Prenasalized consonants occur in a number of natural languages. https://wikivisually.com/wiki/Prenasalized_consonant

When I hear someone pronounce a word that begins with a prenasalized consonant, the nasal portion sounds syllabic to me:


Am I just hearing things, or using incorrect terminology, when I characterize the nasal portions of prenasalized consonants as syllabic?


I agree with Nardog that syllabicity is a phonological concept. It is related to phonetic characteristics of an utterance, but not directly equivalent to them. A comparison might be stress: stress is realized in different ways (in different languages, and sometimes even within a single language), and speakers of one language won't necessarily perceive stress in the same position when listening to utterances in another language.

Therefore, I wouldn't attach much importance to the fact that the nasal sounds syllabic to you or to other non-native speakers. For comparison, Spanish speakers commonly hear an extra syllable at the start of English words starting with /sp/, /st/, /sk/ (e.g. spa, scooter), and Japanese speakers hear an extra syllable at the start of English words starting with /kɹ/ (e.g. crown), but that doesn't imply that the usual English pronunciations of spa and crown should be described as or transcribed as having any more than a single syllable.

Phonological criteria: some examples of possible approaches

Rather, I would say it is correct to characterize the nasal portions of prenasalized consonants as syllabic if they behave as syllabic in the prosody of the language. You could determine this by asking native speakers for their intuitions, examining some other part of the prosodic system of the language that you think is related to syllables (e.g. stress placement, minimal word criteria), or examining usage in poetry. None of those methods is failproof, but any of them would provide a more solid basis than the evidence of your ears.

Phonological theories can get complicated

Unfortunately, saying that syllabicity is phonological doesn't mean much, since there are many different theories of phonology. Some issues might be whether "syllabic" refers to a particular "level" of representation, and whether it is possible for a prenasalized consonant to be non-syllabic at one level but to contain a syllabic nasal component at a more "underlying" level.

For example, this paper says that the Bantu language Babole has [i.mbé.sé] as a realization of underlying /N-bésé/, with an underlyingly syllabic nasal phoneme that shifts the syllabicity onto the preceding epenthetic vowel [i] and then fuses with the following voiced plosive to form the non-syllabic prenasalized onset consonant [mb] ("Phonological Asymmetries of Bantu Nasal Prefixes," by Jonathan Choti, page 39). That example isn't directly relevant to your question, as the resulting prenasalized consonant is not word-initial, but I'm including it as an example of how languages can be analyzed as assigning syllabicity to different segments on an "underlying" vs. "surface" level.

Phonetic criteria: I'm not sure

There might be phonetic criteria that you can use to support an analysis of syllabicity vs. non-syllabicity, but I'm not sure what they would be. Duration is the simplest thing that comes to my mind: all else equal, I would expect a syllabic nasal to have a greater duration greater than a non-syllabic nasal.

The article "Morphological function, syllabic and phonetic form of nasal+plosive combinations in the Bantu language Mpiemo", by Christina Thornell and Mechtild Tronnier, has a section "Phonetic correlates" that says

experience from studies on Japanese and Yoruba shows that a nasal should have a duration of at least 50ms to be perceived as syllabic (Nagano-Madsen, personal communication).

I haven't tried to measure the duration of the nasal phones in the audio that you linked to.

| improve this answer | |
  • I'd expect a syllabic nasal to have an pitch different from that of the following syllable, as compared to the pitch of a non-syllabic. – Greg Lee Jan 11 at 5:35

Syllabicity is a phonological concept, not phonetic. A vowel is a sound that is phonetically a vocoid and phonologically syllabic. A consonant is a sound that is phonetically a contoid and phonologically non-syllabic. A semivowel is a non-syllabic vocoid. A syllabic consonant is a syllabic contoid.

One may not "hear" a syllabicity or lack thereof. There is no acoustic or articulatory difference in [j] vs [i] or [n] vs [n̩] (the chances are [j, n] are shorter in duration than [i, n̩] within the same language, but this does not hold cross-linguistically), just that they are distinguished whenever the notion of syllabicity suits the given analysis of the distribution of sounds in a language.

| improve this answer | |
  • 1
    The IPA has a diacritic for syllabicity. What other phonological concepts are also encoded by the IPA? – James Grossmann Jan 11 at 2:33
  • 1
    @JamesGrossmann Virtually everything the IPA encodes is a phonological concept, or at least a phonologically-informed phonetic concept, save for a few diacritics. See IPA Principles, particularly #2 and #5. – Nardog Jan 11 at 4:04
  • Aside from duration, I've also seen [j] described as representing a vocoid with greater constriction on average than [i], although I don't know how true this is in different contexts. – brass tacks Jan 11 at 4:50
  • @ewawe I'm sure that's the case in many contexts in many languages, but the point is that just because e.g. a [j]-like sound has less constriction than an instance of [i] does not automatically make it not an instance of [j]. – Nardog Jan 11 at 17:04

Prenasalized consonant is a phonological concept, meaning "single segment containing nasality followed by orality". You can't "hear" a difference between a so-called prenasalized consonant and a nasal plus consonant cluster. There are languages which have two kinds of N+C (homorganic) sequences, for example Swahili mbuni "ostrich" and m̩buni "coffee tree" (a fairly famous minimal pair). I suggest working with real speakers from languages that have such consonants, and do not just look at a single language, because their phonetic realization differs substantially from language to language.

| improve this answer | |
  • 1
    In my field methods course as a grad student, our native Tumbuka informant (Patrick Mkandawire) claimed his language had a minimal pair between syllabic and non-syllabic m (although he didn't put it that way). It may have been the pair you mention, though I don't recall. I completely believed him, but despite many repetitions, I couldn't hear it. – Greg Lee Jan 11 at 5:25
  • 1
    @GregLee: In Swahili, I have read that a notable difference between the phonetic realization of prenasalized ᵐb and a sequence of syllabic m + b is that b is implosive [ɓ] (so the cluster with syllabic m is [m̩ɓ]) while prenasalized /ᵐb/ is not implosive (Benji Wald's review of "Swahili phonology reconsidered in a diachronic perspective", by Fidèle Mpiranya"). Wikipedia's consonant table for Tumbuka indicates the same distinction of implosive /ɓ/ vs. non-implosive /ᵐb/ – brass tacks Jan 11 at 7:20
  • Interesting! -- – Greg Lee Jan 11 at 11:22

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.