Some time ago a Russian academic visitor told me that there are no "normal" Russian words with the syllable кы. He said that the only words with the syllable кы are proper names given by non-Russian ethnic groups as well as similarly exceptional words such as rarely used borrowed words for some unique non-Russian stuff. He added that although Russians can sometimes say "кыш" when they accidentally encounter a bear in a forest, the word "кыш" is a single three-letter syllable and thus does not count as containing the syllable "кы."

After the conversation, I performed research in Google and got really puzzled: I found neither "normal" Russian words with the syllable "кы" nor explanations for that.

I cannot see any obvious reason for the syllable "кы" being nonexistent. The letters к and ы are not rarely used at all and are the 11th and 17th most freuqently used Russian letters, respectively, with the Russian alphabet containing 33 letters in total. The letter ы is contained even in most basic pronouns - "ты," "мы," and "вы." I can easily find many words with two-letter syllables ending with ы: дыня, пузыри, бычара, лыжи, ныкать, пырять, сычиный, and so on. Although there are apparently no words with жы and шы, the´latter two syllables are simply prohibited by the well-known orthography rule that prescribes writing them as жи and ши, and I am not aware of any similar orthography rule prohibiting кы.

I do not see any phonetics-related reason either, as the syllable кы is pronounced without any difficulties. At least Turkic people have no difficulties pronouncing it, as can be evidenced by their languages. For example, a girl in Turkish is kız (кыз), which is one of the most frequently used Turkish words. The Turkish word for "red" is "kırmızı" (кырмызы). Some Russian geographic names of Turkic origin are written with the syllable кы, e.g., Кызыл, which is a city in Russia.

At some point I thought that the reason for the nonexistence of the syllable кы in Russian might be that this syllable sounds very similar to some frequently used Russian syllable and is replaced by it to avoid phonetic nuances, but later I dismissed this idea. Perhaps the closest syllable is ки, but it sounds differently enough to be easily distinguished from кы. At least the Turks actively use both (кы) and ki (ки) in their language. For example, "two" in Turkish is "iki" (ики), and I gave a couple of examples with in the paragraph above (kız, kırmızı). After all, if the Russians found it hard to distingsuish between кы and ки, they would probably say and write "киш" instead of "кыш."

My question is this: Are there any "normal" Russian words with the syllable кы, and what is the mysterious reason for their nonexistence or being very few?

I guess that in order to make my question well-defined, I have to define a "normal" word, but it is pretty hard, so I have to appeal to your common sense. A "normal" word is not highly special and, in particular, is not extremely rarely used, is not a proper name of non-Russian origin or its derivative, is not a highly specialized jargon or slang word, is not a rare borrowed word for some unique foreign stuff, etc. A "normal" word is contained in standard dictionaries, is recognized by linguists as a part of the Russian language, is known to almost every Russian, etc. I humbly hope that you get the idea what is a "normal" word. I am especially interested to see Russian words of Slavic origin with the syllable "кы."

  • Comments are not for extended discussion; this conversation has been moved to chat.
    – Quassnoi
    Jun 19, 2019 at 14:20

4 Answers 4


After a couple of days of thinking over it all I have finally decided to put down the whole story of кы, гы, хы the way I understand it. My story will begin with what the answer by Quassnoi begins, with some additions, but then it will continue, from the point when Quassnoi stopped. It will be long, it will require close reading, and it will be very informative and thought-provoking.

From now on, I will use the following notation:

Actual spelling  |  IPA
кы               |  kɨ
гы               |  gɨ
хы               |  xɨ

ки               |  kʲi
ги               |  gʲi
хи               |  xʲi

Phonemes are in slanted brackets: /k, g, x/
Sounds   are in square brackets:  [k, g, x]
Letters  are in angular brackets: <к, г, х>

The earliest attestation of the Slavic languages in writing knew only <кы, гы, хы>, since the then Slavic dialects did not have the distinction between palatalized and non-palatalized /k, g, x/, they had only non-palatalized variants, so no <и> could follow them, since <и> could follow only palatalized consonants, moreover, no front vowel at all could follow them, since all the front vowels caused the palatalization of the preceding consonant. Funny to note, but in the Old Russian, such modern Russian words as Киев and русский were spelled as Кыѥвъ and роуськый, that is with <кы>.

Then, in the 9th century, there appeared Slavic translations of the Bible which contained Greek words which had no Slavic equivalents, and which had front vowels following /k, g, x/ which, naturally, made them sound palatalized [kʲ, gʲ, xʲ]:

кѵпарисъ [kʲ] "cypress", from Greek κυπάρισσος

ангєлъ [gʲ] "angel", from Greek ἄγγελος

хитонъ [xʲ] "tunic, clothes", from Greek χιτών

This change was so revolutionary that in the first Slavic alphabet, Glagolitic, even a special letter for [gʲ] was introduced:

Glagolitic letter G'ervi

Starting with this point in time, we can single out three stages of the co-existence of the velar sounds [k, g, x] with the vowel sounds [i, ɨ]:

  1. First, only the combinations [kɨ, gɨ, xɨ] existed, there were neither combinations [kʲi, gʲi, xʲi], nor did sounds [kʲ, gʲ, xʲ] existed at all.
  2. Then, there still existed the combinations [kɨ, gɨ, xɨ], but there were also the combinations [kʲi, gʲi, xʲi], and also the sounds [kʲ, gʲ, xʲ] could appear before other front vowels, mostly in borrowed words, e.g. кедръ, евангєліє, etc.
  3. Then, all the combinations [kɨ, gɨ, xɨ] changed into [kʲi, gʲi, xʲi], and the combinations [kɨ, gɨ, xɨ] (<кы, гы, хы>) ceased to exist.

On the historic time-line, the edge between stages 2. and 3. coincides with the fundamental structural change in the phonology of Old Russian which was caused by the so called fall of the extra-short (reduced) vowels ŭ and ĭ (Ru. падение редуцированных ъ и ь), a dramatic phonological loss of the two vowels which began in the times of the Common Slavic language and ended in the 11th-12th centuries for Old Russian, with some of its northern dialects having the process finished as late as the 13th century. The two extra-short vowels "fell" in all the Slavic languages about the same time, but with different results which nevertheless changed the phonological inventories of all the Slavic languages and also changed the structure of the syllable in all of them. In Russian, the most obvious result of the fall is the seemingly inexplicable loss of the root vowel in the Nom. vs. Gen. case forms of some nouns, like день – дня or сон – сна, but this otherwise fascinating story is no way in the focus our present discussion.

During all these three stages, the sounds [k, g x] and [kʲ, gʲ, xʲ] were positionally conditioned, i.e. it was the following vowel that defined which of the variants, palatalized or non-palatalized, is used in the given word. That was because the two series of sounds in fact realized just the single series of the velar consonantal phonemes /k, g, x/ that had no such a differentiating feature as palatalized vs. non-palatalized.

As for the sounds [i] and [ɨ], their story was different. Earlier, they used to be the realizations of the two different phonemes /i/ and /ɨ/, but at the edge between stages 2. and 3. these two phonemes merged into one, so that the sounds [i] and [ɨ] began to represent a single phoneme, and the difference between them became positionally conditioned.

Phonetically, this difference was realized so that [i] was possible only after palatalized sounds, and [ɨ] only after non-palatalized sounds.

But! Now, the choice between [i] and [ɨ] became a phonological one. In the phonological sense the choice between [i] and [ɨ] began to be determined not by the degree of palatalization of the consonant sound after which the [i] or [ɨ] appears, but by the characteristics of the consonant phoneme after which [i] or [ɨ] appears – after a palatalized or non-palatalized one. After a "palatalized" phoneme there's [i], after a "non-palatalized" phoneme there's [ɨ].

And also, there is the position at the beginning of the word, and also the position after phonemes lacking the differentiating feature "palatalized vs. non-palatalized", that is after /ʃ, ʒ, tʃ, ts, j/ and also /k, g, x/. So here we again have three positions which the phoneme /i/ can take:

  1. After a "palatalized" consonantal phoneme.
  2. After a "non-palatalized" consonantal phoneme.
  3. After a unit lacking both of those features.

In position 1., /i/ could be realized only as [i]. In position 2., /i/ could be realized only as [ɨ]. And only in position 3. /i/ could be realized as either [i] or [ɨ]. In most cases, in position 3., the phoneme /i/ was realized as [i], and only after /k, g, x/ it was possible for /i/ to be realized as [ɨ], too. So here phonology won over phonetics, and the only outcome of that collision was the choice of only [i] after /k, g, x/, which, in its turn, made /k, g, x/ before [i] turn into [kʲ, gʲ, xʲ].

That is it, the end. You wanted to have it, here it is. A story of the life of some of the utmostly abstract entities humans ever invented, the phonemes. No actual proofs, just logic and common sense.

Source: Касаткин Л.Л., «Современная русская диалектная и литературная фонетика как источник для истории русского языка», М., Наука, 1999

The pages about кы are 194–197, but pages 192–193 can also be of interest for you.

  • Comments are not for extended discussion; this conversation has been moved to chat.
    – Quassnoi
    Jun 19, 2019 at 14:19

The difference between и and ы after consonants in Russian is not phonematic, and neither is the difference between palatalized and non-palatalized к, г, х.

You can probably find some degenerate minimal pairs for the latter, like тот кот / то ткёт, or aforementioned киш / кыш for that matter, but that's about it.

Old Russian did have кы, гы, хы and didn't have ки, ги, хи.

After the fall of the reduced vowels, Russian phonology had a major paradigm shift: vowel quality (front or back) stopped being a distinctive feature, and this role moved to the preceding consonant.

In other words, pairs like мал / мял are pronounced differently not because they have a different vowel, but because they have different consonants before it. А and я are allophones of the same phoneme, and м and мь are two different phonemes.

So now that front and back vowels are not distinct phonemes anymore, Russian had to settle on a way to pronounce к, г, х, which didn't have a phonematic distinction between soft and hard varieties.

Unlike ш, щ, ч, ц (which are either always soft or always hard), к, г, х became soft before и and е and hard before а, о, у.

I don't know why exactly that happened (vowels shaping the rendition of к, г, х and not the other way around), but for some reason it did.

  • Comments are not for extended discussion; this conversation has been moved to chat.
    – Quassnoi
    Jun 19, 2019 at 14:20

You will also hardly find гы, хы syllables. The common quality of these phonemes is that к, г, х are заднеязычные согласные and they followed the same patterns of phonological changes as they (the changes) happened in Old Russian language.

Apparently, about a thousand years ago the situation was quite different:

The full PDF is here

Unfortunately, some formatting has been lost, the 'ять' letter is shown as ' h ' with spaces around it, etc. This can hopefully be fixed but I have no time to investigate at the moment. It is recommended to read the original PDF document instead.

    2. История заднеязычных согласных. 
    а) Заднеязычные согласные долго оставались в стороне от процесса фор-
мирования корреляции по твердости-мягкости. Древнейшим фонетическим из-
менением в этом направлении было изменение сочетаний [кы, гы, хы] в [ки, ги,
хи]. Сочетания заднеязычных не только с [ы], но и с [и] стали возможны после
утраты редуцированных. В северных говорах сочетания ки, ги, хи были воз-
можны и в древнерусский период. В новгородских берестяных грамотах написания
с кы, гы, хы встречаются только в церковных текстах [пакы, но: не моги].
Первые примеры замен кы, гы, хы на ки, ги, хи обнаруживаются на юге в XII в.:
великии (Юрьев.ев. 1120 г.); никии, секира (Добр.ев. 1164 г.); на западе с XIII в.:
княгини, лихии (Гр. 1229 г.); на северо-востоке с XIV в. (Переяслав.ев. 1354 г.,
Москов.ев. 1358 г., ЛЛ 1377 г.).
   Процесс смягчения шел с юга и постепенно стал общерусским. В нем в
известной степени отразилась общая тенденция консонантизма к выравниванию.
К этому времени все согласные кроме к, г, х в результате вторичного смягчения
имели соотносительные пары, и только заднеязычные оставались вне корреляции 
по твердости-мягкости. Кроме того, в древнерусском языке были представлены
заимствования типа: ангел, кипарис, евангелие, где имели место сочетания
заднеязычных с гласными переднего ряда. Это обстоятельство также могло оказать
определенное влияние на этот процесс.
    Как известно, заднеязычные согласные по закону слогового сингармонизма
не могли находиться перед гласными переднего ряда. В этом положении они уже
в праславянский период изменялись либо в мягкие шипящие (I палат.), либо 
в мягкие свистящие (II палат.). Аналогичное изменение происходило и в случаях,
когда заднеязычные испытывали воздействие со стороны *j (j-овая палат.).
    Если результаты I палатализации в современном русском языке сохранились,
то рефлексы II палатализации были в основном устранены. Они представлены
единичными случаями в корнях слов: цена, целый, цеп, цедить, зело, серый,
седой и др. В древнерусском языке результаты II палатализации отражены:
а) в корнях; б) в формах словоизменения (в существительных – друзи, послуси,
ученици; на ноз h , въ руц h ; в глаголах – помози, рьц h те).
    Почему произошла утрата результатов II палатализации? Предпосылки 
к устранению чередований к/ц, г/з, х/с были заложены в природе восточнославянских
    II палатализация отсутствовала в некоторых древнерусских говорах
(древненовгородских и псковских). Здесь заднеязычные могли находиться в одном 
слоге с гласными переднего ряда: къ Лук h , у влдк h , по велик h д(ни) и др.
Современные диалектные материалы показывают, что следов II палатализации нет и
в ряде корневых морфем, напр.диал.: кедилка, кеп, квет, квел. На основании
приведенных данных В.В. Колесов делает следующее заключение:
«…у восточных славян (может быть, не на всей территории) возникли условия,
препятствовавшие последовательному завершению II палатализации заднеязычных».
    Таким образом, по крайней мере в отдельных говорах, появляется возможность
сочетания заднеязычных согласных с гласными и, [ě] дифтонгического происхождения. 
    С другой стороны, сохранение заднеязычных при склонении обеспечивало единый
облик основы во всей парадигме. А в системе любого языка заложена тенденция к
выравниванию основы в пределах парадигмы. Таким образом, эта диалектная особенность
оказалась созвучной общерусской тенденции к выравниванию основы. Поэтому чередования
заднеязычных со свистящими полностью исчезли в грамматических формах. Это произошло
к XIV - XV вв.

There are no inherited native Slavic words in modern Russian with кы because the sound changes prevent this. But there are also no native Slavic words with ф or starting with a-.

Today one can probably expect words with кы among borrowings from Turkic languages, including geographic names or among onomatopoeias.

For instance, "кыпчак" but it can be alternatively spelled "кипчак". Notice, the word has syllable кып rather than кы but I assume the original poster had in mind sequence ки rather than exactly a syllable.

  • Comments are not for extended discussion; this conversation has been moved to chat.
    – Quassnoi
    Jun 20, 2019 at 13:15

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.