How did English end up with a voiced "z" at the end of words, for example in "is", "was", "those"? Does this phenomenon exist in any Indio-European language other than English?

Bonus question: When native French speakers attempt to speak German, do they have a tendency to voice the ending "s"? If they do, this would explain this feature in English.

  • When the letter ‘s’ is after a vowel, another ‘s’, or a voiced consonant, it is pronounced as a /z/ sound. e.g., logs, tubes, beds, moves, clothes, was, becomes, he’s, passes. So. Mostly, it's plural. As opposed to unvoiced. So, the quesiton is: Why is /z/ a plural in English?
    – Lambie
    Commented Jun 10 at 19:12
  • antidote.info/en/blog/reports/… For both voiced and unvoiced ones, plus the irregulars that are still around.
    – Lambie
    Commented Jun 10 at 19:37

1 Answer 1


Yes, it exists in other Indo-European languages.

For example, in the French word française.

It also exists in some other Romance languages like Romanian, in Serbo-Croatian and Ukrainian, in Armenian and Persian.

It may be a bit odd for a Germanic language, though it is found in Yiddish, e.g. frukhtloz,frukhtayz… (cf. modern formal German fruchtlos. Fruchteis). Basically in continental West German, it ended up that in Southern German it’s always s, in Yiddish it’s always z, and in the modern Northern pronunciation of standard German it is z but devoiced to s if not followed by a vowel.

It is not found in Spanish, Portuguese, standard Italian or Russian or some other Slavic languages, even though the words are spellt with -z.

Not sure about Greek, Baltic or Indic languages. My sense is that Greek or Indic languages have it in loans like “jazz” or “khuda hafez”, but not sure if it occurs in core native words and morphological endings like it does in English.

  • 3
    @MWB Unlikely, since many of the words where it’s found in English now are inherited English words, not French loans. Intervocalic fricatives were voiced in quite a few contexts in English (that’s also why we have noun/verb or singular/plural pairs like breath/breathe, knife/knives, etc.). Final /z/ is also reconstructed for Proto-Germanic, but was later lost outside English, either by (a) developing into /r/ (in Norse), or (b) becoming /s/ through final devoicing (West Germanic except English). Commented Jun 3 at 23:35
  • 1
    @JanusBahsJacquet / Adam: Do you happen to know if Dutch, Frisian or Danish have this feature?
    – MWB
    Commented Jun 4 at 6:01
  • 1
    @MWB Dutch – and I believe also Frisian – has final devoicing, so /z/ cannot appear at the end of words (e.g., huis ‘house’, pl. huizen), and Danish has no voiced obstruents (and thus no /z/) at all. Commented Jun 4 at 8:17
  • 1
    @Adam True, but if Wikipedia is anything to go by, that is a later reversal in Eastern Yiddish, possibly by Slavic influence (cf. gut tak in the oldest known Yiddish text, written with qof instead of gimel as gut tag). I don’t know nearly enough about Yiddish to know if the devoicing is present in words where the final obstruent was never part of a final/non-final pair and thus would have no analogical counterpart to be re-voiced by. Commented Jun 4 at 8:22
  • 2
    @MWB French schwa is unstable – in normal speech, it’s usually absent in word-final position, but in careful speech, it’s often present. The recording on Wiktionary definitely says [fʁɑ̃sɛːzə], pronouncing it, but [fʁɑ̃sɛːz] is equally possible, just like both [sεtə] and [sεt] are possible for cette ‘this (f.)’ or even sept ‘seven’. It’s generally, I believe, held to be a non-phonemic, epenthetic vowel in word-final position (cf. this paper). Commented Jun 8 at 17:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.