Time when Russian had no soft consonants

C.S.Hy

Senior Member
Mandarin Chinese
Was there a time when Russian had no SOFT [edited, originally hard. No more connotation to similar edits below] consonants?:D:p

(If you known the answer and do not want to elaborate on it, just say yes or no, I'll appreciate it all the same).

Was there a time when Russian, or its predecessors, had no soft consonants?

Hard, or palatalized, consonants are (or rather the phonological opposition between hard and soft consonants is) regarded as a feature of Russian (or slavic languages?).

But some fragmentary clues seem to be suggesting that there might be a time when Russian, or its predecessors, had no systematic soft consonants: the letter Ь had reputedly had its own sound in history and today in certain circumstances it still produces an extra sound /ɪ/, for example, when coming before an iotated vowel.

Hence it occurs to me that the soft consonants might have derived from their corresponding hard ones under particular conditions.

Is it true or false? Maybe this is a basic and known-to-all matter for students of historical Russian phonetics, but not for me, an utterly beginner of Russian.

What I can think of are: are there other evidences that prove, or disprove (in terms of the literature known so far), the hypothesis? Is there any modern slavic language that has no the opposition between hard and soft consonants? Had it ever had it?

To know yes or no about this question is unlikely to help a lot in Russian learning, but it is still tickling me.
 
Last edited:
  • Awwal12

    Senior Member
    Russian
    Was there a Time when Russian, or its predecessors, had no hard consonants?
    Hard, or palatalized, consonants are regarded as a feature of Russian (or slavic languages?).
    You mean soft, I suppose. Most hard consonants are inherited from the Proto-Slavic state and are mostly intact in Russian (which preserves their original velarization; the labialization is lost, but some scholars claim it never was there in the first place).

    There is still no full consensus about the history of soft (palatalized/palatal) consonants. The old mainstream theory is that Proto-Slavic had a set of soft consonants, which increased over time as the historical palatalizations occured ("ш", "ж" and "ц" are all etymologically soft); in Old Russian around the XI century ALL consonant before front vowels became soft (and Ukrainian has subsequently lost the results due to de-palalatalization); then the fall of the yers turned these soft sounds into actual soft phonemes. Some of the Old Russian soft consonants became hard later, by whole phonemes (e.g. /ш/) or in certain positions (e.g. "ем" < Old Rus. "ѣмь"). In turn, Old Russian velars became soft before /и/ (< /*ы/) in Russian, thus turning Old Russian "кы", "гы", "хы" into formerly impossible "ки", "ги", "хи".

    All in all, soft consonants are a common Slavic feature, even if some modern Slavic languages have lost the opposition between soft and hard consonants entirely (like Slovene), while some spreaded it to almost all kind of consonants (like Russian did).
     
    Last edited:

    Nikined

    Senior Member
    Russian
    All in all, soft consonants are a common Slavic feature, even if some modern Slavic languages have lost the opposition between soft and hard consonants entirely (like Slovene), while some spreaded it to almost all kind of consonants (like Russian did).
    Вы о произношении? В словенском есть же несколько пар твердых и мягких букв
     

    Awwal12

    Senior Member
    Russian
    Вы о произношении? В словенском есть же несколько пар твердых и мягких букв
    Откуда там пары букв?.. Вы со словацким ничего не перепутали? :confused: В словенском нет ни одной пары фонем, которые отличались бы только мягкостью.
     

    Nikined

    Senior Member
    Russian
    Откуда там пары букв?.. Вы со словацким ничего не перепутали? :confused: В словенском нет ни одной пары фонем, которые отличались бы только мягкостью.
    l и lj, n и nj, Ć (на концах фамилий) и Č
     

    Assiduous student

    Banned
    English - UK
    l и lj, n и nj, Ć (на концах фамилий) и Č
    Мало что знаю об этом, но вы уверены в том что есть мягкие л и н в словенском? /lj/ и /nj/ не мякие согласные, а л+й, н+й. Если они мягкие, возможно говорить что у нас в английском тоже есть (в million и onion).
     

    Nikined

    Senior Member
    Russian
    Мало что знаю об этом, но вы уверены в том что есть мягкие л и н в словенском? /lj/ и /nj/ не мякие согласные, а л+й, н+й. Если они мягкие, возможно говорить что у нас в английском тоже есть (в million и onion).
    Да. В югославских языках lj и nj - это именно мягкие согласные, в кириллическом алфавите есть буквы љ и њ, которые за неимением подходящих букв перевели на латинский как lj и nj.
     

    Awwal12

    Senior Member
    Russian
    l и lj, n и nj, Ć (на концах фамилий) и Č
    Unlike in BSC, "lj" and "nj" in Slovene represent simply /lj/ and /nj/ combinations. "Ć" is an orthographic loan from BCS which doesn't represent a separate phoneme either and apparently exists only in BCS surnames (cf. "Vida Tomšič").

    Да. В югославских языках lj и nj - это именно мягкие согласные, в кириллическом алфавите есть буквы љ и њ, которые за неимением подходящих букв перевели на латинский как lj и nj.
    Не существует "югославских языков" как лингвистического таксона. Если боснийский, сербский и хорватский представляют собой в сущности один полицентричный язык, то македонский и словенский тут никаким боком, и если они имеют алфавиты, приближенные к сербохорватской латинице и кириллице соответственно (что понятно с учётом исторических реалий), из этого никак не следует, что они будут читаться так же. Это всё равно что читать чувашский или татарский кириллический текст по правилам русского, апеллируя к "российским языкам".
     

    Nikined

    Senior Member
    Russian
    Unlike in BSC, "lj" and "nj" in Slovene represent simply /lj/ and /nj/ combinations. "Ć" is an orthographic loan from BCS which doesn't represent a separate phoneme either and apparently exists only in BCS surnames (cf. "Vida Tomšič").
    Получается, столица страны, которая везде, где я смотрел, пишется и произносится как "любляна", тоже "не того" происхождения?
    Не существует "югославских языков" как лингвистического таксона. Если боснийский, сербский и хорватский представляют собой в сущности один полицентричный язык, то македонский и словенский тут никаким боком, и если они имеют алфавиты, приближенные к сербохорватской латинице и кириллице соответственно (что понятно с учётом исторических реалий), из этого никак не следует, что они будут читаться так же. Это всё равно что читать чувашский или татарский кириллический текст по правилам русского, апеллируя к "российским языкам".
    Ну с чувашским это совсем загнули. Словенский к одной группе и "подгруппе" принадлежит хотя бы, а не является полностью чуждым языком.
     

    Panceltic

    Senior Member
    Slovenščina
    "Любляна" на литературном словенском произносится "Лъюблъяна". В разговорном языке мы говорим "Лублана" - это касается и других слов, как например pole (polje), popustliv (popustljiv), pluvati (pljuvati) ...

    (На сербском имя города пишется "Љубљана", и произношение как на русском.)
     

    ahvalj

    Senior Member
    The modern Russian soft consonants go back to two earlier kinds of consonants: palatal and palatalized. Palatal consonants (phonetically, those with a stronger palatal articulation, like the Serbo-Croatian nj and lj) arose in all Slavic languages as a result of iotation (that is, the change of consonant + yod) or from velars in vicinity of front vowels. Palatalized consonants (phonetically, those with lighter palatal articulation) arose in most Slavic languages before front vowels. In Slovene and Serbo-Croatian, there seem to be no detectable traces of this second type: for example the Russian конь (with the former palatal n, from *nj) and день (from the former palatalized n, with n softened before the short *i>ь) correspond to коњ but дан.

    In East Slavic, in the last centuries of the first millennium there was a change *el>olo and *il>ъl in the closed syllables (e. g. *telktī>толочи>толочь and *tilkān>тълкѹ>толку), before which the consonants remained hard. Several centuries later, when a change e>o took place in several positions, consonants had already been palatalized, compare for example the Ukrainian льон and дьоготь. That suggests that palatalized consonants arose sometime in the end of the first millennium. Palatal consonants had arisen earlier, and the opposition n : ņ, r : ŗ and l : ļ (as in она : коня, ору: порю, стола : поля) is common Slavic, that is it precedes the formation of a separate Russian language, and in general of East Slavic languages.

    [Corrected the link to дьоготь]
     
    Last edited:

    Awwal12

    Senior Member
    Russian
    The idea of opposition between palatal and palatalized consonants is the most problematic part here. As far as I know, such opposition is typologically doubtful (is there any language where it's actually attested?). The supposition that hard consonants before front vowels were simply losing their hardness (i.e. their (labio)velarization) looks sufficient.

    (Btw, дьоготь may be a loanword from northern dialects; this position is really atypical for labialization in Ukrainian - cf. легкий.)
     
    Last edited:

    Awwal12

    Senior Member
    Russian
    Palatal consonants exist in Slovene and Serbo-Croatian (and partly secondarily in West Slavic), and are directly attested in the Old Church Slavonic orthography, as н҄, л҄ and р҄.
    The trouble is not in the palatal consonants, it's in the suggested contrast between, say, [ɲ] and [nʲ] (I hope I get you right?).
     
    Last edited:

    ahvalj

    Senior Member
    A millennium ago Bulgarian scribes distinguished in writing the consonants in кон҇ь тъ and дьнь тъ; in modern Bulgarian these both merged into in конят and денят. We can't tell how strongly palatalized was n in дьнь тъ, but it seems reasonable to suppose that at some (probably short) period (in Old Church Slavonic itself or somewhat later) it was already somewhat palatalized, but less so than the fully palatal н҇.

    I intentionally don't use the concept of phoneme since it is not that useful in etymology, where positional evolutionary changes often outnumber phonemic ones. If you prefer to speak in these terms, however, then I would say that shortly before the fall of the yers the palatal ɲ was opposed to n, which was (at least outside south-west Slavic) positionally palatalized before front vowels. For the topic question it is important that the Russian soft consonants go back to these two separate types and that one is older than the other, so while palatalized consonants emerged in the course of the history of East Slavic (so that there was time when the Slavic dialects spoken in these lands had no positionally palatalized consonants), the originally palatal ones had already existed when several separate Slavic tribes came here (so that there was not a time when Slavic spoken here had no phonemically palatal consonants).
     

    ahvalj

    Senior Member
    That *rj and *lj disappeared earlier can be seen from that their r and l don't participate in pleophony, so we have *marja>мор҇е>море (not **морое) but *marwijas>моровии>муравей. Thus, when *el>olo prior to development of positional palatalization of the preceding consonant (#12), *rj and *lj did not already exist as consonant clusters: they were palatal phonemes.

    P. S. Interestingly, while independently, the relationship between the Serbo-Croatian palatal lj and nj and the Russian palatalized is replicated, with even almost the same actual pronunciation (before back vowels at least), in the Latvian ļ and ņ vs. the Lithuanian . In modern standard Latvian (except Latgalian, be it regarded as a dialect or a separate language), only ļ and ņ (and ģ, ķ) exist, having developed from *lj and *nj and pronounced as palatals (ŗ has hardened in the first half of the 20th century), whereas in standard Lithuanian every consonant (except j) now exists as a pair of velarized C and palatalized (with even the same fronting effects on the following vowels as in Russian), palatalized consonants having developed from both iotated ones and those positionally palatalized before front vowels. Consider for example vaļa : valia, krāšņu : krosnių [sʲnʲ], in the latter case the plain n in krāsni vs. the positionally automatically palatalized in krosnį.

    The Ukrainian instances of e>o after non-palatals (that is, not й, ч, дж, ш, щ, ж) are all atypical (so both in льон and дьоготь), I just wanted to illustrate that the change took place in an otherwise depalatalizing context.
     
    Last edited:

    Zec

    Senior Member
    Croatian
    Some data from various Kajkavian dialects may be of use. Kajkavian, like Slovene, tends to depalatalize nj [ɲ]and lj [ʎ] although they never decouple into [nj] and [lj] like in Standard Slovene. It also has, atypically for South Slavic, positional palatalization of consonants before front vowels, at least positional palatalization of n and l. This can lead to a merger of n and l with nj and lj before front vowels.

    In my village's dialect, lj is [lʲ], only occasionally fully depalatalizing to a clear [l]. There is palatalization of l before all front vowels, leading to its merger with lj. Before other vowels l is a dark [ɫ] and well distinguished from lj. The consonant nj is however a fully palatal [ɲ]. The consonant n is only palatalized before i, and the resulting [nʲ] doesn't regularly merge with nj [ɲ], only in a handful of words. In a neighboring village's dialect, however, n merges with nj before i as [ɲ] regularly.

    So there is an actual case where [nʲ] is distinguished from [ɲ], but it's apparent this is not a stable situation.
     

    ahvalj

    Senior Member
    So, western South Slavic may have positional palatalization after all ,)

    That may explain the cases like gnjiti in the standard language.

    The question is how old this positional palatalization of only n and l is: from what I know about how such things occur, they almost always affect all the available consonants (Irish, Slavic, Baltic, Mordvinic, instances in Turkic), so perhaps the presence of lj and nj is somehow canalizing the ability of speakers to produce palatalization in only these two consonants. I wonder if the palatal rj survives anywhere dialectally in Slovene and Serbo-Croatian and if a similar positional palatalization is observed for r in this case.
     

    rushalaim

    Senior Member
    русский
    Obviously, linguists are denying Russian language to have more soft sounds (though the Russian language has them) but other languages artificially multiply sounds, for example, metel' (метель, it must be мьетьель) not any "мэтэль", or wesenny (весенний, it must be вьесьенньий) not any "вэсэнный" and more. Russian language has many soft and hard sounds than linguists permitted to have today.
     

    Awwal12

    Senior Member
    Russian
    Obviously, linguists are denying Russian language to have more soft sounds (though the Russian language has them) but other languages artificially multiply sounds, for example, metel' (метель, it must be мьетьель) not any "мэтэль", or wesenny (весенний, it must be вьесьенньий) not any "вэсэнный" and more. Russian language has many soft and hard sounds than linguists permitted to have today.
    I don't understand you, I am afraid.
     

    ahvalj

    Senior Member
    I don't understand you, I am afraid.
    That's a typical outcome of the oddities of the Russian orthography that combines in one letter palatalization/yod and the subsequent vowel. The vast majority of Russian speakers have not a slightest idea that е, ю, я and ё (1) do not denote separate vowels and (2) denote different things postconsonantally and elsewhere. That's the source of those Ekaterinas, Dostoevskys and countless other horrors, not to mention than no transcription marks positional palatalization before e.

    By the way, I once thought this was a local corruption of the Old Church Slavonic system with ѥ, ѩ etc., but no: it has turned out East Slavs wrote this way since the very beginning.
     

    Panceltic

    Senior Member
    Slovenščina
    I wonder if the palatal rj survives anywhere dialectally in Slovene
    Not that I'm aware of. All three palatals (rj, lj, and nj) completely depalatalised in coda position and in front of consonants, and decomposed to r+j, l+j and n+j elsewhere. It is just a quirk of the orthography that lj and nj are still written like this in all positions, whereas we have rj only before vowels.

    Compare: kralj [kral] > kralja [kralja], Kranj [kran] > Kranja [Kranja], car [tsar] > carja [tsarja].

    Dialectally, [lj] is completely depalatalised to [l] in a big part of the country. The realisation of [nj] usually varies between [n], [jn] and even a sort of nasalised [j] in some places. [rj] remains as is everywhere, or even becoming just [r] in areas close to Croatia.
     

    Zec

    Senior Member
    Croatian
    So, western South Slavic may have positional palatalization after all ,)

    That may explain the cases like gnjiti in the standard language.

    The question is how old this positional palatalization of only n and l is: from what I know about how such things occur, they almost always affect all the available consonants (Irish, Slavic, Baltic, Mordvinic, instances in Turkic), so perhaps the presence of lj and nj is somehow canalizing the ability of speakers to produce palatalization in only these two consonants. I wonder if the palatal rj survives anywhere dialectally in Slovene and Serbo-Croatian and if a similar positional palatalization is observed for r in this case.
    Re: the palatalization in gnjiti happens specifically in the cluster gn, a similar palatalization happens to the cluster kn as in knjiga. It does happen only before front vowels and might be old. I've seen it dated to the time of the 2nd palatalization. Supposedly k and g would be palatalized across n like they were across v in clusters kv and gv, but here the palatalization would be transmitted to n while k and g would eventually depalatalize. It doesn't make much sense to me. I'd rather compare it (typologically, I'm wary of assuming any direct influence in this particular case) to palatalizations of cl, gl and gn in Romance.

    As for positional palatalization in Kajkavian, one source mentions palatalization of t and d. It doesn't seem to exist in my dialect, and I'm not sure I've ever heard it elsewhere. In any case the Kajkavian palatalization is younger than the general Slavic one, it definitely postdates the merger of *y and *i and *ъ and *ь, since it happens before all of them indiscriminately.
     

    ahvalj

    Senior Member
    Knjiga is a particular case as it is eventually an oriental loan: in Old Church Slavonic its older form looks like кън҄игы/kъņigy (nominative/accusative plural), thus with a borrowed ņ and a yer between k and ņ.

    I have a manual of Ukrainian published a century ago by a west Ukrainian author, and it distinguishes between two kinds of i: the palatalizing one (written ї) coming from ě and from e lengthened before a disappearing yer (via ie), and a non-palatalizing i (written і) coming from o lengthened before a disappearing yer (via uo>üö>ü>i, intermediate stages alive dialectally), e. g. дҍло>дїло vs. домъ>дім (in its contemporary east Ukrainian and in Soviet standard Ukrainian the non-palatalizing i was abandoned and all paired consonants were automatically palatalized before any i). Thus, there is a core where palatalization is inherited and a body of additional cases where it appeared later by analogy. In principle, the instances of Kajkavian palatalization before y and ъ might have followed a similar path — of course if the palatalization before old front vowels is ancient there.

    The peculiarity of Slovene and Serbo-Croatian is that it seems to be impossible to tell whether they possessed positional palatalization a millennium ago. Most Slavic languages have abandoned most instances of positional palatalization, but there are occasional traces left of the former state of affairs. For example, Czech has ř whenever Russian has рь, both from (bouře, pekař) and from *rʲ (řád, říct, řeka, tři, tvář), which has escaped general Czech depalatalization because of the prior assibilation of rʲ/ŗ.

    And I should correct my statement in #14 that Slovene has palatal consonants: I thought lj and nj retained their palatal character before vowels in the official style.
     

    rushalaim

    Senior Member
    русский
    I don't understand you, I am afraid.
    For example, the word "есть" has initial hard е-sound "JEсть", but why the same letter is for the soft e-sound in "тебя"? There are two different letters for soft and hard и-sound: и and й. Or hard initial я-sound and soft я-sound have just common я-letter, but the word "тебя" would be more correct to write "тебää". Why scientists gave Finnish alphabeth letters for every Finnish sound but still denying Russian sounds have correct letters?
     

    Awwal12

    Senior Member
    Russian
    For example, the word "есть" has initial hard е-sound "JEсть"
    You use some unfamiliar terminology. "Hard" and "soft" are used in Slavistics in relation to consonant quality ("soft" = "palatalized or palatal"; "hard" = "non-palatalized; often velarized instead"). E.g. the hard [s] in "сэр" /ser/ or "вас" /vas/ vs. the soft [sʲ] in "сер" /s'er/ or "Вась" /vas'/.
     
    < Previous | Next >
    Top