Backchannelling

Backchannelling is the listener's production of short verbal or non-verbal signals ("mm-hmm", "yeah", "right", "really?", nods, raised eyebrows) that show engagement during another speaker's turn without claiming the floor. The term was coined by Victor Yngve (1970), who observed that "both the person who has the turn and his partner are simultaneously engaged in both speaking and listening" because a secondary channel, the back channel, carries these short listener messages while the primary channel carries the main speech.

Two-channel model

Yngve's key insight was that conversation is not strict alternation. A single utterance involves two simultaneous channels:

• Primary channel: the current speaker's developing turn
• Back channel: the listener's running commentary of minimal responses that do not take the turn

This distinguishes backchannels from genuine turn-taking moves. "Yeah, I totally agree" with upward intonation claims the floor; "yeah" with a fall-rising tone while the other person keeps speaking does not.

Functions

Maynard (1986, 1989) identifies six common functions, which remain the standard reference taxonomy:

1. Continuer: keep going, I'm following ("mm-hmm", "right", "uh-huh")
2. Signal of understanding: I've grasped what you said ("I see", "OK")
3. Agreement: I concur ("exactly", "yeah")
4. Emotional support / alignment: empathic uptake ("oh no", "poor you")
5. Strong emotional response: surprise, shock, delight ("really?", "wow", "no way")
6. Minor clarification request: a small repair without derailing the turn ("who?", "sorry?")

Non-verbal backchannels (nods, eye contact, laughter, facial expressions) run in parallel and can perform any of these functions on their own or reinforce the verbal signal.

Prosody matters

The same token can signal very different things depending on intonation:

• Fall-rising "yeah↓↑" or "mm↓↑": classic continuer, warm and encouraging
• Flat or slightly falling "yeah", "right", "OK": understanding without strong affect
• Rising "really?", "did you?": surprise or an invitation to elaborate
• Repeated flat "yeah yeah": can read as impatient, dismissive, or even hostile

Teaching backchannels without teaching their prosody risks learners producing the right words with the wrong effect.

Cross-cultural variation

Backchannelling norms differ sharply across languages and speech communities. Japanese aizuchi tend to be more frequent and more overlapping with the speaker's turn than their American English equivalents; American speakers often prefer post-turn rather than mid-turn feedback. The mismatch is a well-documented source of interactional discomfort: the American perceives Japanese backchannels as interruptions, the Japanese perceives the American's larger, later responses as cold or evaluative. ELF research (Wolfartsberger and others) finds that speakers in lingua-franca interactions calibrate backchannel frequency to the interlocutor, not to any single "native" norm.

The practical consequence for ELT: there is no universal "correct" backchannel rate, but L2 learners who under-backchannel in English can come across as bored, incomprehending, or rude. The absence of signal is a signal.

Why learners struggle

• Coursebook gap: transactional dialogues (ordering coffee, asking directions) rarely model backchannels, so learners have no input to notice.
• Invisible to fluent speakers: teachers manage backchannels automatically and may not think to teach them.
• Perceived as "not real language": fillers and listener tokens are often dismissed as peripheral.
• Prosody is hard: even learners who know the tokens may deliver them with flat L1 intonation and misfire on function.

Teaching sequence

A consciousness-raising → controlled → freer progression works well:

1. Notice: play authentic conversation (podcast clip, scripted but natural dialogue) and have learners count or mark listener responses; ask which tokens appear and where.
2. Map function to prosody: match tokens to Maynard's six functions; use pitch arrows or multiple-choice prosody tasks to link tone to meaning.
3. Drill intonation without words: hum the pitch pattern before reintroducing the token.
4. Controlled anecdote practice: teacher tells a short story; one learner must insert a continuer at each pause; partner listens for appropriateness.
5. Mingling anecdotes: learners tell personalised stories in pairs or groups; listeners are graded on backchannel variety, timing, and prosody; recordings reviewed.
6. Contrast with L1: brief discussion of how often and when learners backchannel in their first language, surfacing the calibration learners need to make.

In lesson planning

Backchannels belong to the interactive side of listening that Thornbury flags as a subaim beyond simply decoding the stream of speech. A listening lesson based only on audio recordings cannot practise them; they require live, co-present, turn-holding partners. Plan for at least some classes where the post-listening stage is a productive follow-up that forces learners to listen to each other, with backchannelling as a visible target.

References

• Yngve, V. H. (1970). On getting a word in edgewise. Papers from the 6th Regional Meeting, Chicago Linguistic Society, 567–578.
• Maynard, S. K. (1986). On back-channel behavior in Japanese and English casual conversation. Linguistics, 24(6), 1079–1108.
• Maynard, S. K. (1989). Japanese Conversation: Self-Contextualization through Structure and Interactional Management. Ablex.
• Wolfartsberger, A. (2009). Conflict or cooperation: The use of backchannelling in ELF negotiations. ScienceDirect.
• Cutrone, P. (2005). A case study examining backchannels in conversations between Japanese-British dyads. Multilingua, 24(3), 237–274.