Compound vowel-sounds are called Diphthongs. The words in the following columns illustrate the sixteen simple or elementary vowel-sounds employed in our English speech. Of these sixteen vowel-sounds, eight are long and eight are short.

In studying this list, direct your attention to the vowelsound of each word. The ways in which we represent these sounds in spelling are various, and, from our present point of view, unimportant. Thus, for example, the vowel-sound of fur appears also in herd, firm, work, learn, myrtle: the vowel-sound of pet appears also in head, many, bury, says, heifer, friend, guest.

The Obscure or Natural Vowel a. The vowel-sound exemplified by the a in attend, at the top of the column of short vowels, frequently occurs in English words, but only in unaccented syllables. It may be heard at the end of villa, sudden, cupboard, in the middle of miracle, tendency, harmony, at the beginning and end of America, grammarian, verandah. It is called the Obscure vowel, or the Natural vowel,natural, because the sound is produced with the minimum of effort. Hence boys who get into difficulties over their construing, and orators who are at a loss to proceed with their speeches, occupy the intervals with this sound. This natural vowel bears a closer resemblance in sound to the vowel-sound of putty or fur than to that of pat, with which it has often been wrongly identified.

Observe that many words written with an r at the end are pronounced in the south of England with this Natural vowel in place of the r. Say the words hair, here, poor, our, by themselves or when followed by words beginning with a consonant, and you will hear no r. On the stage and in the comic papers this substitution of the Natural

vowel for is caricatured, when the dandy is represented as saying 'De-ah me! What a bo-ah!' If however final r is followed by a vowel, it has its consonantal sound. Compare, e.g., dear aunt and dear me, poor Ellen and poor Tom.

52. Diphthongs are blends or combinations of two vowel-sounds which are run together in pronunciation. At this point great care is needed not to be misled by the diphthongs of print, æ, æ, neither of which, in our English pronunciation, is a true diphthong at all. The a of Cesar is no diphthong in sound; it is the same as the pure vowelsound of feet. So is the œ of fœtid. The ea in lead, ie in field, ei in receive, are none of them true diphthongs: they are only more or less clumsy ways of showing the length of an elementary vowel-sound.

The true diphthongs in English,-those in which two vowel-sounds are run into one,‚—are five in number: viz.

i in fine: this is a blend of the a in German mann,a sound of a which is extinct in modern English except provincially, and of the i in pit. The blend of the a in father with the i in pit gives us the broader diphthongal sound heard in aye, when we say 'The Ayes have it.'

oi in noise this is a blend of the vowel-sounds in fawn

and pit.

ou in house: this is a blend of the vowel-sounds in father and put.

u in use this is a blend of the vowel-sounds in pit and fool.

In a drawling pronunciation it is possible to detect the elementary vowel-sounds which form the diphthongal blend. Persons of defective education will talk of 'a bee-ootiful baw-ee,' or 'a na-ice ha-use,' when they mean 'a beautiful boy' or 'a nice house.' It should be specially observed that although the i in fine is a single letter, it is diphthongal in sound, and the same is true of the u in use. These diphthongal sounds can be represented in many other ways. Thus i is heard in try, die, dye, sigh, guide, buy, aisle, eye. Oi is expressed by oy in boy, by uoy in buoy. Ow or ough often occurs instead of ou. Diphthongal u is variously written as ue (sue), ui (suit), eu (feud), you, yew, ewe.

Consider next the vowel-sounds of fate and foe. It is undeniable that these are really diphthongal. In each case the vowel with which we start glides into a different vowel with which we close. Thus we pronounce fate as fay-eet or fay-it and foe as fo-oo. If you question this statement run up or down the scale singing fate or foe and note the result. Unless you have been taught singing by a good master, before you have reached half way the vowel-sound which you are producing will be ee or oo. Fay-eet and fo-oo are blends as complete as na-ice or ha-use. And hence some authorities class the vowel-sounds of fate and foe with the true diphthongs. There is some convenience however in placing them in the list of elementary vowel-sounds while recognising that in the standard speech of southern England they have acquired a diphthongal character. For when they occur at the end of a syllable which is followed by another syllable their sound is almost if not quite pure. Thus in fa-tal, la-dy, na-vy, and in foe-man, no-ble, po-ker, the secondary vowel-sound, which is prominent in fate and foe, is scarcely perceptible. Our English tendency to turn long vowels into diphthongs makes it a difficult matter for us to acquire the right pronunciation of such words as été and drôle in French, or geh and so in German, for in French and German these vowel-sounds are pure. But the feat, though difficult, is not impossible.

The reader may have felt surprised at finding in the list of short vowel-sounds the o of pillow. That this o differs from the o in pot is obvious enough, but he may have been inclined to identify it with the o of foe. As we have just seen, the o of foe finishes in the sound of oo: now the oo element is almost inaudible in the o of pillow. This short o occurs only in unaccented syllables, whether at the beginning of a word, as in omit, in the middle, as in proceed, or at the end, as in pillow. To substitute the Natural vowel for this final o and say fella, winda, instead of fellow, window, is a vulgarism.

In dealing with the letter we pointed out that its characteristic trilled sound is heard in English only when the r is followed by a vowel in the next syllable or the next word. And in dealing with the Natural vowel we saw that an untrilled r,-an r followed by a consonant,-is often replaced by this vowel-sound. When we pronounce the word fair, what we really say is fae-a, the Natural vowel taking the place of r. It is only in words such as fair-y, fair-est, car-ing, bear-er, words with a trilled r, that the pure long vowel-sound is heard. A similar substitu, tion takes place when we say beer, boar, boor; our actual pronunciation is bee-a, bo-a, boo-a. In each case we begin with one vowel-sound and end with another. But the blend is not complete. The component parts remain distinct. You will find many lines in Shakespeare in which such words as fire and dear form two syllables, but no actor could make more than one syllable of a word containing a true diphthong,

such as fight and doubt. We may therefore call these combinations Imperfect Diphthongs.

53. We have now enumerated 23 pure consonants, 16 pure vowels, 5 true diphthongs, and 4 imperfect diphthongs. Of our pure vowels two would be placed by some authorities amongst the diphthongs. Adhering however to the scheme adopted in the preceding pages, we give 39 as the sum-total of elementary sounds in English as spoken to-day.

Now if we run over the letters of the alphabet, we shall see that some of them find no place in our classification. The following letters are absent from the list :-c, q, j, x. Why is this?

The letter c is absent because it represents no sound in English not already represented by k, s, or sh. Cat is pronounced precisely as kat would be pronounced, city as sity, special as speshal. Thus the letter c is superfluous.

The letter q occurs only before u and, in combination with it, represents the sound of k+w, a compound, as in queen, or, more rarely, the simple sound of k, as in quay, cheque.

For a different reason we reject the other two letters. They do not stand for simple or elementary sounds at all, but represent compounds. So

j is a combination of d + zh,

X

k+s in excel, or of g+z in exert.

Notice that not only can these sounds be represented by a combination of letters, but they ought to be represented thus. For it is the business of the alphabet to furnish us with separate signs for simple sounds but not for compound sounds. There is no more reason why we should have a shorthand symbol in one letter to express k+s than there is reason why the alphabet should supply

us with a shorthand symbol for and. Such a symbol we do indeed possess in the form &, but we do not regard this symbol as a letter of the alphabet, and nobody but an American humourist would employ it in spelling other words, writing 'h&some' for handsome and 'underst&' for understand. The like criticism applies to the compound sound represented by j. The objection may be raised that, if x is rejected because it can be represented by k + s, we ought to get rid of ƒ because it can be represented by p + h, and that we might spell fife, phiphe, just as we spell philosophy with a ph. But the cases are quite different. The sound of ƒ is not a compound of p+h. It is a simple sound, and it is entitled to a separate letter. It is the use of the ph for f which is open to censure from the alphabetical stand-point. We use the ph because the words containing it come from the Greek, but if we spelt according to sound, the ph would disappear, and we should write filosofy instead of philosophy.

54. The following points connected with the subject of sounds in English deserve attention:

(1) Two mutes of unequal degrees of sharpness and flatness cannot be sounded together in the same syllable; or, if we employ the terms which we saw reason to prefer, a sonant and a surd in juxtaposition cannot be pronounced in the same syllable. We may write them together, but to sound them both as they are written is impossible. It is important to notice this, because sonants and surds often are thus written together, when we form the plurals of nouns or the past tenses of verbs. The ordinary way of making plurals is to add -s to the singular. Now s is a surd mute. Add s to a noun ending in a surd sound, e.g. pat, and the result can be pronounced as it is written, pats. But add s to a noun ending in a sonant sound, e.g. pad, and the result cannot be pronounced as it is written, pads. What we do pronounce is padz, two sonants. We naturally make the ending s give way and turn it into %, instead of preserving the s and changing the last letter of the word into t, as this latter course would alter the meaning of the noun. If we try the experiment with other nouns ending in sonant letters, e.g. hog, slab, we shall find that it is beyond our power to keep the sound of the g or b and at the same time to give its proper sound to the final s. We must say either hogz, slabz, or hocks, slaps. The same principle is seen

W. E. G.