Uniquely among Slavic languages, and unusually among modern Indo-European languages, the Western South Slavic languages (Serbo-Croatian, and apparently some dialects of Slovenian) have a lexical pitch-accent system. That is, the languages permit "phonemic tone, but where only one or two syllables in a word can be phonemically marked for tone, and many words are not marked for tone at all" (Wikipedia). Modern Serbo-Croatian distinguishes rising and falling tone on syllables with long or short vowels (4 accent types).

How did this tonogenesis come about? Was it an innovation in West South Slavic? a sprachbund effect? A reflex of a pitch accent system in Proto-Slavic that was lost everywhere else?

  • Is there a clear distinction between pitch-accent and the alternative(s)? In hasegawa.berkeley.edu/Accent/accent.html it is suggested that even for English, pitch is the most important marker for accent, length being second. I would conjecture that in syllable-timed langauges like Spanish Spanish, the pitch is the far most important marker of accent (with length being out of the picture).
    – dainichi
    May 28, 2012 at 1:51
  • 2
    But in English etc., pitch is not distinctive: there's just one type of stress, so it's not a pitch accent under the definition I'm using. I edited the question to clarify this. May 28, 2012 at 4:33
  • 3
    @dainichi - Yes. There are pairs of words that differ only in whether the pitch-accent is rising or falling. All stressed syllables in Bosnian/Croatian/Serbian are either rising or falling.
    – user780
    May 28, 2012 at 6:23
  • I think we have a misunderstanding because accent has multiple senses. The OP is referring to lexical tone, where there can be minimal pairs with the same phonemes but different tones, whereas some people are reading the question as though the OP were referring to stress. Perhaps somebody knowledgeable can clarify the wording. May 28, 2012 at 8:18
  • 2
    @hippietrail, the OP's use of the term pitch accent is sound, but I submitted an edit that inserted the word lexical before it to make it unambiguously clear that the discussion is about lexical pitch accent systems as opposed to stress systems that mark lexical stress with pitch accents. May 28, 2012 at 14:38

4 Answers 4


The pitch accent system in seen in Southwest Slavic is actually quite old; it actually originated sometime between Proto-Indo-European and Proto-Balto-Slavic. Essentially, certain long vowels in PIE acquired a feature that's just termed "acuteness", but may have been laryngealization, something like the Danish stød. Later on stressed acute vowels developed a rising pitch, leading to the Proto-Balto-Slavic contrast of "acute" (rising pitch) and "circumflex" (falling pitch) stressed vowels.

Southwest Slavic is the only place in the modern Slavic languages where you still see a pitch accent system nowadays (although it's still there in Lithuanian as well), but it was definitely present in all of the Slavic languages early on. Reflexes of this pitch system still survive, for example, in some of modern Russian's stress patterns or in some vowel length alternations in Czech.

  • +1 @voikya! I glossed over the point you make about acuteness. May 28, 2012 at 21:56
  • But the standard Serbo-Croatian tones are not inherited, not even from Proto-Slavic. The rising tone is a result of a stress shift left, and many dialects have older stress patterns which are likely quite older.
    – user4747
    Mar 5, 2020 at 15:53

According to this paper by Marc Greenberg, the pitch accent systems in the Western South Slavic languages are indeed reflexes of an accent system in Proto-Slavic (though, see @voikya's answer to this question for an explanation of acuteness, which is the feature that was relevant in that accent system). At one point all of the Slavic languages had pitch accent systems that were reflexes of this system, but only those in Western South Slavic languages survive today.

Interestingly, he notes that the corresponding accent types in each individual modern language do not necessarily come from the same accentual category in Proto-Slavic. For example, while the falling accents in Slovene and Croatian have the same sources in Proto-Slavic the rising accents in those two languages have different sources. I am not an expert in Slavic languages, but you may find the references in Greenberg's paper useful for further investigation.


Proto-Slavic 'acute' accent has been lost everywhere in Slavic, but the distinction between acute and non-acute has often been preserved as something else, such as difference in length etc.
However, some western south Slavic dialects still preserve the so called neoacute, a high flat tone... opposed to a falling tone, and they 're both Proto-Slavic in origin.
In many dialects there is an additional rising tone (sometimes it's not really rising but low and followed by a high tone in the following syllable), which emerged only in those dialects some time in the 16th century (and this is also found in the standard Croatian, and Serbian languages).
The tonal systems of all dialects are quite complex... involving various rules etc.


What you're describing is the Neo-Štokavian accentuation, the characteristic innovation of the Neo-Štokavian dialects of Bosnian/Croatian/Serbian (including all the current literary forms), which apparently dates to no earlier than the 16th century. Very crudely speaking, Neo-Štokavian falling accents continue earlier accents on initial syllables, whilst Neo-Štokavian rising accents result from a shift where accented inner and final syllables retained their higher pitch, but had the actual stress retracted to the preceding syllable.

By contrast with the marked accentual innovation of Neo-Štokavian, the pitch accent of the Čakavian dialects is apparently an extremely rare (if not unique) partial continuation of the accentual system of Common Slavic, and as such has attracted considerable scientific attention from Slavists.

My apologies if I have bungled any terminology (or even worse, facts)--I'm afraid I'm a mathematician, not a Slavist!

EDIT: I also see, to my embarrassment, that slavist has already said all that needs to be said on the matter. Had I the reputation, I'd have simply added a comment to his answer.

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.