• Dutch
  • Frisian
  • Afrikaans
Show full table of contents
The glides

Glides form the transition between vowels and consonants. They are considered as the close vowels /i/, /y/, and /u/, which are realized as [j], [ɥ], and [w] when not part of the syllable nucleus.


The glides form the transition between vowels and consonants, which is also manifested by the distributional fact that they are always adjacent to a vowel. They are considered here as the close vowels /i/, /y/, and /u/ in so far as these do not have the function of syllable head or head of a diphthong, in which case they are realized as [j], [ɥ], and [w]. So, /i,y,u/ are realized as either [i/y/u] or [j/ɥ/w], depending on their position in the syllable, not on their inherent feature specification.

Having the smallest amount of inherent sonority, i.e. being the most consonant-like vowels, it is no coincidence that the close vowels act as glides. In languages in which there is a regular alternation between vowels and glides, half closed and half open vowels become close when functioning as a glide (see Harris (1985) for Spanish and Booij (1989:326-327) for Icelandic). It is only as the syllable head or as the head of a diphthong that a vowel can fully display its vocalic properties; when not in the head position, there is an automatic decrease in both sonority and duration. The vowels [i]/ [u] and the glides [j]/ [w] have much in common, so that the latter easily derive from the former: [w] is a labio-velar sound, which also holds for the vowel [u], whereas [i] and [j] share their closeness and frontness.

The analysis of glides as underlying close vowels has a number of advantages over considering these as separate segments.

In the first place, it simplifies the underlying segment inventory. Besides, it also accounts for the fact that [i]/ [u] and [j]/ [w] do not, and cannot, stand in phonological opposition to each other.

Secondly, the Obligatory Contour Principle prohibits identical adjacent segments. From this it follows that [j] and [w] can neither precede nor follow [i(:)] and [u(:)], which is confirmed by the facts. In older stages of Frisian, [j] could precede /i:/ and /iə/, to which the old spelling of the words tsjiis cheese, tsjien ten, tsjier chink, and tsjiere to quarrel testify. These words have lost their glide in the course of time and their spelling has changed accordingly, so nowadays they are written as tsiis, tsien, tsier, and tsiere, respectively. It is only in jier year that the sequence [ji] can be heard; this word, however, is generally pronounced as [iər], which renders it indistinguishable from ier vein, blood vessel. Loan words from Dutch may contain [ʃ], as do machine [maʃinə] machine, Chili [ʃili] Chile, China [ʃina] China, and Chinees [ʃine:s] Chinese. The non-native sound [ʃ] is likely to have been reinterpreted as the sequence [sj], which resulted in the forbidden sequence [ji]. Through deletion of [j], the common pronunciation of the above words has become [masine]/ [məsinə], [sili], [sina], and [sine:s]. If [j]/ [w] and [i]/ [u] are looked upon as independent segments, the OCP cannot be invoked for an explanation of the above patterning.

[hide extra information]

Because /y/ is considerably less frequent than /i/ and /u/ in Frisian, [ɥ] is left out of consideration here.

[hide extra information]

The analysis of glides as underlying close vowels implies that falling and rising diphthongs and sequences of a long vowel plus a glide are assumed to have an underlying close vowel. Therefore, the diphthongs [ɛj], [ɔw], [jɪ], [wo] and the long vowel plus glide combination [u:j], for instance, have the underlying representations /ɛi/, /ɔu/, /iɪ/, /uo/, and /u:i/, respectively. For more on diphthongs and long vowel + glide sequences, see complex vowel sequences.

[hide extra information]

It was observed in Fokkema (1940:143) that [j] has a more vowel-like realization when it is preceded by a consonant whereas it has a certain amount of frication when in word-initial position; [w] is also likely to be realized with some frication there. Cohen et al (1959:125) claim that [w] has a rather open realization when it is preceded by labial consonants and a more closed one when following coronals and dorsals. These differences in realization are taken to result from phonetic implementation.

[hide extra information]

In word-final position, the half close long vowels — /e:/, /ø:/, and /o:/ — are realized as [e:j], [ø:ɥ], and [o:w], so with an off-glide (see the realization of the long half close, half open, and open vowels). These realizations are to be viewed as the result of desonorization: the speaker fails to implement the phonological structure of a long monophthong as a steady event. The sonority slightly decreases towards the end, with the inevitable side-effect that the vowel becomes close in its final phase. This also implies that the quality of the off-glide element derives in a straightforward way from the quality of the vowel: the front vowel /e:/ ends in the front glide [j] ( [e:j]), the central vowel /ø:/ in the central glide [ɥ] ( [ø:ɥ]), and the back vowel /o:/ in the back glide [w] ( [o:w]).

The evidence being contradictory, it is not clear whether glides preceding a vowel (a short one, a long one, a diphthong) belong to the syllable onset or to the syllable nucleus. For more on this, see the syllabic affiliation of prevocalic glides.

  • Booij, Geert1989On the representation of diphthongs in FrisianJournal of Linguistics25319-332
  • Cohen, Antonie, Ebeling, C.L., Eringa, P., Fokkema, K. & Holk, A.G.F. van1959Fonologie van het Nederlands en het Fries: Inleiding tot de moderne klankleerMartinus Nijhoff
  • Fokkema, Klaas1940Over de Friese klinkersBundel opstellen van oud-leerlingen aangeboden aan Prof. Dr. C.G.N. de VooysGroningen/BataviaJ.B. Wolters Uitgevers-Maatschappij N.V.140-145
  • Harris, James W1985Spanish diphthongization and stress: a paradox resolvedPhonology Yearbook 231-45