off, people may lose access to news and channels for expression. A community may be left without ways to call a meeting or learn of last-minute changes. Students and workers may be unable to reach classes, platforms, or documents. Health, assistance, and emergency services may also be affected, depending on the duration, scope, and context of the interruption.

The Office of the United Nations High Commissioner for Human Rights has documented that these disruptions can affect education, work, health, and other essential services, alongside expression and assembly. Careful wording matters: not every interruption causes every one of these harms, but a broad disconnection raises the risk that several will occur at once.

Key idea: Internet access is an enabling medium: shutting it down does not restrict one activity alone, but many lawful choices by people who are not part of the stated target.

The problem of harm to third parties

A general order does not distinguish between someone alleged to be causing a threat and someone trying to speak with family, collect a sale, or ask for help. That lack of precision is one of the central problems with broad shutdowns: they inevitably reach third parties unconnected to the conduct the authority says it is addressing.

From a liberal perspective, an open network reduces the need to seek prior permission to speak, associate, and exchange. The power to disconnect it concentrates an exceptional capacity to intervene in millions of individual choices. That is why it is not enough for an authority to invoke terms such as “security” or “public order.” Those may be legitimate aims, but they must be tied to a concrete threat and to a measure whose usefulness can be shown within the rule of law and constitutional limits on political power.

The burden of justifying the restriction lies with the party imposing it. The broader, more indiscriminate, and more prolonged it is, the harder it becomes to explain why people and activities unrelated to the threat had to bear its cost.

How to assess a connectivity restriction

International human rights law offers useful questions. General Comment No. 34 of the Human Rights Committee explains that restrictions on freedom of expression must be provided by law, pursue one of the recognized legitimate aims, and meet the requirements of necessity and proportionality. Its concrete application depends on the jurisdiction, but the test helps organize the analysis:

• Legality: Is there a public, accessible, and sufficiently precise law authorizing the measure?
• Legitimate aim: Does the authority identify a permitted aim rather than rely on a vague justification?
• Necessity: Is there evidence of a real threat and a direct link between that threat and the restriction?
• Proportionality: Are the geographic, technical, and temporal scope related to the harm the measure seeks to prevent?
• Less intrusive alternative: Could the objective have been achieved through a targeted intervention affecting fewer people?
• Non-discrimination: Does the measure fall unevenly on a group without an admissible justification?

According to the High Commissioner’s report, blanket shutdowns cannot be justified under international human rights law. A targeted measure might be defensible only in exceptional circumstances, as a last resort, and when no effective alternative exists. This does not make every dispute identical, but it does reject the idea that an emergency grants unlimited power to disconnect people.

Key idea: A legitimate aim does not make every means legitimate. The authority must explain why a shutdown was necessary and why a less restrictive option was insufficient.

Safeguards before, during, and after a shutdown

An effective limit on power requires more than a statement of principle. It requires procedures that let people know about, challenge, and review the decision. Relevant institutional safeguards include a public order with a defined scope and duration, timely information for the public, independent oversight, and avenues of redress compatible with due process for those affected.

Transparency also makes it possible to evaluate results. Did the measure achieve its announced aim? Which essential services did it interrupt? Was it lifted when its cause disappeared? Without information about the responsible authority, legal basis, and period of application, public scrutiny becomes nearly impossible.

Alternatives should be examined before resorting to disconnection. Depending on the risk, they may include targeted responses, ordinary law enforcement against specific conduct, infrastructure protection, or public emergency communications. Not all will work in every case, but the obligation to consider them prevents treating an entire population’s connectivity as a switch for routine use.

The questions a shutdown must answer

A responsible analysis starts by establishing what happened and avoids calling every failure censorship. It then asks who decided, under what law, in response to what threat, and on what evidence. Finally, it compares the measure with its effects on people outside the alleged target and with the available alternatives.

That process helps distinguish a technical difficulty from an abuse of power. It also recalls something more basic: when a society depends on the network to obtain information, assemble, work, and ask for help, turning it off is not merely a technical decision. It is an intervention in concrete liberties and must therefore be subject to public rules, strict limits, and accountability.