From a general history of the English language, and some remarks upon that subject, I proceed to examin its elements, or the powers of the letters which compose our alphabet.

There are in English, twenty five characters or letters which are the representatives of certain sounds, either simple or combined; a, b, c, d, e, f, g, i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, x, y, z. The English have also the character h, which marks an aspiration or strong breathing, but has very little sound of its own.[G]

Letters, according to the sounds they represent, or the purposes they serve, are very naturally divided into three kinds; vowels, dipthongs, and consonants.

In order to obtain clear ideas of our alphabet, let us attend to the following definitions:

1. A vocal sound, formed by opening the mouth, and by a single position of the organs of speech, is a simple sound or vowel. Most of the vowels in English are capable of being prolonged at pleasure, without varying the position of the organs.

2. No more than one simple sound can be formed by one aperture of the mouth, and one position of the organs of speech. The only difference that can be made with the same position of the organs, is, to prolong and shorten the same sound.

3. Two simple sounds, closely united in pronunciation, or following each other so rapidly that the distinction is scarcely perceptible, form a dipthong. In pronouncing a dipthong, two positions of the parts of the mouth are required.

4. Those letters which are not marks of articulate sounds, but represent indistinct sounds, formed by some contact of the parts of the mouth, or by compressing those parts, check all sound, are denominated consonants.

By the first definition we ascertain the number of vowels in English. In pronouncing each of the letters a¹, a⁴, a³, e¹, o¹, o⁶, u², we observe but one position or aperture of the mouth; the sounds are therefore simple, and the letters are called vowels. The six first sounds are capable of being prolonged at pleasure.

By the second definition, we determine which sounds are the same in quality, and different only in the time of being pronounced. Thus i in fit has the same quality of sound as ee in feet, for both are pronounced with the same disposition of the organs; but the first is the shortest articulation of the sound, and the last, a long or grave articulation. The other vowels have also their short or abrupt sounds; a in late has its short sound in let; a in cart has its short sound in carry; a in fall has its short sound in folly; oo in fool its short sound in full. O is sometimes shortened in common parlance, as in colt; but the distinction between o in coal and colt, seems to be accidental or caused by the final consonant, and not sufficiently settled or important to require a separate consideration.

By the third definition we are enabled to ascertain the dipthongs in our language. The letters i, u and y are usually classed among the vowels; but the first or long sound of each requires, in pronunciation, two positions of the organs of speech, or rather a transition from the position necessary to form one simple sound, to the position necessary to form another simple sound. We begin the sound of i nearly with the same aperture of the glottis, as we do the broad a or aw: The aperture however is not quite so great: We rapidly close the mouth to the position where we pronounce ee, and there stop the sound. This letter is therefore a dipthong. Y has no property but what belongs to i.

U also is not strictly a vowel; nor is it, as it is commonly represented, composed of e and oo. We do not begin the sound in the position necessary to sound ee, as is obvious in the words salute, salubrious, revolution; but with a greater aperture of the mouth and with a position perfectly easy and natural. From that position we pass to the position with which we pronounce oo, and there close the sound.

It must however be observed that when these letters, i, u, are followed by a consonant, the two sounds of the dipthong are not clearly distinguishable. We do not, in fight, hear the sound of ee; nor the sound of oo in cube. The consonant compresses the organs and closes the sound of the word so suddenly, that the ear can distinguish but a simple vocal sound: And notwithstanding these letters are dipthongs, when considered by themselves, yet in combination with consonants, they are often marks of simple sounds or vowels.

The short sound of i and y, is merely short ee. The sound of u in tun, is a separate vowel, which has no affinity to any other sound in the language.[H]

The sound of oi or oy is dipthongal, composed of the third or broad a, and ee. The sound of ou or ow is also dipthongal, compounded of third a and oo. The sound however does not require quite so great an aperture of the mouth as broad a; the position is more natural, and the articulation requires less exertion.

The union of a and w in law, has been very erroneously considered a dipthong. Whatever might have been the ancient pronunciation of these letters (and it is probable that good reasons operated to produce their union) they now exhibit but one simple vocal sound. The same may be observed of ee, oo, au, ai, ea, ei, ie, eo, oa, and perhaps some other combinations, each of which actually exhibits the sound of one letter only, which sound is as simple as that of a or o.[38]

Under the head of dipthongs we may perhaps range wa, we, wo, wi, &c. W has nearly the short sound of oo; for will, dwell are pronounced as if written ooill, dooell. It is a controverted point, whether w should be classed with the vowels or consonants. I shall only observe, that it is pronounced by opening the mouth, without a contact of the parts; altho, in a rapid pronunciation, it approaches to a consonant.[I] It is however very immaterial, whether we class it with the vowels or consonants; as all grammarians agree that its sound is that of oo short. It ought to be named oo or we; which would save children much of the trouble they now experience, in learning its proper sound from that awkward name double u.

The sound of y in the beginning of words, is, by some writers, called a vowel, but by most of them a consonant. Lowth has asserted, that it has every property of a vowel and not one of a consonant. Sheridan considers y in youth, year, &c. as the short ee. But if these writers would attend to the manner in which we pronounce yes, ye, they would acknowlege that y has some property different from ee; for it is very evident that they are not pronounced ee-es, ee-e. The fact is, that in the American pronunciation of y, the root of the tongue is pressed against the upper part of the mouth, above the palate, more closely than it is in pronouncing ee, and not so closely as in pronouncing g hard. The transition however from y to ee or to g, is extremely easy, and hence the mistake that y is short ee, as also the convertibility of y with g.[J] It appears to me that y in the beginning of words, is more clearly a consonant than w.

In many words, i has the power of y consonant; particularly after l and n; as filial, union.

The vowels therefore in English are all heard in the following words; late, half, hall, feet, pool, note, tun, fight, truth. The five first have short sounds or duplicates; which may be heard in let, hat, hot, fit, pull; and the letters i and u are but accidentally vowels. The pure primitive vowels in English are therefore seven.

The dipthongs may be heard in the following words; lie or defy, due, voice or joy, round or now. To these we may add ua in persuade; and perhaps the combinations of w and the vowels, in well, will, &c.

The consonants in English are nineteen; but for want of proper characters, five of them are expressed or marked by double letters. We annex two sounds to th; one to sh; one to ng; and one to si or su, as may be heard in the following words; think, this, shall, bring, confusion or pleasure. These characters should be called eth, esh, eng, ezh; and th should have two names, the aspirate as in think, and the vocal as in this; the latter sound might be distinguished by a small mark drawn thro th. This improvement is so obvious and easy, and would be so convenient for the learners of the language, that I must believe it will soon be introduced.

The consonants may be divided into mutes and semivowels. When a consonant compresses the lips, or the tongue and roof of the mouth, so closely as to check all sound, it is called a perfect mute: Such are p, k, and t, as may be perceived by pronouncing the syllables, ep, ek, et. When the compression of the organs is more gentle and does not stop all sound immediately, the letters are called mutes; such are b, d, and g, as may be perceived by pronouncing the syllables, eb, ed, eg. When a consonant has an imperfect sound, or hissing, which may be continued, after a contact of the organs, it is denominated a semivowel. Of this kind are ef, el, em, en, er, es, ev, ez, eth,[39] eth,[39] esh, ezh, ing. Of these, four are aspirates, ef, es, eth, and esh. The others are vocal, having an imperfect sound.

The whole may be thus arranged.

Perfect mutes—p, k, t.
Mutes—————b, d, g.
vocal, } l, m, n, r, v, z, th,
Semivowels— } zh, ng,
aspirate,} f, s, th, sh.

They may also be classed according to the manner in which they are formed by the organs: Thus, those formed

By the lips, are called labials—b, p, f, v.
By the teeth, are called dentals—d, t, th, z, s, sh, zh.
By the palate, are called palatine—g, k, l, r.
By the nose, are called nasal—m, n, ng.

On the subject of the alphabet, I have this remark further; that for want of a proper knowlege of the powers of sh and th, some material errors in printing have obtained in common practice. Sh are usually united in printing, and generally with propriety, for the combination represents a simple consonant. But in several compound words s and h have been improperly united, where one is silent or where each retains its own power, as in dishonor, dishonest, dishabille, hogshead, household, falsehood, and some others. The union of sh in these words, is embarrassing, especially to children, who are led to pronounce them dish-onor, dish-onest. This error still prevails in printing, except in the last mentioned word, which is sometimes correctly printed falsehood.

Th, tho not united in character, have a tendency to produce, in some words, a wrong pronunciation. For instance, we are very apt to say Wren-tham instead of Wrent-ham. Hotham is also ambiguous; there is nothing in the orthography to direct us, whether to pronounce it Hot-ham or Ho-tham, altho custom decides in favor of the latter.

These remarks show the propriety of attending to our orthography, and of attempting to remove causes of error, when it can be done without much trouble or danger of giving offence.

RULES of PRONUNCIATION.

Having briefly explained the English alphabet, I proceed to the rules of pronunciation.

In pronunciation, two things demand our notice; the proper sounds of the vowels and consonants, and the accent.

In pronouncing both vowels and consonants, the general rule is, that similar combinations of letters should be pronounced alike, except when general custom has decided otherwise. Thus if i in the words, bind, find, mind, has its first sound, it ought to have the same sound in other similar combinations, kind, blind, grind. This is the rule of analogy, the great leading principle that should regulate the construction of all languages. But as languages are not formed at once by system, and are ever exposed to changes, it must necessarily happen that there will be in all languages, some exceptions from any general rule; some departures from the principle of uniformity.

The practice of a nation, when universal or ancient, has, in most cases, the force and authority of law; it implies mutual and general consent, and becomes a rule of propriety. On this ground, some deviations from the analogy of construction and pronunciation must be admitted in all languages. Thus from the analogy already mentioned, wind is an exception; for general practice has determined that i should, in this word, have its second or short sound.[40] Whether this deviation was admitted at first to distinguish this word from the verb to wind, or whether there were other good reasons which cannot now be explored, or whether it was merely the work of ignorance or accident, it is unnecessary to enquire; the common consent of a nation is sufficient to stamp it with propriety.

Another rule in English, which admits of no exception, is, when the accent falls on a vowel, it is long, as o in ho´-ly; but when the accent falls on a consonant, the preceding vowel is short, as in flat´-ter.

It is also a general rule, that when a consonant closes a syllable, the preceding vowel is short, as in fan-cy, habit; altho this rule has its exceptions, as Cam-bridge, dan-ger, and perhaps man-ger.

From this rule, the English except also a²ngel, a²ncient. In this all the standard authors agree, except Kenrick and Burn, who mark a in ancient both long and short. The English pronunciation is followed in the middle and southern states; but the eastern universities have restored these words to the analogy of the language, and give a its second sound. It is presumed that no reason can be given for making these words exceptions to the general rule, but practice; and this is far from being universal, there being many of the best speakers in America, who give a, in the words mentioned, the same sound as in anguish, annals, angelic, antiquity.

The practice of the eastern universities therefore should be encouraged, rather than discountenanced; as it diminishes the number of anomalies. I shall only remark further, that a in these words must formerly have had its third or fourth sound; which is evident from the old orthography; for angel, at least, was spelt like grant, command, &c. aungel, graunt, commaund. In giving a its first sound therefore, the modern English have not only infringed the rule of analogy, but have deviated from former practice.

In the word chamber, a has its fourth sound. It is necessary to remark this; as there are many people in America, who give a its first sound, which is contrary to analogy and to all the English authorities.

With regard to accent, that particular stress of voice which should distinguish some syllable of a word from others, three things are to be considered; the importance of the syllable; the derivation of the word; and the terminating syllable.

The importance of a syllable is discovered by resolving a word into the parts which compose it, or reducing it to its radicals. Thus sensible is derived from sensus in Latin or sense in English. The first syllable therefore is that on which the meaning of the word principally depends; the others being an accessary termination.

The first syllable then is the most important and requires the accent. For the same reason, admire, compare, destroy, &c. have the accent on the second syllable in preference to the first; the last syllables being all derived from verbs, and the first being mere particles.[41]

Another rule for laying the accent of words arises from derivation. Thus all words that take the terminations ing, ful, less, ness, ed, est, ist, ly, retain the accent on the syllable where it is laid in their primitives; as proceed, proceeding, wonder, wonderful, &c.

But the most important article to be considered in the accentuation of words, is the terminating syllable. From the different terminations of words arise various analogies, the most of which are enumerated in the first part of my Institute. The principle which has operated to produce these analogies, is the ease of speaking or the harmony of enunciation. Consequently this principle must take place of all others; and we find that it frequently interferes with the two foregoing rules, and regulates practice in opposition to both.

The general rule, grounded on this principle, is, that words, having the same terminating syllable, have the accent at the same distance from that termination. Thus all words ending in tion, sion, cion, cial, cian, have the accent on the last syllable but one;[42] and this without any regard to derivation or to the number of syllables in the word.

Thus most words in ty, if they consist of more syllables than two, have the accent on the antepenult; as probity, absurdity, probability. I recollect but two exceptions, viz. commonalty, admiralty; the accent of which is laid upon the first syllable, as in their primitives.[43]

But let us observe the force of the last rule, in opposition to the others. Mortal has the accent on the first syllable. Here the first rule takes place, for the first syllable, having mors, death, for its root, is the most important. But the derivative, mortality, conforms to the analogy of words ending in ty and has the accent on the last syllable but two. That the ease or harmony of pronunciation, is the cause of this change of accent, will be evident to any person who shall attempt to pronounce words of this class, with the accent on any other syllable than the antepenult.

Most of these rules admit a few exceptions, which are to be learnt by practice. Custom has made some inroads upon the rules of uniformity, and caprice is ever busy in multiplying anomalies. Still, rules will be of great service in ascertaining and fixing our language; for tho they may not root out old errors, they may prevent the introduction of others.

But besides the principal accent, there is, in most polysyllables, an inferior accent laid on the third or fourth syllable from the principal. Indeed in some words, the two accents are so nearly equal, as to be scarcely distinguishable.

It is denied by some critics that there are more accents than one, in any word. But the composition of words, and the ease of speaking, both require a plurality of accent in a very great number of instances; and our ears inform us that such a plurality actually exists in practice. If a man will assert that in such words as designation, exaltation, there is but one syllable distinguished from the others by a superior stress of voice, he must deny the evidence of sense, and would not listen to argument.

I must however remark that most, if not all syllables, derived from some important word, have some degree of accent:[44] So that in compounds, there are usually as many accents as radicals. Thus in sanctify, which is composed of two radicals, sanctus and fio, we observe two accents; the strongest on the first syllable. The same may be observed in magnanimity, from magnus and animus, in promogeniture, &c. except that in these the principal accent is on the third syllable.

Notwithstanding it is a general rule, that there are as many accents in a word, as radicals, yet one of them at least is frequently removed from the principal syllable, by the analogy of termination, which prevails over all other reasons. Thus in mathematics, the two accents lie on the proper syllables; but in mathematician, the last accent is removed to a less important place. In imperceptible, the principal accent, with propriety, lies on the third syllable, which being derived from a verb (capio) is the most important. The particle im, being the privative, or that syllable which changes the meaning of the whole word from affirmative to negative, becomes important and has some degree of accent. But in the derivative imperceptibility, while the first and third syllables retain an accent, the analogy of termination carries the principal accent to the fifth syllable, which is adventitious and less important than the others.[45]

In many compounds, as, earth-quake, rain-bow, each syllable is pronounced with the stress that belongs to accented syllables; and there is little or no distinction of accent. The reason is obvious: There is no difference in the importance of the syllables; both are equally necessary to convey the idea. By giving one syllable the whole accent, such a word loses its original meaning, or at least its force, as may be observed in the word hussy, a corruption of house-wife; which, from an affectation of a unity of accent, and a hasty pronunciation, has sunk into a low word. From the same ridiculous affectation, work-house is, by some people, pronounced work-us.

On this head, I shall only observe further, that some words of many syllables have three accents; of which we have an example in val'etu'dina'rian.

It has been already remarked that the composition of words, and the ease of speaking, require a plurality of accent. The reason why words of many syllables have two or three accents, is plain to any man that attempts to pronounce them without an accent.

We cannot pronounce more than two unaccented syllables with perfect ease; but four or five can hardly be articulated without an intervening accent. We glide over the unaccented syllables with such rapidity, that we have hardly time to place the organs in a position to articulate them. The difficulty is in proportion to the number: So that after passing over two or three, the voice very naturally rests or falls forceably upon a particular syllable. Hence the words most difficult to be pronounced, are those of four syllables, accented on the first; as figurative, literature, applicable. The difficulty is very great, when the middle syllables abound with consonants, even in trissyllables, as ag'grandize; but is itself a sufficient reason for not accenting the first syllable of such words as acceptable and refractory. When one of the words which have the accent on the first, and three succeeding unaccented syllables, is followed by two or three particles, the passage is weak and often occasions hesitation in a speaker; as "applicable to the affairs of common life."

A remarkable instance of this, we find in Priestley's Preface to Letters to a Philosophical Unbeliever; "Whether of a pleasureable or of a painful nature." In this example there are six weak syllables following each other without interruption, and such passages are not reduceable to any kind of poetic feet. This assemblage of unimportant syllables makes a hiatus in language, which should, as far as possible, be avoided by a writer; for the melody of prose consists in a proper mixture of important and unimportant syllables.[46][K]

DIFFERENCES of PRONUNCIATION and CONTROVERTED POINTS EXAMINED.

Having laid down some general rules reflecting pronunciation, I proceed to examin local differences, and the most material points of controversy on this subject.

In the eastern states, there is a practice prevailing among the body of the people, of prolonging the sound of i in the termination ive. In such words as motive, relative, &c. the people, excepting the more polished part, give i its first sound. This is a local practice, opposed to the general pronunciation of the English on both sides of the Atlantic, sometimes to the rules of accent, and always to derivation. In dissyllables, as motive, active, the genius of our language requires that the accent should be laid on one syllable, and that the other should be short.[47] But by prolonging i in the last, the distinction of accent is totally destroyed.

In polysyllables, which often have two accents, this reason has less force, but the derivation, which is from the French motif, relatif, always requires that i in the termination ive should have the sound of ee short, as in live, give. This is merely the short sound of the French i, and the consequence of the English accent on the first syllable. These reasons, with the authority of the most approved practice, should operate to discountenance the singular drawling pronunciation of the eastern people.[48]

The same reasons are opposed to another local practice of a similar nature in the middle states; where many people pronounce practise, prejudice, with i long. I know of no authority for this beyond the limits of two or three states; and it is clear that the practice is not warranted by any principle in the language.

Another very common error, among the yeomanry of America, and particularly in New England, is the pronouncing of e before r, like a; as marcy for mercy. This mistake must have originated principally in the name of the letter r, which, in most of our school books, is called ar. This single mistake has spread a false pronunciation of several hundred words, among millions of people.[49]

To avoid this disagreeable singularity some fine speakers have run into another extreme, by pronouncing e before r, like u, murcy. This is an error. The true sound of the short e, as in let, is the correct and elegant pronunciation of this letter in all words of this class.

There is a vulgar singularity in the pronunciation of the eastern people, which is very incorrect, and disagreeable to strangers; that of prefixing the sound of i short or e, before the dipthong ow; as kiow, piower or peower. This fault usually occurs after p, c hard, or those other consonants which are formed near the seat of ee in the mouth, or in passing from which to the succeeding vowel, the organs naturally take the position necessary to pronounce ee. But the most awkward countryman pronounces round, ground, &c. with tolerable propriety.

This, with some other peculiarities which prevail among the yeomanry of New England, springs from causes that do not exist, in the same degree, in any other part of America, perhaps not in the world. It may surprize those who have not turned their thoughts to this subject, that I should ascribe the manner of speaking among a people, to the nature of their government and a distribution of their property. Yet it is an undoubted fact that the drawling nasal manner of speaking in New England arises almost solely from these causes.

People of large fortunes, who pride themselves on family distinctions, possess a certain boldness, dignity and independence in their manners, which give a correspondent air to their mode of speaking. Those who are accustomed to command slaves, form a habit of expressing themselves with the tone of authority and decision.

In New England, where there are few slaves and servants, and less family distinctions than in any other part of America, the people are accustomed to address each other with that diffidence, or attention to the opinion of others, which marks a state of equality. Instead of commanding, they advise; instead of saying, with an air of decision, you must; they ask with an air of doubtfulness, is it not best? or give their opinions with an indecisive tone; you had better, I believe. Not possessing that pride and consciousness of superiority which attend birth and fortune, their intercourse with each other is all conducted on the idea of equality, which gives a singular tone to their language and complexion to their manners.

These remarks do not apply to the commercial towns; for people who are conversant with a variety of company lose most of their singularities, and hence well bred people resemble each other in all countries. But the peculiar traits of national character are found in the internal parts of a country, among that class of people who do not travel, nor are tempted by an intercourse with foreigners, to quit their own habits.[50]

Such are the causes of the local peculiarities in pronunciation, which prevail among the country people in New England, and which, to foreigners, are the objects of ridicule. The great error in their manner of speaking proceeds immediately from not opening the mouth sufficiently. Hence words are drawled out in a careless lazy manner, or the sound finds a passage thro the nose.

Nothing can be so disagreeable as that drawling, whining cant that distinguishes a certain class of people; and too much pains cannot be taken to reform the practice.

Great efforts should be made by teachers of schools, to make their pupils open the teeth, and give a full clear sound to every syllable. The beauty of speaking consists in giving each letter and syllable its due proportion of sound, with a prompt articulation.

Thus in order to pronounce cow, power, or gown with propriety, the pupil should be taught, after placing the organs in the position required by the first consonant, to open his mouth wide, before he begins the sound of ow: Otherwise in passing from that position to the aperture necessary to pronounce ow, he will inevitably articulate ee, keow.

A similar method is recommended to those polite speakers who are so fond of imitating the English stage pronunciation as to embrace every singularity, however disagreeable. I refer to the very modern pronunciation of kind, sky, guide, &c. in which we hear the short e before i, keind, or kyind, skey, &c. This is the same barbarous dialect, as the keow and veow of the eastern country people. Yet, strange as it may seem, it is the elegant pronunciation of the fashionable people both in England and America. Even Sheridan, who has laid it down as a rule that i is a dipthong, composed of aw and ee, has prefixed a y short to its sound in several words; as kyind, skyi, gyide, &c. We may with equal propriety prefix e to the dipthong ow, or to o in poll, or to oo in fool, or to any other vowel. It is presumed that the bare mention of such barbarisms will be sufficient to restrain their progress, both in New England and on the British theater.

Some of the southern people, particularly in Virginia, almost omit the sound of r as in ware, there. In the best English pronunciation, the sound of r is much softer than in some of the neighboring languages, particularly the Irish and Spanish; and probably much softer than in the ancient Greek. But there seems to be no good reason for omitting the sound altogether; nor can the omission be defended on the ground, either of good practice or of rules. It seems to be a habit contracted by carelessness.

It is a custom very prevalent in the middle states, even among some well bred people, to pronounce off, soft, drop, crop, with the sound of a, aff, saft, drap, crap. This seems to be a foreign and local dialect; and cannot be advocated by any person who understands correct English. [L]

In the middle states also, many people pronounce a t at the end of once and twice, oncet and twicet. This gross impropriety would not be mentioned, but for its prevalence among a class of very well educated people; particularly in Philadelphia and Baltimore.

Fotch for fetch is very common, in several states, but not among the better classes of people. Cotched for caught is more frequent, and equally barbarous.

Skroud and skrouge for croud, are sometimes heard among people that should be ashamed of the least vulgarism.

Mought for might is heard in most of the states, but not frequently except in a few towns.[M]

Holpe for help I have rarely heard except in Virginia. Tote is local in Virginia and its neighborhood. In meaning it is nearly equivalent to carry. I have taken great pains to discover the etymology of the local terms used in the several states; but this word has yet eluded my diligence.[51]

Chore, a corruption of char, is an English word, still used in many parts of England, as a char-man, a char-woman, but in America, it is perhaps confined to New England. It signifies small domestic jobs of work, and its place cannot be supplied by any other single word in the language.

These local words, and others of less note, are gradually growing into disuse, and will probably be lost: Except such as are necessary in some particular occupation.

The pronunciation of w for v is a prevailing practice in England and America: It is particularly prevalent in Boston and Philadelphia.[52] Many people say weal, wessel, for veal, vessel.

These letters are easily mistaken for each other, and the name of the letter w now used, is a proof that the letter v was formerly called u or oo. The letter in the Roman language had the sound we now give w in will. Via and vinum, pronounced wia, winum, have suffered but a small change of pronunciation in our way, wine. In old English books, down to Shakespear, v was written for the short u, as vp, vnder; for up, under. On the other hand, u was written where we now write v, as uery, euery, for very, every. It seems therefore, that v had formerly the sound of w or oo; and that instead of corrupting the language, the Cockneys in London, and their imitators in America, who say weal, wery, have retained the primitive pronunciation. In confirmation of this opinion, it may be observed that the Danes, who speak a dialect of the Saxon, have no w in their language, but where we write w, they write v, and where we write wh, they invariably write hv; as vind, wind; vej, way; vader, wade; hvad, what; hvide, white; hvi, why. The Germans, whose language is another branch of the same stock, invariably pronounce w as we do v; wall, vall; wir, vir, we; wollen, vollen, will; and v they pronounce as we do f; as vergessen, fergessen, which is the same as the English forget.

The retaining the old sound of v is a proof of the force of custom; but since the nation in general have annexed to it a precise sound, as well as to w, every person should resign his peculiarities for the sake of uniformity.

But there are some points in pronunciation, in which the best informed people differ, both in opinion and practice.

The words shall, quality, quantity, qualify, quandary, quadrant, are differently pronounced by good speakers. Some give a a broad sound, as shol, quolity; and others, its second sound, as in hat. With respect to the four first, almost all the standard writers[53] agree to pronounce a short, as in hat: And this is the stage pronunciation. It is correct, for it is more agreeable to the analogy of the language; that being the proper sound of the English a which is heard in hat or bar. With respect to the two last, authors differ; some give the first, some the second, and others the fifth sound. They all pretend to give us the court pronunciation, and as they differ so widely, we must suppose that eminent speakers differ in practice. In such a case, we can hardly hesitate a moment to call in analogy to decide the question, and give a in all these words, as also in quash, its second sound.[54]

The words either, neither, deceit, conceit, receipt, are generally pronounced, by the eastern people, ither, nither, desate, consate, resate. These are errors; all the standard authors agree to give ei, in these words, the sound of ee. This is the practice in England, in the middle and southern states, and, what is higher authority, analogy warrants the practice. Indeed it is very absurd to pronounce the verb conceive, conceeve, and the noun conceit, consate. Such an inconsistency will hardly find advocates, except among the prejudiced and uninformed.

Importance is, by a few people, pronounced impo¹rtance; with the first sound of o. The reason alleged is, that it is a derivative of import, and o should preserve the same sound it has in the original. It seems however to be affectation, for the standard writers and general practice are opposed to it. Indeed it may be considered as a mere imitation of the French pronunciation of the same word.

Decis-ive for deci-sive is mere affectation.

Reesin for raisin is very prevalent in two or three principal towns in America. One of the standard authors gives us this pronunciation; and another gives us both raisin and reesin. But all the others pronounce the word raisin, with a long; and derivation, analogy and general custom, all decide in favor of the practice.

Leisure is sometimes pronounced leesure, and sometimes lezhure: The latter is the most general pronunciation in America. It is almost singular in its spelling; seizure being the only word in analogy with it; and this is a derivative from seize. The true original orthography of leisure was leasure; this was in analogy with pleasure, measure, and its ancient pronunciation still remains.

Dictionary has been usually pronounced dicsonary; But its derivation from diction, the analogous pronunciation of tion in other cases, and all the standard writers require dicshunary, or dicshonary.

One author of eminence pronounces defile in three syllables, def-i-le. In this he is singular; neither general practice, nor rules warrant the pronunciation; and all the other authorities are against him.

With respect to oblige, authorities differ. The standard writers give us both oblige and obleege, and it is impossible to determine on which side the weight of authority lies. The direct derivation of the word from the French would incline us to prefer obleege, in the analogy of fatigue, machine, antique, pique, marine, oblique, which uniformly preserve the French i or English ee. Yet Chesterfield called this affectation, and it might be so in his age; for the opinions of men are capricious. The English analogy requires i long in oblige; and perhaps this should incline all parties to meet each other on that best principle.

Some people very erroneously pronounce chaise, sha in the singular, and shaze in the plural. The singular number is shaze, and the plural, shazes.

Our modern fashionable speakers accent European on the last syllable but one. This innovation has happened within a few years: I say innovation; for it is a violation of an established principle of the language, that words ending in ean have the accent on the last syllable but two: Witness Mediterra'nean, Pyre'nean, Hercu'lean, subterra'nean. I do not advert to an exception,[55] and why European should be made one, it is difficult to determine. The reason given by some, that e in the penultima represents the Latin dipthong Æ, which was long, is of little weight, opposed to the general practice of a nation, and to an established principle. The standard authors, in this instance, as in all others, where practice is not uniform, very absurdly give both pronunciations, that we may take our choice. As this is a very easy method of getting over difficulties, and passing along without giving offence, so it is a certain way to perpetuate differences in opinion and practice, and to prevent the establishment of any standard. Analogy requires Euro'pean, and this is supported by as good authorities as the other.

Rome is very frequently pronounced Room, and that by people of every class. The authors I have consulted give no light upon this word, except Perry, who directs to that pronunciation. The practice however, is by no means general in America: There are many good speakers who give o its first sound. It seems very absurd to give o its first sound in Romish, Romans, and pronounce it oo in Rome, the radical word. I know of no language in Europe, in which o has not one uniform sound, viz. the sound we give it in rose. It is perhaps the only vowel, in the sound of which all nations agree. In English it has other sounds; but the first is its proper one. A great proportion of people in America have restored the analogy of pronunciation in giving o its first found in Rome; and a desire of uniformity would lead us to extend the practice.[56]

In the pronunciation of arch in many compound words, people are not uniform. The disputed words are archangel, archetype, architecture, architrave, archives. There seems to be no settled principle of analogy, by which the question can be determined. Etymology would require ch, in Greek and Hebrew derivatives, to have uniformly the sound of k; but before most consonants, such a pronunciation is harsh; for which reason it is generally softened into the English ch, as archbishop. But before vowels, as in the words just enumerated, the best practice has decided for the sound of k; and euphony, as well as derivation, favors the decision.[N]

The sound of ch in chart is likewise disputed; and the standard authors are directly opposed to each other. There is as good foreign authority on one side as the other; but in America, ch has generally its soft or English sound. This must perhaps be preferred, contrary to etymology; for we uniformly give ch that sound in charter, which is from the same original; and this also distinguishes the word from cart; a reason which is not without its weight.

There are many people who omit the aspirate in most words which begin with wh; as white, whip, &c. which they pronounce wite, wip. To such it is necessary only to observe, that in the pure English pronunciation, both in Great Britain and New England, for it is exactly the same in both, h is not silent in a single word beginning with wh. In this point our standard authors differ; two of them aspirating the whole of these words, and three, marking h in most of them as mute. But the omission of h seems to be a foreign corruption; for in America, it is not known among the unmixed descendants of the English. Sheridan has here given the true English pronunciation. In this class of words, w is silent in four only, with their derivatives; viz. who, whole, whoop, whore.

One or two authors affect to pronounce human, and about twenty other words beginning with h, as tho they were spelt yuman.[57] This is a gross error. The only word that begins with this sound, is humor, with its derivatives. In the American pronunciation, h is silent in the following, honest, honor, hour, humor, herb, heir, with their derivatives. To these the English add hospital, hostler, humble; but an imitation of these, which some industriously affect, cannot be recommended, as every omission of the aspirate serves to mutilate and weaken the language.

The word yelk is sometimes written yolk and pronounced yoke. But yelk is the most correct orthography, from the Saxon gealkwe; and in this country, it is the general pronunciation.

Ewe is, by the English, often pronounced yo; which is sometimes heard in America. But analogy and the general corresponding practice in this country, with the authority of some of the most accurate writers, decide for yew.

The English speakers of eminence have shortened the vowel in the first syllable of tyranny, zealous, sacrifice, &c. altho in the primitive words, all agree to give the vowel its first sound. This pronunciation has not spread among the people of this country; but our learned men have adopted it; and it seems in some degree to be the genius of our language. In child, clean, holy, &c. we uniformly give the first vowel its long sound; but when a syllable is added, we always shorten it; children, clenly, holyday.

On the other hand, many people in America say pat-ron, mat-ron; whereas the English say either pa-tron or pat-ron, matron or mat-ron; but all agree in saying, pat-ronage. In patriot, patriotism, the English give a its long sound; but a great part of the Americans, its short sound. In all these cases, where people are not uniform, I should prefer the short sound; for it appears to me the most analogous.

Wrath, the English pronounce with the third sound of a or aw; but the Americans almost universally preserve the analogous sound, as in bath, path. This is the correct pronunciation; and why should we reject it for wroth, which is a corruption? If the English practice is erroneous, let it remain so; we have no concern with it: By adhering to our own practice, we preserve a superiority over the English, in those instances, in which ours is guided by rules; and so far ought we to be from conforming to their practice, that they ought rather to conform to ours.

It is disputed whether g should have its hard or soft sound, in homogeneous and heterogeneous: On this question the standard authors are not agreed. The hard sound, as in go, coincides with etymology; but analogy requires the other, as in genius. The same remarks apply to g in phlogiston.

In the middle and southern states, fierce, pierce, tierce, are pronounced feerce, peerce, teerce. To convince the people of the impropriety of this pronunciation, it might be sufficient to inform them, that it is not fashionable on the English theater. For those who want better proofs, before they relinquish their practice, I would observe, that these words are derived to us from the French; fierce, tierce, from fiers, tiers, and pierce from percer. In the two former, the French pronounce both i and e; but it is evident the English originally pronounced e only; for the i was omitted in the spelling of fierce, and was not introduced into pierce till after Spenser wrote.

Chaucer, Knightes Tale 1600.

Canterbury Tales.

Verses to Edmond Spenser.

Pierce is also made to rhime with rehearse. Pope makes it rhime with universe.

Essay on Man, 23.

The rhime in the last quotation, is not unequivocal proof of the pronunciation in Pope's time; but the orthography in Chaucer's and Spenser's writings, are to me satisfactory evidence that e in these words was short. The standard English pronunciation now is ferce, perce, terce, and it is universal in New England. I have only to add, that the sharp abrupt sound of e in the two first words is most happily adapted to express the ideas.

The English pronounce leap, lep; and that in the present tense as well as the past. Some of our American horsemen have learnt the practice; but among other people, it is almost unknown. It is a breach of analogy, at least in the present tense; the American pronunciation, leep, is therefore the most correct and should not be relinquished.

In the fashionable world, heard is pronounced herd or hurd. This was almost unknown in America till the commencement of the late war, and how long it has been the practice in England, I cannot determine. By Chaucer's orthography, one would imagine that it had been handed down from remote antiquity; for he writes herd, herde, and herden.[58] In reading more modern poets, I have rarely found any instance of a verse's closing with this word; so that it is difficult to say what has been the general practice among the learned. But for centuries, the word has been uniformly spelt heard; the verb hear is in analogy with fear, sear, and yet e in the past time and participle has been omitted, as heard, not heared. That herd was not formerly the pronunciation, is probable from this circumstance; the Americans were strangers to it when they came from England, and the body of the people are so to this day.[59] To most people in this country, the English pronunciation appears like affectation, and is adopted only in the capital towns, which are always the most ready to distinguish themselves by an implicit imitation of foreign customs. Analogy requires that we should retain our former practice; for we may as well change feared, seared, into ferd, serd, as to change heard into herd.

Beard is sometimes, but erroneously, pronounced beerd. General practice, both in England and America, requires that e should be pronounced as in were, and I know of no rule opposed to the practice.

Deaf is generally pronounced deef. It is the universal practice in the eastern states; and it is general in the middle and southern; tho some have adopted the English pronunciation, def. The latter is evidently a corruption; for the word is in analogy with leaf and sheaf, and has been from time immemorial. So in Sir William Temple's works, Virg. Ecl.

Leaf and deaf, with a different orthography, are repeatedly made to rhime in Chaucer's works; as in the Wife of Bath's Prologue, L. 6217,

So also line 6249.

This was the orthography of his time, and an almost conclusive evidence that deaf was pronounced deef.[60] This pronunciation is generally retained in America, and analogy requires it.

This dissertation will be closed with one observation, which the reader may have made upon the foregoing criticisms: That in many instances the Americans still adhere to the analogies of the language, where the English have infringed them. So far therefore as the regularity of construction is concerned, we ought to retain our own practice and be our own standards. The English practice is an authority; but considering the force of custom and the caprice of fashion, their practice must be as liable to changes and to errors, as the practice of a well educated yeomanry, who are governed by habits and not easily led astray by novelty. In the instances where we have adhered to analogy, no consideration can warrant us in resigning our practice to the authority of a foreign court, which, thro mere affectation, may have embraced many obvious errors. In doubtful cases, to pay a suitable deference to the opinions of others, is wise and prudent; but to renounce an obvious principle of propriety because others have renounced it, is to carry our complaisance for the faults of the great, much farther than we can justify, and in a nation, it is an act of servility that wants a name.