I heard somewhere that the following consonants can make a syllable on their own (i.e. syllabic consonants): /l/, /m/, /n/, /ŋ/, /r/, /w/, /j/. Of these, I've seen /l/, /m/, /n/, /r/ that make a syllable on their own in English (bottle, prism, button and /r/ in American accents), /ŋ/ in, IIRC, Italian (but I don't remember well). But I've never come across any language that has syllabic /w/ or /j/. I can't even imagine how they can act as syllabic.

Are there any languages that have syllabic /w/ or /j/?

  • 6
    How would that be different from just /i/ or /u/? – Nardog Jun 6 at 8:12
  • 4
    Yeah, syllabic [j] and [w] would be [i] and [u], so yeah many languages do have them :) – Ergative Man Jun 6 at 8:38
  • 2
    @ErgativeMan That seems to contradict the cases mentioned in this question, which shows that there are audible pronunciation differences between semivowels and their corresponding vowels (at least in English). – Aaron Rotenberg Jun 6 at 16:34
  • @ErgativeMan Not at all, they sound completely differently. – Vladimir F Jun 6 at 16:56
  • 1
    /ŋ/ doesn't make its own syllable in standard Italian, but there are such things in some dialects (regional languages), such as the word 'Nduja with a syllabic /n/. – LjL Jun 6 at 18:27

The word "syllabic" as applied to a letter that generally represents a consonant generally means "forms the peak of a syllable". Tashlhiyt (Berber) is said to have many syllabic consonants including /s̩ t̩/, which could in principle include syllabic glides. It is however recognized that this is not a fact of lexical representations, it is a fact about relatively superficial phonetics. At that level, /j w/ do have syllabic variants, which however are usually transcribed with the letters [i, u].

In a very few languages, so-called syllabic consonants have a derivationally deeper pedigree, resulting in minimal pairs, such as Swahili [mbuni] "ostrich" and [m̩buni] "coffee tree". Such contrasts exist in a number of Bantu languages, for example Swahili and Hehe. Even then, though, this is the result of a rule reducing /muC/ to [m̩] ("coffee tree" is /mu-buni/). This is the fate of so-called syllabic consonants in general, until you get to some of the challenging fricative vowels of Chinese languages such as Nantongese which is analyzed (by Ao) as having syllabic /ʒ̩ ʒ̩ʷ z̩ β̩/. In these languages, it is usually a matter of analytic premise that one should treat [z̩] as /z̩/ because there is no direct evidence that [z̩] derives from something else (e.g. /zə/). However, various theories provide indirect "evidence", in that they may prohibit syllabic consonants in underlying forms (typically by saying something to the effect that "the property that defines being a syllable peak does not exist in the lexicon").

The most compelling evidence for "syllabic glides" at a level more abstract that surface transcription would be a minimal pair, as we find in Swahili and a few other Bantu languages. Convincing minimal pair involving [u] and [w̩] or [i] and [j̩] are so far lacking. However, /j w/ do appear as uncontroversial [i u] in Tashlhiyt (and other languages) is exactly the contexts where you find other syllabic consonants. It is a classic observation of phonological theory and theories of features that glides and vowels are "the same" except for one property, the feature "syllabic" (or whatever one calls it).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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