• Dutch
  • Frisian
  • Afrikaans
Show full table of contents
Syllabic obstruents

This topic is about the occurrence and phonological behaviour of syllabic obstruents in Frisian. They appear to occur in a context which is far more disparate than that of syllabic sonorant consonants, whereas they show little interaction with phonological constraints. This is why their occurrence is interpreted as a side-effect of phonetic implementation.


Syllabic obstruents also occur in Frisian, witness the examples in (1):

Example 1

Examples of the occurrence of syllabic obstruents
harssens /hasəns/ [hass̩] nut, conk
jou dy tas ris /tɔs əs/ [tɔzs̩] [tɔss̩] [tɔzz̩] give (me) that bag
is ommers ek sa /ɪs oməs/ [z̩oməz] that reminds me, of course
is oars /ɪs oəz/ [z̩o.əs] is different
is uzes /ɪs y:zəs/ [z̩y:zəs] is ours
of sa /ɔf sa/ [f̩sa] or so
of net /ɔf nɛt/ [v̩nɛt] or not
och nee /ɔx ne:/ [ɣ̩̩ne:] oh, no
moast ris sjen /mat+st əs/ [mass̩] look at this!
jasses /jasəs/ [jass̩] ugh!
fansels /fɔnsɛls/ [f̩sɛls] of course!
ferdomme /fərdomə/ [f̩domə] damn(ed)
sekreet /sikre:t/ [s̩kre:t] (dirty) swine

Syllabic obstruents appear to be fricatives, notably /s/. The reason for this may be that, as opposed to plosives, which are characterized as -continuant segments, fricatives have the feature +continuant, which can be prolonged in pronunciation. A syllabic consonant has a longer duration than a non-syllabic one, which only fits in with the feature +continuant. Bell (1978:182-183) formulates the implicational universals concerning syllabic obstruents below:

Bell's implicational universals concerning syllabic obstruents:

  1. If a language contains syllabic stops or affricates, it contains syllabic fricatives;
  2. if a language possesses syllabic obstruents, it possesses syllabic [s] or [š], given that it has non-syllabic [s] or [š].
From the above it emerges that the unmarked syllabic obstruent is the fricative [s̩].
[hide extra information]

Riemersma (1979:90, note 90) mentions the word kapot /kapɔt/ broken; worn out, dead tired, for which he considers the pronunciation [k̩pɔt], so with the syllabic plosive /k/, to be possible. Be this as it may, [k̩pɔt] is not a common realization.

The syllabic obstruents in (1) violate some well-established phonological constraints of Frisian. In cases like is ommers ek sa /ɪs oməs/ [z̩oməz] that reminds me, of course, och nee /ɔx ne:/ [ɣ̩̩ne:] oh, no, and of net /ɔf nɛt/ [v̩nɛt] or not, a word begins with a voiced fricative and in a case like jou dy tas ris /tɔs əs/ [tɔzs̩] [tɔss̩] [tɔzz̩] give (me) that bag, a word either ends in a voiced obstruent ( [tɔzz̩]) or it contains a sequence of obstruents with opposing specifications for the feature voice( [tɔzs̩]). Since 1) a word in Frisian cannot begin with a voiced fricative (see the obstruents: the fricatives), since 2), due to Final Devoicing (see final devoicing), a word cannot end in a voiced obstruent, and since 3) the obstruents in a cluster must have the same voice specification (see word-final sequences of two obstruents), the above seems to imply that syllabic obstruents fall outside the scope of Frisian phonology proper.

As to the latter, there are several reasons why the occurrence of syllabic obstruents and syllabic sonorant consonants must be analyzed in different ways. In the first place, the sequence schwa + sonorant consonant underlies a syllabic sonorant consonant (see evidence that syllabic consonants derive from /ə/ + consonant). But a syllabic obstruent, so it seems, can derive from a full vowel preceding an obstruent (whether or not there is a preliminary stage of vowel reduction), as shown by examples like is oars /ɪs oəz/ [z̩o.əs] is different and of sa /ɔf sa/ [f̩sa] or so.

Secondly, a syllable headed by a sonorant consonant must have an onset (see the onset condition), whereas a syllabic obstruent can occur at the very beginning of a word, as in och nee /ɔx ne:/ [ɣne:] oh, no.

Thirdly, syllabic sonorant /n/ does not undergo regressive place assimilation (see regressive place assimilation), but examples like is ommers ek sa /ɪs oməs/ [z̩oməz] that reminds me, of course and of net /ɔf nɛt/ [v̩nɛt] or not show that a syllabic fricative does undergo normal regressive voice assimilation (see regressive voice assimilation: type 2).

Fourthly, there seems to be an intimate connection between the syllabification of obstruents and fast and/or casual speech, whereas the occurrence of syllabic sonorant consonants is not impossible in slower and more careful speech (though it is less common there).

The occurrence of syllabic obstruents is to be understood as a side-effect of phonetic implementation. The processes ensuing from the latter are characterized by their gradient effect, their variability, their dependence on speech rate, and their lack of interaction with the phonology (McCarthy (1986:250). A phonological structure must be interpreted phonetically. A syllabic obstruent results from the failure to interpret a vowel in a phonetically appropriate way, that is to say, so that the air can freely escape the oral cavity. This is perceived as the syllabification of the obstruent preceding or following the vowel. It need not come as a surprise therefore that the context in which syllabic obstruents show up is far more disparate than the context of syllabic sonorant consonants. When articulating quickly and/or carelessly, the phonological structure of every vowel can remain without a proper phonetic interpretation. This implies that it is not necessary to assume that, in order for syllabic obstruents to arise, a full vowel must reduce to schwa first; it may well be sufficient for the full vowel in question to be unstressed.

  • Bell, Alan1978Syllabic ConsonantsGreenberg, Joseph H. (ed.)Universals of Human Language2: PhonologyStanfordStanford University Press
  • McCarthy, John J1986OCP Effects: Gemination and AntigeminationLinguistic Inquiry17207-263
  • Riemersma, Tr1979Sylabysjerring, nazzeljerring, assymyljerringLjouwertKoperative Utjowerij