vestigating the correctness of these conclusions by attending to the positions of the organs while enouncing the vowel elements, some may experience a little difficulty from failing to regard two important facts, which here deserve notice. First, in the English language, in particular, some of the simple sounds receive a modification in some instances from the action of the lips and tongue. This is true of the element oo, when distinctly and fully uttered. In undertaking to form it, the lips will be protruded and the breath will be forced through them in a circular form. The English 00, when accented and not followed by a mute, receives from this conjoint action of the lips a peculiar roundness and distinctness. This, however, is not essential to the element. It may be perfectly formed without any action of the lips, and is actually so formed in unaccented syllables and when followed by a mute, as in bistoury, root. The same is true of the short i, as heard in pin. The sides of the tongue are drawn up, sometimes, against the teeth, giving the element something of a consonantal quality. Both the and oo, when commencing a syllable, have this peculiarity, as in one, wo, your, al-ien. Indeed, there is a strong propensity in the formation of all or nearly all the vowel elements, when they are to be made prominent, as in pure and accented syllables, to imitate the action of the articulating organs. But this, it should be remarked, is not essential.

Again, some of the English vowels are diphthongal. Of these, one is always so; others only occasionally. The long i in pine is always so;-the organs in the larynx evidently moving, in forming it, from a position near that in which the a in father is formed to that in which short i is produced. The elements occasionally diphthongal are a in fame, which commences with a sound peculiar to itself, and terminates with that of e in mete; and o in bone, which commences with the sound of o in colt, and ends with that of oo.

We are now prepared to solve a problem which has exceedingly puzzled English orthoëpists. It respects the power of the vowels in unaccented syllables. Concerning these, Mr. Dupon

confidence can be placed in the derivations of words from one language to another, when this relation of the sounds is entirely overlooked, and the signs are regarded exclusively? If languages were transmitted by writing, and not by speech, this would be safe; but not otherwise.

ceau has well remarked: "There is nothing so difficult for the ear to take hold of and correctly to discriminate, as the short sounds of the English unaccented vowels. The principal characteristics of our language are strength and rapidity. The voice does not act by pressure on accented syllables, as it does in the Italian and Spanish, resting upon them awhile so as to fall gently on those that are unaccented and give them their correct articulation, but strikes with sudden force on the accented vowel, and, impelled by the momentum which it gives to itself, rolls on rapidly through the unaccented syllables to where it is obliged to renew its stroke. Hence our accented vowels are in general short, and those unaccented are passed over with so much quickness that the vocal organ does not dwell upon them long enough to enable a common ear to catch their precise sound; and it perceives only an indistinct vibration, a small vacant space, as it were, between the consonants, like the sheva of the Hebrews and the French e feminine." "This sheva the English phonologists have almost uniformly represented by u short." "Thus altar, cancer, honor, martyr, when their pronunciation is to be explained, will be spelled, for demonstration's sake, altur, cansur, honur, martur, as if the vowel sound of the last syllable in all of them were the same. But the similarity is nothing, in my opinion, but a deception produced in the ear by the rapidity of the voice passing over the unaccented vowel." These observations, although perhaps somewhat vague, are in the main, and so far as they have meaning, correct, and attest the accuracy of the author's ear. It is, certainly, a decided mark of vulgarity to confound the vowel sounds in such cases. Yet, it is true, that these sounds in unaccented syllables, are not as fully and distinctly articulated, as they are when under the accent. The following observations, it is believed, will explain the manner in which the pronunciation should be given.

In the first place, as has been seen, some of our vowel sounds are occasionally diphthongal. These, in the quick enunciation of unaccented syllables, lose one of their constituents and become simple or monophthongal. Again, although the vowel sounds are, as has been observed, in a sense independent of the articulating organs, properly so called, yet, in the English lan

* Amer. Phil. Trans. ubi sup.

guage, to help out the sounds and make them more distinct, those organs are sometimes called into action. This is always the case when the vowels are in pure and accented syllables, as in aw-ful, no-tion. In aw, for instance, the mouth is opened wide, and its cavity very much arched. In the same element in the word inauspicious, however, the cavity is not enlarged more than it is in uttering the other elements of the word. Similar remarks are applicable to, perhaps, all of the other vowel sounds. Thus we have the general law, that each of these elements has both a simple and unarticulated, and also an imperfectly articulated power; the former occurring in rapid enunciation, the latter in pure and accented syllables, and, also, in impure, unaccented syllables, when the enunciation is slow and distinct. The peculiar distinctive force of the element, however, remains the same in both cases. And correct speakers will never, therefore, confound the vowel sounds in the last syllables of such words as altar, cancer, honor, murmur, petal, level, carol.

These remarks will apply to most cases of this description. There is, however, one other class of syllables where the peculiarity of the pronunciation is to be explained on another principle. Few speakers, who regard at all correctness of pronunciation, for instance, would give the e in the final syllable in government its proper sound as heard in met. It is not, however, entirely silent, as is the case in such words as listen, heaven. There is a sound distinctly perceptible between the m and the n. What it is, may easily be explained on referring to the fact before mentioned, that, in passing from one element to another in the same syllable, the voice continues to flow out uninterruptedly. In this particular case, it is evident that between the m and the n, the organs separate; the voice is unimpeded in its passage out of the mouth, and consequently the sound must possess a vowel character. Yet it cannot be any proper element of the language; it is at most only an approximation to one. It certainly is the furthest possible from the element u in but. The same is true in every syllable where, in the transition from one letter to another, the articulating organs separate from all contact with any part of the mouth. If the voice continue to issue, it is evident it must bear the character of a vowel sound. Such is the case in the final syllables of words like tremble, terrible. In the last syllable of tendon, the movement of the or

gans, in dropping the tip of the tongue after the d is formed, to give the vowel sound, and then raising it again to form then, is easily perceived. It is plain, also, that this vowel sound is neither short o nor u, nor any other of the proper vowel ele

ments.

One other fact in relation to the vowel elements deserves to be noticed. It is that some of them are affected by quantity; and that when protracted, the sound is somewhat more open than when short. Thus the broad a sound in inaugurate is longer and somewhat more open than in inauspicious; in nor than in not. While in mock, cross, lost, and the like, it is of a medium quantity, being neither so long as in mawkish, nor so short as in rock. The element oo, likewise, is longer and more open in pool than it is in took.

In the English language, as in many others, the vowel elements are often found in combination in the same syllable. The oo and the short i thus, as has been before intimated, frequently precede other vowels. But for some of these vowel compounds, sometimes, but improperly, called diphthongs, we have peculiar characters appropriated; as the u long in tube, which is composed of short i and oo as heard in took; ou, as in route, compounded of a in father and oo; and oi in toil, compounded of a in all and short i in pin. The slide in passing from one organic position to the other in the formation of these compound sounds, it should be observed, gives them an effect upon the ear somewhat different from what would be produced by the two constituents alone. Being thus compounded of two other elements, they are not regarded as elements themselves, as is i in pine; since this is supposed to have for its first constituent a sound different from that of any other element.

To this class of elements belong, as another species, the mutes represented by h and wh. Although there may appear to be some impropriety in the name, if the etymology be regarded, yet both the principle of classification which we have adopted, and, also, considerations of convenience, sustain us in denominating these vowel-mutes. They consist of mere aspiration, and are formed, like the vowels proper, without any contact between the articulating organs and other parts of the mouth. They are, in fact, the vowels u and oo with the vocality suppressed; and cannot in whispering be distinguished from these vowels, except that, perhaps, the breath is sometimes more forcibly ex

pelled in them, probably from habit, than in the corresponding vowels. It is obvious that nothing forbids the indefinite multiplication of the vowel mutes in a language, but the difficulty of distinguishing them, which is greater here than in the case of the vocals. It will not appear strange, either, that different languages should select different vowel mutes from those which, are foune in our own language.

We have before remarked, that within certain limitations which were mentioned, the number of possible vowel elements may vary ad libitum. The vowel sounds attained in the way described may each be doubled, by causing the vocalized breath to pass through the nostrils instead of the lips. We may have thus a set of pure vowels, and a corresponding set of nasal vowels. The French language has, in fact, four of these nasal vowels, expressed by an, in, on, and un. That they are but the common vowels nasalized, is evident from the fact that they are formed from the others simply by causing the breath to pass through the nostrils, in a manner precisely similar to that in which the m, the n, and the ng are formed from the b, the d, and the hard g respectively.†

The other class of alphabetical elements, denominated consonants, are susceptible of a subdivision into two species. The first consists of those in forming which the articulating organs, by being brought into contact with various parts of the mouth, but partially obstruct the passage of the breath through the lips; the other of those which entirely occlude it. The first may hence be denominated the partial, the last, perfect consonants.

Of the partial consonants, the English language has none but those which are formed by the action of the two articulating organs, the tongue and the lips. Palatal partial consonants are, however, found as mutes in various languages, as the

* We are aware that the wh, as heard in when, is by some regarded as compounded of the aspirate h and the vowel oo. But a diversity of independent considerations, all leading to the same result, force us to the view presented in the text. We think this will be admitted by those who will carefully observe the position of the organs and the sound when pronouncing when, both audibly and in a whisper, and also when uttering the supposed combination h-wen.

Have we not the vowel e in pen, nasalized in the collo. quial eh?