Basic Revision for EFL Students

The Human Speech Apparatus and The GB Consonants & Vowels


Human Speech Apparatus

Fig. 1

The sounds of speech are produced by organs whose primary functions enable us to breathe and take in food. Almost all of them are made by the passage of a stream of air form the lungs up through the windpipe, through the valve called the larynx popularly known as the voice-box. The bulge at the front men's throats indicative of the presence of the larynx is traditionally called the Adam's apple. The airstream passes on out through the mouth and/or the nose. The larynx contains the vocal folds formerly usually called much less appropriately the vocal cords. Any sound to which the vibrations of the vocal folds contribute is known as voiced: thirty-five of the forty-four phonemes or word-distinguishing sounds of the analysis adopted of the General British pronunciation of English ('GB') in this book are canonically (ie ideally or characteristically though not necessarily most frequently) voiced; nine are normally voiceless. The pitch of speech sounds is controlled mainly by varying the tension of the vocal folds as they vibrate. The part of the larynx where they are situated and/or the space between them is known as the glottis: when they are brought together and then 'snapped' apart the resultant sound is known as the glottal plosive or glottal stop. This last expression has to an amazing degree, for what is after all a technical term from the vocabulary of the science of phonetics, gained wide popular currency since its first appearance in Henry Sweet's History of English Sounds in 1888. A strong form of this plosive occurs in coughing. The glottal stop is not a phoneme in English but nevertheless a common sound in emphatic speech and in a variety of other ways. Its use to replace a / t / between a vowel and a following unstressed further vowel eg as in [beʔə] for better, though very common in British dialects (eg at Bristol, London, Leeds and Glasgow) is very unfashionable when used as in such words. Non-native learners of English may often prefix it excessively frequently to words beginning with vowels, producing an unpleasantly jerky effect. This is particularly noticeable where native English speakers would use a linking /r/. When a part of the glottis is vibrating much more slowly than the rest an effect is produced called 'creaky' voice. This is heard from many speakers at the lowest range of their voices or as a hesitation signal (especially when straining to come out with the next words). When air passes audibly through the glottis without subsequent stronger friction or vibration of the vocal folds we hear the usual form of /h/ which is widely referred to, though not entirely satisfactorily, as the English glottal fricative consonant. It is also sometimes known by the term voiceless vocoid and by some authorities classified as a voiceless approximant. In English it invariably precedes a vowel and its articulation takes the form of a voiceless version of that following vowel.

The space above the larynx at the back of the mouth is called the pharynx: the walls of the pharynx can be contracted to produce the tense, tight 'pharyngeal' voice quality which is sometimes to be heard accompanying the English vowel / ӕ /. After passing through the pharynx the air can pass either through the mouth or through the nasal cavity or through both of these. Only three English phonemes are nasals /m, n, ŋ/. The others are all prevented from being so because to articulate them we shut off the nasal cavity by raising the soft palate, ie the movable back part of the roof of the mouth. The soft palate, with which only the back of the tongue can make contact, has the less usual Latin name 'velum': sounds made there are referred to by the corresponding adjective 'velar'. There are only three English phonemes with  tongue-to-velum contact: /k, g, ŋ /, though for the closer back vowels / u: / and / ʊ /, and for /w/ the back part of the tongue is raised towards the soft palate, as it is also for the so-called 'dark' varieties of / l /.

The middle of the soft palate ends in a small tip of flesh called the uvula. This is not used in articulating the general forms of English, but it is heard from some dialect speakers in the far northeast of England and from other speakers in various places as an individual peculiarity.

In front of the soft palate is the hard palate: no English phoneme is characterised by contact of the tongue here but the tongue is raised near to the hard palate in making the palatal approximant /j/ and the four post-alveolar (also called palato-alveolar) phonemes / ʧ, ʤ, ʃ, ʒ/. These last four phonemes, besides involving a general raising of the middle of the tongue towards the hard palate, have contact of the forepart of the tongue with the ridge which lies to the front of the palate and immediately behind the upper teeth. This is known as the alveolar ridge (or simply the teeth-ridge). It is the part of the roof of the mouth to which the forepart of the tongue lies opposite when the whole tongue is at rest. It is therefore the easiest and not surprisingly the most frequent place for the tongue to approach or touch in making consonants. Almost half (eleven) of the English consonant phonemes are alveolar, viz /t, d, ʧ, ʤ, s, z, ʃ, ʒ, n, l, r/. Frequency counts of speech sounds in samples of GB have shown more alveolar consonant occurrences than for all the other kinds taken together.

The final pair of English lingual phonemes /θ/ and /ð/ are made by most speakers with the tip of the tongue brought forward to the upper front teeth. These are the English dental fricative consonants They may take stop (plosive and affricate) allophones in certain prosodic situations, eg sometimes in thanks and there pronounced emphatically, when they are perfectly distinct from English /t/ and /d/. Sequences of two identical non-affricate consonants do occur occasionally in English but always either involve two morphemes eg in words like night-time, guide-dog, pen-name, plainness or result from an assimilation which would probably shock purists if they became aware of its existence as in eg /`ӕbbətaɪz/ advertise (with labiodental b's) or / rɪ`zemml/ resemble.

Of the English phonemes articulated essentially at the lips one pair is labiodental, ie made canonically by raising the lower lip to the upper teeth: /f, v/. This may well be blended with a bilabial articulatory posture without sounding abnormal especially in bilabial contexts.

The way the lips are held is an important element in the formation of many speech sounds. There are two main types of posture of the lips that we must take into account — rounded and neutral or unrounded. When the lips are rounded the corners of the mouth are usually drawn inward to some extent. Various degrees of rounding are characteristic of English back vowels — in general the closer the tighter the rounding. However, many if not most speakers have less than tight rounding of their /uː/ and / ʊ / leaving them with usually less vigorous rounding than they use for their opener back vowel / ɔː / and the consonants / ʃ, ʒ, ʧ & ʤ /. Rounding also characterises the bilabial approximant consonant /w/ which is always velarised, ie involves raising of the back of the tongue. Considerable degrees of rounding also normally accompany the five alveolar phonemes whose location of articulation is strictly speaking to the rear or the alveolar ridge viz / ʧ, ʤ, ʃ, ʒ & r/. In the case of this last consonant there is very frequently strong lip action in emphatic articulation but by no means invariably involving lip-rounding. Such articulations are clearly best described as labialised. It's unfortunate that writers on phonetics so often use this term as a mere synonym for rounding.

This completes our list of articulatory manners and locations.

As well as by manner and position of articulation, speech sounds may be classified by the effect of their articulation upon the airstream which is utilised in their production. Among the types found in English are eight stops /p, b, t, d, k, g, ʧ , ʤ /. When a stop consonant is made the airstream is held up by a closure which is complete, firm and long enough for pressure to build up behind it. Each stop has the three stages: approach, hold and release. If the release is rapid a sort of explosive effect is produced and the stop is termed a 'plosive'. Taps and trills have closures which are relatively firm but too brief to produce compression. If a stop is released with a noticeable puff of air, ie is followed by a sort of [h] it is termed aspirated. GB voiceless stops (including /ʧ/) are usually markedly so articulated when they begin stressed syllables. If the release is so slow that the organs producing the stop are retained close enough to each other for some sustained friction to be heard as the air rushes out between them the resultant sound is termed an affricate. There are two such English phonemes viz / ʧ & ʤ /. Affricated sequences (not usually classed as phonemes) are produced when /t/ or /d/ are followed immediately in the same syllable by /r/ or /j/ in try, dry, tune, due. Most General British speakers no longer regularly differentiate dune and June etc nowadays. (Most GA speakers have no /j/ in dune.)

The most numerous type of consonant in English is produced by close narrowing of the airstream so that audible friction is heard. These, /f, v; θ, ð; s, z; ʃ, ʒ & h/, belong to the fricative division of the general class of continuant speech sounds. The most important difference between the fricative pairs / θ, ð/ and /s, z/ is that whereas the dentals have a somewhat (laterally) contracted and protracted general tongue posture, the forepart being rather flat, the alveolars have a less contracted etc posture, the forepart having a shallow groove running down the centre from back to front. The post-alveolar fricatives /ʃ, ʒ/ are produced like the alveolars except for raising of the middle part of the tongue and more extensive grooving.

To refer collectively to the alveolar and post-alveolar fricatives and the post-alveolar affricates the term sibilants is sometimes useful. These are / ʧ, ʤ; s z; ʃ, ʒ/.

The phonemes called the plosives, the fricatives and the affricates are often termed collectively the 'obstruents'. The seven characteristically-voiced consonantal phonemes /m, n, ŋ, l, r, j, w/ together with the vocalics (vowels and diphthongs) may be collectively termed the 'resonants'. The group of five consonants which can most readily become syllabic, viz /m, n, ŋ, l/ and /r/ are sometimes referred to as the 'sonorants'.

Three of these sonorants are made in ways exactly corresponding to our three pairs of plosives /p, b; t, d; k, g/ but differ from them in that the airstream is never held up (and so there is certainly no compression) but is allowed, by the lowering of the soft palate, to make its way out through the nasal cavity. These are the bilabial, alveolar and velar nasal consonants /m, n, ŋ/.

We have noted that a fricative consonant is produced by approximating two articulating organs so closely that when air passes between them with average force it produces audible friction. When either this approximation or the breath-force is reduced, vibration of the vocal folds produces a series of resonants that are  termed 'approximants'. Physically this category overlaps with the closer vowels but approximant is applied to relatively unsustained sounds which are not central to their syllable whereas a vowel is essentially at the centre of its syllable. The terms semivowel and vowel glide have been applied to some of them but the expression approximant is preferable if only as being more comprehensive. There is an approximant corresponding to every fricative except the non-buccal (not made at the mouth) fricative [h]: phonemes characterised as voiced fricatives can be expected to have approximant allophones in some of their weakest realisations. This is certainly true of English /v/ and /ð/. The three English approximants are /r, j/ and /w/. The term 'vocoid' may be used to refer collectively to those sounds which are vowel-like in that they involve no central oral obstruction of the passage of the airstream viz vowels, approximants and /h/.

There are two consonantal phonemes /l/ and /r/ in English which owe their characteristic qualities principally to the fact that the tongue is very considerably contracted in their articulation. It can be convenient to call them collectively 'contractives'. They both involve closest narrowing of the airstream at the dental/alveolar range. The first of them /l/ is contracted mainly from side to side hence its label 'lateral' (which word alone is very often used to identify it). The forepart of the tongue makes (light or firm, momentary or prolonged) contact usually with the alveolar ridge, and may often be plainly seen to have assumed a rather wedge-shaped configuration. The lateral contractive is usually a resonant but if the air is expelled with considerable force, as for instance after stressed voiceless plosives, a fricative allophone may be heard. The other English contractive consonant /r/ has its main contraction from back to front so it can be termed the 'longitudinal contractive'. The tip and edges of the tongue are slightly curled up to produce a rather cupped or spoon-shaped posture. It too has closest narrowing of the airstream usually at the alveolar ridge and rather more to the rear of the ridge as an effect of the drawing back of the forepart of the tongue to produce the contraction. For this reason it is often labelled 'post-alveolar'. It is characteristically an approximant but is invariably fricative when it follows /t/ or /d/ in the same syllable. It is also quite often fricative after other obstruents. It is usually voiced but loses its voicing if influenced by a preceding voiceless obstruent. It is often syllabic eg in temporary /temprri/. Occasionally, when a GB speaker uses a specially vigorous enunciation he may possibly trill or more often make what is called a alternatively a tap articulation for an /r/ but these variants — which are not used by all speakers in any case — need not be cultivated by the EFL learner. This tapped allophone of /r/ is often heard when /r/ intervenes between a short stressed vowel and another vowel eg in the word 'very': here its effect is to sound rather 'vigorous'. In such situations and as a very common allophone immediately after /θ/ or /ð/ it sounds fairly unremarkable. However, if a tapped articulation is used in situations where a specially vigorous manner is not appropriate, the effect produced is either of affectation (the actor Noel Coward provided a good example of this) or dialect influence. In Scotland and over a good deal of the north of England strongly tapped types of /r/ are very common articulations, notably in Liverpool and much of Yorkshire. If the tongue-tip is curved further back than is usual in GB a characteristically hollow sound is heard which is termed retroflex. Many people in southwest England have such an /r/ and many Americans use varieties of it.

The remaining two consonants of GB are also approximants. For the palatal approximant /j/ the middle of the tongue makes a brief movement towards the hard palate. It thus passes through the area of /ɪ/ or of /eɪ/ and /i/. In the same way the rounded velar approximant /w/ usually passes through the area of /ɔ/ or of /ʊ/ and /u/.

Of the twenty-four English consonantal phonemes sixteen are members of the eight pairs distinguished from each other by being what we may very conveniently call soft and sharp though in the phonetic literature they are generally termed rather unsatisfactorily voiced or lenis (ie weakly articulated) for /b, d, g, ʤ, v, ð, z, ʒ/ on the one hand, and similarly voiceless or fortis (ie strongly articulated) for /p, t, k, ʧ, f, θ, s, ʃ/ on the other. These common terms are also unsatisfactory because so-called fortis consonants may well often be quite weakly articulated and lenis ones may receive quite strong articulation.

We may summarise the English system of consonants as follows

The English Consonant Phonemes

1. /p/ as in pen. A generally sharp bilabial plosive (aspirated when syllable-initial and stressed).

2. /b/ as in bad. A generally soft bilabial plosive.

3. /t/ as in it tea. A generally sharp alveolar plosive (aspirated when syllable-initial and stressed).

4. /d/ as in it did. A generally soft alveolar plosive.

5. /k/ as in it cat. A generally sharp velar plosive (aspirated when syllable-initial and stressed).

6. /g/ as in it get. A generally soft velar plosive.

7. / ʧ / as in chin. A generally sharp palato-alveolar affricate

(aspirated when syllable-initial and stressed).

8. / ʤ / as in it June. A generally soft palato-alveolar affricate.

9. /f/ as in four. A generally sharp labio-dental fricative.

10. /v/ as in very. A generally soft labio-dental fricative.

11. /θ/ as in thin. A generally sharp dental fricative.

12. /ð/ as in then. A generally soft dental fricative.

13. /s/ as in see. A generally sharp alveolar fricative.

14. /z/ as in zoo. A generally soft alveolar fricative.

15. /ʃ/ as in she. A generally sharp palato-alveolar fricative.

16. /ʒ/ as in vision. A generally soft palato-alveolar fricative.

17. /h/ as in how. A glottal fricative.

18. /m/ as in map. A bilabial nasal.

19. /n/ as in new. An alveolar nasal.

20. /ŋ/ as in sing. A velar nasal.

21. /l/ as in leg. An (alveolar) lateral contractive. A rather velarised allophone is used before consonants, except /j/), at the ends of syllables, and especially when it is syllabic.)

22. /r/ as in red. An (alveolar) longitudinal contractive. (Usually an approximant but sometimes syllabic and often fricative, voiced, notably after /d/, and voiceless, notably after /t/, when both consonants belong to the same syllable.)

23. /j/ as in you. A palatal approximant.

24. /w/ as in wet. A rounded velar approximant.

GB Vowels and Diphthongs

Actual Limits of the Vowel Area
Fig. 2 The True Limits of the Vowel Area

Cardinal Vowels
Fig. 3 Primary and Secondary Cardinal Vowels

Jone's Vowel Diagrams
Fig. 4 Two Diagrams from the Daniel Jones Outline of English Phonetics


Fig. 5

Fig. 6

Fig. 7

Fig. 8
Titles for Vertical Divisions
Fig. 9
Titles for Horizontal Divisions
Fig. 10
Vocalic Sounds aka Vowels and Diphthongs etc
The vowel system of any other language one studies will hardly ever be the same as one’s native vowel system and will often differ from it greatly. For most of those who study English as an additional language the vowels will constitute one of their greatest pronunciation problems because English has a much more complex system of vowels than most languages.
  The study of vowel diagrams in language learning is very simple but very important. They visualise for the student the contrasts which must be maintained within the vowel system of the target language and the points of danger where the target language contains a vowel contrast not present in the mother tongue. Compared with listeners who can recognise the basic vowel contrasts of English, those whose ears are under-trained in this respect are working very much harder — subconsciously most of the time — sorting out the messages from the ‘scrambled’ versions of the signals which are the only ones they are capable of perceiving. These speakers are also tiresome to talk with at length because their listeners have constantly to unscramble the inefficient signals they give out.    

The Nature of Vowel Sounds
 When we recognise and distinguish vowel sounds what we are doing is very similar to recognising a note, or better a chord, being played on one musical instrument rather than another. In the case of musical instruments their characteristic timbres are mainly due to the shape of the cavity in which a column of air has been set in vibration. Much the same goes for the continuous cavity which is known as the vocal tract, extending from the larynx to the lips and, when the soft palate is lowered, including the nasal cavity. The change from one vowel to another is effected by changing the shape of this tract, principally by altering the position of the tongue — whose great mobility allows us to do so very quickly.
     The reason for there being two-note chords is that the mouth and throat cavities function to some degree separately. When the back of the tongue is raised highest the middle and forepart are automatically held lower than the back and there is therefore a maximum-volume mouth cavity. If the forepart of the tongue is raised highest the cavity in front of it has minimum volume and the throat cavity now extends up over the lowered back of tongue. If the middle of the tongue is raised highest there is usually of course lowering of the forepart and of the back of the tongue. Since these adjustments of one part of the tongue relative to the rest may be taken to be fairly automatic we only need to know which of its three main divisions is highest to know also the posture of the other two and therefore of the configuration of the whole tongue. This circumstance makes it feasible for us to treat the position of the highest part of the tongue as an index to the shapes of the front and back cavities and consequently of the quality of the vowel produced. We can thus with excellent effect use two-dimensional diagrams to represent vowel qualities to considerable degrees of precision.
The Cardinal Vowels
II.2  The usual type of vowel diagram employed is that of the IPA (the International Phonetic Association). It was devised in the second decade of the last century by the late Daniel Jones (1881-1967). He based it on x-ray photographs of the positions of his tongue taken during the articulation of four selected vowels. The two most fundamental of these were the ones which could be produced with the greatest precision without reliance upon auditory memory or comparison. One was obtained by raising the tongue so high and so far forward that (with average breath-pressure) any further movement would produce a consonant. The other was obtained by lowering and retracting the tongue as far as possible without producing a consonant. Starting from these two he produced, by auditory judgement alone, two further sets of three vowels. The first set, cardinals 2 to 4, were made by lowering the front of the tongue through what he perceived to be three equal successive intervals; the others, cardinals 6 to 8, were arrived at by raising the back of the tongue three successive stages which he judged to maintain the same auditory intervals as for cardinal vowels 1 to 4. The x-ray photographs on which the diagram was based were of the basic pair (1 and 5) and the lowest of the front series (4) and highest of the back series (8). When the highest points of the tongue for these four vowels were plotted a figure could be drawn to show their relationships: on this were placed the remaining four cardinal vowels in positions to conform with their auditory relationships. As such a curve-sided figure was difficult to draw, a form of diagram was adopted with straight sides. Diagrams of this shape featured notably in Jones’s influential An Outline of English Phonetics in 1932.
The cardinal vowels were chosen ‘upon the principle that no two of them are so near to each other as to be incapable of distinguishing words’ (IPA Principles 1949 p.4). Conversely to represent vowel phonemes to a very much greater degree of precision than the interval between adjacent cardinals is undesirable. Even a highly trained phonetician cannot pinpoint a vowel more precisely than within for example one-twenty-fifth of the distance from open to close. The untrained listener is generally not likely to notice vowel variation involving less than about a sixth of this distance.  
 
  In 1932 in his Outline of English Phonetics Jones included side by side with his schematised (straight-sided) allegedly more ‘accurate’ diagram a ‘Simplified Chart of English Vowels for use in practical teaching of the language’. In his later works such as the ‘re-written’ edition of The Pronunciation of English (1950 etc) and even in his theoretical treatise The Phoneme (1950) Jones himself used this further simplified shape in preference to the earlier one. It is virtually the only shape now in use and was ultimately made the official International Phonetic Association version in the 1989 revision of the International Phonetic Association’s chart of authorised symbols etc.     

 The relationships of the vowel positions to each other and to the whole diagram are much more easily grasped and remembered if they are submitted to a high degree of schematisation. For present purposes the choice of size for the vowel position indicator implies an area of range, (‘Dots’ in the manner of geometrical drawings, which imply position but no dimension, have been consciously avoided).
  We employ a ‘grid’ within the diagram containing about 30 ‘slots’ most of them of approximately the same area as the 11 regular square interstices on its right-hand side. Besides tongue position, lip-posture (rounded or unrounded) is shown by placing the vowel symbol within a square or circle. Squares are also used to indicate diphthongs which begin and end unrounded, circles those which begin and end rounded. A D-shape indicates a diphthong beginning unrounded but ending rounded, its reverse one beginning rounded and ending unrounded.    
  These diagrams are designed to bring out the broad differences between mother tongue and target language and to avoid unrealistically minute distinctions which may cause learners to underestimate the adequacy of their powers of discrimination and so discourage them.    
  Because the diagram is designed to suggest the greater advancement of the front of the tongue for closer than for open vowels only eleven of the thirty-four interstices are regular squares. The number of vowel indicator centrings available if we choose to centre vowel indicators (squares or circles) only midway between sets of parallel lines (except in so far as the lines on the left of the diagram are not parallel) is about 100. This does not mean that this system suggests a hundred absolute vowel contrasts because all areas with adjacent indicator areas overlap, laterally adjacent 50%, diagonally adjacent 25%. In fact about 30 basic vowel types are suggested by such a diagram.    
  We have standardised the size of all the squares and circles on that of each of the completely regular squares on the right hand side of the diagram. The vowel areas indicated by our squares and circles may be taken as representing a moderate proportion of the range of variation of the phoneme in the ordinary speech of an individual speaker.
  The amount of further variation possible is still considerable but differs in degree and direction for each phoneme and would necessitate inconveniently complex diagrams. Particularly in syllables with least prominence, it is possible for phonemic oppositions to become completely neutralised, though it should be remembered that tongue-position coincidence alone may not be sufficient to produce neutralisation: other features including lip-posture and length regularly preserve phonemic distinctions.
  The choice between representing any particular vowel phoneme articulation range by either of two adjacent indicators at least along the vertical axes can safely be made on the basis of convenience. The three articulations e , ɜː and ɔ on Fig. 1 could be made one degree lower without representing at all unsuitable targets for the learner. The indicator positions in this diagram of the English simple vowels could be shifted one degree in any direction possible within the diagram without representing a totally unacceptable quality for the phoneme in question. Vowel areas of at least twice the present size might have been employed in most cases. Only a bout ten of the thirty possible two-degree shifts would coincide with other vowels.
 We may summarise the GB system of simple vowels as follows
The Characteristically Monophthongal GB Vowel Phonemes    
 1. / i: / as in see. When relatively short, a semi-half-close front-centralised simple vowel: otherwise often realised as a very narrow front-closing diphthong [ij].  See Fig 1. A rhythmically distinct typically short and never diphthongal weak allophone is represented as /i/.
2. / ɪ / as in six. A half-close, front-central simple vowel. Usually relatively short. Always checked in mainstream GB usage.
 3. / e / as in ten. A mid front simple vowel. Usually relatively short. Always     checked ie followed in its syllable by a consonant.
 4. / ӕ / as in hat. A retracted and/or lowered semi-half-open front simple vowel. Usually short before sharp consonants (though not always so so in eg that) but otherwise often fully long (eg in bad, etc).
 5. / ɑː / as in arm. An open back-centralised-to-central simple vowel. Usually     relatively long.
 6. / ɒ / as in got. An open back slightly-rounded simple vowel. Usually     relatively short. Always checked
 7. / ɔː / as in law. A mid back medium-rounded simple vowel. Usually     relatively long.
 8. / ʊ / as in put. A half-close back-central fairly well rounded simple vowel. Usually relatively short. Always checked in mainstream GB usage. Among many speakers, especially younger people, this vowel tends increasingly to be considerably centred and almost or entirely unrounded.
9. / u: / as in too. When relatively short, an approximately semi-half-close back-centralised moderately rounded simple vowel. Otherwise, very often, realised as a very narrow back-closing diphthong [uw]. See Fig 1. Usually more central after / ʧ , ʤ / and /j/.
10. / ᴧ / as in cup. A semi-half-open front centralised-to-central simple vowel. Usually relatively short. Always checked.
11. / ɜː / as in fur. A mid-central simple vowel. Usually relatively long.
12. / ə / as in banana. A mid-central simple vowel. Usually relatively – often very – short. Slightly more open in final unchecked syllables.

     The classification of /iː/ and /uː/ along with the simple long vowels, although they are diphthongal probably in the majority of their occurrences, is justifiable on two counts. Firstly, they are certainly much more often monophthongal than the other two narrow closing diphthongs /eɪ/ and /əʊ/. Secondly, native speakers of English without phonetic training are usually quite unconscious of any movement involved in making / i: / and / u: / though they very possibly may be so in the case of / eɪ / and / əʊ /.    
  The American phonetician Kenneth L. Pike preferred to transcribe /eɪ/ and /əʊ/ with the single symbols /e/ and /o/ for American English chiefly because he found that very many of his beginning students were not conscious of these sounds as diphthongal. The Kenyon and Knott Pronouncing Dictionary of American English (1944) showed the same preference.  Others, beginning with Henry Sweet, have preferred such symbolisations as /ij, uw, ej, ow/ etc.    

  The twelve vowel phonemes described above are referred to as simple vowels because in their characteristic forms they have no obviously noticeable change of quality such as is produced if there is considerable movement of the tongue during their articulation.    
  Complex vowels or diphthongs, on the other hand, have in their characteristic forms such movement and quality change taking place within the limits of a single syllable. Syllables are constituents of words uttered with a single effort of articulation. There are in GB five closing and three centring traditionally recognised diphthongs. They are as follows:    

The Characteristically Diphthongal GB Vocalic Phonemes

13. / eɪ / as in day. A narrow front-closing diphthong starting about mid         front.
14. / əʊ / as in old. A narrow back-closing diphthong starting about mid     central.
15. / aɪ / as in five. A fairly wide front-closing diphthong starting about semi-half-open front centralised-to-central. Very often narrow or monophthongal before /ə/.    
16. / aʊ / as in now. A fairly wide back-closing diphthong starting about semi-half-open back centralised-to-central. Very often narrow or monophthongal before /ə/.
17. / ɔɪ / as in boy. A fairly wide front-closing diphthong starting about semi-half-open back. It begins slightly rounded. Often narrow or monophthongal before /ə/.    
18. / ɪə / as in near. Earlier a centring diphthong starting approximately semi-half-close to half-close front centralised-to-central. In the last century often narrowed to a long simple vowel at least before consonants and when unstressed, but now increasingly monophthongal in any situation especially among younger speakers. 
19. / eə / as in hair. A centring diphthong starting about half-open front. In the last century it was latterly generally realised as a long simple vowel  before consonants, when unstressed, and when stressed but in a structural word. At present the monophthongal version has become so nearly universal among middle-aged and young speakers that it is arguably more suitable to re-classify the phoneme as a GB simple vowel. The ODP (Oxford Dictionary of Pronunciation for Current English 2001) uses the notation /ɛː/ for it as does the leading EFL textbook Practical Phonetics and Phonology by B. S. Collins and I. Mees.
20. / ʊə /  as in pure. In the last century this phoneme was mainly a centring diphthong starting about semi-half-close to half-close with back- centralised-to-central medium rounding but often narrowed to a long simple vowel before consonants and when unstressed. In the present century among younger speakers it has begun to be widely monophthongised in all situations also among a monority losing its rounding and becoming so open as to resemble so much the mainly unstressable schwa phoneme /ə/ as to merge with it.

It is quite an adequate description of a GB diphthong to say where it begins, whether narrow or wide, and in which of the four areas close-front, close-central, close-back and central it terminates by referring to it as front-closing etc or centring. Diphthongs generally begin in approximately the same place as one or other of the simple vowels of the language is to be found.

 The diphthongs of any language generally constitute a much less complicated and less stable system than its simple vowels. The exact number of diphthongs a language may be said to possess is, as we see in respect of English, often open to debate. One reason for this is that some phonemes may as we have seen be equally validly designated as long vowels or as diphthongs. Another is that there may be doubt as to whether two successive vowel sounds should be considered as separate phonemes or as constituting a diphthong. The sequence in the English word ruin is a case in point. A third important reason for lack of certainty in classifying diphthongs is that the number of words in which a particular diphthong occurs may be so very limited as to make it difficult to come to a decision on its status. Also the words may be so rarely used or so exclusively learnèd in character that it is doubtful whether they can properly be recognised as ‘naturalised’.    
  The most useful purpose for studying diagrams of the English diphthongs is to appreciate the relationships especially of their starting-points and lip-conditions relative to the positions and lip-conditions of the English simple vowels.
  Only /oʊ/ of the General American diphthongs differs noticeably from the usual GB value in having for many speakers a rounded and more back beginning. The most common GB type is, however quite common among Americans; a version beginning front of centre is very rare there and becoming increasingly conspicuous in the UK. The diphthongs traditionally represented in GB as /ɪə/ and /ʊə/ are by American phoneticians usually analysed as the sequences /iːə / and /uːə/ when they do not correspond to r-spellings as in /ɪr/, /ʊr/. Such an analysis would be perfectly reasonable for GB. Since in GA pairs of words like Mary and merry are not usually distinct, GA /er/ corresponds to both /eər/ and /er/ of GB.    

Lip Conditions and Phonetic Correction    
II.12  Although the characteristic lip-posture for a target vowel may be rounded, a teacher should be careful never to let the appearance of students’ lips prompt comment. The sound quality alone should be the criterion. There are compensating adjustments possible within the vocal tract by which many speakers are able with visibly unrounded lips to produce rounded sound quality. The Japanese high back vowel seems by many speakers to be made somewhat auditorily rounded without being obviously visibly so.     

Vowel Practice Sentences
 1. 'Each of  'these| is 'equally 'easy to re`peat.
 2. It’s in `ink, `isn’t it?
 3. 'Fred gets his `head wet.
 4. He 'has that bad `back of his.
 5. Father was `calm at the `ˏstart.
 6. Tom’s got a 'lot of 'long `jobs to do to`ˏmorrow.
 7. `George| poured `water all over it.
 8. He 'wouldn’t even `look at a good `ˏbook.
 9. We `soon `ˏknew | it was our `duty to do it.
10. `Someone up a`ˏbove | 'must be having `fun.
11 `ˏShirley | was Herbert’s 'first  `girl-friend.
12. They `had ba`nanas about a `ˏminute or so a˚go.
13. They’d 'waited and `waited |for 'days and `days.
14. `Oh, `no. `Don’t go home by `ˏboat, Joe.
15. 'Why is it 'tied `quite so `tight?
16. 'How had they 'found `out about it?
17. `ˏJoyce 'coyly a 'voids | employing `boys.
18. It’s not `nearly as `serious as they `ˏfear.
19. I `daresay  `ˏpears | are 'fairly  `scarce  `ˏthere.
20. `During a `European `ˏtour, | he di'scovered the `cure.