As far as I understand, "ч" is always soft. The letter "ь" usually denotes that the preceding letter is soft.

These two facts together seem to imply that the combination "чь" (just like "щь") would/should always be redundant, and thus never used or replaced by "ч".

However, one sees this letter combination in the pronouns "чьё/чья/чьи" -- I have not seen it anywhere else in the Russian language although I do not presume to have a wide vocabulary.

Is the reason why this letter combination exists in these three words historical? Or does it actually affect the pronunciation of the words?

(Say compared to "чё/ча/чи" -- of course because of the spelling rule "чя" is illegal but such a combination would be redundant anyway since the vowel following "ч" is always soft.)

Sort of related: "чё?" or "чо?" - which is the correct spelling?

6 Answers 6


Not redundant, does affect. The soft mark is the reason чьи is pronounced [tɕji] and not [tɕi].

Basically, ь between a consonant and a vowel is pronounced [j] — an actual separate iotic consonant, as opposed to just the preceding one getting palatalised: пьяный ['pʲja-] vs. пятна ['pʲa-]. The principle works with the ь-unaffected sibilants too.

Now the soft mark supposedly isn't itself "pronounced" but rather "resets" the so-called modifier letters (е/ё, ю, я), causing them to iotate instead of palatalise, but with a word like чьи (or, say, бульон), the [j] can only really be regarded as the pronunciation of ь itself.

(Of course it's a little more complicated from a historical viewpoint: in obsolete and dialectal pronunciations, и can also be a "modifier letter" with a [ji] pronunciation in the same positions as where я would be [ja], etc., while ьо being pronounced [jo] is more recent and more random.)

  • I would add that "ьо" is restricted to several loanwords, mostly in ьон corresponding to French ion or illon (лосьон, бульон, почтальон, etc.) but also sometimes word-finally for other Romance languages (каудильо, rarely ниньо). It is not productive (id est. words that are loaned nowadays usually don't take ьо).
    – Viridianus
    Aug 31, 2016 at 12:33
  • @Viridianus I'm not at all sure it's not productive. If the Spanish word alloza, "green almond", were loaned tomorrow, what would its spelling be? "Альёса", rather than "альоса"? I don't think so. Aug 31, 2016 at 12:44
  • 1
    @Nikolay_Ershov I would be almost certain it would be альёса (leaving apart the incredibility of making loanword for изумрудный).
    – Viridianus
    Sep 1, 2016 at 23:53
  • 1
    I guess my confusion was that I don't really know how to distinguish clearly between a "y"/[j] sound and palatalization [ ' ]. I understand conceptually and physically why they are different, how I should move my mouth and tongue and such -- when I try to actually make them sound different I don't hear much of a change though. But just because I can't pronounce the difference/need to practice pronouncing the difference, doesn't mean that it isn't there or that I should ignore it. Thank you for the clear explanation re-introducing this mental distinction -- I appreciate it. Sep 26, 2016 at 15:34

These vowels are iotized after ь but not after consonants. So the soft sign serves to indicate iotation.

чья = [ч'й'а]

чьё = [ч'й'о]

By contrast,

чё = [ч'о]


We use "ь" in чья,чьи,чьё (pronounced differently from чё) to mark the sound [j] like in "й", so it's not redundant (чё--чьё).Examples : мужичьё,дурачьё,ничья, but чёлка,плечо,влечёт. Чё is spelt both ways(чо,чё) By the way,we also use ь with other consonants for the same reason пьяный--пятница.


You probably know the difference between и and й. Think like е/ь relates the same way и/й does. Try to tell "чеё", feel the difference of "чё". Then make "е" as short as possible. The same is right (in most cases) for ы/ъ.


It has nothing to do with letter "ч". "ьё" always produces "йо", "ья" - "йа" and "ьи" - "йи". For example, "лукоморье" -> "лукамор'йэ", "ульи" -> "ул'йи" etc.

For "ьё" or "ья" one may explain it as "opening a syllable", so it's pronounced as if there were two separate words, e.g. "спросонья" -> "спросонь + я". But for "ьи" it doesn't work that way and the reason why "ьи" is always pronounced as "йи" now is purely historical.


The [j] element in чья, чьё, чьи also makes sense from a structural standpoint - it's the same [j] that's at the end of чей. The -е- in чей/чья etc. can be present or absent because of a historical sound change.

  • 1
    "е" is not actually absent, it still there in form of "short e" - "ь". Too bad that nowadays the historical roots of ь and ъ are forgotten and the letters are studied as if they were some incomprehensible and transcendental "signs". Sep 20, 2016 at 9:12
  • @yaapelsinko Well, actually the linguistic theory of language can and must be synchronic (i. e. not explain facts of a state of the language using its history). However, as Zaliznyak has shown in "Russian Nominal Declension", the е/zero alternation can (and must) be predicted in synchronic model.
    – Viridianus
    Sep 29, 2016 at 20:46

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.