About Syllables

Segments (that is consonants and vowels) are grouped into syllables. We have an intuitive grasp of what this means, and we can count syllables by tapping a finger as the word is pronounced: The number of syllables in some words in English though, depends on our pronunciation: how many syllables are in the words 'fire' and 'fail'?

This has nothing to do with the meaning of the word: it is strictly an observation about sound. All of the words given as examples so far have only one morpheme (or 'unit of meaning') each.

In English, each word has a syllable that is stressed: that sounds louder and is longer than other syllables.

Some words, depending on length, have a second syllable that is also stressed although not as much as the first. This is called 'secondary stress'.

Syllables can be grouped into feet. this is important in some kinds of poetry where we count not only syllables per line, but feet per line. A good poet can make the counts work out while having the lines read naturally.

Dr. Seuss was a good poet., the stressed syllables here are capitalized. Notice how they are equally distributed in the text: two unstressed syllables followed by one stressed one. To read this well, the reader never has to place stress on a function word or on a syllable that is normally unstressed.

Here are two lines of poetry which have the same number of syllables and the same rhyme as the good Dr.'s, but are much harder to read. They're hard to read because the stressed syllables are all in the wrong places. English is a stress-timed language and that means we like to keep stressed syllables from being next to each other when we can. We'll even shift the stress in some cases. Trying to keep stressed syllables apart means also that you can smash all sorts of other words between the stressed points in a sentence. Vowels in those words often reduce to schwa. All syllables are not spoken equally, and if you do give each syllable in a sentence the same amount of stress and make them all equally long, you sound quite unnatural.

Without getting overly technical, we could say that a syllable is a beat. Think of them as peaks of sonority within a sentence. The vowel, in the middle, is the loudest part of the syllable. Consonants before and after each vowel are less loud. The word 'slink', for example, begins with a voiceless fricative, gets a little bit louder for the liquid [l], and reaches its loudest point at the vowel. Then, the nasal is less loud followed by the release of the [k] which is not very loud at all. Simplifying slightly, this is what English syllables do. The vowels are in the center with liquids and nasals next to them and the fricatives and stops on the outside.

Phonotactics is the term for rules that govern how many and what kind of consonants can group around a vowel in a particular language and rules that govern what can go where in a syllable. Phonotactics is how we know what is a possible word in a particular language:

For example, The forms below, as transcribed, are all impossible in English. If a native English speaker tries to say them, they will almost certainly 'fix' them by epenthesis, by making one segment unnaturally long, or by deletion.

Those are probably hard to pronounce in most languages. Here are few things specific to English:

In English, ŋ ENG is restricted to the ends of syllables. It just is: there is no good reason for this. The other nasals, [m] and [n] have no such restriction.

English also lacks [tl] clusters word initially, although the sequence [tl] is permitted in the middle of words if [t] belongs to the end of one syllable and [l] to the beginning of the next. Note that this restriction does not hold for the other voiceless stops however:

Our speech production system takes phonotactics seriously. We have trouble pronouncing words from other languages with other phonotactic constraints. so we epenthesize, delete segments or change segments to make the word fit a pattern we are used to producing.

Think of syllables as having an onset: (consonants before the vowel) a nucleus ()the vowel) and a coda (consonants after the vowel.

Using these ideas, we can express some observations about English more succinctly:

English allows lots of segments into codas and onsets: transcribe the word strengths ... (Don't forget the epenthetic [k]. Now abstract it a little: put a C where ever there is a consonant (any consonant) and a V for any vowel and you get:

CCCVCCCC

English allows three consonants into an onset and four into a coda. As we've seen, not just any consonant can occur in any order though. Try to work out what the rules are for the restrictions on consonants in English syllable onsets.

How we think of syllables:
A little game:

Get the process? Here's another. Don't think about it; just do it. Good. Now, do this one:

This is a pretty good linguistics party game: Depending on how syllables work in your native language, you will either say: 'pat' or 'pet'. Amaze your friends by predicting what people will say! Most native English speakers, and speakers of languages with complex syllable codas say 'pat'. Most Japanese speakers say 'pet'. This probably has lots to do with Japanese orthography and with Japanese syllable structure. Japanese syllable structure is quite different than English, allowing only one initial consonant followed by a glide, a long or short vowel, and a final nasal. CGVVN (The story of Japanese syllables is slightly more complicated than represented here: if you want to learn more, google 'mora').

In Japanese, the onset and nucleus are grouped together. In english, the opposite: nucleus and coda go together and we call that a rhyme'.

Japanese speaker: pa-t

English speaker p-at.

Maximum syllables

The next task is to determine what the maximum syllable of a language is (assuming for the sake of present pedagogy that you can't just look this up...)

One way is to just come up with lots and lots of one-syllable words in the language until you hit upon the 'biggest one'. Another is to start dividing words into syllables and see how far that takes you:

Set aside the rules you may have learned about hyphenation and think about sounds only.

We take the vowel as the center of each syllable. Then, we group consonants around those vowels to form syllables. Some consonants may come before and some after each vowel.

We begin by transcribing a word and that word may as well be 'phonology'
[fʕnalədʒi]
Now, we want to divide the word into syllables. four syllables to be exact.

It's pretty apparent that the initial [f] segment has to go with that first vowel. That is, the [f] goes in the onset of the first syllable. After that, though, we have choices about where the [n] goes. Two possibilities:
[fʕ / na
or
[fʕn / a

The way to keep things organized (and the way that is coherent with cross linguistic data and psycholinguistic evidence) is to put as much into each onset as possible. Our word would then look like this:
[fʕ / na / lə / dʒi]

But when do you stop putting things into onsets? What about a word like 'linguist'? should it be
[lI / ŋgwIst]
with all those consonants in the middle of the word in the onset of the second syllable?

A handy test: if the syllable is OK to start a word, then it's an OK syllable. In English, I can't make a word like [ŋgwIst], so that's not a good syllable: that Eng can't be in the onset. So, I divide the word like this:
[LIŋ / gwIst]

To figure out the mightiest (that is maximum) syllable possible in a language, do that kind of analysis on words in the language. When you find the most impressive syllable, you'll want to abstract it a little as we did above with 'strengths'. Substitute a capital C (to mean any consonant) and a V to mean any vowel. Linguist becomes
CVC / CCVCC

Some languages (like Hawaiian) permit only a single consonant in the onset and no consonants in codas at all. Their maximum syllable is CV. Some languages permit two consonants in the onset and no codas. Their max syllable is CCV. Some languages permit one consonant in the onset and a single consonant in the coda like this CVC.

The important take home for language teaching is this: English permits complex onsets and complex codas. speakers of languages which do not permit this will have to learn how to pronounce these words, and they will take a variety of paths toward that goal.

Updated: September, 2020