VOWELS, CONSONANTS AND SYLLABLES IN ENGLISH:
An English Teaching Perspective

David S Taylor
School of Education
University of Leeds
source: http://www.leeds.ac.uk/

Abstract

The terms vowel, consonant, and syllable are crucial in all discussions of English pronunciation. Despite this, the meaning of these terms is often not clear. In particular, it is frequently unclear whether they are being used in a phonetic or a phonological sense. Both phonetically and phonologically there are problems of definition. This paper explores the meaning of these terms, and points out that, as teachers of English, we can only make sense of the phenomena to which these terms refer if we take into account both the phonetic and the phonological aspects, while at the same time carefully distinguishing between them. 

TOPICS concepts of vowel, consonant, & syllable phonetic vs. phonological - def. differ vowels - how many?  5 vowel letters, 12v 3 ways to look at vowels & consonants    what they are, ...do, orthography vowels- syllabic and non-syllabic syllable continuous vs. discrete distinguish between the phonetic and the phonological aspects three ways to look at vowels:  vowels - vocoid, contoid, consonant syllabic consonants how many vowels in English?  Its not five

Discussions of English pronunciation, particularly from a teaching and learning point of view, often suffer from a failure to distinguish sufficiently between phonetics and phonology and to consider the relation between the two. For example, the well-known vowel and consonant charts reproduced in most treatments of English pronunciation often do not make it clear whether the sounds referred to are phonetic or phonological entities. In the British and European tradition there has perhaps been a tendency to adopt a too phonetically-oriented approach, while the American tradition has perhaps paid too much attention to phonology (see Taylor 1994). Yet it is obvious that most phenomena we come across in dealing with English pronunciation can only be properly understood with reference to both phonetics and phonology. Nowhere is this more clear than in the case of vowels, consonants and syllables. It is the intention in this paper to explore these problematic terms and what they refer to from the point of view of what English teachers need to know about them.

The terms vowel and consonant are fairly familiar, and probably anyone who can read and write has come across them. This very familiarity is in itself a problem, however, because we all tend to assume that we know what vowels and consonants are. But when we come to consider the matter more closely we quickly realize that this is far from being the case. This can easily be shown by asking an English speaker, How many vowels there are in English? By far the commonest response to this question is 'five'. A moment's reflection, however, will show that the matter is by no means as straightforward as this. It is clear that the answer 'five' refers to the vowel letters of the Roman alphabet used to write English. It is equally clear that there are more than five vowel sounds in English. 

Here then is a first difficulty. In common use and understanding the terms vowel and consonant tend to refer to the letters of the alphabet, in other words to writing. The letters do of course represent sounds, but not in any straightforward way, as can be seen from the fact that we have only five vowel letters to represent about twenty English vowel sounds. In talking about vowels and consonants, therefore, we must first be careful to distinguish clearly between sounds and letters, and between speaking and writing.
 

Unfortunately, this is not the only complication. As pointed out above, we can look at speech sounds from the point of view of phonetics or from the point of view of phonology. We can look at vowels and consonants from the point of view of what they are (phonetics), or from the point of view of what they do (phonology). This is in addition to the point of view of orthography or writing just mentioned. There are thus three possible ways of looking at vowels and consonants, and it is not always easy to keep them separate. It is no wonder that there has been a great deal of confusion and that the two terms, which are very old (they were first used by the ancient Greeks), have been extremely difficult to define. The confusion is heightened in English especially by the fact that the three categories of vowel and consonant do not necessarily coincide. That is, a sound which is phonetically a consonant may behave phonologically like a vowel (we shall see some examples of this when discussing syllable structure). Similarly, an orthographic consonant (e.g. y) may represent what is, phonetically or phonologically speaking, a vowel. In many languages there is a reasonable match of all three categories, and these complications do not arise to the same extent.

An obvious solution is to use different terms for phonetic, phonological, and orthographic vowels and consonants, and a number of terms have been proposed. Perhaps the best are those of Pike (1943), who uses the terms vocoid and contoid for phonetic vowels and consonants respectively, and syllabic and non-syllabic for vowels and consonants in the phonological sense, from the point of view of their function in the syllable. Using these terms in combination enables us to give an exact phonological and phonetic description of any given segment. The phonetic description will of course remain constant, while the phonological description may vary from language to language and perhaps also from context to context within the same language. Thus a phonetic vowel is a vocoid. When a phonetic vowel corresponds with a phonological vowel, the most usual case, we have a syllabic vocoid. The normal vowels of English are syllabic vocoids. Sometimes, however, a phonetic vowel behaves phonologically like a consonant and then we have a non-syllabic vocoid, such as /j/ or /w/ in English. ([j] and [w] are vocoids according to Pike's strict phonetic definition.) Similarly, a phonetic consonant is a contoid. Normally a contoid is also a phonological consonant, and so we have a non-syllabic contoid. English consonants are normally non-syllabic contoids, in other words consonants from both phonetic and phonological points of view. In the occasional cases where a phonetic consonant behaves as a phonological vowel we have a syllabic contoid. In English, /l/ and /n/ sometimes behave like this.

These terms of Pike have unfortunately not gained the widespread acceptance they deserve. Note, incidentally, that Pike has nothing to say about orthographic vowels and consonants. Likewise, we shall say no more here about the orthographic aspect of vowels and consonants, but we shall look at the phonetic and phonological aspects, considering them first in terms of nature or form and then in terms of their function, which essentially means their function within the structure of the syllable, the third of our three terms.

Vowel, Consonant, Syllable - Difficult to define
If vowels are obstruction free, then where is the obstrution for h, y, and w?

It must be admitted that, even in purely phonetic terms, truly satisfactory definitions of the terms vowel and consonant have still not yet been arrived at. No matter what definition is used, there are always some sounds which do not fit easily in either category. The commonest definition states that vowels are sounds that have no obstruction to impede the flow of air through the mouth as it passes out from the larynx and eventually out through the lips. All other sounds are consonants. While this takes care of the vast majority of sounds, about which there is little doubt over which category they fall into, at least in English, some problem cases readily spring to mind. Where is the obstruction in [h], generally admitted to be a consonant in English? Do [y] /j/ and [w] really have obstructions, despite the fact that they are usually classified as consonants and appear as such on consonant charts? (Incidentally, as we shall see, adopting Pike's terminological system and strict phonetic definitions would get round this problem.) A number of other questions are begged by this definition, but we shall adopt it nevertheless, and assume that all sounds can readily be classified as either consonants or vowels in the phonetic sense. The technical term for an obstruction is stricture, and it follows from this definition that a consonant is a sound produced with a stricture. This means that consonants can be most easily described in terms of articulation, which is why we have the familiar consonant chart with its columns for places of articulation and its rows for manners of articulation. On the other hand, vowels, having no stricture, are more easily described in terms of their auditory relationships, that is, how they sound in relation to one another, in other words within a basically auditory framework. Hence the Cardinal Vowel system and the well-known Daniel Jones diagram, with its vowels at auditorily equidistant points.   Cardinal Vowels (trapezoid)

We can in principle define a syllable from a phonetic point of view or from a phonological point of view. (We leave out here any consideration of orthographic syllables, although it is important to point out that the principles of syllable division in writing are quite different from those in phonetics and phonology.) Again, a syllable is something that we all think we know. Certainly, it is something that most people seem to recognize, in all the  languages of the world.
 

Syllable division in writing is  quite different from those in phonetics and phonology

Many writing systems are based on syllables. If people are asked to say how many syllables there are in a given word, there is generally a broad measure of agreement. But if we ask where one syllable begins and one ends, then there is a great deal of disagreement. To take a common example, the English word extra /"ekstr«/ is generally agreed to have two syllables. But where does the first end and the second begin? Do we divide it as /"e + kstr«/, /"ek + str«/, /"eks + tr«/, /"ekst + r«/, or /"ekstr + « /? There is no easy or obvious way of deciding. It is problems such as these that have made it very difficult to find a satisfactory phonetic definition of the syllable, and indeed it seems that it is not very helpful to pursue this line, either in terms of articulation, or in terms of acoustic factors such as the relative sonority or prominence of the different elements of the syllable (Roach 1991:67-68, Giegerich 1992:130-137). We seem to be much more influenced in our judgements about syllables by our view of the structure of the words concerned. That is, we take account of wider linguistic considerations to do with the morphological or phonological structure of the word. For example, let us take one of the comparatively rare cases where there is disagreement about the number of syllables in an English word. If we ask how many syllables there are in the word hire, we are likely to get the answer "one". If we ask how many syllables there are in the word higher, the answer is more likely to be 'two'. 
 

How do you divide the word, EXTRA, into syllables?   "e      + kstr« "ek    + str« "eks   + tr« "ekst  + r« "ekstr + «

Yet for nearly all speakers the pronunciation of the two words is identical (generally /"haI«/). Clearly people are influenced by their view of the word as consisting of one unit (hire) or two (higher - high + er). It seems that we can make more sense of the notion 'syllable' if we consider it purely from the point of view of its function as a whole in such domains as stress, rhythm and intonation (and also, to a certain extent, as the above example shows, morphology), and from the point of view of its structure, in terms of the phonological elements that make it up. Thus we need to look at the syllable from a phonological rather than a phonetic point of view. For the purposes of the present paper we shall consider the phonological structure of the English syllable, looking at the possible combinations of (phonological) vowels and consonants to be found in legitimate English syllables. The following account is intended to give a clear overview, possibly at the risk of some oversimplification, for the benefit of practising English teachers

Every syllable must, by definition, contain a syllabic element, or phonological vowel. This can be represented by the symbol V. In addition to the V there may be one or more non-syllabic elements, or phonological consonants, which can be represented by the symbol C. Of course, from the phonetic point of view, as we have already seen, the V may sometimes take the form of a consonant (a contoid in Pike's terms), just as the C may take the form of a vowel (a vocoid). It is important to bear in mind, therefore, that the symbols C and V refer to consonants and vowels in the phonological sense (respectively non-syllabic and syllabic elements in Pike's terms).

The V is obviously the central, essential part of the syllable. The Cs may occur either at the beginning or at the end of the syllable, on either side of the V. Syllables which have a C or Cs at the end are called closed syllables. Those which have no C at the end are called open syllables. All languages have a central V element in their syllables, but they vary greatly in the combinations of Cs that they allow at the beginning and end of syllables. Some languages have only open syllables, that is, no Cs are allowed at the end. These languages often also allow only one C at the beginning. The syllable structure for such languages could be represented by the formula (C)V, where the brackets indicate that the C may or may not be present. This is obviously a relatively simple syllable structure. It is quite common in the languages of the world. The structure of English syllables is much more complicated. Even a brief reflection will bring to light examples of syllables beginning with up to three Cs, e.g. spray /spreI/, strike /straIk/, scream /skri:m/. A little more reflection will turn up examples of syllables with up to four Cs at the end, e.g. glimpsed /glImpst/, twelfths /twelfTs/. Accordingly, we can summarize the possible structure of English syllables with the formula (CCC)V(CCCC). Of course, this does not meant that we can have any combination of Cs at the beginning or end of a syllable. An examination of syllables beginning with three Cs in English will quickly reveal that the first of the three must be /s/, the second must be /p/, /t/, or /k/, and the third must /l/, /r/, /j/, or /w/. Even within this restricted set not all combinations are possible (there is no /stl/ or /spw/, for instance). Restrictions of a similar kind apply to the combinations of Cs found at the end of syllables. Various other restrictions are found. The phoneme /h/ is not found at the end of a syllable, while /N/ is not found at the beginning. The combination /ps/ is found in final position, but not in initial position. There are many other constraints of a similar nature. We will not go into the details of these constraints here. It is quite an interesting exercise for teachers to try and work them out for themselves. Those who lack the patience to do this can look up the tables of phonotactic possibilities (this is the technical term for the permitted combinations) in English in Gimson (1989:243-256).

English orthography can be bewildering

Combinations of Cs are called clusters, and these can cause severe difficulties for some learners of English. The reason is not hard to find, for, as we have seen, languages vary widely in their syllable structure and hence in the number and kind of clusters that they allow. Probably most languages have a somewhat simpler syllable structure than English. Speakers of such languages, when they come to learn English, face a bewildering array of unfamiliar clusters. There may also be the problem of familiar sounds in unfamiliar positions. As we mentioned above, some languages do not have Cs (consonants) at the end of syllables. In this case learners must overcome a tendency to make all syllables end with a V (vowels), with consequent disastrous effects on the rhythm of their speech in English. The problem is not necessarily with the production of the individual sounds, which may well exist in the mother tongue of the learner, but with pronouncing them in unusual combinations or positions. Clusters of stop consonants, for instance, have special features which make the pronunciation of stops in clusters different in some ways from their pronunciation when they occur on their own. Difficulties with clusters can lead to a lack of fluency which is very disturbing to the listener. It is very helpful, therefore, if teachers are aware of English syllable structure. This will enable them to foresee difficulties and deal with them as they arise.

We have been taking a phonological view of the syllable, without so far taking into account the exact phonetic nature of the C and V elements, although we have noted the possibility that the syllabic elements, or phonological vowels, may be realized by segments or speech sounds that are, from the phonetic point of view, consonants (contoids), and that, conversely, the non-syllabic elements, or phonological consonants, may be realized by segments that are phonetically speaking vowels (vocoids). The so-called semi-vowels /j/ and /w/ are a good example of the latter case. Indeed their traditional name gives some indication of the difficulty caused by the schizophrenic character of these sounds, whose phonological function belies their phonetic nature. Here Pike's terminology goes some way to getting round the difficulty. The term non-syllabic vocoid effectively removes any ambiguity and makes it easier to come to grips with the situation, which is made more tolerable by the fact that in most accounts of English phonetics and phonology /j/ and /w/ never have a syllabic function. (They are always Cs.) On the other hand, some segments or speech sounds which are phonetically consonants (contoids) have a dual role in English, sometimes behaving non-syllabically, like phonological consonants (Cs), while at other times behaving syllabically, like phonological vowels (Vs). This is commonly found with /l/ and /n/. In lot /l�t/ and not /n�t/, for example, /l/ and /n/ are both phonetically and phonologically consonants (non-syllabic contoids), whereas in bottle /"b�tl/ and button /"b¿tn/ they take on a syllabic role as the central and indeed only element of their syllable (syllabic contoids). The syllable structure of these words is therefore CVCV. This may sound like a formidable complication, but in practice it does not seem to cause great difficulty.

Learners may avoid syllabic /l/ and /n/ by inserting a schwa vowel // to take on the syllabic function, thus enabling the /l/ and /n/ to revert to their non-syllabic role. In this case, in the examples above we would have /"b�t«l/ and /"b¿t«n/ respectively (CVCVC). It must be acknowledged, however, that this is not a natural pronunciation for most native speakers of English.
 

Learners must be alert to the presence of syllabic L and N

Learners must therefore be alert to the presence of syllabic /l/ and /n/ if they are not to have difficulty in understanding the speech of native speakers (or indeed the speech of many non-native speakers with an L1 different from their own). This is all the more reason why teachers should have a good knowledge of such phenomena which arise from the phonological syllable structure of English.

There is no space here to pursue this topic further. Indeed, to do so would, from a teacher's point of view, probably be counterproductive. Teachers need accurate information about the important phenomena of vowels, consonants and syllables in English, but they also need information they can use. They need clear and unambiguous information, as far as possible. What they do not need is overdetailed descriptions where the status of the entities discussed is ambiguous. This paper has tried to provide a clear framework with enough detail to enable teachers to make sense of the phenomena, especially by showing how essential it is to distinguish between the phonetic and the phonological aspects of the case.

NOTES

Pike's definition is different. While in some respects clearer and more rigorous, it creates other problems, about the nature of a stricture, for example. See Abercrombie (1967:79-80)

2 For an extremely lucid account of this whole debate, including Pike's contribution to it, see Abercrombie (1967:79-80)

REFERENCES

Abercrombie, David (1967): Elements of General Phonetics. Edinburgh: Edinburgh University Press.

Giegerich, Heinz (1992): English Phonology. Cambridge: Cambridge University Press.

Gimson, A. C. (1989): An Introduction to the Pronunciation of English. Fourth edition, revised by Susan Ramsaran. London: Edward Arnold.

Pike, Kenneth L. (1943): Phonetics. Ann Arbor: University of Michigan Press.

Roach, Peter (1991): English phonetics and phonology. Second edition. Cambridge: Cambridge University Press

Taylor, David S. (1994): "What do EFL teachers need to know about pronunciation?" In Lewis, J. Windsor (ed.) Studies in general and English phonetics in honour of Professor J. D. O'Connor. London: Routledge.