Voicing refers to whether the vocal folds (vocal cords) in the larynx vibrate during the production of a speech sound. When they vibrate, the sound is voiced; when they do not, the sound is voiceless. This binary distinction is one of the three defining parameters for consonant classification, alongside place and manner of articulation (Roach, 2009; Ladefoged & Johnson, 2015).
English consonants often come in voiceless/voiced pairs that share the same place and manner but differ only in voicing:
| Voiceless | Voiced | Place | Manner | Examples |
|---|---|---|---|---|
| /p/ | /b/ | Bilabial | Plosive | pat/bat |
| /t/ | /d/ | Alveolar | Plosive | ten/den |
| /k/ | /ɡ/ | Velar | Plosive | coat/goat |
| /f/ | /v/ | Labiodental | Fricative | fan/van |
| /θ/ | /ð/ | Dental | Fricative | thin/then |
| /s/ | /z/ | Alveolar | Fricative | sip/zip |
| /ʃ/ | /ʒ/ | Post-alveolar | Fricative | ship/measure |
| /tʃ/ | /dʒ/ | Post-alveolar | Affricate | chin/gin |
Some consonants are inherently voiced with no voiceless partner in English: the nasals /m n ŋ/, the lateral /l/, and the approximants /r w j/. The glottal fricative /h/ is voiceless with no voiced partner.
The voicing distinction in plosives is more accurately described through voice onset time: the interval between the release of the plosive closure and the onset of vocal fold vibration (Lisker & Abramson, 1964). In English:
Many languages (French, Spanish, Arabic) use pre-voicing (negative VOT) for their "voiced" plosives and short-lag VOT for "voiceless" ones. This means a French speaker's /b/ has earlier voicing than an English /b/, and their /p/ sounds like English /b/ to English ears because it lacks aspiration. The distinction is not simply "vibration vs. no vibration" but a continuum of timing.
Voicing plays a grammatical role in English:
High functional load. Voicing contrasts distinguish hundreds of minimal pairs in English. Errors in voicing frequently cause misunderstanding: price/prize, half/have, back/bag. Voicing has one of the highest functional loads of any phonological feature in English.
Final devoicing. Speakers of German, Dutch, Russian, Polish, Turkish, Vietnamese, and many other languages systematically devoice final consonants in their L1. They transfer this to English, producing dog as [dɒk], bed as [bet], five as [faɪf]. This is one of the most widespread and persistent pronunciation errors across L1 groups. Teachers should prioritize final voicing contrasts.
The aspiration trap. As noted under VOT, many learners' "voicing problem" with initial plosives is actually an aspiration problem. A Korean or Thai speaker may voice initial plosives correctly but fail to aspirate voiceless ones, causing /p t k/ to be heard as /b d ɡ/. The teaching fix is not "add vibration" but "add aspiration": a puff of air. Holding a piece of paper in front of the mouth to show the air burst is a classic technique (Kelly, 2000).
Self-diagnosis technique. Students can feel voicing by placing fingers on the throat while producing sustained sounds: /sssss/ (no vibration) vs /zzzzz/ (vibration). This kinesthetic awareness is the foundation for all voicing work.