The Right to Be Unavailable

A green dot is not a yes

Somewhere right now, a small green dot says that someone is available. It means their laptop is open. It does not mean they have time, attention or energy. It does not mean this is their working day, that your request outranks what they are doing, or that they want to talk.

Yet we read it that way. And once we do, something shifts. Being reachable becomes being noticed. Being noticed becomes being available. Being available becomes being expected to answer. And not answering becomes something you have to explain.

Technical availability is a system state. Human availability is a form of consent. We should stop treating the first as evidence of the second.

I noticed this in myself on an ordinary evening. I had opened my laptop to finish one small thing — ten minutes, no more. My status turned green. Within a minute, another request arrived, and I felt an almost automatic pull to answer it.

Nobody had done anything wrong. The message was reasonable. What surprised me was the feeling: as if, by being visible, I had agreed to something. I had chosen to give ten minutes to one task. The system had made that choice look as though I had offered my attention to everyone.

Consent to work on one thing is not consent to be available for everything.

Nobody wrote it down

Most organisations never instruct anyone to answer at night. They don't need to.

When Eurofound asked workers in four European countries why they replied to work messages outside their contracted hours, 75% said it was expected. 61% feared a negative impact if they didn't. Half expected it to help their career.¹ More than eight in ten had received such messages in a typical week, and almost nine in ten had replied to at least some.

What struck me most was this: companies with a right-to-disconnect policy did not see fewer out-of-hours messages, or fewer replies. Their people did report a better work–life balance and less stress. That tells me the real issue is not whether a message can arrive. It is whether someone is obliged to respond — and what happens to them if they don't.

Availability is a question of power and culture before it is a question of settings.

The cost you cannot see in the output

It is tempting to say that people can simply ignore a notification. The research is less forgiving. Merely receiving a phone notification can disrupt performance on a demanding task, even when the person does not pick up the phone.²

And when work is interrupted, people often keep their output stable by working faster — while reporting more stress, frustration and time pressure.³ The urge to reply quickly to work messages is linked to poorer recovery.⁴

That is why always-on cultures can look healthy on a dashboard. The work still gets done. The price is paid somewhere else: in evenings, in recovery that doesn't happen, in attention that never quite comes back.

My flexibility is not your obligation

I don't want a world in which nobody may send a message after five o'clock. Flexibility is one of the real gifts of digital work: being able to choose when and where the work gets done.

The problem begins when my flexibility quietly becomes your availability. When I send at 22:00 because it suits me, and you feel you must answer at 22:05 because it came from me.

So the right to be unavailable is not a right to stop others from communicating. It is the principle that technical reachability must not automatically create an obligation of immediate human response. A message can be asynchronous only if the sender genuinely gives up control over when the recipient pays attention to it. Otherwise the medium is asynchronous, but the culture is not.

Rest does not need a return on investment

Much of the conversation about digital boundaries is framed as productivity: switch off so you perform better tomorrow. I understand why. But it concedes too much.

Older traditions make a different argument. In the Torah, Sabbath rest extends beyond the individual to servants, strangers and animals, and it is tied to the memory of liberation from slavery.⁵ Rest is not a privilege for whoever has enough autonomy to claim it; it is shared across hierarchy. The Qur'an calls believers, at the call to Friday prayer, to leave trade.⁶ Commerce pauses because something else is given priority.

I am careful not to turn these practices into productivity tips; their meaning is their own. What they show me is that a boundary is strongest when the time outside it does not have to justify itself in the language of output. Rest does not need to make Monday more efficient to matter.

What the law does — and doesn't — say

There is no international human right to ignore your messages. But there is a long-standing right to rest: the International Covenant on Economic, Social and Cultural Rights recognises rest, leisure and reasonable limitation of working hours.⁷

France is often said to have “banned work email after hours”. It did something more interesting. It made the right to disconnect, and the regulation of digital tools, a subject organisations must negotiate — and where there is no agreement, employers must write a charter and train people in reasonable use.⁸ The law does not draw the line for everyone. It requires organisations to govern it.

That feels right to me. Digital reachability does not erase the right to rest. It just means someone has to decide, explicitly, where the boundary sits.

When the agent is always on

AI agents add a new layer. They can acknowledge, draft, schedule and act while the person responsible is asleep. That can protect people, by absorbing routine requests. It can also create false availability: an organisation that looks permanently responsive, while no human has actually looked at the question.

I think we will need to distinguish three states clearly: a human is available; an agent is available; a human judgement is required. AI should be allowed to extend service availability. It should not silently extend claims about human availability.

What I would ask for

1. Name response times, not just channels. What is emergency, what is same-day, what can wait until tomorrow?
2. Make urgency costly enough to stay meaningful. A deliberate urgent override, used rarely, protects both focus and emergencies.⁹
3. Promise no reprisal. Outside agreed availability, not replying is not a performance issue.
4. Lead by sending later. Senior people scheduling non-urgent messages for working hours changes more than any policy.
5. Show expectations, not surveillance. “Routine replies tomorrow” is more honest than a green dot.
6. Look at workload. No setting fixes a job that does not fit inside working time.

Reachable, and still free

I want to be reachable by the people I work with and care about. I also want my attention to remain something I give, not something that is taken because the system says it can be.

Back to that evening: the laptop was open, the status was green, and the request was waiting. Nothing about the technology was wrong. What needed to change was what I let it mean — and what we, together, agree it means.

The dot can stay green. The reply is still a choice.

Sources

The full evidence base is on the Human Limits research page.

1. Eurofound (2023). Right to disconnect: Implementation and impact at company level. eurofound.europa.eu. ↩
2. Stothart, C., Mitchum, A. & Yehnert, C. (2015). The attentional cost of receiving a cell phone notification. doi:10.1037/xhp0000100. ↩
3. Mark, G., Gudith, D. & Klocke, U. (2008). The cost of interrupted work: More speed and stress. doi:10.1145/1357054.1357072. ↩
4. Barber, L. K. & Santuzzi, A. M. (2015). Please respond ASAP: Workplace telepressure and employee recovery. doi:10.1037/a0038278. ↩
5. Exodus 20:8–11; Deuteronomy 5:12–15. Sefaria. ↩
6. Qur'an 62:9. quran.com. ↩
7. International Covenant on Economic, Social and Cultural Rights, Article 7. ohchr.org. ↩
8. Code du travail (France), Article L2242-17, 7°. legifrance.gouv.fr. ↩
9. Slack. Pause your Slack notifications. slack.com. ↩