Async Is a Political Position

Thesis

Synchronous communication is a power structure. It rewards real-time fluency, audible confidence, and colocated time zones. Async-first writing is a deliberate redistribution of who gets to shape the outcome. Treating the default medium as a personality preference hides the hierarchy it installs.

Context

On Nightbind a platform working group met every Tuesday for eleven weeks. The same three voices took the airtime. A senior engineer in a later time zone wrote strong objections in Slack after each call: reasoned, dated, ignored as “already decided in the meeting.” By sprint six we shipped a caching layer the async objectors had flagged. By sprint eight we rewrote it. Disagreement was not the cost. The cost was that disagreement was structurally illegible until it became a rewrite.

I have been the person in the room. I have also been the person three hours behind, drafting the better answer after the call ended. Same skill set, different political outcome, determined by medium rather than merit. No reorg was on the table. The available move was procedural: RFC docs before implementation lock. That is a political move even when you file it under process.

This is a different claim from async-as-accessibility, which is about nervous systems and load. This one is about who is allowed to count. You can run a written culture that is still a clique. The medium is necessary. It is not sufficient.

Mechanism

Meetings optimize for performance, not truth. Spoken debate rewards speed, volume, and social comfort. The person who needs forty-five quiet minutes to name the fatal edge case does not get those minutes in a fifty-minute slot with six attendees and a hard stop. Priya Parker notes that gatherings inherit hidden purposes, often status display wearing alignment as a costume. When the hidden purpose is “look engaged,” the artifact is attendance. Decision quality is optional.¹

Async redistributes the floor. An RFC in GitHub Discussions gives the slow thinker, the non-native speaker, the introvert, and the caregiver the same surface as the loudest extrovert: equal column width, equal timestamp, equal permalink. Responses can be judged on content rather than on who got there first. Jason Fried and DHH built Basecamp’s operating model on this premise. Writing is the primary coordination layer because speech is ephemeral and political. The document can be quoted. The vibe in the room cannot.²

Default medium is default hierarchy. Sync-first teams hire for sync competence: timezone overlap, meeting stamina, performative certainty. Async-first teams can hire for written clarity and recruit across continents without filtering for who can stomach six hours of Zoom a week. The team you assemble is downstream of the communication politics you normalize. You will then blame “culture fit” for a filter you installed in the calendar.

“Already decided” is a power move. When leadership treats the meeting as the decision venue and async comment as commentary, they encode a two-tier system: present bodies count, absent bodies annotate. The fix is procedural. No implementation lock until the RFC window closes. Dissent logged in the ADR. Explicit disagree-and-commit with names attached. Disagreement that survives that process is real alignment. Silence in a room is not. Irving Janis documented how cohesive groups suppress dissent to preserve harmony. Fast meeting consensus is often that mechanism with better snacks.³

Async is not automatically democratic. Written culture can favor eloquent writers over domain experts, or people with uninterrupted writing time over parents on fragmented schedules. You still need readable templates, review norms, and leaders who respond to written pushback without coding it as “not a team player.” Changing the medium without changing the incentive just moves the clique into markdown.

Calendar invites are allocation. Recurring syncs consume the same budget as headcount. They just do not appear on the spreadsheet. When I audited Nightbind’s standing meetings against output, three weekly forums existed mainly so a director could “stay close to the work.” Moving status to written updates returned twelve engineer-hours per week. Nobody missed the room. They missed the guilt of being visibly busy.

Tradeoffs

Legitimacy versus velocity. RFC cycles add days. They also prevent the rewrite sprints that follow faux-consensus. Measure cost in total calendar time to a stable ship, not time to first merge. The caching layer that shipped in sprint six was fast. The rewrite in sprint eight was the actual invoice.

Relationship versus transaction. Some trust builds only in real time: conflict repair, sensitive feedback, celebration. Async-first teams budget sync for that work instead of spending it on status readouts that belong in a doc. The political claim is about defaults, not a ban on voices in a room.

Crisis override. Outages and security incidents still get bridges. Defaults are the politics. Absolutes are the exception log.

When sync-first is rational. Small colocated teams with high trust and low turnover can afford oral culture until they scale or hire remotely. The inflection point is the first hire who cannot attend 9am standup without waking at midnight. After that, “we just talk it out” is a filter, whether you meant it as one or not.

Close

Adopt async on purpose, with an explicit decision protocol, or inherit sync’s power map by accident. The team you end up with, and the architecture they ship, will reflect whichever default you chose.

The meeting will still protect the loudest person in the room unless you give everyone else a room that does not require a microphone.

— JV · Dark Heart Labs.

References

Priya Parker, The Art of Gathering (Riverhead, 2018). Meeting design encodes purpose and power, including who is centered and who is excluded by the format. ↩
Jason Fried and David Heinemeier Hansson, Remote: Office Not Required, and Basecamp’s published communication principles. Writing over meetings as an organizational stance, not a personal preference. ↩
Irving L. Janis, groupthink research (Yale, 1970s–1980s). Cohesive groups suppress dissent to preserve harmony — the psychology behind fast meeting consensus that written dissent can still catch. ↩