Verifying information during an internet shutdown means falling back on methods that never depended on the network in the first place: named human sources, physical confirmation, radio cross-checks and a written log of what you know and how you know it. The tools change. The standard does not.
This matters because shutdowns do not stop information — they only stop good information. Rumour travels perfectly well by word of mouth. What disappears is the ability to check it quickly, which is precisely the capacity a newsroom needs most during the events that trigger shutdowns.
What an internet shutdown actually breaks
It helps to be specific, because the loss is uneven. Reverse image search, social media monitoring, messaging apps, cloud storage and remote collaboration all stop. So does the ability to reach a colleague in the next region to ask a simple question.
What survives is more than people expect. Shortwave and FM radio continue. SMS often continues, at least intermittently, because operators may throttle data while leaving voice and text running. Physical movement continues. Institutional relationships — with a clinic, a school, a district officer, a church or mosque — continue, and those are the relationships that carry verification when nothing else does.
Organisations that track these disruptions, including NetBlocks and the #KeepItOn coalition, publish measurement data after the fact. That record is valuable for advocacy, but it arrives too late to help a newsroom working through the shutdown itself.
Build the fallback before you need it
The single most useful preparation is a contact tree written on paper. Not exported from a phone. Written down, duplicated, and held by more than one person.
A workable tree has three layers:
- Anchors — people who are physically fixed and institutionally visible. A clinic nurse, a head teacher, a market association secretary. They are reachable at a known place, and their information is tied to something observable.
- Relays — people who travel routinely between areas. Drivers, traders, health outreach workers. They carry confirmation across distances that the network no longer covers.
- Broadcasters — whoever can put a corrected fact in front of thousands of people at once. In most of the places this matters, that is a community radio station.
Agree in advance what each layer is being asked for. Anchors confirm what they have personally seen. Relays carry messages and note what they observed en route. Broadcasters do not originate claims; they repeat only what has cleared the standard below.
The verification standard that survives offline
Strip verification back to its irreducible core and you get three questions, none of which require connectivity:
- Who saw this directly? Not who told you. Who was present. If nobody in the chain was present, you have a rumour with a long tail, not a report.
- Can a second person, unconnected to the first, confirm the same detail? Independence is the whole point. Two people repeating the same market gossip is one source, not two.
- What physical trace would exist if this were true — and is it there? A road closure leaves stopped vehicles. A clinic influx leaves a full waiting room. If the trace should exist and does not, the claim is weaker than it sounds.
Write the answers down as you get them. During a shutdown you will be handling many partial claims at once, and memory is not a filing system. A simple ledger — claim, source, what they saw, time, status — is enough, and it becomes the record you publish from once connectivity returns.
Working with radio when radio is what is left
Community stations are the most under-used verification asset in a shutdown. They already have the audience, the reach and the trust. What they usually lack is a verification protocol they can apply under pressure.
Three practices make the difference. First, separate confirmed information from unconfirmed reports explicitly on air, in plain language, every time. Second, state what you do not know as clearly as what you do — an honest gap is more useful than a confident guess. Third, schedule correction slots rather than waiting for corrections to feel newsworthy; a fixed slot normalises the act of correcting.
If you are setting up these relationships from scratch, our field guide to community radio partnerships covers the agreements worth making before a crisis rather than during one.
What to do the moment connectivity returns
The window immediately after restoration is the most valuable and the most wasted. Everyone rushes to catch up on the feed. The more useful move is to publish your ledger.
Post what you verified while offline, what you could not verify, and what turned out to be false. This does three things at once: it corrects the record while the audience is still paying attention, it demonstrates method rather than asserting authority, and it produces the documentation that advocacy organisations need. Groups such as CIPESA and Paradigm Initiative build cases from exactly this kind of contemporaneous, ground-level record.
That handover from newsroom to advocacy is where most of the value leaks away. We look at how to close it in digital rights advocacy after the shutdown ends.
Common questions
Can you verify anything at all without the internet?
Yes. Every verification standard in use today predates the internet. What connectivity changed is speed and reach, not the underlying logic of independent confirmation.
Is a VPN a solution?
Sometimes, and only partly. VPNs can route around some filtering, but they do nothing when data service is cut entirely, and in some jurisdictions their use carries legal risk. Treat them as one option in a plan, never as the plan.
What is the most common verification failure during a shutdown?
Circular sourcing. A claim moves through five people and comes back sounding like independent confirmation. Asking "who saw this directly" is the cheapest defence available.
If your organisation is building this capacity, get in touch — this is the kind of protocol that works far better designed in advance than assembled under pressure.