Syllables are units that consist--in most cases--of a vowel surrounded by one or more consonants. The syllable unit generally plays a role in a number of patterns within each language, including the patterns for the placement of stress. In this section we will look at some of the factors involved in how words divide into syllables.
Note: in some cases, the center or 'nucleus' of a syllable can be a consonant (usually a sonorant consonant).
Each language will set some individual own constraints on how words are divided into syllables. Specifically, the language will contain rules or templates which specify:
(i) How many consonants can go before and after the vowel; and
(ii) Which specific consonants can appear in which part of the syllable.
As an extreme example of a restriction on how many consonants appear in the syllable, we note that some languages allow syllables that consist maximally of one consonant followed by one vowel, i.e. a 'CV' syllable. One language with maximally CV syllables is Hawaiian, where words consist entirely of long strings of 'CV' sequences.
An example of restrictions on which consonants can appear in specific parts of the syllable would be Japanese. Japanese allows CVC syllables, but only specifically nasal consonants (and one other type of sound) are allowed in the post-vocalic position. In general, restrictions on content are more severe for the material after the vowel than for the material before the vowel within a syllable.
Terminology Note: the technical term for the material before the vowel (or other nucleus) in a syllable is 'onset'. The term for material after the vowel (or other nucleus) is 'coda'.
Using
the terms 'onset' and 'coda', how would you formulate the observation that 'material
before the vowel in the syllable tends to have fewer restrictions than the material
that follows the vowel in the syllable'? [click
here for answer 1]
2.3 General Constraints
In addition to language-specific constraints, two general cross-linguistic principles govern division into syllables. These are summarized, in a somewhat simplified form, below:
(a) Maximize onsets (rather than codas)
(b) Sonority always increases in onsets. Therefore, you cannot have [sonorant]+[non-sonorant] sequences in onsets
(c) Sonority always decreases in codas. Therefore, You cannot have [non-sonorant]+[sonorant] sequences in codas
For example, the principle of maximizing onsets tells us that the word 'annoy' will syllabify with the [n] in the onset of the second syllable (and not in the coda of the first) as in shown below:
syll 1 syll 2
annoy = [ ə nOɪ ]
In principle, English would have allowed us to put the [n] into the coda of the first syllable ([ən] is a well-formed English syllable); however, because of the general principle of maximizing onsets, the [n] instead goes into the onset of syllable 2. Other restrictions can, of course, block a consonant from going into an onset; the 'maximize onset' principle should be understood to mean 'put as many consonants into the onset as you can, barring other constraints'. (Note: maximization of onsets is a strong cross-linguistic tendency; however, there are some languages that maximize codas.)
One effect of saying that sonority increases in onsets is to rule out syllables of these forms: *[nka], *[mta], *[rga]. In each of those cases, a sonorant consonant appears before a non-sonorant consonant in the onset, which is impossible. Within words, the requirement that sonority increases in onsets will force syllabification of a word like 'Cancun' as shown below:
syll 1 syll 2
Cancun = [ kæn kʊn ]
Note that we do not put the [n] into the second syllable, and make the second syllable *[nkʊn] (even though that would maximize the onset); the reason is that [n] is sonorant, where-as [k] is not.
Decreasing sonority in the coda has a similar effect, except in reverse. Syllables like these will be ruled out: *[akn], *[akr], etc.
Note: Although our discussion has been limited to combinations of sonorant and non-sonorant consonants, some linguists have attempted to class all sounds, including fricatives and stops, onto a graduated 'scale' of increasing sonority; this is called the 'sonority hierarchy', which is used to derive many more subtle syllable constraints on syllabification.
How would you divide the words (i) 'ego', (ii) 'conquer' and (iii) 'argue' into syllables? What principles determine each division? [click here for answer 2]
[Link : Information on the Sonority Scale]
[Link : Further discussion of syllables]
[Link : Further discussion of syllables]