• Dutch
  • Frisian
  • Afrikaans
Show full table of contents
Complex onset: sequences of obstruents and liquids

According to the Complex Onset Sonority Constraint (see complex onsets) the segments making up a complex onset cannot belong to adjacent sonority classes. Taking into consideration the Sonority Sequencing Constraint (see complex obnsets), the unmarked word-initial sequence of two consonants is an obstruent followed by a liquid. This yields a complex onset with a maximal sonority contrast between its members. This topic deals with the word-initial obstruent-liquid sequences in Frisian.


Most of the logically possible combinations of an obstruent and a liquid, going from left to right, are instantiated in Frisian, as exemplified in the following table (non-occurring sequences are starred):

Table 1: Examples of words with the unmarked initial sequence of obstruent plus liquid
Obstruent plus /r/ Obstruent plus /l/
prom /prom/ plum plak /plak/ place, spot
brea /brɪə/ rye bread blom /blom/ flower
trang /traŋ/ rancid /*tl-/
drip /drɪp/ drip, drop /*dl-/
krop /krɔp/ crop, maw klam /klam/ clammy
grûn /ɡrun/ ground glei /ɡlaj/ randy; rank
frett(e) /frɛt/ to gorge flym /flim/ lancet
slak /slak/ snail, slug /*sr-/
wrak /vrak/ wreck /*vl-/

The onsets /tl-/, /dl-/, and /vl-/ do not occur. In many languages, /tr-/ and /dr-/ are possible onsets, whereas /tl-/ and /dl-/ are not. Universally, tautomorphemic sequences of consonants with the same specification for the feature continuant are less favoured than those in which the consonants have different specifications (see Clements (1990:316)). This might explain the impossibility of /tl-/ and /dl-/, provided that it can be made plausible that /l/ is -continuant. The specification of laterals with respect to the feature continuant − i.e. whether they should be characterized as either +continuant or -continuant − is a long-standing phonological issue (see also the liquids).

The status of /v/ with respect to the feature continuant is unclear as well. On the one hand, it is a +continuant, on the other hand a -continuant segment (see the labial fricatives).

The fact that both /v/ and /l/ may act as +continuant and -continuant segments may render the impossibility of  the initial sequence /vl-/ understandable, but it does not offer a full explanation. For one thing, /f/ and /s/ are +continuant segments, whereas /l/ may also act as such; yet, /fl-/ and /sl-/ are fine. Therefore, the impossibility of the word-initial consonant sequences /tl-/, /dl-/, and /vl-/ needs an explicit statement in the grammar of Frisian.

It should be noted, though, that the initial sequences [tl-] and [dl-] do occur, be it only in word-medial position in complex words. Some examples are wrantlich /vrɔntəl+əɣ/ /(vrɔn).(tləx)./ resentful; peevish, handler /hɔndəl+ər/ /(hɔn).(dlər)./ trader, dealer, and noedlik /nuəd+lək/ /(nuə).(dlək)./ hazardous, risky, perilous. The Syllable Contact Constraint is likely to be the motivation behind this syllabification. According to this constraint, the contact between two syllables is better to the extent that the first one ends more sonorously and the second one begins less sonorously (Murray (1983), Clements (1990:319-320)). In /(vrɔnt).(ləx)./, /(hɔnd).(lər)./, and /(nu.əd).(lək)./, the first syllable ends in [-t] or [-d] and the second one begins with [l-], hence the syllable contact is bad. In /(vrɔn).(tləx)./, /(hɔn).(dlər)./, and /(nu.ə).(dlək)./, on the other hand, the first syllable is open or ends in a sonorant consonant and the second one begins with [t-] or [d-], which provides for a much better contact.

  • Clements, George N1990The role of the sonority cycle in core syllabificationPapers in Laboratory Phonology1Cambridge University Press283-333
  • Clements, George N1990The role of the sonority cycle in core syllabificationPapers in Laboratory Phonology1Cambridge University Press283-333
  • Murray, R.W. & Vennemann, T1983Sound change and syllable structure in Germanic phonologyLanguage59514-528