---
title: "Turn-taking (conversation analysis)"
type: "entity"
entity_kind: "concept"
status: "hub"
canonical_name: "Turn-taking"
aliases: ["turn-taking systematics","conversational turn-taking"]
first_seen: "2026-07-30T00:00:00.000Z"
writer_model: "claude-sonnet-5"
connects_to: ["conversation analysis","Sacks, Schegloff & Jefferson 1974","Tanya Stivers","voice AI latency budget","queueing-theory abandonment (adjacent, not the same mechanism)"]
---


The conversation-analysis concept that the timing of ordinary talk is governed by a systematic mechanism for allocating who speaks next, not by chance — first formalized by Sacks, Schegloff & Jefferson (1974), "A simplest systematics for the organization of turn-taking for conversation," *Language* 50. Entered the vault via [[claim-stivers-2009-cross-linguistic-turn-taking-gap-208ms-mean]], which instrumentally measured turn-transition timing (question-to-response gap) across ten languages and found a narrow, near-universal window — mode 0-200ms, cross-linguistic mean +208ms. That timing window is now the empirical anchor for the human-perception side of [[claim-voice-ai-sub-second-conversational-latency-budget]], which argues voice-AI systems must fit responses inside roughly the same window or read as unnatural.

## References
- [[claim-stivers-2009-cross-linguistic-turn-taking-gap-208ms-mean]] · [[claim-voice-ai-sub-second-conversational-latency-budget]]
- Captures: 10-inbox/raw/2026-07-29-what-are-the-primary-sourced-thresholds-for-human.md