Is "kids" pronounced [kidz] (the same /z/ of "zebra", voiced) or [kids] (the same /s/ of "star", unvoiced) ? I have always heard and pronounced the latter, but I have just seen the former in Wiktionary. Oddly, I hear [kids] in the audio available in that wiki.

Reference: https://en.m.wiktionary.org/wiki/kids

  • Comments are not for extended discussion; this conversation has been moved to chat.
    – ColleenV
    Commented Sep 14, 2019 at 22:24

3 Answers 3


There is no contrast between /s/ and /z/ in this position (word-final, or syllable-final, after an obstruent consonant). But "kids" is typically transcribed with the phoneme /z/.* Note that in English, the typical pronunciations of the words "adze", "adds" and "ads" all sound the same.

"Voiced" obstruents in English may be devoiced, or partially devoiced

Keep in mind that the phoneme /z/ is not always pronounced as a fully voiced [z]. In word-final position, /z/ may be less phonetically voiced (or maybe even completely devoiced), but it is still distinguishable from /s/ because syllables ending in /s/ have shortened (or "clipped") vowels, but syllables ending in /z/ do not. The following words have a clipped vowel: kit, kits, kiss. The following words do not have a clipped vowel: kid, kids, fizz.

I don't know that much about the exact phonetic nature of devoicing or partial voicing. Here is a paper about the topic (which I haven't read) that I found when I did a brief search: "Variability in the implementation of voicing in American English obstruents", by Lisa Davidson.

See the following ELU questions:

The reasons "kids" is analyzed as having /dz/ and not /ds/

I think it's a bit difficult to definitively establish that kids ends in /z/, but there are a few lines of argument that could be taken.

The plural/genitive suffix contains /z/ rather than /s/ in environments where there is a contrast

In all environments where a /z/ vs. /s/ contrast is phonologically possible in English (after a vowel or after an approximant) the plural and genitive suffixes (as well as the third-person singular suffix) contain /z/. The following minimal or near-minimal pairs demonstrate this:

  • bays (bay-s) /beız/ vs. base /beıs/
  • bars (bar-s) /bɑɹz/ vs. parse /pɑɹs/
  • bells (bell-s) /bɛlz/ vs. else /ɛls/
  • tens (ten-s) /tɛnz/ vs. tense /tɛns/

Based on this, it can be argued that /z/ is the basic form of the suffixes, and /s/ (as well as /ɨz/) is a less basic form whose use requires some explanation. The use of /s/ after /p t k/, and after /f θ/ when they do not alternate with voiced /v ð/ in the plural, can be explained as an assimilation at the phonemic level to the voicelessness of the preceding segment. Since the segment /d/ is not voiceless as the phonemic level, it wouldn't be expected to cause assimilation to voicelessness, so this analysis predicts that words like kids should surface with unchanged /z/.

*/dz/ and /ds/ can contrast before a vowel, although the functional load is low

English speakers don't necessarily think of "kids" as ending in a /z/ sound. I do think that most would identify Z as a possible alternative spelling of the sound. In contexts where creative spelling is used, such as advertisements, "z" shows up as an alternative spelling of -s (e.g. "Kidz Bop") although it is not strictly limited to contexts where the suffix is actually pronounced as [z~z̥] (we also see Z spellings in "Bratz" and "Flipz", which just have [s]).

The strongest argument that I can think of in favor of analyzing kids (and so on) as ending in /dz/ is that there is a contrast word-medially between /dz/, as in sudsy, and /ds/, as in Hudson, and the word-final cluster -ds (as in suds) don't sound like /ds/ when a vowel is placed after it. The contrast between /dz/ and /ds/ is not especially robust: it wouldn't shock me to hear a native English speaker pronounce "Hudson" in a way that sounded like either "Hu/dz/on," with progressive assimilation, or like "Hu/ts/on", with regressive assimilation. That said, I wouldn't consider either of these a standard phonemic realization of the consonant sequence in the middle of the name: on the phonemic level, I would say that /ds/ is a possible sequence that contrasts with both /dz/ and /ts/.

A possible weakness of my argument in the preceding paragraph is the matter of syllabification. Some linguists think that empty onsets are disfavored in syllabification: applying this principle would result in the syllabifications su[d.z]y and Hu[d.s]on. In that case, the contrast is fairly clear (although there might still be an issue of whether the final syllable of Hudson has a different level of stress than the final syllable of sudsy). But other linguists believe that empty syllable onsets occur more extensively in English phonology, which raises the possibility that sudsy and Hudson might be syllabified differently: suds.y vs. Hud.son. If we accept the syllabification suds.y, and view syllabification in general as a possible factor in creating minimal pairs for pronunciation, then I don't think it's possible to establish with this kind of argument that the consonant after the /d/ is /z/ rather than /s/.

  • Is it possible to be partially voiced? I always thought that voicing was a binary feature. Commented Sep 14, 2019 at 21:03
  • @AlanEvangelista: Phonetically, voicing can be partial: a sound can start out voiced and then become voiceless partway through. Phonologically, English only makes a two-way distinction between voiced and unvoiced. This distinction can be modeled in phonology in various ways. "Binary feature" is terminology used in some phonological theories, but not in others.
    – sumelic
    Commented Sep 14, 2019 at 21:06
  • This is really new to me. So if "kids" is pronounced with a partly voiced [z], could I transcript it phonetically as [kɪdzs] ? That I am able to hear in the Wiktionary audio for that word! My whole problem is that I listen an odd "S" at the end of the word. Commented Sep 14, 2019 at 21:12
  • 1
    @AlanEvangelista: Yes, that is one possible phonetic transcription. Another transcription that people use is [z̥]. The ring diacritic officially means "voiceless", so technically that's a transcription of "voiceless z" rather than of "partially voiced z", but it can be interpreted as the latter. Also, I should say that partial voicing is just one option in this context. As I said, /s/ and /z/ are not contrastive here, so different speakers could use more or less voicing. As long as the vowel earlier in the syllable is not shortened, the distinction with "kits" will be maintained
    – sumelic
    Commented Sep 14, 2019 at 21:21
  • 2
    @Lambie: The IPA letter /z/ represents a consonant sound categorized as voiced according to the official definitions of the IPA. There is no rule saying that a voiced consonant must be followed by an unvoiced consonant; "s is unvoiced as it comes after an voiced consonant and is pronounced like the letter z" doesn't make sense. Since none of your comments or posts on this page have made sense to me, and you haven't explained your apparently idiosyncratic definitions of the terms "voiced" and "unvoiced", I don't know how productive it will be to continue this discussion.
    – sumelic
    Commented Sep 17, 2019 at 0:57

There is a rule to determine whether a final s is pronounced with an 's' or a 'z' sound.

If the final sound in the base of the word is voiced, we use the voiced alveolar sibilant /z/.

If the last sound in the base is an unvoiced consonant, we use /s/.

  • Could you please confirm if the audio in the Wiktionary link in my question says /z/? I need to know if the audio is bad or if I am unable to hear the /z/. Commented Sep 13, 2019 at 19:27
  • 2
    I don;t know where you live, Lambie, but right here we say the final s of cuts and of bears differently. Commented Sep 13, 2019 at 19:46
  • 3
    I agree with Michael. At least in Midwestern and Southern American English, "kids" most definitely ends with a voiced /z/ as in "zip." This is also what I hear in the clip.
    – TypeIA
    Commented Sep 13, 2019 at 19:50
  • 1
    I think you are just being contrary, Lambie. Commented Sep 13, 2019 at 19:57
  • 2
    I'm definitely not a linguist, but I'm quite certain that I and other native speakers that I know pronounce "kids" with the final sound as in "zip," and "cuts" (Michael's example) with the final sound as in "sun." No links or use of terminology can change that fact, so in my variety English, what you are saying is plainly inaccurate.
    – TypeIA
    Commented Sep 13, 2019 at 19:59

Wikipedia English plurals

For all other words (i.e. words ending in vowels or voiced non-sibilants) [like kid] the regular plural adds /z/, represented orthographically by -s:

boy boys /bɔɪz/ girl girls /ɡɜːrlz/

chair chairs /tʃɛərz/

That is the rule. the letter d in kid is a voiced non-sibilant.

kids = /ˈkɪdz/ phonetic transcription, IPA, British and American English.

The d in kids is voiced, the s is voiceless.

Spanish and Portuguese speakers have a hard time with this because there is no such ending in those languages.

Other words that end this way:

  • words
  • deeds
  • feeds
  • aphids
  • Tsk, tsk, tsk. d.v.ers
    – Lambie
    Commented Aug 22, 2020 at 20:55

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .