r? [OVH] LUiVAUlUJ ^UJlUX^ BOUGHT WITH THE INCOME FROM THE SAGE ENDOWMENT FUND THE GIFT OF X891 A-JJ^.§/:?...M^ /^/^Jja 3 1924 022, 373 041 »» Cornell University Library The original of tliis book is in tlie Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924022373041 NINTH THOUSAND. NOVELLO, EWER AND CO.'S MUSIC PRIMERS. Edited by Sir JOHN STAINER. SPEECH IN SONG BEING THE SIiNGER'S PRONOUNCING PRIMER OF THE PRINCIPAL EUROPEAN LANGUAGES FOR WHICH VOCAL MUSIC IS USUALLY COMPOSED ALEXANDER J. ELLIS LITT.D., F.R.S., F.S.A. "" AUTHOR OF "early ENGLISH PRONUNCIATION," TRANSLATOR OF PROFESSOR HELMHOLTZ'S WORK " ON THE SENSATIONS OF TONE." PRICE TWO SHILLINGS. In Paper Boards, Two Shillings and Sixpence. iQQ LONDON &■ NEW YORK NOVELLO, EWER AND CO. LONDON : NOVELLO, EWER AND CO., PRINTERS. A PRIMER OF SPEECH IN SONG. I.— INTRODUCTION. %* Pronounce the word Primer to rhyme with simmer, and not with rhymer. It is an old English word, liber f rim arius, and is not formed from the verb to prime. Reference to any part of this book is made to the quarter page, the figure indicating the page, and the letters a, b, c, d the quarter. Thus, "p. 26c," means "p. 26, third quarter." The Index of Elementary Sounds in (Received) English, German, Italian, Spanish, and French, and the Index to Musical Examples and Exercises, and the Contents, are placed at the end. The object of the following pages is to teach Singers, as dis- tinguished from Speakers, how to pronounce, so as to render their words audible, and at the same time interfere as little as possible with the flow of the music. To do this it is necessary to study the nature of each speech sound individually with reference to its musical capabilities and peculiarities, and also to study the mode in which speech sounds should be connected. This again necessitates some knowledge of the physical con- stitution of sound, and how it is affected by the disposition, and especially the alteration of the disposition, of the vocal organs. Incidentally, of course, it will be necessary to touch upon " received" and "discredited" pronunciations, but no attempt will be made to prescribe or recommend a complete scheme of English pronunciation. That would far exceed the limits of a Primer, and those who require it must consult a pronouncing dictionary, or undergo special training. In order better to understand the relation of speech sounds, attention will not be confined to those in English, but, considering that singers have constantly to deal with German, Italian, Spanish, and French, the few extra sounds which occur in those languages will also be examined and explained. Exercises are interspersed and added for illustration and self-practice, but are always especially adapted to the needs of the singer. Many more exercises adapted for class teaching, and full examples of German, Italian and French printed phonetically, with alphabetical keys to those languages, will be four.d in my " Pronunciation for Singers " (Curwen, London). SPEECH IN SONG. [Sect. II. II.— SHORT KEY TO GLOSSIC. In order to write of speech sounds briefly, it is necessary to have some systematic notation. The following method of writing sounds, called Glossic, which was invented by the author for the purpose of writing all the English dialects on the basis of existing English uses of letters, will answer every purpose that is required. Full explanations follow. (See Section XVI., " Index of Elemen- tary Sounds in (Received) Enghsh, German, Italian, Spanish, and French," for references to the special explanations of each symbol.) ENGLISH SOUNDS. [The large capital letters indicate the special symbols illustrated. The ■words are in Glossic spelling, and under each is placed the word in ordinary spelling, with the letters which illustrate the sounds italicised.'] bEEt be«t bAIt bait bAA baa kAUt cawght kOAt coat kOOt coot nIt knit nEt net nAt gnat nOt knot nUt nut fUOt foot hEIt height fOIl foil fOUl fowl fEUd fewd Yai _yea Pee /lea Wai way Bee 6ee WHai whey ToA toe Hai hay DOA fl!oe CHest chest Jest jest Kaip cape Gaip g-ape Fei /ie Vei wie THiN thin DHen then Seel seal Zeel zeal RUSH rash ROoZH rouge eaR ear R'ING ring eaRR'ing Lai earring la.y Mai may Nai way siNG siw^ ADDITIONAL FOREIGN SOUNDS. [G. German, I. Italian, F. French.'] bAEt bete F. vAEN' ■vin F. daaKH dach G. V'AAL wahl G. lAHsh lache F. AHN' an F. EiKY'Hu eiche G. paaLY'aa pag-Zia I. nAO no I. OAN' on F. taaGHu tag'e G. BEOZAONY' besog'we F. dUE dH F. OEN' un F. zeeGY'Hu siege G. fEO few F vOEf ve»f F. Sect. II. 111.] NATURE OF MUSICAL SOUND. 5 A turned period ( • ) placed after a vowel or consonant in any syllable indicates " stress ; " that is, force. After the vowel it shews the vowel to be long, as bea-ting hee-ting. After the consonant it shews the vowel to be short, as batt-ing bat'ing. Every vowel may be long or short. Many other signs employed to indicate peculiar actions or sounds will be explained as the necessity for using them arises. III.— THE NATURE OF MUSICAL SOUND. Sound. — Sound is a sensation due to a " vibration" of the air acting through the drumskin of the ear upon the terminations of the nerves of hearing in the internal ear. Simple and Compound Vibration. — The vibration is said to be " simple " when each particle of air moves regularly to and fro like a pendulum. Any other vibration is called " compound," because it can be produced by the excitement of several simple vibrations at the same time. Simple Tone, and Compound Tone or Note and Noise. — A simple vibration produces a " simple tone." A compound vibration produces a " compound sound," which is a " note" or "musical sound" when every " double vibration " (or whole movement both ways) takes place in the same time ; it is other- wise a " noise." The ear, by means of the numerous terminations of the nerve of hearing, "analyses" a compound sound, whether "note" or "noise," into the simple tones corresponding to the simple vibrations of which the compound vibration is, or may be, compounded. Every compound sound, whether note or noise, is therefore heard as if a series of simple tones were sounded in its place, and must, of course, be treated as a collection of simple tones. Those simple tones which collectively produce the audible effect of a note or compound sound are called its " partials." Resonance. — Take an ordinary C or A tuning fork and an ordinary drinking tumbler, strike the fork and apply the end of it to the table in the usual way : the result is a slightly compound note. A note produced on a piano, violin, harmonium, or the human voice, is greatly compound. This means that the tuning fork note has few, and the other notes have many partials. Strike the fork again, and hold it with the flat side steadily over the mouth of the tumbler. It will generally be found that the tuning fork is inaudible. Then take the cover of a book, or even the outstretched hand, and gradually cover over more and more of the aperture of the tumbler. It will be found that the sound gradually increases in power as the aperture diminishes up to a certain amount, and then becomes weaker again, and that the deeper fork requires the smaller aperture to bring out its tone. 5 SPEECH IN SONG. [Sbct. lU Simple and Compound Tones, produced. — The tone thus heard is a " simple tone," and should be immediately compared with the " compound tone " obtained by bringing the handle of the fork down upon the table. The experiment is in every respect fundamental, and has a special bearing on the production and modification of all speech sounds. It should therefore be tried with jars of various shapes and sizes, such as prune bottles, large and small preserve pots, &c. When the jar is large, not only will no sound result from partly covering the mouth, but it will be necessary to pour in water to diminish the quantity of air in the jar. It will be quite impossible to acquire a proper conception of what follows unless the experiment here indicated is actually tried till the student is familiar with it. All other experiments and exercises suggested below should also be care- fully tried by those who wish to derive benefit from this Primer. It is only by so doing that the student can properly appreciate the meaning of the annexed remarks, which are exclusively practical. Exciting Tone; Reinforced or Quenched Partial; Re- sonance Cavity. — The note of the tuning fork in all these cases is the " exciting tone," and the simple tone which results is " the reinforced partial," which is adapted to set inclosed air of that particular size and shape in powerful vibration, and inclosed air of a slightly different size and shape into less powerful vibration, and to leave practically unaffected inclosed air of very different size and shape. The inclosure in which the air is held is called the " resonance cavity," and we see that every resonance cavity answers to a given simple tone and to tones slightly different from it, " reinforcing " the first strongly, and the second some- ivhat, but " quenches " simple tones which are considerably different. The same compound note may be made to produce different simple tones (as the lowest partial and its Octave on the tuning fork,) by varying the size and shape of the resonance cavity. Absolute and Relative Pitch. — By the " pitch " of a simple tone, or of a note that is a compound tone, is meant the number of " double " simple or compound vibrations to and fro (not in one direction only, the latter being distinguished as " single " vibrations) which each particle of air makes in a second. This number is extremely difficult to determine with perfect exactness. But the " relative pitch " of two notes — that is, the fraction that one pitch is of the other — can be much more easily determined by the property that two lengths of the same string are to each other inversely as the pitches of the notes produced by plucking or bowing them. If we stop a violin string at half its length, the free part vibrates twice as fast, producing the Octave of the original note. If only one-third is free, it vibrates three times as fast, producing the Twelfth or Octave Fifth of the original Sect. III.] NATURE OF MUSICAL SOUND. note. Similarly one-fourth the length vibrates four times as fast, giving the double Octave ; one-fifth the length vibrates five times as fast, giving the Seventeenth or double Octave major Third ; one-sixth the length vibrates six times as fast, giving the double Octave Fifth ; one-seventh the length vibrates seven times as fast, giving the double Octave harmonic (or sub-minor) Seventh ; one-eighth the length vibrates eight times as fast, giving the triple Octave, and so on. Harmonics and Partials. — Those "compound" notes which vibrate 2, 3, 4, 5, 6, 7, 8, &c. (observe the omission of i) times as fast as a given note are called its and, 3rd, 4th, &c. " har- monics." Those " simple " tones which vibrate i, 2, 3, 4, 5, 6, 7, 8, &c. (observe the insertion of i) times as fast as the given "compound" note are called its ist, 2nd, 3rd, 4th, &c. "partials," because by means of those simple tones the original compound tone can be reproduced. But it cannot be reproduced from its "harmonics." The and, 3rd, 4th, &c. partials (omitting the ist) are also frequently called " overtones," an English word formed from Helmholtz's " Obertoene," or " upper tones," which was a contraction for his " Obertheiltoene " or " Oberparzialtoene," that is, "upper part tones," or " upper partial tones." On account of many confusions that have arisen from the use of the word " overtone " it will not be employed in this Primer. The com- pound " harmonics " of the lowest note of the violoncello are as follows, where the notes marked * do not exist on the piano, and those marked f are much sharper on the piano. Those marked ^ are flatter and those marked ^ are sharper than recognised notes. The pitch of the lowest note is assumed to be 66, it varies in practice between 64 and 6g. The numbers give the pitches of the lowest partials of each note, which are also the pitches of the simple partials of the lowest note of the series, and these again are identical with the lowest partials of the harmonics. Ex. I. -r±- i iSi 5 isz 5 Pitch Absolute 66 Relative i 132 2 198 3 264 4 330 5 396 6 462 7 i ,+ fe ^ Sec. ^ 594 9 660 10 726 II 792 12 858 13 924 14 990 15 1056 16 8 SPFECH IN SONG. [Sect. III. Loudness. — The "loudness" or" force" of a simple tone depends mainly on the extent to which the particles of the ^ir vibrate to and fro while the tone is sounding, but also upon the quantity of air which it puts in motion. A tuning fork when struck and held freely in the air is scarcely audible, because of the small quantity of air which it puts in motion. Held before a proper resonance cavity, one of the partials of the note of the fork puts the small quantity of air in that cavity into very strong vibration, and this being communicated to the outer air that particular partial is heard, but the others are not. Put with its handle on the table the fork communicates its compound vibrations to the whole board of the table, which again communicates them to a large mass of air, and hence all its partials are heard. This is the action of a soundboard. On the violin and violoncello this soundboard consists of a box with two holes, the air in which will not resound specially to every note of the instrument (p. 316), but the whole compound vibrations of the string (hardly audible by itself) are communicated to the whole sides of the box, and thence both to the air without and to the small quantity of air within it, and this last reacts powerfully on the external air. Quality of Tone. — The " quality" of a note depends on thft relative loudness of its partials. This variable loudness of different partials, which makes the difference between the sounds of different instruments and different vowels, is produced by the different forms of resonance cavity to which the vibrating body is applied. Such cavities are the tubes of wind instruments, in- cluding the wood, the brass, and the organ ; while the vibrating body is the air itself blown across an opening (flutes and organ flue-pipes), a single " reed " (" striking " in organs and clarinets, or " free " in harmoniums and concertinas), a double reed (in oboes), or the lip-reed for^ned by the lips themselves with the mouthpiece of brass instruments, and the so-called " vocal chords " for the human voice. The various shapes and lengths of the resonance cavities in all these instruments are principally intended to modify the " original quality of tone" of the vibrating body, to make it more pleasant, more harmonious, more musical, or more characteristic. Beats. — "Beats" are the source of noise, unmusical character, ©r dissonance. Two simple tones having nearly the same pitch, so that five times the pitch of the higher is less than six times the pitch of the lower — that is, so that they make with each other an interval less than a minor Third — do not sound on smoothly together, but break up into alternations of loud tones and silences, which when rapid are like little taps of a hammer, and are hence appropriately called "beats." Take one tuning fork, strike it, and roll its handle between the finger and thumb, and you hear a result like beats. Take two C forks which have not Sect. III. IV.] SPEECH AND SONG. 9 quite the same pitch (and these are very easy to find at any music shop, the difficulty being to find two which have precisely the same pitch) and hold them over a cavity which resounds to either of them separately, and the beats will be heard beautifully. In compound tones the beats are not so distinctly heard, and when, as often happens if a Fifth or Third is out of tune, the beats only exist between higher partials, while the louder lower partials do not beat at all, the silences become covered up by the non-beating continuous tones, and are difficult to hear. Such beats of the Fifths are best observed on an harmonium, as ordinarily tuned. When once clearly distinguished they are apt to be extremely annoying. Even when only one compound note is produced, provided that it is 'composed of more than eight partials, distinct beats will arise, producing a harshness or crispness or tartness, or even a screaming or thumping, eifect when very powerful, which is sometimes most unmusical and unpleasant. The lower notes of an harmonium show this well, producing the " roughness " known as " reediness." These beats become extremely fast in the higher notes, and are not observed. The Organs of Speech as a Musical Instrument. — As the organs of Speech are essentially a musical instrument, having an exciting vibratory body (the vocal chords), and a series of resonance cavities between them and the open air, which are "variable" (like the tube of the trombone, only to a much greater extent), and which, in each form they assume, materially modify the "original quality of tone" of the exciting vocal chords, it is impossible to understand the precise connection of the " positions " of the vocal, organs with the various sounds of speech, without being tolerably familiar with the facts here in- dicated, which were in the main first published by Professor Helmholtz in 1862, and are now generally admitted to be correct. For a fuller exposition, with an application to musical scales and chords, see my " Basis of Music " (Hodgson, London). IV.— SPEECH AND SONG. Relation of Speech to Song. — Speech is produced under very different circumstances from song. We all dislike to hear a person sing in his speech. It is only for comic purposes that an attempt is made to speak in song. But it is necessary that a singer should utter significant sounds in order that the emotions excited by the music should find their explanation in the words. A song when sung to laa (all italic words representing sounds are in Glossic characters) produces a totally different effect from 10 SPEECH IN SONG. [Sect. IV what it does when sung to words even of an unknown language. The principal differences between speech and sound are as follows : — Difference in Pitch. — In speech, pitch is uncertain, unsustained, and " gliding " through several pitches but slightly differing from each other. In song, pitch is (or should be) quite certain, sustained, and rarely gliding off to another pitch. Occasionally this is permitted in portamento poa-r'taamain-toa, but then the passage is from one definite sustained pitch to another. Difference in Compass. — In speech the change of pitch is usually confined within narrow hmits, seldom exceeding a Fifth, though occasionally reaching an Octave. In song a very great variety and compass of pitch is usual, generally exceeding an Octave and a half, and frequently touching notes much higher and much lower than those which would be attempted in speech. The consequence is that in song, speech sounds have to be at- tempted under totally different conditions from those, under which they originally arose and are generally employed in language. Difference in Musical Capability. — Speech is full of sounds which will not produce a good musical tone at all, but contain slight beats and dissonances, producing hums, murmurs, buzzes, and hisses (of which the last bear some resemblance to the noise made about the mouth hole of a flute), together with absolute breaks or silences. Nothing of this can be sung. The result is noises or pauses, which separate singable sounds, but still must be produced by the singer under pain of being unintelligible. No one, for example, would recognise " Thus shall I sing my song of praise " in dhu al ei ing mi ong ov prai, although without doubt these'sounds are more musical than dhus shal ei sing mei song ov praiz. Difference in Relative Length. — Speech sounds may indeed be uttered quickly or slowly, but the relative pace is dictated by emotion, or prescribed within the limits of a word by the meaning itself. Thus the vocal part of : heed, aid, awed, bode, pool heed, aid, aud, boad, pool, is necessarily longer than that of: hid, add, odd, bud, pull hid, ad, od, bud, puol. In singing the necessities of melody do away with all this relation of length. Words naturally incapable of being sustained, have to be produced on a long note; words naturally incapable of being shortened, have to be produced on a short note. Difference in Musical Quality. — Speech is far from being wholly musical, even in its most musical parts — the vowels. Song should be wholly musical in the vowels, while the con- sonants, or unmusical parts, which cannot be dispensed with, should be reduced to a minimum, without being extinguished. The vowels even are not equally well adapted for all pitches, and it is impossible to sing them without modification. The singei Sect. IV. v.] PRODUCTION OF SPEECH SOUNDS. II then has to learn how and to what extent the vowel element of speech may be modified, and how the consonant or unmusical element can be made distinct, with the least possible injury to both music and meaning. v.— PRODUCTION OF SPEECH SOUNDS. Larynx and Glottis. — The musical element of speech con- sists of "voice," the unmusical of " flatus " _^arfM5. These are produced as follows. In the throat the hard projection known as " Adam's apple " corresponds to the edge of a nearly triangular gristly box called the larynx lar'-ingks, forming the upper end of the windpipe, and closed with a lid called the " epiglottis " ep'iglot'is. Across the lower part of this box lie two highly elastic bands, called " the vocal chords," united together at the front part, and at the back part attached severally to two pieces of gristle called the "arytenoids" ar'-itee-noidz. Extend the two first fingers of the left hand, doubling down the others and holding the back of the hand upwards, rest the tips of these two fingers (which represent the vocal chords,) on the back of the lowest joints of the two first fingers of the right hand, the other joints and fingers being bent down. These right hand fingers represent the arytenoids. By opening or closing the right fingers the left fingers can be made to open and close. When open there is a lozenge-like hole between them. This corresponds to the " glottis " glot'is, or opening between the vocal chords and arytenoids. The larger part, between the fingers of the left hand, corresponds to " the elastic glottis ; " and the shorter part, between the right fingers, corresponds to " the cartilaginous kaa-rtilaj-inus glottis." Flatus; Whisper; Voice. — When we "breathe" or "blow," flatus issues from our lungs, and the glottis is wide open. When we "whisper," the glottis is almost quite closed, but, as the air passes through, it causes the edges of the vocal chords to vibrate faintly, producing the first approach to a musical sound, which is, however, accompanied with a large escape of air that has not been thrown into musical vibration. When we utter " voice," the edges of the glottis are brought quite close together, and the air driven from the lungs forces them apart, allowing a puff to pass, after which the chords close again, owing to their own elasticity, and then they are again forced open by the breath, and so on. Thus a continued and very rapid succession of puffs passes through the larynx, setting all the air between the glottis and the outer air into vibration, so powerfully that, even if there were no cavities to pass, a very distinct and extremely compound musical sound would result. This would be " the original quality of 12 SPEECH IN SONG. [Sect. V tone," which is subsequently modified by the resonance cavities. This original quality, of course, varies much under different cir- cumstances, according to the will and emotion of the speaker, but those particular modifications, of which we shall have to speak, are made after the vibrations have passed as waves of sound beyond the larynx. Original Quality of Tone. — This is due to several causes : first, the nature and powers of the vocal chords, which differ in different individuals ; secondly, the extent to which the chords are forced open by the air, which passes through in a puff; thirdly, the extent to which the chords themselves vibrate, which is very variable ; fourthly, the accuracy with which the chords close by their own elasticity after each puff, which has a very material effect on the cleanness of the separation of the puffs, and hence on the number and nature of the partial tones, and differs much in individuals ; fifthly, the mode in which the wind from the larynx is brought to act upon them ; and, lastly (because in what follows no modifications will be considered which do not occur after the stream of air has left the larynx), to the resonance within the larynx itself. The last-named resonance is due, first, to the very variable cavity immediately above the vocal chords, called the "ventricle of Morgagni " ven'trikl ov Moar'gaa-ny'ee, the upper part of which is bounded by a fold of mucous membrane called the " false chords," and, secondly, to a funnel shaped cavity extending from the false chords to the epiglottis. The extent and resonance of this last cavity can be greatly modified by bringing down the epiglottis (or lid of the larynx), which is furnished with a pad on the under part, and is generally wide open for high notes, nearly closed for low notes, and tightly closed for " holding the breath." Modifying Resonance Cavities. — The following are the most important : — 1. The "lower pharynx" far'-ingks, or fleshy bag at the back of the mouth, leading from the larynx into the mouth, and bounded at its upper part by the uvula eu-veulu, or little ♦ongue seen to hang from the further end of the soft palate. The uvula is generally pressed against the back wall of the pharynx in speaking, to prevent the air from passing through the nose. The pharynx can be contracted and widened by muscular effort, and shortened by driving the larynx higher or lower. It is generally short and contracted for high, and long and widened for low notes. 2. The " arches of the palate," or passage leading from the pharynx into the mouth, bounded by two arches, which can be easily seen, between which lie the "tonsils." These arches can be more or less contracted, and can hence much modify the pitch of the resonance. ShcT. v.] PRODUCTION OF SPEECH SOUNDS. I3 3. The " mouth," bounded at the back by the arch of the palate, on the roof by the soft and hard palate, in the front by the hard teeth and soft lips (which can be variably, and each differently, opened and closed), on the sides by hard separable teeth and soft continuous cheeks, and at the bottom bj' the tongue, which is capable of entirely changing the shape and resonance of this cavity by its power of extension, contraction, expansion, folding up, turning in different directions, vibrating, touching or approaching different parts of the palate and teeth, either with penetrable or air-tight closure. The tongue can also be brought close up against the arches of the palate, or against the upper side teeth and hard palate, so as to exclude any wind from passing into the mouth from the pharynx. The lips can also be closed so tightly as to prevent any air from passing, and can likewise vibrate. The uvula can also vibrate. The resonance of the mouth can, therefore, be changed to a wonderful extent, and these changes are of the greatest importance to the speaker and singer. 4. By relaxing the pressure of the back of the uvula on the pharynx, the pear-shaped cavity of the " upper pharynx " is brought into action. In this cavity are two holes, called the " back nostrils," which lead into very compUcated resonance cavities above the hard palate, full of little plates of bone and gristle, called the " nasal passages," and terminating in the front nostrils at the end of the nose. The resonance in these cavities is very peculiar, and not much under control. It can be varied considerably by altering the tightness of the pressure of the uvula against the back of the pharynx, and by combination with resonance in the mouth as more or less obstructed and /imited by the tongue and lips. Physems. — The " bellows-actions " of the lungs or " physems" fei-semz are very important. The bellows have only two external apertures, the lips and nostrils, which lead into one passage, the windpipe. To feed the lungs with air, therefore, their cavity has to be enlarged, either by the relaxation of the compressing muscles or by a muscular effort, which can be best appreciated by drawing very gradually a deep breath through the nose only. It is necessary that the epiglottis and glottis should be wide open, and that the nasal passages should not be obstructed, as otherwise a noise of " snuffling " would be occa- sioned. Breath is drawn most rapidly through the mouth, but if any of the passages are only partially open, or not open suffi- ciently for the suddenness of the inspiration, " gasping " and other noises ensue. When the lungs are filled, or nearly so, the muscles between the ribs (chiefly used by women), and the muscles of the diaphragm dei-ufram or muscular layer sepa- rating the lungs from the parts below (chiefly used hy men), 14- SPEECH IN SONG, [Sect. V. VI. compress the wind gently from the lungs through the windpipe. Gentleness and steadiness are the chief points to be attended to in physems ; jerked and spasmodic actions are, however, occa- sionally employed. The rate at which the wind escapes depends mainly on the degree of closure of the vocal chords. When they are wide open it is rapid, when they are closed for vowels or singing it is gentle. Apply the back of the hand close to the lips while singing laa, and observe how faint the wind is. Suddenly open the glottis, but continue to breathe without altering any other part of the position of the vocal organs, and observe that a large quantity of breath falls instantly on the hand. The different effects of singing laa, lee, loo should also be noted, and the apparently much greater quantity of wind in the lee than in laa, and in loo than in lee, should be observed, and traced to the different forms of the delivering aperture, which for loo concen- trates the air as in blowing out of the narrow end of a funnel, whereas in laa we blow out of the wide end. The "management of the breath " forms an important part of elocution and singing, but does not enter into the course of this book. Implosion. — This consists in producing a sound by closing up the external apertures of the lips and nose, and driving the larynx suddenly up, so as to compress the air in the mouth This is stated on good authority to be used in Saxony in place of initial p or b, and t or d, and may, perhaps, be used for the York- shire definite article, usually written, " t' man" (in Glossic, °d maan). It will not be further alluded to, as it is quite unsingable. Clicks. — These arise from suddenly separating two parts of the mouth without drawing breath, as in our familiar " t't," or "tut" (c in Glossic), and " cl'ck," spoken with one side of the tongue only, to incite a horse (a; in Glossic). These and other clicks are used as parts of words in the Caffre and Hottentot lan- guages, and have been written as c, x in books in those languages. Recoil. — A very faint form of click, here written p°, t°, k°, occurs at the end of words occasionally in English, as hap°, hat°, hak°, which is a mild form of the " recoil," written hap°h, hat°h, hak°h, in which on loosening contact a small portion of flatus is driven out. Felt with the back of the hand p°h will be found strongest and k°h weakest, for the same reason that made the wind from loo strongest, and that from laa weakest. Both click and recoil are entirely unsingable. VI.— GLOTTIDS. Glottid. — Pronounce glot-id.. This is the action of the glottis in commencing any sound. °H. " Open " Glottid (Flatus). — The glottis is open as for SacT. VI.] GLOTTIDS. I5 breathing. No musical sound is heard. If the parts of the mouth are in the position for any vowel, as ee, aa, 00, a sound as of blowing occurs, and this is called " flated ee, aa, 00," and is written °ee, °aa, °oo. Each of these flated sounds differs from the others by the imperfectly excited resonance of the corresponding cavities. All such sounds are perfectly unsingable. °H'. "Contracted" Glottid (Whisper). — The glottis is par- tially closed, but there is a considerable chink left, so that there is rather a "fluttering" than a regular vibration of the edges of the vocal chords, and even to produce this a forcible expulsion of breath is necessary, which is very exhausting. Taking the vowels ee, aa, 00, the whispered effect is written °'ee, °'aa, °'oo. Observe the much more distinct resonance for these than for the flated °ee, °aa, °oo, and the much greater distance at which they can be heard. Contrast the effects, °ee, °'ee; °ee, °'ee, &c., by passing rapidly from one to the other. Observe the entire absence of voice in °ee, and its partial presence in °'ee, but that, although the latter can be made effective in speaking, even to large audiences (in " stage-whispers"), it is quite useless to the singer. H'. "Close" Glottid (Voice).^ — Theedgesof the vocal chords are brought close together, so as to meet completely down the middle line, and not to overlap. This is the only position in which distinct musical "voice" can be produced. As it is supposed to be inherent in every symbol of a "voiced" sound, as ee, aa, 00, it is not marked before them. Contrast °ee, °'ee, ee; °aa, °'aa, aa; °oo, °'oo, 00, by uttering them in rapid succession, but each distinctly separate. Observe the ease with which the first and last of each set are produced, and the exhausting nature of the middle sound. Singing Registers. — There are several kinds of voice, known as "registers" rej-isterz, arising from the different degrees of "tension," or stretching of the vocal chords, necessary to produce notes at different pitches. They are commonly called " chest," " falsetto," and " head " voices. These names are founded upon certain feelings excited in the singer, but do not express any fact in the action of the vocal chords. The matter has recently been much investigated, and has, in particular, been made the subject of examination on a large scale by the Tonic Solfaists. The latter name the registers : (i) "thick," as due to the vibration of the vocal chords through their whole thickness, causing the whole body to vibrate with it, and even the " chest " at times to resound sympathetically ; (2) " thin," as due to very tightly stretched chords, of which only the thin edges vibrate thoroughly ; (3) " small," because apparently only a portion of the chords nearest the point of junction really vibrates, and to secure their action the other part is not closed at all, so that there is great i6 SPEECH IN SONG. [Sect. Vt waste of air and consequent fatigue in using this register. Both tbe thick and thin registers are subdivided into "lower" and " upper." In Mr. Curwen's " Standard Course " (pp. 32, 33, 66, 67, 105-107) will be found many practical remarks on this subject; the following diagram has been adapted from page 106 of that work. The pitch of the notes may be considered as the " French pitch." The compasses of the voices — B, bass; Bt, barytone; T, tenor; C, counter-tenor; A, contralto; M, mezzo or second soprano; S, first soprano — are indicated by these letters, a following hyphen, as B-, shewing the beginning, and a preceding hyphen, as -B, shewing the end of the usual compass in average voices. The Thin and Small Registers can be cultivated downwards in men and women respectively, and the Upper Thick can be cultivated upwards in women. But the thick lines in the diagram mark the constant natural breaks. The Lower Thick is marked in semi- breves, the Upper Thick in minims, the Lower Thin in crotchets, Upper Thin in quavers, and Small in semiquavers. Great break constant. Ex. 2. Lower Thick. zrsz i Upper Thick. ^ IS2I 1221 IS =221 M--B S--Bt Small break constant. _ . Small. Upper Thin. .^ ^. 1^ B- Bt- T- LowER Thin. =fc^]^ m -T -c -M -S Common Registers. — The Upper Thick is common in one i note C() I - to B, Bt, T, C, A, M and S, and when each of these voices sing this particular note, the quality of tone is really identical. The two following notes -^ are common to Bt, T, C, A, M and S voices. They are also identical in quality in all three voices. I have heard the expe- riment tried successfully. Speaking Registers. — In speaking, basses always employ the Lower Thick ; tenors generally the Upper Thick, the Lower Thick being gruff and the Lower Thin squeaking; contraltos and Sect. VI.] GLOTTIDS. I7 sopranos usually speak in the Lower Thin, as, being women, they think the Thick too " manly." These limitations of the speaking voice must be borne in mind in estimating the diffi- culties of bringing out common spoken sounds in the singing, where only one voice, the bass, habitually keeps within one register. Also in the change of register there is so much alteration of " original quality of tone " that there is great difficulty in preserv- ing the vowel sound intact in passing from one register to the other, so that the modifying positions require to be slightly altered. Natural Vowel. — The pure original quality of tone of the voice in any register cannot be heard, because it is impossible to avoid modifying the voice by the resonance chambers through which it passes to the outer air. The vowel which seems least coloured or modified is the u in but, at the ordinary middle pitch of the speaker's usual speaking register. The sign h' is used for any short or indistinct sound resembling this, when the precise form of the modifying cavity cannot be determined. J, jH. "Gradual" Glottid. — The speaker begins with his glottis wide open as for °h, and closes it as he goes on through the contracted form for °h' to the close form h'. Hence, in saying jee, he not only says °ee, °'ee, ee in succession, but fills up the gaps between them by intermediate sounds. This produces a " glide" from flatus to voice, very short indeed, but generally quite per- ceptible. When the initial flatus is particularly strong and offensive we write jh. The effect is slovenly and inelegant, and should be carefully avoided, especially by singers, to whose voice it gives an unpleasant breathiness. All German words, however, written with an initial 's'" before a vowel, though supposed to be spoken with z, in which the voice should be set on at once, usually commence with a gradual glottid, when not follow- ing a vowel in a preceding word, as ' sie,' ]zee, rather than zee; but this flatus falls out before the second z in ' sie sehen,' jzee zai-en. In this case the word begins with a very brief s, which glides on to z, and we may write szee zai-en. German singers generally make this sz so distinct that it was from them that I first learned the habit. It is, however, unnecessary for intelli- gibility, inelegant, and unmusical, and hence should be avoided. As a voice-sound maybe commenced, so it maybe also finished by a gradual glottid, the vocal chords being gradually separated. Thus eej ends decidedly with °ee, and "'«« is touched on the way. In the case of strongly pronounced final vowels this effect is not infrequently heard in English, especially with eej and ooj. But that is more common in Denmark and Iceland, where the change oi°ee, °oo into yh, wh, or kw'h sometimes takes place. But the gradual glottid is extremely common for v, dh, z, zh' final in English, when a word is final, as five, feivj, feivf; breathe, breedhi, breedhth ; sins, sinz-i, sinzs (distinguished from since, B »8 SPEECH IN SONG. (.Sect. VI. sins, merely by the absence of a short intermediate z in the last) ; cage, kdi-ji, that is, very nearly kai-jsh (more accurately kai-jsh', or kai-dy'zh'sh', as will appear subsequently). So common is this pronunciation that it is very difficult for an Englishman to avoid it. The buzz of the v, dh, z, zh' final is so harsh that it is annoying to the speaker to continue it, and, as he does not like to cut the letter short, he opens his glottis almost imme- diately, and /, th, s, sh' result. Even in the middle of woids or sentences this occurs, as in " his foes amazing close his eyes alive," often pronounced as hizs foa-zs amai-zsing kloa-zs hizs eizs ulei-vf. In singing, this is often quite distressing to the ean The position of an s or zs is generally painfully evident in psalm- singing. This must be sedulously avoided by the singer. Let him take the nonsense-verse just given, and sing it to such an air as " God save the Queen " over and over again, quickly and slowly; in different keys, so as to reach a high and low part of his voice, and stop every time he hears the least trace of a hissed s. He should make the z quite pure, at first very long, and at last very short, though always distinctly audible. Ex. 3. Male voices will sing an Octave lower. s: =F ^ m^ hiz foa-z mai - zing kloa'Z hiz foa-z u * ^. ^- kloa-: hiz eiz lei'V. And so on for the next four lines, repeating the first six syllables continually, and ending wifh the four last. This will be found a very difficult exercise, but it should be thoroughly mastered. J "Clear" Glottid. — The cross indicates the cutting off of all flatus. The vocal chords are brought into the precise position required for voice and to the precise tension for pitch, and all the modifying cavities are duly arranged before the least wind is driven through the lungs. Hence the voice begins " with a clean edge," so to speak. This is the only glottid which should be used by a singer for setting on the voice. Similarly, he should always break off the voice in the same way without dropping into flatus ; thus, eyes, jeizf, not jeizj. It is good practice to sing feef, jeef, pef ; jaa^, fauf, fauf ; foof, foof, foo}, all in one breath, pausing between each sound for about the same length of time as the sound itself lasts. The glottis, if possible, should be kept steadily closed during the whole time that the exercise is sung, during the silence as well as during the sounding, and the sudden stop and commencement should be due to a sudden Sbct. VI. VII.J PHYSEMS. I9 cessation and renewal of the wind, which, however, should take place without the slightest jerk, so that the loudness may remain uniform. Then sing the artificial words, fzeezf, }saazf, jzoozf till there is no difficulty in beginning and ending the buzz with the clean edge of the clear glottid. This will complete the exercise on hiz foa-z, &c. The clear glottid will not be written, but should be applied initially and finally when the vowel does not run on to a preceding or following consonant. ; " Check" Glottid. — The passage of the breath through the larynx is totally obstructed by the tight closing of the vocal chords and the dropping of the epiglottis, which, by its pad, presses on the false chords and prevents any wind passing. This check is u«ed in very staccato staahkaa-toa passages to give great suddenness to the commencement of the vowel-tone, and also great suddenness to its conclusion. The effect thus produced is quite different from the clean edge of the clear glottid. The latter is like simple " separate bowing " of each note on the violin. The check is similar to the intermittent jerked bowing by which the staccato is usually produced. The check is not used as an element of speech in English, Italian, or French, in all of which languages it is considered very disagreeable. It constantly occurs in German speech, when a consonant has to be clearly separated from the preceding vowel, or at the begin- ning of syllables, as: eine unausstehbare erinnerung ;aaynw ;uon-;aawssh'tai-baa-r'u ;er';in-ur'uong, (which Englishmen are apt to pronounce as ei-nu uon-ousshtai-baar'u u-r'in-ur'uong, having a totally different effect). German orthoepists do not distinguish the clear and check glottids. In Arabic the check glottid is called haam-zaa, and has an especial sign. It was probably used in all Semitic languages. It is supposed to have been the "spiritus lenis " spir'-itus lee-nis of the Greeks, but it has totally disappeared from modern Greek. ;H. Plated Check. — This represents a sudden explosion of flatus after the air has broken through the check. This explosion is particularly disagreeable, and should be carefully avoided. Other Glottids. — Two of these may be just mentioned. The Danish " croak" ,r, a kind of trill formed by a rattle of the cartilaginous glottis, much resembling the Arabic " bleat " or :aayn, which is much stronger. They are both accompanied by the voice of the true vocal chords. VII.— PHYSEMS. Physem. — Pronounce fei-sem. This is so called from theGreek word for a bellows (which we ca.\\/eusee-ma, and modern Greeks fee-~ seemaa), a.nd indicates a peculiarbellows-action ofthelungs(p.i3c). 20 SPEECH IN SONG. [Sect. VII. Experiments with Artificial Physems. — Procure a child's small hollow indiarubber ball with a hole in it, costing twopence or three- pence at a toy-shop, and an ordinary shortish clay pipe costing one halfpenny at a tobacconist's. Break the tube of the pipe off till it is thick enough entirely to fill the hole of the ball quite air- tight. The ball represents the lungs, the tube of the pipe the windpipe or trachea {travkiu as an English word ; as Greek we call it trakei-u, as Latin trukee-u, as modern Greek traakhee-aa), and the bowl the cavity of the mouth. The fingers of the left hand, used to compress the ball, act as the intercostal in-terkos-tel muscles and diaphragm. The bowl may be held steady between the fingers of the right hand, and the right thumb used to cOver the opening more or less, as the lips cover the opening of the mouth. This gives a convenient model for studying the action of flatus when the vocal chords are open. The back of the hand, the cheek, or the flame of a small taper act as sensitive bodies to note the force of the expelled wind. First remove the pipe. Compress the ball suddenly. There will be a little, but very little sound. Remove the compression suddenly. The ball recovers its form and the air rushes in, but still there is very little sound, except from the final expansion of the indiarubber, which may be prevented by not allowing the ball to recover its form completely. Next stop the hole of the ball with a finger of the left hand, while the ball is full. Com- press the ball gradually, greatly condensing the air, and remove the finger suddenly. The air escapes with an " explosion " such as follows a check glottid (;), when the glottis and epiglottis are opened suddenly, and the lungs compressed. In the case of the ball the air escapes with a kind of " spit," which is very audible. Next insert the piece of pipe without the bowl, so that the opening of the ball is elongated, forming a slight resonance cavity. Compress the ball as before. Observe the much greater force of the wind, which is not immediately dispersed at the exit, but concentrated into a stream. Observe also that the sound of the air rushing in and out is decidedly more perceptible than before, shewing that the resonance is beginning to tell. When it is in this state, observe the action of the stream of air on the flame of the candle. Direct it towards the flame just above the wick and hold the tube quite steady with the right hand, as it will otherwise wabble impracticably. Compress gradually. Observe that the flame moves away and recovers itself, because it is impossible to blow with perfect steadiness. Observe that an increased steady rapidity causes the greater deflection of the flame, which returns to its usual position as the rapidity diminishes. This corresponds to crescendo kraishain-doa and diminuendo deemee'no&ain-doa in singing. Sect. VII.] PHYSEMS. ■*> 21 Now try a set of rather sudden crescendos separated by very short pauses. This is best managed by holding and partly com- pressing the ball with the thumb and three fingers of the left hand, while by moving the left forefinger rapidly up and down, without quitting the surface of the ball, you compress the air with a sudden "jerk." A succession of "puffs" is thus pro- duced, and the flame of the candle will be seen to oscillate. This resembles, on a large and rude scale, the puffs produced by the elastic action of the vocal chords when the air from the lungs forces them open and they return by their own elasticity (as the ball recovers itself) ; while the vibratory motion of the flame corre.- sponds to the vibratory motion of the air which is set up beyond the larynx, and, when fast enough (that is, not less than fourteen — practically not less than thirty-two — times in a second), con- stitutes musical sound. Now, holding the tube quite steady with the thumb and three, fingers of the right hand, press the forefinger on the extremity of the tube, so as not to press the tube in, but to prevent the passage of the air. Compress the ball rapidly, and suddenly remove the finger. We have again an " explosion," still more resembling that after a check glottid, and less like a " spit," on account of the direction of the stream of air and the resonance of the tube. When, without first stopping the end of the tube, we compress the ball suddenly, we do not produce so strong an effect. Slight compressions, with end stopped for a very short time, and the closure suddenly removed, should also be studied, by the ear. And, without stopping, the repellent and attractive effect of sudden compression and sudden recovery of the ball should be studied on the flame of a taper. Next insert the part of the pipe which ends with the bowl. The former experiments with the flame should be tried and the new results contrasted with the old. The effect on the flame will be stronger than before ; but the most noteworthy difference is in the resonance, which is very much stronger, so that even a small sudden compression of the ball produces a sound, and the suction, which is due to the elastic recovery of the ball, produces almost as loud a sound. To shew that this is due to resonance, holding the tube steady, close the bowl with the thumb about half-way, and observe that the sound produced by sudden compression is evidently much flatter. Contrast the sounds by rapidly pro- ducing one and the other. Next close up three-quarters of the bowl, and observe that the sound is very much flatter than before. In ail these experiments, which are merely rude illustrations of operations performed under only roughly similar circum- stances by the speaker and singer, we must recollect that very small portions of air are sent through a very narrow tube, which 22 SPEECH IN SONG. [Sect. VII. issues by a small opening into the bowl, and that it is only at its extremity, and not in its interior (as in hissing and whistling), that the bowl (or mouth) can be modified at all. But any one who is at the trouble of repeating them will have a much clearer notion of the action of a physem than by merely reading the description, and the knowledge thus gained is of great value to the singer. H°h. " Oral Flated Jerk." — Physem produced by a sudden compression of the lungs while the chords are wide apart, and the mouth open as for u in but. The effect is much the same as in " panting," but of course is much weaker in speech. If the resonance cavity is to be specified, then it must be written in place of h. Thus h°ee is jerked flatus through the resonance cavity of ee. H°Ji. " Nasal Flated Jerk." — The mouth is closed by the back of the tongue, and the jerk is made through the nose, as in the very common operation of nose-blowing, for which, however, the resonance is generally increased by com- pressing the nostrils, and some people are able to trumpet grandly. H°hj "Inspired Oral," and H°^hj "Inspired Nasal Flated Jerk." — These will oe understood from the former experiments and descriptions. Hp "Gradual Glottid Jerk," or "Flated Aspirate." — Such a sign as hpe differs from lee, already described, by beginning with hiee in place of °ee ; that is, with jerked instead of quiet flatus. The flatus thus becomes very evident. This produces the effect commonly called the aspirate as-pir'et, rough breathing, or " splritus asper " spir'-itus as-per, as heard in Germany and Scotland, and very generally in England. It is said also to occur in the South of France. It is not used in Italian, Spanish, oi received French pronunciation, and even in England it is continually omitted by provincial speakers, and often softened to the next following form (or clear glottid jerk) by careful speakers. The omission of the aspirate at the beginning of words is considered almost a crime in educated society, except in a few words, which have decreased in number even within the last fifty years, and in these the insertion of the aspirate is as great a sin. Even now elderly people, of unimpeachable social position and education, call herb, erb; hospital, os'pitel; humoui, eu-mer; humble, um'bl (the two last are given as the " proper " pro- nunciations by Smart, and the first and last, but not the second or third, by Cull, in Ogilvie's Dictionaries). The aspirate is doubtful in ' hotel,' although pronouncing dictionaries insist on it, but ' hostel ' always has it, and ' hostler ' never — indeed the word is sometimes even written ' ostler ' or ' osier,' and is generally called os-ler. Doubtless many persons think of it as Sect. VIM PHYSEMS. 23 au'sler, as they pronounce it, meaning the servant at an inn who takes care of what they often call airsez, that is, 'horses.' Any one who inserts an aspirate in heir air, honest on-est, honour on-er, and their compounds and derivatives, when speaking, or omits the ' h ' when writing, is considered out of the pale of civilisation. In ' perhaps ' and ' at home ' usage differs much. Very educated people, who move in good society, say pr'aps, u-toa-m, or u-toa-wm; and per-hjap-s, at hpa-m are seldom said without the consciousness of having done a good action. In 'exhaust,' 'exhort,' and even 'exhibit,' 'exhibition,' 'ex- hume,' and their derivatives and compounds, many force them- selves to use /j^. Thus Smart and Cull both write egz-hjib'it, eks-hjibish'un ; but the change of eks into egz in the first, and the absence of accent on the second, point to egzib-it, eksibislreii as the analogic forms; and similarly egzairst, egzau-rt, egzeirm. When the stress falls on the syllable beginning with ' h,' the aspirate is, however, generally saved, as in behest bi-hjes-t, abhor, abhoTTence ab-hjau'r, ab-hior''ens, (not u-baw, u-bawens). In such rare words as ' vehement,' ' Behemoth,' the aspirate may be pardoned, as vee-hiiment, Bee-h^imoth ; but analogy points to vee-imenf, Bee'imuth, and in fact vee-ument is most usually said. When ' h ' begins a word and an unaccented syllable at the same time, it takes the indefinite article ' an,' as 'an habitual drunkard,' 'an harmonious combination,' 'an harangue,' ' an harpoon,' ' an hermaphrodite flower,' ' an hexameter verse,' ' an historical account,' ' an hotel,' and some speakers are inclined to omit the ' h ' in consequence. This is not at all necessary. The voice rests on the an prior to bringing out the next syllable, which allows of the easier forma- tion of A/ ; thus, an- hjabit-euel drungk-erd, an' -hjaarmoa-nius kom-binai-shen, an- hjur'ang-, an- hjaarpoo-n, an- hjer- maf-roadeit flower, an- hjeksam-iter vers, an- h]istor'-ikel u-kou-nt, ati; hpatel-. A custom has lately grown up of writing ' a ' in place of ' an ' in such cases. It is then almost necessary to pronounce it ai, emphatically, as ai hjur'ang-, as the usual u hjur'ang- would be far too harsh. In singing such combinations are not very likely to occur, but when they are met with an is far preferable for the voice. Hf. " Clear Glottid Jerk," or " Voiced Aspirate." — The organs being disposed as for the clear glottid, the wind com- mences with a jerk, and hence the voice commences with a jerk, without being preceded by any flatus. This form of " aspirate " is by no means uncommon in England, and it is the only form which is suitable to a singer, because no flatus can produce anything but an interruption to a musical sound. Distin- guish ' air, hair,' as ai-r, hfai-r. Exercise singing ee, hpe ; aa, hfaa; 00, hpo, each pair repeated many times, in one 24 SPEECH IN SONG. [Sect. VII. breath, and on the same pitch, and then on very different pitches. Thus : — i Ex. 4. r=r=f -(=^ W±: ^ ^ ? ee hfee ee aa hjaa aa 00 hfoo 00 hpe ee hpe hjaa aahfaa hfoo 00 h)oo ee hfee ee aa hfaa aa 00 h}00 00 hjee ee hpe hjaa aa hfaa hfoo 00 hfoo ee aa 00 and then with repetitions thus : — • i ^s 5 £ hfee hfee hfee hpo hfoo hfoo hfaa hfaa hfaa hfaahfaahfaa hfee hfee hfee hfoo hfoo hfoo hfoo hfoo hfOO hfaahfaahfaa hfee hfee hfee ^^. hfee hfee hfee hfoo hfoo hfoo hfaa hfaa hfaa hfoo hfaa hfee and the same without the aspirate. The student should distinctly feel as well as hear the jerk, and be satisfied that he produces no flatus at all. A fellow student placed at a distance should indicate by a silent gesture with the hand v/hether he hears the jerk, and whether he hears any flatus. The exercise must be varied, and the two students should sing and listen alternately. H. "Common Aspirate." — ^When the simple symbol ^ is used in Glossic writing, at the beginning of words, or after a hyphen when following a consonant or a, it indicates either hj or hf at the pleasure of the reader. But the singer should always take it as hf. The -symbols hj, hf, are not used except in theoretical discussions. Hence h is the only physem which is usually written. After a consonant, as in yh, wh, th, dh, kh, gh, sh, zh, Ih, rh, and after a vowel, as ah, the h only modifies the meaning of the preceding letter, in a way which will be shortly explained. It is sometimes quite impossible to pronounce h^j and then hf must of course be used. Thus the Irish and Indian jerked pro- nunciations after voiced consonants such as beam b-hfaim are not pronounced h'b-hiarm (that is, nearly ub-hjai'm) either in Ireland or India. Hence, if such a combination as b-h is found at the beginning of a syllable, b-hf is meant. On the contrary, pea. p-hai, may be p-hjai or simply pjai; and so for other mute or flated consonants. 'H. "Wheeze Physem." — This is a well known Arabic sound, which may be just mentioned. The wheeze is occasioned by forcing flatus, without voice, through the cartilaginous glottis. It is therefore closely related to the bleat faayn. It occurs in the name of Mahomet Moo'haam'tnaad. Sect. VIII.] VOICE, FLATUS, SILENCE, GLIDE, SLUR, BRE.4K. 25 VIIL— VOICE, FLATUS, SILENCE, GLIDE, SLUR, BREAK. Voice and Flatus. — As we have already seen (p. iic,) h' or " voice," and°A or " flatus," are the two great divisions of speech sounds considered independently, regarding whisper °h' &s merely " imperfect voice." They give rise to numerous subdivisions, depending upon their "permanent" modifying resonance cavities. The voice-sounds are alone capable of being sung. All flatus is mere interruption, so far as song is concerned. Silence, Glides, and Slurs. — There is a third division, con- sisting of " silent positions " written Ji. Of course " silence " can neither be spoken nor sung. But the silent positions Ji become " effective " when the speaker produces voice h' or flatus °h while changing from these silent positions through others which are not silent, and which bring into use " continuously altering," insteadof " permanent," resonance cavities. It is clear that these altering positions will not occur only in passing from a silent position, but must always occur in passing from one sustained position to another, provided either voice or flatus be produced during the change. Thus, on, the violin we may play two notes on the same string perfectly distinct from each other, with no pause and yet no connection, by stopping first with one finger and then, without raising it, bring down another in advance of it while continuing to bow. But the player might also slide his first finger down from one position to the other, while continuing to bow. He would thus produce the same initial and final notes as before, but they would now be connected by a continuously altering series of notes, producing a peculiar effect called glissato gleessaa-toa, or glissade gleesaad. Similarly the voice may sing two notes at different pitches quite distinctly from each other, or may run up through a continuously changing series of, pitches, from the lower to the upper, in a method known as portamento poa-r'taamain-toa. I call Ijoth these effects " pitch glides;" and inasmuch as notes played or sung at different pitches shew different qualities of tone, which fade into each other during the glide, we have " quality glides " also. Similarly in passing generally from one voice sound or flatus sound to another (voice to voice, voice to flatus, flatus to voice, or flatus to flatus), or from one silent position to voice or flatus, or conversely, there will arise a series of " glides," which form the " cement," as it were, which connects the isolated sounds into one whole. These are theoretically represented by + (read '.' glide") between the symbols of the extreme sounds; practically it is possible to dispense with this sign. An. imperfect form of the glide, where sound is never dis- 26 SPEECH IN SONG. [Sect. Vlll. continued, but merely weakened between the extremities, is here called a " slur," from its partial resemblance to a musical slur or legato laigaa-toa effect. This is theoretically symbolised by ~ (read "slur"), as a kind of broT^en +. When there is absolute silence, or absence of all connection, between the two there is said to be a " break," which is symbolised by ... (read " break "). Elements of Speech. — We have to consider in order — 1. Forms of "voice," which in its formless or amorphous umairrfiis condition is h'. 2. Forms of " flatus," which in its formless or amorphous con- dition is °h. 3. Forms of " silence " (produced by completely blocking the passage of either voice or flatus), which in its formless or amor- phous condition is indicated by jt. 4. Forms of " glide "" from any one of the first three to any other, symbolised gei.erally by -f between the symbols of the other forms, under which we may include, as particular cases of growing imperfection, the " slur" (-^) and " break " ( ... ). It is evident that the principal business of the singer lies with the voice and the voice part of glides, on both of which more or less singable sounds can be produced. But, as the singer has to make his words inteUigible, he must not only learn how to pronounce the various voiced forms and voiced glides correctly, but also how to connect them with flated and silent forms in such a way as to make the latter well felt and understood by the audience, with as little interruption as possible to the music. This will be the main object of the following pages. Transition from Voice to Silence. — It would therefore seem to be most natural to begin with what was easiest to sing, and to descend by degrees to what is most difficult, then to turn to what cannot be sung at all, but has a sound, and to end by the silent positions which suppress all sound. Taking only some of the best-known sounds, this would give the following order : — Voiced Sounds. — Singable. Vowels AA EE 00 /Vocals L M N R' m Buzzes Z V c Sonants B D G Co § i Flated Sounds,. — Unsingable. Hisses S F Silent Positions. — Inaudible. VMutes ... P T K After which the connections between any two classes — that is, the " glides " and " slurs " — and even the " breaks " would have to be studied. Sect. VIH.] VOICE, FLATUS, SILENCE, GLIDE, SLUR, BREAK. 27 All voice tones are original qualities of tone modified by definite forms of the resonance cavities between the larynx and the air. If the forms change continuously the voice tone changes continuously in quality, and generally, unless care is taken with the vocal chords, in pitch also. The fixed forms give vowels, vocals, buzzes, and sonants, according to the degree of musical character in the resonance. The vowels have very various degrees of musical character under particular circum- stances ; some are as bad as vocals, and can be made even to approach nearly to buzzes. But, as a general rule, the vowels possess a musical character which does not belong to any of the other forms above mentioned ; and this has led to the usual classification of speech sounds into vowels and consonants. The latter word implies that the effects could not be produced except in conjunction with vowels. This is not correct, as the following exercise will shew, justifying the preceding division of the voiced consonants according to their musical character. Experiments on the Musical Character of Vowels, Vocals, Buzzes, and Sonants. — Assuming a general know- ledge of the common English sounds, sing the following passage from " See the conquering hero comes " to the notes here given, which have been transposed to be within male compasses when taken an Octave lower ; the same speech sounds will be sung to the second line as to the first. i Ex.5. ii s w "S?" ee aa 00 I I m ee aa 00 I tn ng ee aa 00 I tn ng b b S S ee aa 00 I m ng r' b g ee aa 00 I m ng V b ee aa 00 I m ng ee aa 00 I m ee aa 00 I m ng ng r' r' V V b b g g ee aa 00 I m ng b g iS SPEECH IN SONG. [Sect. Vlli, This exercise requires considerable care to execute satisfac- torily, but even an imperfect attempt will convey more informa- tion than a lengthened description, respecting the different effects of vowels, and the great distinction between them and the voiced con- sonants. A few directions may be of service. Sing the sounds as clearly, firmly, and musically as possible, as if you wished to address a large audience ; a little private note, as it were, for your own use, would only give you the pitches, not the qualities of to»^e at their best, which is just the point to be discovered. Vowels. — Lead off with ee. It will be found very difficult to preserve its character purely on the opening g' and the notes below it, but it will come out brightest on the c". On the low c it will become very " reedy," and decidedly change in quality. The whole passage will sound thin and unpleasant, even when sung by a practised voice. Changing to the aa, observe how the whole effect alters, and how rich the tone is, and how comparatively clear and fixed the character of the vowel is through the whole Octave. On closer observation the vowel will be found to "thicken " below g' and to "thin" perceptibly at c", losing much of its quality, and becoming " fluty." The change to oo is very great indeed. The whole richness of the aa line disappears. The sound is altogether " fluty," or, in the male voices, approaching the sounds emitted by the wide- . stopped flue-pipes of an organ. The oo will come out best on the notes below g", and will have to be considerably altered in quality to be produced well at c". Vocals. — Some difficulty will be felt at first in singing the following lines without any vowel at all. But if we observe such words as : little rhythm lit-l rith-m, we shall find that the last syllables have really no vowel, but that the vocals I, m can be sustained and sung on for any time. The effect, however, even of these, the two best vocals, is unpleasant, and forms a strong contrast to either ee or oo, with which they should be compared. The ng does not occur as a separate syllable in English. To secure singing it without a vowel, keep the mouth shut, but beware then of falling into the much more agreeable m or n. Perhaps it is best to begin with such a word as sung, and prolong the ng, at first with the mouth open, and then to shut the mouth while the ng continues to sound, observing that opening and closing the mouth makes no real difference in the sound, unless the position of the tongue is also altered. The tone of ng is harsh in the extreme, and no one would willingly sing on it except for experiment. These exercises will convince the singer that he should avoid sustaining such sounds as I, m, n, ng when singing. The trill r' presents peculiar difficulties for Some persons, but if the student is unable to trill an r' he must skip this line. It Sect. VIII.] VOICE, FLATUS, SILENCE, GLIDE, SLUR, BREAK. 29 will not do to substitute for it the usual untrilled r of the south of England in : heard her murmur herd her mer-mer, for that is simply a dull vowel. It will be impossible to take the slurs properly on r', so that the notes must be sung separately to a repeated r'. Buzzes. — The buzzes z, v must be carefully held, or' they will fall into the hisses s,f, which have indeed an independent sound, but are wholly unmusical (p. i8a, b, c). The effect of prolonging the z, V is not very dissimilar from that of playing the delightful schoolboy instrument, a piece of thin paper wrapped round a small-tooth comb. These prolonged buzzes are very disagree- able to the singer as well as to the listener. The low notes are especially difficult to bring out. Sonants. — It will be some time perhaps before the student can bring out a note on b at all. The mouth must be kept resolutely shut from the first, to prevent the furtive escape of a vowel. The passage to the nose must also be kept closed by the uvula, so that no m may appear. The sound is very much smothered, and soon comes to an absolute pause, but may be a little prolonged by swelling out the cheeks, as in blowing. After each note it will be found necessary to relieve the mouth of its condensed air by allowing some to escape by the nose. Of course no note can be sustained for its proper time, and slurs are impossible. The g is still more difficult than the b, because the resonance cavity of the mouth is reduced to a minimum. Practical Division of the Subject. — Although this experi- ment fully justifies the preceding graduated scale from vowels to sonants, yet, as a matter of practice, it is very inconvenient to separate the voiced and flated consonants in studying the generation of their sounds. Again, although the mutes have absolutely no sound, yet their positions are so exceedingly well defined, from the complete contacts which close every means of escape for the air, that they become the simplest forms to which all the other consonant positions can be referred. But it is im- possible to consider the mutes properly without having regard to the glides by which they manifest their action. In the following pages then such an order will be chosen, without reference to exact theory, as will tend to make the explanation and compre- hension of their sounds and their relations most easy. This leads to the consideration of the most marked positions first, referring the others to them, and noting some at least of the most important ghdes by the way. Mirror and Probe. — Those who wish to study the positions of the vocal organs, so far as can be done without special instru- ments, are recommended to procure a very small mirror (a " doll's iookin g- glass " from a toy-shop will do, or even a broken piece of looking-glass, the principal conditions being that it should be 30 SPEECH IN SONG. [Sect. VIII. IX. very small, not more than two to two and a half inches either way, so as not to exclude the daylight) and a long bone knitting- needle, with a knob at the end. These will be referred to as the "mirror" and "probe." IX.— VOWELS IN GENERAL. Vowel Defined. — A vowel may be defined as a " fully musical " modification of an original quality of tone which is produced in the larynx, by means of definite forms of the resonant cavities above the larynx. VovyEL Genera. — Every change of form in these cavities, and every change in the original quality of tone, consequently produces different vowels. Each different person, at each different pitch that he uses, produces a different vowel. Individual vowel qualities are therefore infinite in number. But the experience of speakers has grouped them, more or less carefully, into several kinds or genera jen-ur'u, the individuals within which are practically identified. Great differences of opinion exist as to the proper number of these groups. Any one hearing or trying to pronounce a new language for the first time will be sure to confuse groups of vocal sounds which the native accurately distinguishes. The English vowels, for example, present peculiar difficulties to all foreigners, and the differences are instantly detected by English ears, although generally the resemblances are accurate enough to prevent confusion of thought. The vowels of the English dialects are sharply distinguished by peasants in neighbouring villages, even when all the speakers are unable to catch the received pronunciation of educated speakers. Persons who have become familiarised with a mispronunciation are generally totally unaware that they use it, and can seldom be made conscious of it without great difficulty. All this must be carefully borne in mind. We can then understand that as vowels have naturally been accommodated to the speaking, and not the singing voice, the mere fact of singing produces an alteration o< more or less importance ; yet nations are ready to identify spoken and sung vowels if not too diverse. But the great variety of pitch in which vowels have to be sung materially changes their character, as we have already learned by -Ex. 5 (p. 27c), and hence produces still further diversities between these and the spoken vowels. Vowel quality is best perceived in rapid succes- sion. All vowels are spoken in a very short time indeed, and when lengthened are apt, especially in English, to glide off to a totally different pitch and quality. In singing they have to be sustained at a steady pitch and with a steady quality, and the ear consequently rapidly fails to recognise any particular vowel. S*CT. IX.l VOWELS IN GENERAL. 3> Vowel Resonance Cavities. — None of the resonance cavities, except perhaps the larynx itself, are capable of resounding to a very low tone. According to Koenig, who does not differ materially from Helmholtz, the resonance cavities of the principal vowels differ by a whole Octave, and he puts them at the follow- ing, where •* indicates flatter than on the piano: — Ex. 6. *»W i ^ ^^ Octave JZ t higher. s 00 Vibrations 224 OA 448 AA 896 AI 1792 EE 3584 It is obvious therefore that, if we heard only those particular partials which the resonance cavities reinforced or weakened, we should hear very little indeed. The resonance cavities then appear to act much in the same way as the resonant cavity of a violin. This presents two principal resonances, one between ^ and 1^ and one between and: ^ The reso- nance for the violoncello is between For the and viola the two resonances are each about a tone lower than for the violin. The effect of these resonances is to reinforce certain partials of certain notes. Thus only the Octaves of i 5*=^^=^ the three lowest notes of the violin, can be reinforced by the upper resonance tone. The notes %^ J —— I i them- selves, and not their Octaves, will be reinforced by the lower resonance tone, while i?> . J J J^ - will be reinforced by the higher resonance tone (see my translation of Helmholtz, p. 86). None of the other notes will be directly affected. But all the other notes are heard, and well heard, in consequence of the vibration of the walls of the box, communicated directly to the outer air, and indirectly through the air in the box, by which it is strengthened. Similarly the strong resonance of the lower SPEECH IN SONG. [Sect. IX tones of the human voice, communicated to the whole upper part of the body (which may be felt by the hand even to vibrate with it), acts directly on the outer air, and also indirectly, but essen- tially, through the resonance cavities, by which it reaches the outer air, and which materially influence its quality of tone. Male and Female Vowel Qualities. — But a large portion of the voice of a man is, so to speak, unvowelised. I have often listened to male public speakers and noticed how much sound they produced which was not differentiated into vowel quality, and which therefore rather diminished the intelligibility of their utterances. The part really effective seemed to lie, in general, at a high pitch, and to sound so weak and thin, that the wonder was how it was perceived at all, and that it was no wonder that ears required long practice before they could apprehend such minute and fleeting differences. On the other hand, this did not strike me so much in those female speakers who used their Lower Thin Registers. With them the voice always appears to be more thoroughly vowelised, principally, I think, from the absence of the deep partials. The subject has not been suffi- ciently investigated, but it has struck me that this may be one reason why it is difficult for one of either sex to acquire speech sounds from one of the other sex. Influence of Register on Female Vowel Quality. — Madame Seller (" Altes und Neues iiber die Ausbildung des Gesangorgans," 1861, p. 52, which differs slightly from the English edition, 1871, p. 102) gives her own great experience as to the effect of pitch on the vowel qualities of the Female Voice, which may be condensed thus, introducing a mark to show the change of register (p. 17^), which obviously affects the continuity of vowel change. i Ex. 7. Upper Thick Register. Below. Lower Thin Register, 3 ^MtJ-^^^^^ q= TJ =sr Favourable to 00 oa ah an all vowels Upper Thin Register. m to Small Register, ^ ^=£ ee The change into the Upper Thin makes one of the thin vowels, te, most prominent. But the entrance upon the Small Register Sect. IX.] VOWELS IN GENERAL. 33 makes the favourable vowel change from ee to aa, whence it proceeds to ee. Below c' all the vowels are muffled, and approach oa in character. Women students should verify this table by trying to sing each of the vowels here marked throughout so much of this scale as they can accompHsh, and noting on what part it sounds best in their voice. They should also note whether there seems to be any retrogression when the Thick Register is cultivated upwards to g' or a' , and also when the Small Register is cultivated down to e" or d". Male Vowel Quality. — For Male Voices Helmholtz (see my translation, p. iii) gives as the result of his own observations, which had not been very extensive, that " vowels speak best when their characteristic tone is a httle higher than that of the note sung, and those speak next best which have a characteristic tone that is the second or third partial of the note sung." The results may be given as follows, the minims marking the charac- teristic notes of the vowels according to Helmholtz, and the crotchets the notes sung. The last note was at the extremity of his own falsetto. ="•« ■ b«- m i i W. m=^ ■^ ^^-F-=^^=^=^ Favourable to oo Willis's and Helmholtz's Experiments on Vowel Reson- ance CAViTiES.^That vowels are really qualities of tone, due generally to resonance, has been shewn in various ways. Pro- fessor Willis many years ago, by applying a reed to one end of an extensible tube, found that, as the latter varied in length from about one-third of an inch to four and three-quarter inches, the ear caught a change- of vowel through ee, ai, aa, au to oa. This experiment I witnessed many years ago, when repeated by the late Sir Charles Wheatstone. The change had to be made rapidly to the proper lengths of tube, otherwise the intermediate glide rendered the distinction difficult to hear. When any length of tube was retained for too long a time, the vowel quality became an indistinct drone, like u in but. Similarly, Professor Helmholtz (ibid. p. 117, where Willis's results are given) applied to the same reed different resonance cavities tuned to different pitches, and found that when they were tuned to, Ex. g. 1 mi ^^- they gave 00 i \Ut- W s ^ 34 SPEECH IN SONTt. rSr.CT. IX. where a' is the a in ask a'sk, as subsequently defined, he also got ae, oe, at, ee, by applying spherical resonance cavities into which tubes varying from two and a third to four inches were fixed, in order to give two resonances. Helmholtz's Experiments on Artificial Vowel Qualities. — Herr Georg Appunn, of Hanau, has made an instrument with organ pipes by which the vowel sounds can be produced with great exactness. This is highly spoken of by Professor Helmholtz, who, however, used a compHcated apparatus of tuning-forks kept [Continue text on p. 35, Ex. 10. See explanation on p. 35a. w -^ No. of Partial 00 OA AU AE AI / mf P P P ^ PP PP P P P f ■^- ^- P ff f P «4>t ^f- ff ff m -if ff '\% •r^ Ex. II. W: ^ No. of Partial 00 OA AA AE AI I / mf 7nf mf it / mf f mf P tnf f «Jl«. ^Jt / P f ff ff ff ff '^'e^^ ff ■f Sect. IX 1 VOWELS IN GENERAL. 35 in motion by electricity, and placed before resonance cavities to reinforce them, while an apparatus for partially or entirely closing the mouth of any cavity enabled him to enfeeble or entirely destroy the tone of any fork. The table on p. 34 is adapted from the condensation of his results, derived from two sets of forks, which I gave in a note to my translation on p. 124. The horizontal lines of notes give the pitches of the forks employed, which were exactly tuned as partials. The vertical line of notes gives the characteristic pitches of the vowels written against them according to Helmholtz (ibid. p. no, excepting the reson- ance d" assigned to au, for which see p. io5), the ae and ai having two resonances. The usual musical signs/, mf, p, pp in the same line with any vowel indicate the relative loudness of the partials by which the vowel in the same line was produced. Probable Variety of Vowel Resonances. — This will give some idea of what the nature of vowels really is. A great deal more has to be done to the subject before the results are per- fectly consistent and satisfactory. Thus Professor Helmholtz admits only a single resonance for 00, oa, aa, and a double one for ae, ai, ee, oe, ue ; but Professor Graham Bell, of Boston, U.S. (inventor of the Speaking Telephone), was able to shew me two resonances for all the vowels (ibid. p. 107, note), and the numerous cavities already noticed lead us to suppose that in all cases numerous resonances come into action, and unite in producing the vowel quality. Bell's Nomenclature for Vowel Resonance Cavities. — In describing a vowel it will therefore be in the first place neces- sary to describe the form of its resonance cavities. We can describe only those which we can see and feel, and hence the description must be rough. But it would seem as if there were some agreement among the connecting cavities, so that an adjust- ment of the larger cavities entails the approximate adjustment of the others, for it is certain that children born deaf, or rendered . deaf before they could speak, have been enabled by such descrip- tions to pronounce the vowels not only intelligibly but with great accuracy, as I have myself witnessed. I shall adopt the nomen- clature of Mr. Melville Bell, the inventor of Visible Speech — which his son, Professor Graham Bell, already mentioned, has intro- duced extensively into the Deaf and Dumb Asylums of the United States — because it is the most satisfactory with which I am acquainted, and I have used it for many years. Tongue Modifications of Interior of Mouth Cavity. — The principal resonance is in the mouth, supposing the nasal resonance to be cut off by the uvula. This is limited internally by the tongue, and externally by the lips. Three parts of the upper surface of the tongue must be distinguished: the "back;" the " front," or middle portion ; and the " point," or " tip-" 36 SPEECH IN SONG. [Sect. IX Either the back or front may be raised towards the palate, at various heights, of which three are distinguished as " low," " mid," and " high." And both back and front may be raised together to those three heights, giving " mixed " results. This will be found to furnish nine distinct positions of the tongue limiting the interior of the mouth. Of course there is a great variety of intermediate positions, which it will not be necessary to notice here, as this is a practical treatise and will deal only with sounds which the singer of English, German, Italian, Spanish, and French songs ought to know familiarly. Lip Modifications of Aperture of Mouth Cavity. — As we have constantly seen that the shading of the external aperture greatly modifies resonance, we are prepared to distinguish four principal arrangements of the lips : the first " open " (always assumed to exist unless the contrary be expressed) in which no part of either lip is in contact with the other, and the corners of the mouth have a distinct vertical cross-line in place of an angle ; the second " low round," in which the comers of the mouth are brought together, as well as a small adjacent portion of the lips (in English this opening of the lips accompanies the "low" positions of the tongue); third, the "mid round," where the lips are much more closed — fully half-way from the corners to the middle line on each side (this opening of the lips usually accompanies " mid " positions of the tongue) ; fourth, the " high round," where the central opening between the lips is very small indeed (this opening usually accompanies " high " positions of the tongue). Complete closure of course cuts off all sound, and, although it occurs for real sounds, as the sonant b and the nasal m, does not belong to the vowel group at all. The " open " and " rounded " forms of the mouth, each " rounding" being appropriated as above, give eighteen forms to the oral resonance cavity. Throat Resonance Cavity. — The other cavity connected with the mouth that can be felt to alter, is the lower pharynx, which may be called the " throat." This alters, not only by contraction and relaxation, or "narrowing" and "widening," but by the rising and falling of the larynx. Mr. Bell considers the " nar- row" position to be " primary," and to accompany all the others if not particularly named, but the "wide" has to be specified. The high and low positions of the larynx he has not specially regarded, considering them to accompany the " primary " and " wide " forms respectively. Many objections have been taken to this introduction of the resonance cavity of the lower pharynx as a characteristic in the classification of vowels, but I have not yet seen my way to a better hypothesis, and this seems to answer all the necessary conditions, although alterations in the original quality of tone must not be neglected ; but the last-named altera- SSCT. IX.] VOWELS IN GENERAL. 37 tions appear to depend upon circumstances that we are as yet unable to specify. The two pharyngal /Mr'«»g-^«Z positions apply to every one of the eighteen purely oral forms, giving thirty-six vowel resonance cavities, each of which of course may have several sub-forms. Experiments on Nose Resonance Cavities. — In addition to these the nasal passages may be opened, and allow the voice to pass through both mouth and nose. The quantity of air which passes through the nose may be regulated by the distance through which the uvula is withdrawn. This action of the uvula should be well understood. Open the mouth widely while facing a window, throwing back the head to admit full daylight on the back of the mouth (artificial light is insufficient), and apply a mirror. To keep down the tongue, think of the vowel aa or au. You will see the arch of the palate, with the uvula hanging from its middle like a prolonged keystone. Breathe quietly through the nose only. The tongue immediately rises, and entirely con- ceals the arch and the uvula. Its back is brought into air tight contact with the soft palate, so that the passage of air through the mouth is entirely cut off, and if the mirror is held close under the nostrils so that all breath from the nose passes over the back of the mirror, while the front faces the open mouth, the glass will remain perfectly unsullied. Now hold the mirror so as to watch the action of the tongue and uvula. Draw breath through the mouth, the tongue sinks ; expel it through the nose, the tongue rises. Inspire and expire by the nose only, the tongue remains fixed above ; inspire and expire by both mouth and nose, the uvula advances gently at each expiration. Expire with a jerk, the uvula is jerked forward, together with all the folds of the " soft palate," and the point of the uvula is thrown upwards, owing to the rapid passage of air both behind and below the uvula. Make two or three quiet respirations as before, and then sud- denly sing aa to a short staccato fiote at an easy pitch, and proceed with the quiet respiration. Do this until you distinctly observe that every time aa is thus sung there is formed a sudden dimple or saucer shaped depression in the uvula just below the arch of the palate, arising from bringing that part of the uvula suddenly against the back wall of the pharynx to close the approach to the nasal passage. The point of the uvula remains quite free. Repeat as before, but sing aa quietly and steadily, and observe the persistent continuance of the dimple. Then attempt to give aa a nasal "twang," written ,aa, and observe that the dimple disappears at the same moment. Alternate aa, fla, aa, flo. several times, and observe the continual formation and disap- pearance of the dimple. As the head is thrown back all the time, naturally tending to remove the uvula from the pharynx, it is clear that a muscular action is absolutely necessary to 38 SPEECH IN SONG. [Sect. IX. close the nasal passage. There will be some difficulty in keep- ing the tongue sufficiently low to allow the action to be seen when saying ,aa, because it instinctively rises to block the pas- sage to the mouth. But for the French word an (ahn' in Glossic), where the uvula is more advanced than for aa and the tongue is lower, the effect is more easily seen on repeating ah, ahn', &c. While saying aa, fla, or ahn\ completely close the nostrils with the fingers. Observe that the sound of aa remains absolutely unchanged, but that both fl.a and ahn' are altered, not into aa, ah, but into a different sort of nasal twang, resulting from reson- ance within the nasal passages, which are now open behind but closed in front. These "twangs" are serious faults in singing, and are here spoken of to be understood and avoided by sedulous practice with the mirror, if, as is not unfrequent, there is a slight tendency to pronounce aa with a nasal twang. But the greater degree of nasality «', forming the well-known French "orinasal" vowels, will have to be studied hereafter. Of course at least thirty-six such vowels are quite possible, but only four need be practised. Classification of Vowels by Height of Tongue, Width OF Throat, Rounding, and Nasality. — Although the front of the tongue may be low, the tongue itself will be higher than the high mixed tongue, while the low mixed tongue (p. 36a) is higher than the high back tongue. We are therefore able to arrange our vowels according to the height of the tongue, which agrees to a great extent with the pitch of the resonance, the highest tongue iContinue text on p. 39. VOWELS SYSTEMATICALLY ARRANGED. See explanation on p. sgfl. 2. ffq" Tonoue I. II. III. IV. V. Primary • Wide Round Wide round Orinasal I 2 3 High Front Mid Front Low Front EE AI ae I E A ui EO UE OE * * oen' aen' 4 5 6" High Mixed Mid Mixed Low Mixed * U * i' a' e' * * * * * * * 7 8 9 High Back Mid Back Low Back uu' uu ua u' AA ah 00 OA AU uo AO * can' ahn' Sect. IX.] VOWELS IN GENERAL. 39 producing the highest resonance pitch. This gives the diagrann on p. 38c, in which the large capital letters represent the vowels recognised in received English speech, the small capital letters the additional vowels in German, Italian, Spanish, French, or provincial English, and the small letters certain other vowels which it will be necessary to touch upon incidentally, while * marks the shape of the resonance cavities for vowels which it will not be necessary to consider at all. Trilinear Arrangement according to Vowel Effects. — But a mere arrangement like this by resonance cavities does not sufficiently shew the relations of effect. This is far more intel- ligibly conveyed by a trilinear (or triangular) arrangement like the following, in which five of the previous incidental vowels, ui, uu', i', e', ua, being very unusual, are omitted for convenience. The Speaker's Vowel Trigram. EE. I AI E ae A a' /UE EG OE u' U UU AA 00 UO OA AO O AU ah^ This indicates that if we proceed from left to right on the top line from EE to a' we seem to be continually approaching AA ; and if we descend to AA in the middle line, and then to ah in the lowest hne, and continue the series after passing through AA, we seem at each step to get further and further from AA, though in another direction, and then finally reach 00. If from either EE or 00 we go to ue, we arrive at a vowel which partakes of both the characters of EE and OO, and still more of I and UO (the " wide " forms of the former) ; and if we then go to the right we reach AA by another path, which, as far as each of the vowels ue, EC, oE are concerned, seems to lie in some respect "between" the paths of the vowels above and below it. This is not quite so marked for the next three vowels, u', U, uu. All this will be better understood after the treatment of each of the twenty-six oral vowels in the table, including the incidental ones, in ten groups in the following order : — EE, I, i'— AI, E, AE— A, A'— A', AA, ah— AU, O, AO— AO, OA— UO, 00— ui, UE— Eo, oE— uu', u', U, e', uu, ua, where it will be observed that each of the vowels a' and AO forms part of two groups, the reason for which will appear hereafter. Characteristic Vowels. — All the vowels in capitals and small capitals in the last list will have to be well mastered by the speaker. But the singer will have to master only ten of Ihern. 40 SPEECH IN SONG. [Sect. IX. X besides the orinasals, representing the ten groups marked off above, in a manner to be hereafter explained. These may be arranged thus : — The Singer's Vowel Trigram. ^UE OE U — AA (uO OA AU-^ Each of these vowels represents a group, of which the indivi- duals are generally brought out by pitch independently of the will of the singer, as will be presently explained. Length of Vowels. — The length of vowels is very important in spoken English and German, although it is not well determined in spoken Italian, Spanish, and French. But the singer has to take his vowels short or long according to the will of the com- poser. The sole means of distinguishing the occurrence of long and short vowels consists in the mode of running the vowel on to the consonant, that is, the nature of the glides, as will be explained hereafter. X.— VOWELS IN PARTICULAR. Oral Vowel Series. / Group.— ££, I, I'. EE. — Front or middle part of tongue high, very near to the hard palate, but sufficiently far from it to admit of the passage of a strong ribbon of air. Point of the tongue directed down and completely concealed from view in the mirror by the lower teeth. Back of the tongue probably nearly parallel to the uvula but not touching it. Sides of tongue at the back part pressing against both upper and lower teeth, and against the lower side teeth to within a very little distance of the point. The cavity of the mouth is therefore like a very short rather flat funnel, of which the tube runs at the back of the tongue into the pharynx. Such a tube has a very high resonance, as may be seen in Ex. 6 (p. 31a), The lips are wide open, so that the disposition of the tongue can be well seen in the mirror. The " probe " (p. 30a) should also be used, with the pointed end inserted above the tongue as far as it will go. Measure the length of the part inserted, from the upper teeth to the point (in my case an inch and three-quarters), and feel out the way in which all side passage of the air ia S"T. X.] VOWELS IN PARTICULAR. Stopped by the pressure of the tongue against the teeth. The throat is primary, or narrow (p. 36^), and the larynx high ; the epiglottis is as widely open as possible. The voice is set on with a clear attack or glottid (p. i8c). Examples of words containing the sound. English : final, we wee, ye yee, he hee, pea pee, be bee, tea tee, key kee, kefee, thee dhee, sea see see, she shee, lea Lee Legh Leigh lee, me mee, knee nee ; before a consonant, eat eet, each eech, eke eek, eve eev, ease eez, eel eel, e'en een ; between consonants, heap heep, weep weep, peep peep, deep deep, keep keep, sheep sheep, leap /ee/>, reap r'eep, glebe g-Zeei, heat Aeef, wheat wheel, peat ^ee^, beat beet, cheat c^ee^, feet feet, seat s««f, sheet s^eef, leet /eef, meet meet, neat wee^, heed heed, weed weerf, bead 6eerf, deed deed, keyed keed, feed feed, seed see^, lead /e«rf, read reed r'eed, mead weerf, need knead need, weak lefeeA, peak pique peek, beak 6eei, teak iee^, cheek cAe«^, seek seek, leak Z«e^, reek wreak reek, meek jwee^. German : long, sie zee, ihn een, lieb Zee/), sieht ze«f, mir meer' ; short, ich eeky'h, bitte beet-u. Italian : middle length si see, di 3-- aj ai e e ae as ai e ae ai e ae !ij at ai e ae ai e ae ai e ae ai e ae A ' Group. — A , A '. A. — The position is the same as for ae (p. 446), but the throat is wide. The vowel is almost peculiar to the southern and eastern pronunciation of English. But it occurs in Scotch occa- sionally, in Dutch, and in Hungarian, as a substitute for short e or ae, and it has been recognised in Lap, Ostiac, Gaelic, Portu- guese, Catalan (Roman), and Rhetian. It is very common as a short vowel in English closed syllables, as, hap, hat, had, hack hak, hag, has haz, gas, gash, sallow sal-oa, Iamb lam, tan, sang. In the town of Bath, Ba-th locally, it occurs as long, and all cases of long aa there become long a, as card kaa-rd, locally ka-rd, whereas cord kau-rd is locally kaa-rd. The lengthened sound is about the most disagreeable vowel possible for a singer. If ae resembles the bleat of a young lamb, a may be likened to the bleat of an old ewe. The singer . should verify its character by the last exercises (Ex. 14, 15) on ai, e, ae, adding a, and treating it in the same way. It will be felt that this is a sound which every singer should avoid, if he has a reasonable excuse to do so. Foreigners naturally regard a&s a. very " broad " pronunciation of their own ae, from which few succeed in distinguishing it. Some un- polished speakers will speak of a cab as kaeb, instead of kab, and call bank baengk, in place of bangk. Even refined speakers will thengk or thaengk, instead of thangk-ing. Throughout the midland and northern English counties a is replaced by aa, and this has such a boorish effect that no singer should venture on it, though after all aa is a better and more in- telHgible substitute than ae. The real substitute is the following A'. — Tongue in the mid mixed position, the same as for m._ It is lower than for a. Use tht mirror and say ai, aa. For ai the back of the mouth is quite invisible, because the tongue blocks up the passage. For aa the whole inside of the mouth, the arches, and the uvula are well seen. For ae or a just the very top of the arch is visible, so that in passing from ai to a the tongue descends much less than in passing from a to aa. This should be verified with the mirror. Between the a and aa heights of the tongue, several could be interposed, and about midway lies that for a'. The tongue is nearly flat, lying about the height of the top of the lower teeth, and quite free from the upper teeth. The top of the 48 SPEECH IN SONG. [Sem. X arch is well seen. This is generally assumed to be the most natural position of the mouth, and u (hereafter considered, p. 62), is hence often called " the natural vowel," although of course all vowels in constant use are equally natural. But for u the throat is "narrow," for a' it is wide. The effect of a' is that of a very fine aa, or a rather broadened a, and has none of the boorish character of aa. In the southwest and east of England this a' is commonly substituted for a, and it is sometimes difficult to dis- tinguish whether a or a' is said. Before the sounds/, v, th, s, m, and « diversities of pronunciation occur. Compare balm, calm, palm, psalm; calf, calve; half, halve ; chaff, staff ; graft, shaft; lass, pass ; ask, bask ; asp, clasp ; cast, fast ; bath, lath, path, wrath ; chance, dance ; plant, aunt, haunt, jaunt, gauntlet. It is most usual to say baa-m, kawm, saa-m, but many say ba'-m, ka'-m, sa'-m, and not a few ba-m, ka-m, sa-m; no one says bam, ham, sam, with a short vowel, because these words have different meanings, written " bam, cam, Sam." Similarly " halve" is never hav, though " half" is occasionally haf. But in all the other words you may hear from different speakers a, a', aa short or long, as pas-, pa's-, paas-, and pa-s, pa'-s, paa-s. For some time past the aa'- as paa-s, laa-f has been cried down, although in a language so deficient in long aa- as English it is a pity to . lose it, especially to the singer. The short aa, however, is allowed, as paas-, laaf-, but the sound most usually heard from refined speakers is a' long or short, &S pa'-s, la'-f, or pa's-, la'f-. The real short a is seldom heard short in these words, except from refined speakers in Leeds and other Northern towns where it maybe due to a reaction from the prevailing a^ 1 ^f— 1 ^~ — 1 v* } -'-^ r • . « t. aa aa a % aa aa aa at I ai I aa ac I at I i n Ex.17. ^ ^^E^ 5=:P ^ ■•-*H- ^z3l zi^:r=g. S^= -^=^ qiii=^= aa aa aa aa aa aa aa aa aa aa aa aa Ex. 18. aa aa aa aa aa aa aa aa aa aaaaaaaaaa Sect. X.l VOWELS IN PARTICULAR. 5 1 The rapid detached notes in the last example are particularly difficult to bring out well, with a good quality of tone, and with a perfectly clear glottid before and after each aa, thus faaf. It would be found much easier to sing them with laa, or even taa. The exercise is, however, on aa only. A U Group— ^ U,0,AO. Rounded Vowels. — All the preceding vowels are spoken with an open mouth, from the widest opening and acutest resonance EE, to the widest opening and gravest resonance AH. To deepen the resonance it becomes necessary to close, or, as it is called " round," the lips more or less, and this is the chief characteristic of this and the four following groups. AU. The tongue is disposed in the same position as for ah, but the throat is narrow, and the lips are slightly contracted at the corners, thus diminishing the aperture very slightly, and of course flattening the resonance. As already remarked there is a great tendency to pass into au from ah in singing downwards. The vowel au is very common in English, as in : awe au, daw dau, caw kau, law lau, gnaw nau, draw drau, laud, in all of which words care must be taken not to let the least suspicion of an ' r ' occur ; on the other hand r' may be used in : or au (before a consonant, but aur', or or' before a vowel), nor nau, naur', nor' (under the same circumstances), lord lau-d. These words may be always sung as au-r', nau-r', lawr'd. The introduction of a hint at 'r' (the meaning of which expression will appear hereafter) is never allowed where ' r' does not occur in writing. The follow- ing are more common words of the same two sorts, but the singing forms alone are given, i) Without R, dau-b, awed au-d, cough ytaw/ (sometimes kof-), hawk hau'k, hall haul hairl, fallen fau- In (one syllable, not fau-l-n), salt sawlt (often solt), awn au-n, hau-nt or haa-nt, ought aught au-t, broth br'au-th or br'oth. Some speakers call off au-f, coffee hau-fi, office awfis, often au-fn, dog dau-g, cross krau-s, and even long lau-ng occurs in America, pronunciations ridiculed or despised by those who persist in saying o/-, kof-i, of-is, of-n, dog-, kros-, long-. As far as I have observed, the use of dau-g, lau-ng is not common, au-fis is more usual, and au-f, kau-fi, au-fn, kraus very common indeed. 2) With R, drawer drau-r', orb au-r'b, scorch skau-r'ch, scorn skau-r'n, dwarf dwau-r'f, horns hau-r'nz, horse hau-r's (hau-s is always used in speech) north nau-r'th. The sound of au is quite unknown in German, Italian, Spanish and French. Foreigners usually confuse it with aa or ah, and occasionally with ao. It occasions them great difficulty. In English itself it seems to be a recently developed sound replacing aaw or ao. 52 SPEECH IN SONG. fSncT. X. O. The mouth and lips precisely as for au, but the throat wide. As many persons do not hear the difference between au and 0, the following contrasts should be studied. Both the words in au and o should be first pronounced short, and then long, those in o being drawled, and it should be observed that though the preceding and following consonants are the same, the words never sound alike. This will serve to separate the sounds dis- tinctly in the mind. Awed odd au-d od-, pawed pod pau d pod', pawned pond pau-nd pond, sawed sod sawd sod, hawed hod hau-d hod, haul holiday hau-l hol-idai, maul Moll mau-l Mol, stalk stock stau-k stok, wall wallow wau-l wol'oa, auricle oracle au-r'ikl, or'-a'kl (spoken or'-u-kl), awn on awn on, yawn yon yau-n yon, gnawed nod nau-d nod, wrought rot r'awt r'ot, taught totter tairt tot-er. The vowel o is unknown in received German, Italian, Spanish and French, where it is replaced by short ao or short oa. In the North of Germany, o occurs as a local pronunciation. Before r' in German, Italian, Spanish and French all Englishmen at first naturally hear the foreign ao as their own au or o, and it is generally some time before they can even perceive, and still longer before they can acquire the dis- tinction. In the same way foreigners hear English au as aa, and English o as ao. AO. This vowel has a very different position from aw. The tongue lies exactly as for aa, and the throat is wide. The lips are in the "mid-round" position, that is, are half closed, so that the resonance is much deeper. Sound aa and close the lips rapidly to the mid-round form and then open them rapidly again. The alternation of aa ao aa ao will be produced. The motion of the lips should be rapid, or intermediate vowel qualities will destroy the contrast. The vowel ao occurs indeed in English, but only before an r in a long syllable, as forming part of a diphthong to be hereafter explained, and then it is an alteration of oa ; the glossic symbol oa-r really means ao-ur\ as will be explained, and bearing this in mind the vowel is properly heard in oar ore oa-r, door doa-r, floor floa-r, roar r^oa-r, yore yoa-r, hoard horde hoa-rd, mourn moa-rn (also called moo-rn), course coarse koa-rs, hoarse hoa-rs (distinguish from horse hau-rs.) The sounds should be kept apart from oa-er, doa-er, Jioa-er, r'oa-er, yoa-er, hoa-erd, moa-ern, koa-ers, hoa-ers, which may be heard from elderly people. But the modern custom of speech is to reduce ao to au, and get rid of the u, in whole or in part, so that ao-ur' reduces to aur', and before a consonant r' is omitted, though it should be made just audible in singing. The above words then are commonly heard as au or au-er,dau'ox dau-er, flau or flau-er, r'au or r'auer, yau or yau-er, hau'd or kau-erd,mau-n (so thai 'mourn' is not distinguished from 'morn') or mau-ern (which makes the distinction,) kaus or katrers. Smt. X.] VOWELS IN PARTICULAR. 53 hau-s (so that ' hoarse ' is not distinguished from ' horse ') or hau-ers (which restores the distinction). On account of the great prevalence of the sound of ao in German, and French, where it replaces English o, and in Italian and Spanish, where it is a marked sound of middle length, (the only sound of o in Spanish), and also for another reason given in the next group, it is recom- mended that Englishmen should study the production of ao which has a very beautiful quality of tone, much superior to au. The exercises on ao will be taken with the next group to which it properly belongs. AU,0. Exercise as follows. First try the compass as before. Observe that au and o give an excellent low resonance, but that on the higher parts of the scale the sound begins to become too " fluty" and that o becomes more easy than au. Compare nod- on a low and gnawed nau-d on a high pitch. This change from au to o is quite immaterial except in low pitches where o resembles oa too closely. Hence it is better always to think of singing au, and to leave the pitch to settle the rest. Sing all the preceding exercises 12 to 18 on au. Be very particular with the upper notes. OA Group— ^O, OA. AO. This was described in the last group. Tongue for an, mid-round lips, throat wide. OA. This differs from AO only in having the throat narrow, which however greatly alters its character. OA-W, OAW, OW. As ai tended to vanish into i by raising the tongue, producing ai-y, so oa tends to vanish into 00 by closing the lips while proceeding with speech, producing oa'w In London and its vicinity this is so marked that many speakers cannot sustain oa at all, but run off with a glide to 00 or a sound very like it, producing almost the effect of the diphthong ou, in the form oaw, (which differs from oa-w in having a short initial vowel), or even o!£', just as the ai-y vanish degenerates into aiy, ey, aey, and then becomes confused with ei. Observe that the same speakers who pronounce what they mean for ai, oa, in a way which sounds almost like ei, ou to those who do not use the vanish so distinctly, are far from confusing ai, oa with ei, ou to their own ears, because instead of calling my cow mei kou, as the others would, they broaden them into maay kaaw, or even use maay kaew. These errors are mentioned only to be de- tected and avoided. The use of the vanish is very prevalent when oa ends a syllable at the end of a sentence, as : do you go ? no. oh ! doo eu goa-w ? noa-w. oa-w ! and the two last words are especially apt to sink into noaw, oaw. Compare also bow boa-w, doe dough doa'w, though dhoa-w (Scotchmen say thao, with th and without the vanish) foe foa-w, blow bloa-w, slow 54 SPEECH IN SONG. ISect. X. sloa-w, sow sew soa-w, mow moww. But when a consonant follows, this is not so common, so far as I have observed, although many orthoepists as Smart, Cull, Bell, make oa-w universal. The tendency towards the oo effect is of course greatest before k, g and p, b, and least for t, d. Thus choke choa-k or choa-wk, (which however I cannot say without an effort,) rogue r'oa-g or roa-wg, hope hoa-p or hoa-wp, robe r'oa-b or r'oa-wb, home hoa-m or hoa-wm. But mote moat moa-t, mode moa-d, close kloa-s kloa-z, most moa-st, oa-f, have not nearly so much tendency to 00. Before Z, as sole soal soul soa-l, hole whole hoa-l, the slight u sound which often precedes an I, is apt to give a diphthongal effect, and there is just a tendency to close the lips a little more towards the end of oa, but the tongue remains quite still, so that no real sound of oo is reached. O'A. Another not unfrequent mispronunciation of oa is of just the reverse nature. In oa-w the lips begin in the proper position, but are brought too close as the vowel proceeds. In 6a, as it is written, the mouth is quite open to begin with (producing uu), and the lips close up quickly to the position. This may frequently be heard in the interjection oh! 6a that is uu + oa-, where the + represents a glide. Sometimes even the two effects are united, by the lips closing too much thus producing 6a-w. AO, OA. The changing character of the vowel is however prejudicial to the effect of the note, and hence singers are recommended in English to aim at singing ao, from which there is no tendency to run off into oo, because there are no bad habits to be overcome. In German however, oa' long and ao short should be used for long and short ' o ' as : ohne holtz oa-nu haol-ts. In Italian, oa is always used in unaccented syllables, and both oa and ao are used in accented syllables, but where oa and where ao occurs can be learned only from a dictionary. In Petronj's, the ao is always indicated by a cir- cumflex, as volto (arch, arcade) vaol-toa, volto (face) voal-toa. In Spanish ao only is used. In French oa is generally- long and ao short, as in beau boa', botte baot (not bot.) Exercise with the compass as before. Observe that difficulties in ascending are felt more strongly than for au, o, but that still a very tolerable quality can be got out for ao even at a high pitch, but that this is not so easy for oa. For very high pitches it will be found convenient to open the mouth wide and slightly contract the arches of the palate, which helps to produce somewhat of the same effect as rounding the lips, and at these pitches makes the tone more sonorous. For the middle and lower notes ao is generally very fine, much pleasanter than au. The same exer- cises 12 and 13 can be taken as for ee observing that it is now the high notes which have to be regarded rather than the low. S=CT. X.] VOWELS IN PARTICULAR. The exercises for the other vowels should also be tried, and the following scraps. Ex. ig ^„ ^y. - -^ / it... p H <<,- -#■ ■ it M ' 1 r • 1 c S * n m 1 r. s . r 1 ■ r ^]J (1 \^ t> ao oa ao oa ao oa ao oa ao oa ao oa ao oa ao oa Ex. 20 m PJ- »,(./».-, m /-J _ 1- *^ It' ' 1 — \^ Id 1 f " 1 1 ao oa ao oa 20 oa ao oa ao ao oa oa ao ao oa oa ao oa 1 ao oa UO Group— [/O, 00. UO. The back of the tongue, as distinguished from the front, is raised nearly into the position which we saw it assume for excluding the breath from the mouth (p. 376), that is, nearly into the k position, but not so high, and of course not so as to exclude the passage of the breath. For aa, ao, oa the back of the tongue was of mid height, and the whole tongue nearly level but low. For UO the tongue rises at the back only, and the back is thus higher than for aa, but is still rather low. The lips are closed to the "high-round" form, so that the aperture of the mouth is very small, making the pitch of the resonance low. The position of the tongue is invisible in the mirror on account of the closure of the lips, and must be felt out with the probe. The throat is wide. This wideness of the throat gives the vowel much thick- ness, and makes it somewhat resemble oa. The sound is not frequent in English, but yet it occurs in several very common words, and it is always short, except in a few cases to be con sidered hereafter, p. 761^. Thus good guod, could kuod, should shuod, would wood wnod, book huok, look luok, took tuok, bull buol, full fuol, bush buosh, push puosh, foot fuot. This vowel is generally sounded for German short u, as bund huont, but it does not occur in Italian, Spanish or French. Examples will be given on the next page. 00. This differs from uo only in having the throat narrow, though the mouth is also frequently brought closer. The sound is much finer than uo, but does not occur short in closed syllables in English. Compare English pull puol, pool poo-l, with French poule pool- and observe how much finer the sound appears to be in the last word than in the first. UO-W, O'O. In some parts of the country the sound begins with uo and runs up to 00, which may be written iww. In other sfi SPEECH IN SONG. [Sect. X. parts the vowel begins with the mouth open, producing the vowel uu', and then closes rapidly to the high-round form. This effect is written 60, compare 6a, p. 546, and also p. 796. Of course all these effects should be carefully avoided. 00, UO, UO' . Sing 00 in the middle of the compass. The quality of tone is very fluty. Proceed upwards. It is soon difficult to get out any decent quality of tone at all. I have heard a whole class of singers suddenly collapse on such a passage as Ex. 19 (p. 55n), which is intended for sopranos and an Octave lower for tenors, and may be taken an Octave and a Fifth lower for basses thus Ex. 21. . — . F • ' /-N. it n ^ • m i # m . 1 pj. 3lt _ - 1 r 1 V— ' r! • 1 • J p 1 1 H . 1 1 w 1/ 1 00 00 00 00 00 00 00 00 UO UO UO UO UO UO UO UO The UO succeeded better than the 00, but all power of bringing out a musical tone at all ceased on the highest note. This diffi- culty was overcome to a great extent by instructing the singers to keep their mouths quite wide open for the upper notes, and produce the effect of closing the lips by contracting the arches of the palate. Write the result u6. The vowel quality is of course not true, but it is recognisable, and a good musical note can be got out on u6, much better than on uo, at pitches for which 00 fails utterly. 00, AO, A A. On the lower notes from the middle of the compass downwards the 00 acts very well, but it should be directly contrasted with ao in these positions, which will be found much fuller and rounder. The student is recommended to con- trast such effects as Ex. 23. fe ^ by singing each note first to 00, then to ao, and then to aa. Although 00 comes out very well on the low notes, it will be found poor and even "woolly" in some voices, in comparison with ao and aa. Change to ai and««, or even e and i and observe how greatly superior 00 is to either of the latter in the low notes. The necessity for careful voice training on the extreme vowels ee or i and 00 or uo, at extremes of the compass of the voice, will thus become apparent. UE Group— ?77, UE. Intermediate Vowels. — The preceding seven groups complete the vowel series as known to received English, with one exception Sect. XJ VOWELS IN PARTICULAR. 57 (group U). It completes the journey I, E, A' — AA — AU, OA, UO, produced by gradually lowering the front of the tongue from I to AA, leaving the lips open, and then gradually raising the back of the tongue and at the same time gradually closing the' lips from AU to UO. We now enter upon series in which the tongue is the same as in the I and E groups, but the lips, instead of being open as for them, are closed as for the UO and OA groups respectively. They are not known in Italian or Spanish at all, and in English only provincially, but they are extremely common in German and French. UI. This is not a generally acknowledged vowel. The throat is narrow as for ee and the tongue is in precisely the same posi- tion, but the lips are closed as for oo. This is the sound of ' u ' in Germany, where the lips are often opened, so that the sound becomes simple ee, as echer' for uvber' iiber, sh'tee-lu for sh'tui'lu stiihle, a pronunciation known over two-thirds of the country, but which is always regarded as incorrect. There is no occasion for Englishmen to attempt to distinguish ui from the next vowel ue. UE. The tongue and lips are in precisely the same positions as for ui, but the throat is wide. This is the well-known French ' u.' It is therefore made immediately by saying i and while uttering that vowel, closing the lips to the position for oo (or " high rounding " them). The French never confuse this sound with ee, English people are very apt to confuse it with their eu, and after r' with 00, calling rue r'oo for r'ue, because eu is not used after r' in received English pronunciation. As a provincial sound it occurs in Devonshire and Norfolk, replacing both eu and oo, and also in Scotland. Those who wish to pronounce French decently must study the sound well. The common rules, " put your tongue for ee and try to say oo," or " put your lips for oo and try to say ee," will generally lead to the right sound. It may be always used for German ui. Although this is not a particularly bright or pleasing vowel, it is very easily sung at all pitches, with not such great altera- tions of effect as either ee or oo. But on jumping from a high to a low note, as in Ex. 12 and 13 (p. 436) the change of quality will be very perceptible. The vowel should first be tried through the compass, p. 166, and then on all the preceding exercises. OE Group— EO, OE. This group resembles the last in its double formation by tongue and lips. It is thoroughly un-English, but is heard occasionally in Scotch. EO. Tongue as for ai, and throat also narrow, lips as for oa. One sound of French ' eu ' and German ' o.' 58 SPEECH IN SONG. [Sect. X. OE. The same position of tongue and lips as for eo, but the throat wide. Another sound of French ' eu ' and German ' 6.' French and German Examples. — Neither French nor Ger- mans are very particular in distinguishing these two sounds, and even many orthoepists in France and Germany do not recognise them as different. The difference consists especially in- the throat being narrow for eo as for ai, and wide for oe as for e, and when we consider how few English orthoepists distinguish ai and e we cannot be surprised at foreigners confusing eo and oe. Those who do distinguish them make eo, French deux deo, feu feo, neveu neoveo, and generally je zkeo, le leo, me meo, though some give oe in these words. German Goethe Geo-tu, ohl eo'Z, hofe heo-fu, hohe heo-u, hohnen heo-nen, hoflich heo-jieehy'h, always long, often confused with ai in Germany. oe, French peur^oe)'', seul soel, nenl noef, peuple poepleo (the eo very short), boeuf boef, both long and short. German, bocke hoek-u, holle hoel-u, rockchen roek-ky'hen, kopfe kcep-fu, knopfen knoep-fen; always short, often confused with e in Germany. EO, OE. AI, E. Exercises. Sing to compass, observe where the distinctive character of the vowel fails. The sound is never very bright and pleasant. It has most of the bad qualities of ai, e, and these are only partially improved by some of the good qualities of oa, ao. Englishmen are apt to confuse the sound when long with er and when short with u, calling deux der, feu fer like 'fir, fur' Goethe Ger-tu, holle hul'u &c. This should be avoided. When the sound is well known, there will be felt to be a great difference between the sounds oi er and eo or oe. Both eo and oe lose effect in upper and also in lower notes. It is only in the middle notes that the effect is good. Try the exer- cises on pp. 46^?, 47a, 55a, for both eo and oe, and contrast the effects by singing ai eo on each note of the compass, and also e oe, and then eo oe. It will be found best to aim at singing the wide oe and let the pitch decide when eo occurs. The following passage will contrast the middle and lower tones on the compass. Those who cannot take the low g may sing an Octave higher, as shewn by the small note. Ex. 23. ^ te ^ ifes m 5 5 *=* ati eo oe eo oe eo oe eo oe eo eo eo eo oe oe oe oe eo eo oe oe Sect. XJ VOWELS IN PARTICULAR. 59 17 Group— [7 1/', U', U, E', UU, UA. Varieties of the Natural or Ill-Defined Vowel. — These are unpleasant, indistinct sounds, some of them very easy and familiar, even in accented syllables, to speakers in the south of England, and found in weak, or unaccented syllables, all over the country, Scotland, and North of Germany. But the sounds are extremely difficult to Frenchmen and Italians, either in weak or strong syllables and to many of our Northern English speakers in strong ones. Even those English speakers who use a vowel of this group, vary its sound considerably by different modes of produc- tion and do not hear the difference of the sounds produced. Hence although singers should know how they are all produced, they need only aim at one, selecting that one with which their organs are most familiar. I3ut the fine sound of u, like the fine sound a', has generally the best effect. The thicker sounds uu, e', however, naturally occur at low pitches, and the thinner uu' u' at high pitches. UU'. This is produced by placing all the organs, except the lips, in the position for 00. The lips must be quite wide open. The sound is very bad and quite unacknowledged, though it sometimes occurs in the defective pronunciation of 00 as 60 alread} referred to (p. 56a). It may occur naturally for u in high passages but it should be avoided rather than cultivated. Those who are unfamiliar with it, can best produce it by saying 00 and then while speaking, or better, while singing, suddenly opening the lips quite wide. The alteration of vowel quality is very conspicuous, and it will be found difficult to preserve the same pitch. U'. This bears the same relation to uu' as uo does to 00. It is obtained by saying or singing mo with a wide throat, and suddenly opening the lips. This is a vowel which is very useful to the singer, and deserves cultivation, as a refinement on M,to be used where u would bethought coarse and no distinct vowel is admis- sible. It never occurs in strong syllables. English Final -A. — Many persons consider u' to be the proper vowel to be used in weak syllables, in cases where u seems to me more common in speech, and a' has been recommended for singing, provided 'a' is now written (p. 48^), thus China Chei-nu or Chei-nu' in ordinary speech and Chei-na' in singing and declamation. German Final -E. — The common German final -e, as in eine gute alte Dame aaynu goo-tu aal-tu daa-mu, is usually pronounced with a vowel which is either u or u', and which I find it quite sufficient to think of and pronounce in either way. Some Germans however, especially German teachers, insist upon 6o SPEECH IN SONG. tSECT. X. saying ai more or less distinctly. This is like the use of a' for u' in English eidee'a' for eidee-u'. Of course aaynai goo-tai aal'tai daa-niai, would be intelligible in German, but when I heard it for the first time in Franzensbad, after two years' residence in Dresden, it had an excessively stiff, strange and orthographical effect to my ears. Substitution of C/'L, C7W, U'M for Vocal L, A^, M. — But the chief use of u' is in such words as able avbl, open oa-pn, rhythm rith-m, where no vowel at all is used in speech before the vocals I, m, n. The effect of singing on the vocals, as already observed (p. 28c), is far from good. The late well known basso Henry Philips in singing the popular song of " The sea! the sea! the open sea!" insisted on singing the word " open " as oa-pn, without any vowel at all in the second syllable, thus Ex. 24 0- hi ! . !/ , \>Yi n »r ? t-b h '> t. « . m J » •^ iTT'^ 17 Q s • « ^ ~ * • . <■ • 1 lvL» « 1 J • Philips dhu see ! dhu see ! dhu oa-p - n see ! (oa - pu'-n oa- pu'-n oa - pu'-n oa - pu'-n. Exercises ai - bu''l ai - bu'-l ai - bu'-l ai - bu'-l. [ka z- u •m ka z- u'-m ha z- u'-m kaz- u'- m. The effect of this vocal n after the fine oa was ear-rending, as I have more than once witnessed. But Philips contended that either oa--pen-, or oa--pun- would be wrong, and he refused to sing either. In this he was right, but not in supposing that his only choice was the singularly disagreeable oa-p-n. If the singer will take oa-pu'n, he will find that u' may be -prolonged without falling into er, and that it gives a very decent quality of tone, in- voluntarily thickened perhaps by the low note into e', but by no means disagreeably worse than oa. Of course the composer did wrong to give the last syllable of ' open ' to a long note, but the singer should know how to make the best of such sacrifices of speech to melody. Thus all the syllables in I, m, n, should be sung with an interposed u'. I have found it a useful exercise to sing Ex. 24 to open oa-pu'n, four times repeated, as indicated in the second line of text, throwing the accent on to the -u'n in each case, and similarly for able ai-b-u'l, chasm kaz--u'm, as shewn in the last two lines of text, carefully avoiding -em. The vowel is of course lengthened and accented, as is indicated by the turned period in oapu'-n &c. Final Unaccented -AL, -AN, -AM &c. — In those numer- ous final unaccented syllables containing -al, -el, -an, -en, -on, -am, -om, &c, it is the custom in speech to make the vowel Sect. X.] VOWELS IN PARTICULAR. 6l extremely indistinct. Thus 'ideal' is not clearly eidecal, or eidee-a'l, or eidee-ul, or eidee-el, while it is certainly not eidee-l ; 'guidance' is not gevdans, gevda'ns, gerduns, gevdens ; nor is ' passion ' quite pash-on, pash-un, or pash-en. The termi- nations of ' innocent infant ' are identical, and an attempt to distinguish them as in-oasen-t in-fa'nt would be un-English, passing muster when a foreigner sings English, but merely oithographical helplessness in an Englishman. In all these cases I write in ordinary glossic -el, -en, -em, laying down the rule that these combinations of letters in weak syllables have these . indistinct sounds, and that when we give them the distinct values, we should write -eZ-, -em', -en-, because we always lay some stress on them in this case. Thus ' Clapham ' is Klap-em not Klap-am, Klap-a'm, Klap-um, Klap-erm, Klap-em-. But the singer wants more than this. He must know what vowel to sing, and the best vowel for him to aim at is u', thus eidee-u'l, gei-du'ns, pash-u'n, in-oasu'nt in-fu'nt. The two sounds n', u differ principally in raising the back of the tongue for the first ; and the singer should distinctly feel the back of his tongue rise, and should get out a good clear note. This should be practised by false accents, as already recommended for ai-bu'l, oap-u'n, kaz-u'm. Such a passage as the following, given in four parts for all four voices, will shew the nature of such practice. Each part will sing the same nonsense line of words here written in glossic only : — Treble, Alto. Tenor {8ve lower), Bass. Ex. 25. |^fe=^ ver i W ^^ vizh ME^ ^ ^ y^ — ^ ^^ nu'-l sim - hu'-l fei du'-m haul- ^ u'-ji hrist - yu'-n wuom- 62 SPEECH IN SONG. rSECT. X. -Tfc ^ f^^ g^^^ =tc=p: ;2iz!i 5 nu'-l sen-tru'-l toa tu''l ee--kwu'-l maur-tu''l. fci fe Pn=P ■= ^-i--- A-- tj u'-n mil--yu''n pa'sh--u'-n ^ _ _ , ^ nav-shu''n dee'-mu'-n. ^ S ^ ^?^=P= ± -^1 ; thu'-mfree-du'-m hing--du'-m spek'-tru' •mfoa' -r'u' •»!• ^53 m * [7. The tongue is in precisely the same position as for a', and is " mid-mixed," lying nearly flat and so that its tip is just concealed by the lower teeth, which however do not touch it. The throat is narrow. This is usually considered to be the " natural vowel," that is, the sound that would be heard if the vocal chords were exposed directly to the air, without the intervention of any resonance cavities. Of course that is all imaginary. Moreover it is difficult to call that a " natural " vowel, of which so many forms exist, uu', u', u, e', uu, ua, which are all extremely difficult to foreigners, and by no means easy to distinguish even by Englishmen. For examples, see the next page, under U, UU. E'. This has a "low-mixed" position of the tongue, that is both back and front are on a level and low, but the front is not so low as for oo, while the back is a little higher. The tongue is therefore level, and slightly lower altogether than for a'. The throat is wide. According to Mr. Melville Bell this is the sound of the Italic letters in prefer, earnest, firm, m^yrth, guerdon, as pronounced by speakers who distinguish this vowel from that, indicated by the Italic letters in word, jowrney, furnish, which he considers to be uu. These forms will be considered on the next page. The singer has no occasion to practise e' as an independent sound. UU. The tongue lies in the position for aa, but the throat is narrow. Grasp the throat lightly above the larynx and sing aa, and then change to the nearest approach to the word ' err ' that you can make, without moving the tongue, and hence without sounding an /. This is uu. The throat instantly narrows, as is Sect. X.] VOWELS IN PARTICULAR. 63 shewn by the expansion of the muscles. Observe that this ex- pansion is not due to force of singing, for aa may be sung very loudly in middle pitch without any such muscular expansion, that is, without any such contraction of the throat, whereas uu cannot be sung even softly without it. U, UU. The vowels u, uu bear to each other the same relation as a', aa. And in speech there is the same relation of fineness. To those who say rub, such, snuf, dr'um, hunt, punch, the sound of ruub, suuch, snuuf, dr'uiim, huunt, puunch, is as coarse and disagreeable as is the sound of laugh laaf, pass^aas, path paath, to those who say la'f, pa's, pa'th. On the other hand, ears accustomed to mm, aa, find u, a' " finical." C7\ In South Lancashire where aa short is used even in man maun, the vowel u or uu is replaced by a still " thicker " vowel u, which may be produced very nearly by putting the tongue and throat into the proper position for uu, and the lips in the position for 00. The resulting vowel il sometimes strikes a stranger as uo, sometimes as uu, sometimes as short oa. It is very extensively used in the North and Middle of England, and seems to have been the intermediate sound through which early English uo passed into the present uu and u. ER, UR. Mr. Bell distinguishes such words as " earn urn," by calling the first e\^rn, and the second uu^/n, where ,,r represents a defective r' which will be explained hereafter, and may at present not be distinguished from r. I hear these words usually pronounced u-n, with the long vowel w, in the South of England. Of course provincially great varieties occur, on which I cannot enter. But it would not suffice to write u-n because it would not indicate that after pronouncing «■ and before coming to M a very slight trilled r' may be pronounced, even in the South of England, and that it is often pronounced in public declamation, or that any -of the substitutes for r', such as ,/, may be inserted. This might be indicated by u-rn, with r unmarked, but then there is so much prejudice of eye, that if this were written for 'earn' people would be struggling to pronounce some deeper and illegitimate sound as uu-n. Now to say hee must uu-n un uu-n for ' he must earn an urn ' would be almost uninteUigible, and would be thought " rustic." But either hee must u-n un uu-n, or hee must u-n un u-n would be intelligible, and I find are best conveyed by writing hee must er'n un ur-n and hee must er-n an er-n. Hence the symbols er-, ur- when they occur, must be understood to mean long u-, uu-, with permission to trill the r slightly after them. But er, ur, are short, with permission to trill the r slightly, and, as will be hereafter explained, do not glide up so closely to the following consonant as simple « and uu would do. At the same time I do not recommend either speaker or singer to attempt to make such distinctions 64 SPEECH IN SONG. [Sect. X. as wn or e'-n for ' earn ' and uu-n for ' urn.' The following pairs of words, notwithstanding their extreme difference in ordinary spelling, may be all pronounced identically, with what I have explained to be the meaning of er. Thus kernel colonel ker-nel, pearl purl perl, pertinence purtenance pertinens, pervade purveyed pervai-d, circle surcle ser-kl, kerb curb .kerb, firs furze fer-z, earn urn er-n, serf surf ser-f. Of course Scotchmen, even the politest Edinburgh speakers, will pronounce differently and probably use eV'*, er'-, or aer'- in the first and uuf- in the second word of each pair respectively, .keeping the vowels and trilling the r' strongly, but this is thoroughly un-metropolitan, so far as England is concerned. Especially when unaccented and final, only one sound is used, which is still written er because the r may be trilled, but which, when the r is not trilled, as is usual in the best Southern English, is not distinguishable from u or u'. Thus finer fei-ner, is generally quite /erww or fei-nu' , rhyming absolutely to China Chei-nu or Chei-nu'. But there are these differences. To say Chei-na' is "elegant" (= pedantic) and admissible, to say fei-na' is " vulgar " to the last extent (though used by "exquisites"). To sa.y fei-nu-r' with a trilled r' is "elegant" (pedantic) but when any one says Chei-nu-r' with a trilled r', he is looked upon as beyond the pale of civilisation. It is to indi- cate these strong opinions that I write Chei-nu or Chei-na', but fineTfei-7ier, labour lai-ber, murmur mermer, elixir eelik-ser &c. ER. Exercise in singing er, that is, w, to the compass as usual. Observe carefully how it alters its effect. At the same time observe in the mirror how the tongue alters its position with the pitch, passing from uu', or at least u' at a high pitch, to u at a medium pitch, then, when reaching a lower pitch, to e' and uu, and finally, at the very lowest depths of the voice, to a vowel written UA, for which the tongue is in the position for ah, but the throat is narrow, so that u, uu, ua bear to each other the same xelations as a', aa, ah. U. These natural passages of the u vowel into all its related sounds will convince the singer that he has only to aim at u in all cases, and leave the rest to the modifications produced by pitch, making an effort to produce the best quality of tone and a vowel quality resembling u. All the previous exercises may be taken on this vowel. The Ten Oral Vowels. — This completes the oral vowel series so far as it needs be studied. The singer must recollect that it is only vowels on which he has a chance to produce good clear bright and resonant qualities of tone, and that very few of the vowels really lend themselves to this purpose. Hence in the above exercises I have endeavoured to make him feel how far he fcBcT. XJ VOWELS IN PARTICULAR. 65 may modify them, and what is the meaning of reducing all the vowels which he aims at singing to the ten forms, i, e, a' — aa — au, oa, uo — ue, oe, u, leaving difference of pitch to do the rest. Orinasal Vowel Series— AEN', AHN', OAN', OEN'. N'. The English syllables am, aam, oam, um, and an, aan, oan, un and also ang, aang, oang, ung, end with a nasal reso- nance, preceded by an oral vowel. The nasal resonance also partly takes place in the mouth, but no breath passes the lips. In the French vin vaen', vent vahn', vont voan', un oen', on the contrary the breath passes through the mouth as well as the nose. This is signified by writing the oral vowels, ae, ah, oa, oe, for which the tongue throat and lips are disposed, and writing n' after them to shew that the uvula is advanced so as to let the air resound in the upper pharynx and nasal passages. The advancing of the uvula and the resonance of the upper pharynx considerably modify the effect of the oral vowels before the nasal resonance comes into play. It is consequently not possible to assign the corresponding oral vowel accurately. For the French nasals, the oral vowels ae, ah, oa, oe, were selected as the probable base, because the French themselves refer their sounds to these oral vowels, which are the nearest they possess in an unnasalised form. But to an English ear a, 0, oa, u seem nearer, so that to such an ear the above words would be more naturally represented by van', von', voan', un'. Three of these vowels a, o, u, however, are unknown to Frenchmen, so that the English ear is very proljably mistaken. It must be remembered that n' does not indicate any sound or any consonant, but simply the addition of two resonance chambers to that of the oral vowel, one of which consists in the very complicated nasal galleries above the hard palate. N', NG. Englishmen are apt to confound aen', ahn', oan', oen' with their syllables ang, ong, dang, ung, in fact they often confound ahn' and oan' completely and use ong for both. Now this is intelligible, because, as ng does not occur in French, these syllables could not be meant for anything else. But such a pronunciation is outlandish and disagreeable to the last degree. To understand the great difference, sing ang, lengthening the ng, and suddenly check the breath through the nose by pinching the nostrils. The sound almost immediately dies away just the same as in ag. On the other hand on singing aen', and pinching the nostrils, although the sound is shghtly modified by having a closed instead of an open nasal resonance, it continues to issue through the mouth, and can only be stopped by very tightly compressing the lips with the hand. If however we leave the nostrils open and close the lips with the hand, the result is not an unmodified ang, but a much injured aen', which is never. E 65 SPEECH rN SONG. [Sect. X,' theless distinctly an aen'. This shews that aen' is really an orinasal vowel. Try similar experiments with the other orinasal vowels. The results are very different. Though pinching the nostrils materially affects aen', oen', it leaves them quite recognisable ; but it renders ahn' very bad indeed, and scarcely intelligible, and almost entirely destroys the sound of oan'. On the other hand covering the lips with the hand in all cases reduces the nasal vowel to a hum, just as if the lips were closed, notwithstanding the different positions of the tongue. AEN\ AHN', OAN', OEN'. The following are examples of these four vowels, which cannot be well learned except after frequently hearing them pronounced by natives. If possible get a Frenchman to pronounce each example six times over, and do not attempt to imitate till he has thus treated all the words under one vowel. When the ear has thus grown accustomed to the sound, try to imitate it by means of the directions already given. Never mind errors in other parts of the words. If the true sounds of these four vowels are acquired correctly in one hour the student will have done remarkably well. aen\ fin faim faen', soudain soodaen', pin pain paen' , timbre taen'breo, dessein daisaen', bien byaen' , tiens tyaen', soin swaen', temoin taimwaen' . ahn'. dans dahn' , quand hahn', entends ahn'tahn', chant shahn' , dansant dahn'sahn' , silence seelahn's, sanglant sahn'glahn', enfuni aim' f ahn' ,wente vahn't, Jean Z h ahn', ca.mpa.gnek ahn' pa any'. oan'. bon boan' , reponse raipoan's, soup9on soopsoan', fonde foan'dai, allons aaloan', ombre oan'breo, tomber toan'bai, non noan', compagne koan'pciany'. oen'. un oen', chacun shaakoen', brun br'oen', parfum paar'foen\ a jeun aa zhoen'. The nasal quality of tone is generally very disagreeable to English ears, hence there is much difficulty in making the orinasal vowels pleasantly. A little practice will shew that very different degrees of nasality can be given^to the French vowels, by altering the balance between the nasal and oral resonances through the action of the uvula. The further the uvula is thrown forward the more nasal the effect. Watch this on a%«', where on account of the low position of the tongue, the uvula is fully visible, and make an effort to produce an extremely nasal effect. Aim finally at making the vowels as fully oral and as slightly nasal as is com- patible with recognition. In practising these vowels on the scale of the compass, (p. i66) similar effects of pitch will be perceived, as in the oral vowels, and hence all the other exercises should be tried on them. The modified forms not existing as acknowledged speech sounds are not provided with symbols. The singer has always to aim at aen\ ahn', oan', oen' , and leave the rest to pitch. Sect. XI-l DIPHTHONGS AND TRIPHTHONGS. 67 XL— VOWEL GLIDES AND SLURS, DIPHTHONGS AND TRIPHTHONGS. Experiments on Vowel Glides and Slurs. — The general nature of glides has been already explained (pp. 25-6). The peculiarity of vowel ghdes is that they begin and end with a clear vowel. Sing i w first with a. clear separation between them, then legato laigaa-toa eh: i =l=fe V still aa - ee however separating the sounds by at least a " slur," that is, keeping any intermediate emission of breath, which occurs while passing from the aa to the ee position, so gentle that no vocal sound can be heard. Some continual emission of breath must be made or else the vowels will be separated by the clear glottid (p. i8c) thus foa, pe. Sing then g.aa-i-ee with a clear glottid to the first vowel con portamento only. Lastly sing i ^ ^ con portamento koan poa-r'- taamain'toa, with a "carrying on" of the voice, so that it is not only heard at aa and ee, but during the change from aa to ee, producing the glide aa + ee. Repeat this experiment, and observe that as aa goes into the glide ( + ), and the glide ( + ) leads on to the ee, a sound of ei is heard. Also observe that by keeping aa strong but gradually diminishing its length the sound of ei comes out more clearly, and that the lengthening of the final ee or i does not affect the result. Again keep aa long and strong, make the glide ( + ) forcibly, but end on a gentle short ee or i. The effect of a broad ei still remains, but it is not like a spoken ei. It is however the singer's ei where the chief prominence is given to the good vowel aa and to the distinctive glide, and the least to the bad vowel ee or i. These^several effects may be collected thus. Ex. 26. separatamente. i ^ legato. con port. 1-. — : ^- jaa '•fe )aa-i-ee iaa-\-ee can port. f r t con port. f P^ =^=i faa+ee faa-\-ee 68 SPEECH IN SONG. tSscT. XI. By practising these in succession the mode in which one form of the common diphthong e_i as in high hei, is produced, will be easily understood. Diphthongs. — Diphthongs arise then from gKdes between two vowels, one of which is short. If both are long the ear recognises two syllables. Both may be short, but that is seldom the case in strong syllables. The glide must be strong in English and Ger- man diphthongs. It is often weakened to a slur in Italian Spanish and French. " -% The Nine Groups of Diphthongs. — The glide is of course most conspicuous when it occurs between vowels in widely differ- ent positions, as aa- + ee, aa' + ue, aa' + oo, ee + aa', ue+aa', oo+aa-, but it often occurs between more closely connected positions, as aa-+u, au- + u, ue + ee\ &c. It is convenient to divide diphthongs into nine classes ; three of diphthongs ending and three beginning with weak ee, ue, oo ; one ending with weak u; one ending with a " vanish " and one commencing indeterminately and hence styled " inchoant " iti'koaent. The slurred are most conveniently considered in connection with the glided diphthongs. The weak ee, ue,oo in diphthongs are expressed at pleasure by ee, ue, do, or more conveniently hy y, wy', w, either before or after the vowel, thus aaee, aaue, aado or aay, aawy' aaw. If the first element is long in strong syllables, the accent mark is placed after it, as aa-y^ aa'wy', aa"W ; if it is short the accent is placed after the second element, as aay, aawy'-, aaw In case of the n group, it or r is always used. For the groups oi " vanishes " peculiar marks will be employed. In these diph- thongs a true jc or yh, and w or wh, or even gw'h is often sounded, when final. This may be shewn by doubling the final con- sonant, as aayy, aaww. The singer will of course avoid such an unnecessary lopping short of a vocal element. Group I. Weak Final EE or /, or +ee, written -Y. EI. This is an ambiguous symbol used to represent any one of various similar sounds in cases where it is found inconvenient to analj'se them completely. Provincially over the whole of England at least two sounds of this diphthong are distinguished in different words. In received Southern English only one form is intentional, and although several are really used, the same speaker himself generally uses only one (his own peculiar one), except in the wDrd ' aye' (called aay, aay or even ahy, auy when counting votes, sometimes ai, aiy, but always now distinguished from ' I, eye ' which are uy, a'y among the most refined speakers, and uuy, aay and even ahy among broader speakers). Both uuy and ahy sound very unrefined. The speaker must be careful not to refine too much, however, for ay,aey,ey, are very bad, and «y especially so. Local pronunciations distinguish uy and aay, Sect. XI.] DIPHTHONGS AND TRIPHTHONGS. 69 and in Edinburgh words like : pipe, bribe, bite bide, dike, fife, blithe, dice, pile have ey, while words in which ei is final or pre- cedes a vowel, or the plural -z and the verbal -d, have ay (with a not aa) as : cry cried cries size pies trial buyer. All these words must have one single diphthong in England. The student must also guard against nasalising the first element, especially before n, and saying mflayn mine, for example. In speaking, the first element is always very short, so short indeed that there is a difficulty in determining whether the usual ei is uy or a'y. With the first element lengthened u-y is strange, and inadmis- sible like the provincial uu-y, e'-y, but a'-y is admissible ; not so aa-y. In singing, aa-y that is aa' + i, as lately analysed (p. 67), is the proper sound. The note is sustained on aa as long as the music requires, till just at the end, when it glides smartly and clearly on to i or ee. But a good effect is also produced by changing when possible from aa to a' and making a' the prin- cipal vowel, before gliding to i, thus aa-^a' ■\-i. This almost gives the effect of sustaining the diphthong, which is of course impossible. In Germany aay, ahy and occasionally ah-y, are used ; but in some districts aey is also employed where ' ei ' is written. Singers should always use aay. In Italian the diphthong is slurred as vai vaa-^ee, and the effect of a diphthong is almost lost, coming out almost like two separate vowels. Compare eye a'y, ahi aa-r-ee. The symbol aay is, however, retained for convenience. The same occurs in Spanish. In French these diphthongs are found only when generated by the modem pronunciation of the "Imouillee" el mooyyai properly ly' as simple j. Thus gouvernail goovaer'naay , iaiWiv faayyeer', m6daille maidaay ; where some speakers still use aayly' or aaly'. In hair aa-eer', there are distinctly two syllables. AI-Y, is the English "vanish" (p. 44c), but it more properly belongs to Group VIII., p. 78a. In French aey occurs from the change oily' into y as conseil koan'saey. 01 has two English sounds. It is any or auy when final, as boy bauy, jauy, or before the derived -d, -z as enjoyed enjauyd, enjoys enjauyz. But before s it is generally oy, which has a more refined effect as oyster oyster not auyster. It should never be pronounced as ei in the following words, where ei used to be polite in Queen Anne's time and is now vulgar,' anoint, ointment, oil, boil, broil, coil, foil, foist, froise, groin hoist, join, joint, joist (joist not jeis) loin, moil, point, poise, poison, soil, spoil ; and also : destroy, decoy, loyal, royal, voyage ; all of which must have oy or auy. In singing auy or auy will 70 SPEECH IN SONG. [Sect. XI be used, the notes being sung on au till just at the last when the vowel glides rapidly into i. Distinguish clearly, even in singing, between cawing kawing and coin kauyn, his jaw is low hiz jau iz loa, and his joys low hiz jauyz loa. In North Germany ' eu, au ' are generally oy, as eule oylu, freund fr'oynt, leute laute loytu, and this is the only sound that a singer should trouble himself with ; but ah-y is common and even oewy', aowy\ aawy' are said to be heard, but they seem to me theoretical, for during three years' residence in Germany, although I was aware of the theory, I never noticed them in practice. In French, the diphthong aoy occurs occasionally when ■ oy ' precedes a vowel, as royaume r'aoyyoam, voyons vaoyyoan', but the more usual pronunciation is r'waayoam, vwaayoan' . OOY, does not occur in English. In German it occurs in the unusual word pfui 1 pfooy. In Italian it becomes slurred as usual in the common word lui loo-~ee, written looy, and treated as one syllable. The French ' oui ' is 6o-\-ee, both vowels being very short, but the stress varies so that to an English ear, it sometimes sounds ooy and sometimes wee, see Group VI., p. 74^. UEY is also a diphthong ue-^ee with variable stress, and hence sometimes sounds like uey and at other times like wy'ee. To produce it, the tongue is placed in the position for ee at first, and then the lips are rounded for 00. On beginning to speak, this gives ui or ue to begin with, which may or may not have the chief stress. Then without altering the position of the tongue or ceasing to utter voice, open the lips smartly, and ee or i results, with a previous glide. This should be well practised, as it is a very common French diphthong: lui luey or Iwy'ee, (as in all the other examples, but notice the great difference between this and Italian lui looy or Iwee, because English people constantly mispronounce the French in that way,) puis puey (or pwy'ee, and constantly mispronounced pee by French people,) puiser puey- zai, puisque pueyskee, bruit hr'uey (be careful that the r' does not change ue into 00,) tuilerie tueylr'ee, cuiller kueyyaer', cuir kueyr', fuite fueyt, suivant sueyvahn', juillet zhueyyaet, ruisseau r'ueysoa. This diphthong does not occur in German. OEY is also a common diphthong in French. It occurs in ceil oey at present, or oeyly' or oely', accueillir aakoeyyeer' , recueil r'eokoey. Englishmen have to guard against pronouncing the ' 1 ' in the first word, and confusing oey with uy or uuy, (a form of ei), and hence calling coup d'oeil koodeil instead of hoodoey. The other word", are often mispronounced akwlyeer, rek-eul, which should be carefully avoided. Group II. Weak Final tJE or +ue, written -WY'. There are no acknowledged diphthongs of this class, but in Devonshire, where ue is substituted for 00, the usual ou is e'-wy', Sect, XI] DIPHTHONGS AND TRIPHTHONGS. 7' as I seem to hear it from natives, but uirwy' or oe-wy', according to other appreciations. It is an extremely difficult diphthong to analyse, and the second element differs slightly from ne. And as already mentioned (p. 70a) German theorists insist on con- sidering ' eu, au,' (which the)' therefore spell eii, aii) to be aawy'. Group III. Weak Final 00, or -\-od, written -W. OU is used in Glossic to represent a diphthong which has even more varieties of sound than ei. As the singer's ei is aay, derived from aa+ee, as shewn on p. 67, so the singer's ou is aaw derived in a precisely similar manner from aa+00. But in aay the pitch naturally rises and in aaw it naturally falls; hence the Example 26 on that page is now better written Ex. 37. separatamente. legato. con portamente.. —I ^^^ ^=± w^^ I^iC jaa foo faa-T-00 faa-^-oo faa-\-oo faa+00 The great difficulty in these diphthongs is to preserve the pitch, and hence they should all be tried on single notes at different pitches as j? ! \) J — . ! and a careful watch must be main- =^i= aaw aaw tained over the voice to see that the pitch does not fall. The commonest refined spoken sounds of English ou are uw, a'w, and uuw aaw ahw have a broad effect, while uu-w or aa-w are quite unendurable. The nasalised flaw must also be avoided. A common provincial habit is to round the lips at the begin- ning of the diphthong, and hence say oaw, aow, which must also be strictly avoided. Some speakers refine too much, and pro- duce the London and North Kent ew (not eu observe, but e + 00), which occasionally falls into, the frightful Norfolk and Lanca- shire aew or aw (not au observe, but a + 00). All these are not only inadmissible in speech, but are bad for singing. In German, aaw only is heard, as in haus haaws, which is much broader than English house huws or ha'ws. In Italian the vowels are slurred but the diphthong nevei. forms more than one syllable, as in Laura Laa-i-6o-r'aa or, as we may still write, Laawr'aa (not Lau-r'a as in English,) fraude fr'aawdai. (Sometimes 'au' in Italian represents two syllables, as paura />art-T-oo'r'a or paaoo'r'aa.) Spanish is similar. 72 SPEECH IN SONG. [Sect. XI. In French the diphthong is like the Italian except that about an equal stress is laid on each vowel, though the second is shorter, but it occurs only in foreign words as caoutchouc kaaooshoo, written kaawshoo. OA'W. This is the English vanish (p. 53c), when it reaches a distinct form, which is rarely the case. Generally the tongue is left in the same position, and the lips are somewhat more closed. This will bring it into Group VIII., p. 786. Occasionally how- ever the oa is shortened, and a full 00 sounded after it, produc- ing oaw, which then has the effect of an ou form. Singers should very carefully avoid this form. EEW, will be spoken of under EU in Group IV. Group IV. Weak Initial EE or ee-\-, written Y-. At the beginning of a word in English we cannot have such a diphthong, because the weak initial ee degenerates at once into the consonant jy. The same is true for German. But in Italian and French there is no true y consonant, so that all these diphthongs, although written with y and w in Glossic, have a real vowel ee or do at the beginning. And in English and German, singers are counselled always to sing this ee or do, as the effect to the listener will be nearly identical with the consonants y and w. After a consonant in the same syllable, various tricks are played in speech with the object of saving time, producing the " palatalised " and " labialised " consonants hereafter con- sidered, but the singer is recommended to cling to his vocal ee and 00 as a weak initial, because of their vocal effect. EU. This is the principal English form in this class. It is ee-{-oo-, with the first element short and weak, the glide sharp and distinct, and the last element strong, and of variable length. This is how the singer should always produce it. But the speaker has several varieties. Initially in youth euth it is yoo-tk, with a true consonant, but in yew, use eu, euz often ywo-, yioo-z, with an interposed brief i or ee, more generally i so that ' you yew ' are distinguished as yoo, yioo. After a consonant as tune teun, dew deu, the tioon dioo form is generally used, except when t, d are " palatalised " into ty\ dy\ as ty'oon dy''oo, in which case there is much danger of falling into choon, joo. Indeed in the common terminations, -tue -ture, -dure, when un- accented, this is the commonest custom even among refined speakers: as virtue ver-teu = ver'tioo comTnonly ver'choo, nature nai-teur =navtioor commonly nai-choor nai-cher, verdure verdeur — ver'dwor commonly vevjoor, vevjer (like ' verger ' !) Let the singer always sing Widh vevdioor kla'd and not Widh ver-jer Had. The same palatalisation changes s, z into sh, zh, and in that they are now fixed, although the i is sometimes retained, Sect. Xl.J DIPHTHONGS AND TRIPHTHONGS. 73 as sure shoor ot shtoor, never sioor, or seewr, which is nearer the present pronunciation of sewer seewer, a word which used to be called shoar some forty years ago. This eew = ee- + 00, is not an uncommon mispronunciation, which the singer should care- fully avoid, thus Susan Soo-zen or Sioo-zen, never Seewzen, (except in the comic song of Prit-i Seewzen) truth tr'ooth, never either tr'iooth or tr'eewth. Again pleasure plezh-eur=plezh-ioor, or plezh'er, fissure fish-eur =fish-ioor or fislver (the same as 'fisher'!). The singer should guard against the change of 00 into 100, which properly belongs to Suffolk and even Norfolk (though in both ue or lue seems to be the fundamental sound,) as too too not tioo, do doo not dioo or deew, afternoon aaf-ternoo-n not aaf'ternloon. In such words as monument monevment the sound is monioo -i-ment not mon-i-ydd-T-ment. The singer how- ever need not trouble himself with these distinctions, but should keep to 100 always, where eu would be written in glossic. The loo should be retained in singing even after an aspirate thus hue heu = hioo, human hetrmen = hloo-men, although yhoo, yhoo-men are the most refined pronunciations ; yoo, yoo-men should be carefully avoided. In Italian the sound is more dis- tinctly eeoo and not 100, as piii peu, that is pee + oo-, with a true glide, not a slur. YAA, YAE, are common enough in Italian, as piano pyaa-noa =pee + aa'noa, with a true glide and not a slur, and so for other cases as miei myaey=tnee + ae-^ ee, }en yae-r'ee—ee + ae-r'ee. Several similar diphthongs are more or less retained in English after the i has already influenced the preceding consonant, thus fuchsia feushyu =fioo-shiu, though some say feushu, Asia Ai-shu always, but Asian Ai-shyen = Arshiu'n (some say Ai'zMu'n, Ai-ziu'n) and never Arshen, but. Asiatic Ai-shi-at-ik [orAi-si-, Ai-zi-,) usual eu-zheuel=yioo-zhioo-u'l in slow speaking, often yoo-zhl, and ususilly yoo-zhu-li. This pronunciation of eu for u being comparatively modern, and the change of consonant being still more modern (so that although sh, zh for 5, z are generally acknowledged, ch,j for t, d are not,) usage, which has had many phases during the last 300 years, is not even yet fixed. Group V. Weak Initial UE or w+, written WY'-. WY'EE = ue+ee, is the common French diphthong described under its other form uey, in Group Ij on p. 70c. WY'AEN'=ue+aen\ occurs in a few French words as Juin Zhwy'aen', quinquag6sime kwy'aen'kwaazhaiseem. By attending to the former directions, and first taking the ue long, and gradually shortening it, the singer will have no difficulty with these curious diphthongs, which occur in French only. 74 SPEECH IN SONG. [StcT. XI. Group VI. Weak. Initial 00 or 66 + , writien W-. In English the oo of these diphthongs at the beginning of a word is always converted into w, as in one wun, which was originally aa-n, and was still ao-n as late as Shakspere's time (compare Two Gent, of Ver, act 2, sc. i, first three speeches, where 'on,' and 'one' sound near enough alike to pun on.) The first English pro- nouncing vocabulary gives wan, an importation from the West of England, and originally oo + a'w. But the singer will always ling oo + MW and not wun, as much more vocal. German, Italian Spanish and French have no w, so that when w is written before a vowel in the Glossic of these languages it means 66 + . After a consonant in English, the w as in ' dwell ' is pronounced as 66+, thus d66+el\ or else the consonant is " labialised " as dw'el. The singer will always give doo+el. In this sense WEE, WI, WAI WE, WAU WO, WEI occur in English as in tweak tweek, twin, twain, twelfth, dwarf dwaurf, twaddle twod-l, twine twein. The kw- of ' qu-,' as in queen kween, is usually a labialised k as kw'een. WAI, WAA, WAO are common in Italian, as in questo kwais'toa, qunato kwaan-toa, uomo wao'moa, uovo wao-voa Suono swao'noa. Here the w is really never used even initially, but a quite distinct 66+ao &c. is employed. Compare Italian ■uomo' and English 'warm,' 6d+ao-moa, wau-rm. In such cases as Italian quale k66+aa-lai, English' qualm kw'aa-m, with the labialised k, the contrast is great in speaking. But in glossic we write kwaadai, kwaa-m, with waa in both cases, except when we wish to draw attention to the difference, and the singer will always use 66+ for W, even in English. WAA, WAE. Waa is very common in French. It used to be wae, and traces of this still remain. In speaking French, the w labialises a preceding consonant as much as possible, thus toi tw'aa, doit dw'aa, foi fw'aa, soi sw'aa. But in singing the pure 66 + , written w, should always be given, remembering that the French have no sound of w, and have even much difficulty in acquiring it. Ex. oie waa = 66 + aa, po61e pwael- and often pwaal, {which is properly the pronunciation of 'poll,') fouet fwaet orfwaat. WAEN', arises from nasalising the old sound wae. Thus soin swaen' never swahn', temoin taimwaen' , moins mwaen\ but moine mwaan. WEE. This is also 66 + ee, but is sometimes oo + ee or ooy, which see in Group I. (p. 706). The word ' oui ' must be carefully distinguished from the English 'we.' Compare oui, oui, we, 00-i-ee, 66 + ee, wee. But both the last words are written wee in Glossic. In German w is replaced by v\ and even 'quelle' is pronounced kv'ael'u. None of the 66+ forms seem to occur in German. Sect. XI.] DIPHTHONGS AND TRIPHTHONGSr 75 Group VII. Murmur Diphthongs, or Weak Final U, that IS, +lt, FREQUENTLY WRITTEN -R. In the English dialects these diphthongs are largely developed, especially with ee and oo, or rather * and uo for the first element, thus look lv-\-uk or li-uk, rope ruo-+np or ruo-up. When the u is very indistinct indeed, it is written h' as li-h'k ruo-h'p. When the first element is short as reason rik'-zn it is difficult to detect that there is a real diphthong. These diphthongs in a more developed form as ai-ad, ai-do &c. seem to have existed in the Anglo-Saxon times. In received English speech they have all disappeared, but a new series has been developed by means of the letter ' r.' This fell into a mere elevation of the tongue, called the " point-rise " and written ,,r, being an imper- fect form of the southwestern English " reverted "/, which is really trilled, the bottom part of the tip of the tongue being pre- sented to the palate. In the rise, however, the tip is merely raised a little, so that the position of the tongue is very like that for u, with the point somewhat higher. Hence the effect of this „r on the preceding vowel was nearly to add u after it. In modern received speech it has become quite so, in the words 'ear air oar poor ire our,' the 'r' having entirely disappeared when no vowel follows. In the case of ' are, or,' when no vowel follows, even the u has disappeared in general, except when the speaker is anxious to shew he means ' are ' and not ' ah ! ', ' or ' and not ' awe.' But power to add an r' is maintained, and in public speaking this added r' is often rather faintly trilled, though the previous diphthong is not abandoned. When a vowel follows, the trilled r' is a necessity, as it is after er, pp. 63-4, but the diphthong is maintained. We have therefore four regu- lar and two occasional " murmur" diphthongs in English, with three regular murmur triphthongs. These are a great peculiarity of Southern English. They do not occur in the Scotch pronun- ciation of English, in which r' is well developed, nor, when a vowel follows, in American English, although in general the r' has sunk to a mere rise ,,r. These diphthongs are unknown in German, though two of the triphthongs occur. In Italian and French, and even in Spanish, notwithstanding its ,,r, they are completely unknown. We write these with an unmarked r, to shew that the diphthongs and triphthongs arise from r, and that a trilled r' may be added. In German where this is compulsory, ur' is written. The singer should always trill the r' slightly. EER, EERR', in the words ear ee-r, beer bier bee-r, deer dear dee-r, fierce/ee-rs, fears fee-rz, and then in earring ee-rrHng, -hearing hee-rr'ing, you fear it eu fee'rr'it. The complete analysis of eer is v + u, and of eerr' is v-\-u-¥r'. The vowel is always 76 SPEECH IN SONG. [Sbct. XL perceptibly " widened." But the natation eer is retained as most suggestive. Scotch people are apt to say hee-rHng, ekspee-r'iens, and Americans hee\^ring, ekspee\fiens, both without any diph- thong, and this marks them out at once as un-English, AIR, AIRR', in the words : air ere e'er heir eyre avr, bare bear &arr, pear pare pair pai-r ; and also in: airing ai-rr'ing, bearing bai-rr'ing, ia.hy favrr'y. The full analysis of the sounds ai-r, airr' is e-+u, e-+u-i-r', the first vowel being perceptibly widened. The omission of the diphthong is un-English or vulgar, as the Scotch vai-r'i for vavrr'i, and Dickens's Sairey Gamp, Savr'i for Sai-rr'i. OAR, OARR', in the words : ore oar oa-r, door doa-r, more moa-r, shore shoa-r, porch poa-rch, soared sword soa-rd, mourn moa-rn, court koa-rt, and also in : flooring floa-rr'ing, roaring roa'rr'ing, glory gloa'rr'i. The full analysis of oa-r oa-rr' is ao' + a., ao-+u+r' ; the vowel oa being perceptibly widened to ao. This is the only occasion on which ao- occurs in received pro- nunciation, and it is being rapidly supplanted by ««•, which again generally entails a loss of the diphthong, so that oa-r becomes au-u and o&en au,doa-rbeconies dau-u a.nd often dau. I have some- times been unable to distinguish ' Mr. Shore ' from ' Mr. Shaw,' ' court ' from ' caught,' both being shau, kau-t ; ' glory,' which some older speakers call gloa-r'i, incorrectly, continually dege- nerates into glau-r'i or even dlau-r'i. The singer is recom- mended to avoid these extremes and say oa-r=ao-+u. But so rapidly has the sound oi au'+ii, or au- gained the ascendant, that very few know, without the assistance of a pronouncing dictionary where to say oa-r = ao-+u, and where to say au- or au- + u. Thus ' mourn ' and ' morn ' both fall into mau-n, instead of being distinguished as mao-un, mau-un, written moa-rn, mau-rn, and the two meanings of ' form ' are no longer distinguished as foa-rm fau-rm. When therefore the singer is in doubt he should take au-r, as it is a less mistake to say au-r for oa-r than to say oa-r for au-r. OOR, 00i?i?' in the words: boo-r, moo-r,poo-r, sureshoo-r ; and in : boorish boo-rr'ish, mooring moo-rr'ing, poorer poo-rr'er. The full analysis of oor, oorr' is uo-+u, uo-+u-i-r', the vowel being again perceptibly widened. Here again moo-r'ing, poo-r'er' must not be said. AAR, AARR'. The aar is a very ill developed diphthong in : are aa-r, garb gaa-rb, large laa-rj, barm baa-rm (not usually distinguished from balm baa-m,) arms aa-rmz (not usually distinguished from alms aa-mz, unless the latter is pro- nounced a'-mz but no one objects to dropping I in ' balm alms,') farther faa-rdher (not usually distinguished from father faa-dher) ; and in starry staa-rr'i (usually staa-r'i,) don't jar it doa-ntjaa--r'-it. Hence arise numerous errors, as: the idea-r Sect. XI.J DIPHTHONGS AND TRIPHTHONGS. 77 of it dhi eidee'u-r'ovit, is papa-r out ? iz pupaa'-r'out. The in- tention is to say pupaa-u, and this induces a euphonic r', which however is not tolerated. The singer is recommended to trill the r' slightly wherever 'r' is written, and to avoid either it or the H where 'r' is not written. This rule is really pedantic, but it will contribute to intelligibility, and save all imputation of vulgarity. AUR, AURR'. The aur is also a very ill developed diph- thong, in: abhor ab-hau-r, or au-r (before a vowel awr' when strong, and or' when weak, before a consonant au when weak) nor nawr (follows the same law as ' or,') drawer drau-r (a sliding box, drawer, in two syllables, one who draws) ; lord lau-rd (not usually distinct from laud lawd.) Before a consonant weak aur sinks into au as ornate autiai-t, organic augan-ik (compare Augustus Augus-tus, and see that the words com- mence identically) ; and in ' abhorring,' properly ab-hau-rr'ing few persons say anything but ab-hau-r'ing, when they do not say ab-hor'-ing, as they must say ab-hor'ent abhorrent. EUR, EURR'. These contain triphthongs, and are com- pletely analysed into i+uo'+u, t-\-uo--\-u-i-r' after a consonant, the I becoming y when initial, as in mature maturing muteu-r, mutewrr'ing, that is mutioo-u, tnutioo-ur'ing. Such words as hewer heu-er=Moo-u or yhoo-u, ieyf&x fewer have two syllables. EIR, EIRR\ is also a triphthong, = «-i-!i, and ei~u-^r\ where ei has any one of its values (p. b'&d), in the words ire ei-r, iron ei-rn, mire mei-r, and also in tiring tei-rr'ing, hirer hei-rr'er. So natural is the tendency to insert an u and even dwell upon it, that many speakers introduce an additional syllable, and do not distinguish hire hei-r from higher hei-er, lyre lei-r from liar lei-er, &c. It is a much less fault to say hei-ur'ing than hei-r'ing for hei-rr'ing. In German the two syllables are written, as leier lei-er' (lyre,) geier gei-er'. In English also it is written, in fiery fei-ur'i. In Shakspere ei-r is generally dissyllabic as ei-er, but modern poets require it to be monosyllabic. OUR, OURR' is a similar triphthong = om^m, ou-^u-^r', in the words bower bou-r or bou-er, but bowery bou-ur'i, sour sou-r, sourish sou-rr'ish not sou-ur'ish, lour lou-r, louring lou-rr'ing not lou-ur'ing, although usage is not at all decided. Singers must slightly trill the r' and follow the poet or composer in using one or two syllables, but they must never omit the triphthong, never say sowr'ish and so on. Group VIII. Vanish Diphthongs. Vanishing. — This class prevails very widely indeed in dialectal speech and in many languages, but it is necessary here to confine attention principally to received speech. There seems to be a dislike in speech to maintain one position of the organs for any length of time, and hence after a very minute fraction of a second 7^ SPEECH IN SONG. [Sect. Xi it is changed, rather indefinitely, to some other one. Whethei the other one would be easier to hold if produced isolatedly, is by no means clear. The final position is very difficult to determine, Orthoepists have been, as usual very ready to determine it off- hand, but my own observations lead me to think that their indica- tions are a very rough approximation to the fact. These final vanishes are indicated by an acute accent as they are all formed by a closer position, either a more raised tongue, or closer lips. AT. After saying ai, the tongue rises, not by any means into the position of ee, but sufficiently so to lead those who keep ai or e steady (as Scotch, Italian and French) to feel that ai tapers towards ee. Hence the form ai-y is usually employed in Glossic in place of ai. See further on p. 44c. OA'. After saying oa the lips close slightly leaving the tongue in the same position, but the effect is sufficiently like 00 to lead those who keep oa pure, to affirm that od tapers towards 00. Hence it is commonly written oa-w in place of od in Glossic. See further on p. 53c. AA', AU' consist also in raising the tongue slightly and narrowing the throat, so that the voice passes off to an indistinct sound something approaching m. Hence, some, as Dr. Rush, consider ad, au to be really aa-ii au-u, or murmur diphthongs. Group Vn. Sometimes this actually happens, but it is then considered that r' is added, and it is put down as a gross vulgar- ism, by those who do not " trill their r's " and hence cannot properly judge of the real action. In point of fact the vanishes of ad, au are different, because in the latter the lips are slightly closed (" low rounding") and in the former they are wide open. EE' 00' are also vanishes, not usually noted, the tongue rising very nearly, but not quite into the position of Y and W, so that the ending is dull and sudden. Although ai od are defended by orthoepists, and certainly occur in good pronunciation, under circumstances previously detailed, this is not so much the case with ad, au, and still less with e'e 06. Singers should not use any of them. It is the object of the singer to produce a clear bright uniform tone, and hence vowel, for the whole of the note to which it is sung. The diph- thongs of groups I, II and III are bad enough, because it is difficult to preserve pitch and impossible to preserve quality at the end of the sound. But in those cases there is an unavoidable neces- sity. To say aa-s for both ice eis aays and house hous haaws, would be almost unintelligible, though this actually occurs in dialects. But to avoid the vanishes is not only not to be unin- telligible, but is to adopt a pronunciation usually esteemed "elegant" — whatever that may mean. Hence once for all the singer should sing clearly «• not ai, ao' not od, aa not ad, aw not au, and the sustainable ee, 00, and not the maimed «l, 06. S£cr. XI.) DIPHTHONGS AND TRIPHTHONGS. 79 Group IX. Inchoant Diphthongs. Inchoance. — The phenomenon of the vanish sometimes occurs dialectally in inverse order. Instead of ending indeterminately, the spealcer begins indeterminately and works up to his proper position. The resulting diphthongs, therefore, are termed in- choant in'koaent or " commencing " and undeveloped, Latin ' inchoans.' But in this case also, as well as in the former, the organs pass from an opener to a closer position. All the forms are dialectal and are here mentioned only to be recognised and avoided both by speaker and singer. They are indicated bj' an acute accent on the first of the two letters indicating the sound. E'E, a common Cumberland, Westmorland, Lancashire and Derbyshire sound, where the speaker begins with an indeter- minate sound nearer to ai, but not reaching it, and proceeds to a very fine ee. The speaker believes he is saying ee, but his ee sometimes reminds a stranger of ey, and seems to have been the first step of the transition from the old ee to the modern ei, in words where " long i " is written. U'O is a similar change for uo or oo, but the sign uo is used because the symbol 60 is wanted for another and commoner case. In iio, (also written uo-w, ^^■w,) the 00 is begun with a deeper indeterminate vowel, more like uo or u, (p. 636) and ends in a fine 00. The speaker believes that he says 00 all the time. A 'O, begins with au and during the pronunciation of au the lips close to a position nearly that of ao. This sound is not pre- cisely reached. The effect is almost that of the diphthong auw, which however is not produced. It occurs in South Herefordshire. O'O, O'A, E'O, O'E, all point to another singular dialectal action, whence a number of curious changes may be produced. The speaker begins the vowels 00, oa, eo, oe with his mouth too open, sometimes wide open, and closes it gradually but quickly as he continues to speak, so that the effect approximates to iiu'-i-oo-, uu-\-oa-, ai+eo-, e-\-oe-, but no such determinate initial vowels are produced, and the speaker believes he uses the final 'owels only. The 60 is widely used, especially in South Lanca- shire and North Derbyshire, and passes into loo on the one hand and uuw on the other, in neighbouring dialects. It is evidently one of the transition forms between 00 and ue on the one hand, as in French, and East and South West dialectal English, and between old oo and modern ou on the other, thus out was formerly cot. Triphthongs. The class of murmur triphthongs, as eur, eir, our has been already explained (p. 776, c). Others are produced by beginning with weak ee, ue, 00 proceeding to some strong vowel, as in So SPEECH IN SONG. fSECT. XI. XIl groups IV. V. VI., and then proceeding to another weak ee, ue, or 00, as in Groups I. II. III. That is they are of the types yaay yaawy' yaaw, wy'aay, wy'aawy', wy'aaw, waay waawy' waaw where the usual symbolisation is employed. They occur in dialects, but also in Italian as miei myaey~mee-^ae--^ee, puoi pwaoy = poo-i-ao'-^ee, suoi swaoy &c. These must be sung carefully as analysed. Italian has also many other slurred triph- thongs in singing, because i) the final vowel of one word, 2) such a monosyllable as i, e e, a a, o 6, ee, ai, ae, aa, ao and 3) the initial vowel of the following word a];e slurred together in singing, or the same thing happens with a simple vowel and diphthong. The following lines • from Tasso's GKrusalemme Liberata vv. 5-8. 80, illustrate both these points, and also the generation of slurred diphthongs in the same way in Italian poetry. E invan 1' Inferno a lui s' oppose, e invano ai-r-eenvaa'n leenfaer''noa-T-aal looy soappao'zai-i-ai-i-eaivaa'noa S' armo d' Asia e di Libia il popol misto, saarmaod daa'zyaa-r-aid dee Lee'byaa-i-eel pao'poal mees'toa Che il Ciel gli dift favore, e sotto ai santi hai-i-eel chyael ly'ee dyaif faavoa'rai,-T-ai soat'toa-r-aay saan'tee Segni ridusse i suoi compagni erranti. Sarny'ee reedoos'sai-i-ee swaoy koampaa'ny'ee-haer'r'aan'tce. Di Guelfo, e i chiari antichi esempi apprende. Dee Gwael'foa,-i-ai-i-ee kyaa'r'ee-i-aantee'kje-i-aizaem'pee-i-aappraen'dai. With such slurs there is no difficulty in bringing the three vowels on to one note, each being clearly audible, but of course the alteration in quality of tone injures the effect of the note. In the above example, notice the duplication of the following con- sonant after 'a, e, armo, diS'. This is universal in the best Tuscan pronunciation, and also occurs after ' cio,' and all words ending in 'a, i, b, 6, u.' It arises from the reappearance of an omitted consonant as an assimilated consonant. XII.— CONSONANTS AND THEIR GLIDES IN GENERAL. Total Interruption of Voice, the fundamental form OF Consonant. — The nature of the transition from vowel to consonant has been already considered (pp. 27-g.) The charac- teristic of a consonant; is the " more or less interruption " of voice sound, and hence more or less injury done to its musical Sect. XII.] CONSONANTS AND THEIR GLIDES IN GENERAL. 8l value, amounting in some cases to its complete extinction. This last total interruption is therefore the most characteristic form of consonant, to which all others can be referred. Mutes. — Total interruption ensues when the nasal passage is closed by the uvula, and (i) the lips are tightly closed, or (2) the point of the tongue is wedged against the palate or teeth, and its sides against the teeth, so as to check all passage of air, or (3) the front of the tongue is brought tightly against the palate and the sides against the teeth, (a class not usually considered but forming an essential point in the old Indian Alphabet, and in Mr. Melville Bell's "Visible Speech"), or else (4) the back of the tongue is brought against the soft palate and arches so as to form a complete plug to the passage into the mouth. These give the principal " mutes " and determine the four great classes of consonants. Sonants. — But although these positions absolutely check the sound, their action is not immediate, for if the voice be forced through the larynx, there will still be a resonance for some little time of a character determined by the four positions just described. Hence arise the " sonants," p. 296. To check the voice imme- diately, the larynx itself or the lungs must be attacked ; the chords must either be opened, producing a silent condensation of air, (disagreeable to the speaker) or the whole glottis must be closed by the check glottid ( ; ) p. iga, or simply there must be no compression of the lungs at all. Glides off Mutes and Sonants. — When by this means there is no air, of course all positions would be silent, and the assumption of any position could only be shewn by the glide oif it. These glides, unpreceded by any sustained sound, but passing immediately from absolute silence to sound, as a ball begins to roll down a slope from rest when its support is removed, are therefore the determiners of the nature of the absolutely " mute " positions assumed. The effect of these glides can be modified by the way in which the voice is set on, in other words by the physem and glottid employed, and this will have to be duly expressed when it cannot be assumed from the habits of the language. Thus paa, taa, kaa would have the initial glides from the mute positions, of the same nature in English, Italian, Spanish and French, but they are generally different in German, and both kinds occur and have special symbols in the Sanscrit and derived Indian languages. Continuants. — These complete closures may however be more or less extensively impaired or broken through centrally or laterally, and then breath can be emitted for any length of time, either as voice or flatus. As in this case a continuous sound is audible I proposed many years ago (in 1844,) to call these consonants " continuants " a term now generally adopted. 82 SPEECH IN SONG. [Sect. XIL Central Hisses and Buzzes, Steady and Trilled. — The rupture of the mute closure is "central" or in the direction of a line from the uvula to the central point in each lip, and of two kinds, " steady " or " trilled," according as the sides of the orifice or floor of the cavity are steady or trerfiulous. The result for flatus in the " steady central continuants " is a " hiss," which is entirely unmusical unless it can be formed into a whistle, to which wh approaches, having a small back aperture above the "high back" tongue and a small lip aperture less than the " high rounding" of oo, indeed both back and front aperture bear a close resemblance to those employed for oo, but are smaller, so that the mouth in some sort resembles a bird call, consisting of a small flat cylinder with a perforation at either end. But though we may " whistle a tune," we must not whistle speech, and still less whistle singing. All hisses are therefore unmusical, and act as interruptions to the flow of song. They must be heard, for a suffi- cient length of time to have their individual character appreciated', because the mere glides without the hiss might be more or less confounded with those from mute positions, and words would be rendered unintelligible, but they should be held for the smallest time consistent with intelligibility. These "steady" central positions are also susceptible of a voice resonance, but it is very bad and imperfect, and produces the effect of a mixture of hiss and voice, which is like the " buzz" of an insect or rather worse, p. 29a. Hence the voice does not like to dwell upon it, and either Isegins or ends it with flatus, considerably complicating the; phenomenon see p. 17c, d. The "trilled" central continuants also produce a peculiar sharp tremulous hiss, comparable on a small scale to a railway guard's whistle, which reproduces trills by the action of a small pea, dancing about in the lower part of the whistle. But the trilled positions are much better suited for voice than for flatus, and while the hissed trills are scarcely used in speech, the voiced trilled central continuants, or voiced trills simply, can give as much musical effect as the tremolo stops on an organ, having the character of " beats of interruption " (as distinct from the usual musical "beats of interference.") Through the / we hear a vowel continually interrupted and always modified, by the up and down " flapping " of the point of the tongue. Lateral Hisses and Buzzes, Tremulous. — Continuants of the other kind release the breath on the two sides (or at least on one side) and check it in the middle, and are hence called " laterals." As the loose sides of the tongue necessarily vibrate, laterals are always " tremulous." The hisses for flatus are not nearly so sharp as before, being generally much modified by the moisture of the mouth (as in the Welsh Hall 'Ihaa'lh), and are little used. But the voiced laterals, or " murmurs " are the most vocal of any of the oral consonants, p. 28c, and approach veiy Sect. XII.] CONSONANTS AND THEIR GLIDES IN GENERAL. 83 nearly indeed to the vowels, forming syllables by themselves in speech as in able ai-bl, but not musical enough for the singer who has been already recommended to sing the approximate ai-bu'l, p. 60a, Nasal Snorts and Hums. — These comprise all classes of oral consonants for which the nd^al passages are closed. But we may release the mute positions by simply advancing the uvula, and thus allowing the air to pass through the nose with- out in any respect altering the perfect closure of the mute positions. The resulting "nasal" consonants, are not well adapted for flatus, which when forced through these positions somewhat resembles a very faint shadowy imitation of a snort or snuffle. On the other hand the voice resounds admirably in the nose and mouth, and m, where the whole mouth is at liberty, is constantly chosen for " humming " on. The progressive deterioration through n to ng, as the oral resonance diminishes is very marked, p. 28c, and even m and n are not fit for placing in immediate contrast with vowels, p. 60a. Systematic Arrangement of Consonants. — The above are all the classes of consonants we have to consider, but the indi- viduals in each group are numerous, and if we include extra- European languages are very difficult to understand or arrange. Prince Louis Lucien Bonaparte in my " Early English Pro- nunciation," part iv.,pp. 1353-7, enumerates and arranges 310 con- sonants, and he has subsequently informed me that he has found many new ones in the Caucasian languages. On the next page will be found a systematic arrangement of so many as are here mentioned. As in case of the vowel arrangement on p. 38, the large capitals signify English consonants, the small capitals the additional German, Italian, Spanish, and French consonants, and the small letters others to which it is convenient to allude incidentally, but only the first two classes have to be fully studied. The reader who misses the common English CH, J, should understand that they are contractions for ty'sh' and dy'sh' to be explained hereafter. Pitch of Consonant Resonance Cavities according to Mme. Seiler.— All these consonants have resonance cavities like the vowels, and these cavities have certain pitches which can be excited, by smacking the lips for p, or separating the tongue from the palate for t, k, or by the hisses in the continuants. Mme. Emma Seiler Zevler', (Voice in Speaking, translated from the German by Dr. W. H. Furness, Philadelphia, 1875), and Dr. Oscar Wolf V'aolf (Sprache und Ohr, Brunswick, 1871) have endeavoured to ascertain the pitch of these resonances, without considering the action of the voice at all. Consequently they both regard p, t, k as sounds, because their resonance cavities will produce sounds when the air in them is properly excited, and u < D O z o H > -^1 " fg H % O o K O H w -a III M a\ ■a _ B «) £213 00 >^ >'^ s v t5 H z o Ah W D O z o H r^ C ID HP -S-S =Sp^ £>J •§2 IT) Th :S5 fO ^? fc> a: (M 1-) •a n D M 0. m ^ ^ p-f= •i^ 09 •| » a tj '-' «^ c o U to ?N -« -2 >-^ :dd .d ^ Sect. XII.J CONSONANTS AND THEIR GLIDES IN GENERAL. 85 those resonances modify the vowel at the beginning of the glide, and by their differences make the audible distinction between any two of these consonants. Hence it is important to know the pitch of these resonances, and I annex Mme. Seller's determina- tions, giving to each note (which is only approximative,) the exact number of vibrations which she assigns. Owing to the peculiarities of German pronunciation, the contacts are made looser for the sonants than the mutes, and it is in fact merely by this looseness that many Germans distinguish p and b. The tightness and looseness seems to affect the resonance, and the lower note is always given to the looser contact. The two forms assigned in some cases shew the limits of the pitch of resonance observed, and the chord to sh indicates three different resonances heard at the same time. No resonance is given for the Other consonants, but th is stated to have the resonance for t united with one nearly the same as for s. Observe that the follow- ing are the pitches to which the cavities of the mouth formed in producing the consonants named, speak or resound best, and are 7iot the pitches of the consonants themselves, which having either no sound or else no musical sound at all, have necessarily na pitch. Ex. 38. Vib. 320346 576 616 L. limit. 720 768 .^limit. mm 862 £ '-limit. -m 5^^ B G K. D Vib. 3700 ?iva. i &va. : limit. :t 2328 Sva. -(^ ?>va. 'Srlimit. 65 72i V 3666 i --2328 ;5=i378 Kr: 5 R' 'R GY'H KY'H SH Comparison of the Pitches assigned by Mme. Seiler TO THE Vowel and Consonant Resonance Cavities. — It will be convenient to contrast these resonances with those which Mme. Seiler assigns to the vowels, as these differ materially from those assigned by Koenig or Helmholtz (as given above pp. 31 and 34), indicating that the three speakers all used different vowels. Mme. Seiler's are as follows, and it must be again remembered that the following are not the pitches of the 86 SPEECH IN SONG. [Sect. XII. vowels, but the pitches to which the forms of the cavities assumed for these vowels best resound. Those for the consonants are in crotchets. Ex. 39. V i ^^ Sf=> \» £e^ fe S"=^ zfad 00 OA OE UE AE P g f AI EE Nature of Glides from and to Consonants. — The peculiarity of mute and flated consonants is that having no vocal character themselves they affect a preceding or following vowel, or even flatus, so as to entirely alter its character and meaning. They present us with a vowel having a glide on each side, say +aa + , which does not lead up to or from any other vowel. And although this is not precisely the case for the voiced consonants, especially the vocals, or R', L, M, N and NG class, yet practically, and especially for the singer, these consonants also offer no resting place for the human voice. Hence to the singer the ghde is the all important part of the consonant, and any noise which the consonant may possess, though indispensable for recognition, is only an interruption for the singer to ride over as well as he can. And for this purpose he must make his glides sharp and clear avoiding all unnecessary introduction of flatus, and his indispensable hisses, buzzes, murmurs and hums, which cannot be sung, must be all as short as is consistent with recognition. Initial Glide from a Mute. — In passing from a mute as p, t, k, to a vowel, the glottis should be from the first arranged for the clear glottid, thus pfaa (p. i8c), not for the gradual glottid piaa (p. 176), because that would allow of beginning the sound of aa by a flatus, not indeed the same as in jaa, because that is °aa (p. 15a), or flatus through a fixed position, whereas in piaa we have a flatus through a gliding position, namely from that of p (affected by its resonance) to at least an intermediate position between that of p and that of aa. Still less should we make that flatus prominent as pihaa, which some persons think correct even in English, and is customary in Germany. Some go so far as to jerk it, producing p-hjaa. There is also a trick of jerking the vowel without flatus, as p-hjaa, which produces an unpleasant effect, but which may be heard after all consonants, especially from Irish speakers. Another fault is to make the contact too tight giving a feeling that it is dwelled upon as ppaa, in which case the vowel comes out with a little " pop." This is Sejt. XII.j consonants AND THEIR GLIDES IN GENERAL. 87 more apt to be the case if the speaker relies for the perfect silence of his mute upon using, the check glottid (;) and actually resisting air driven from the lungs, instead of using the clear glottid whicii there is no attempt to force, because p;aa is likely to produce an explosion of flatus before the vowel is touched, thus p;haa (p. i(jd). There is only one form of " initial glide " from a mute to a vowel for a singer. The contact should not be tight 01 strained, the clear glottid ? should be employed, but the breath should not act upon the chords till the moment the contact is loosened, the two actions should take place absolutely at the same time, — if the opening take place first there will be a loss of a distinctive part of the glide ; if the voice is set on first, there will be a brief resonance in the cavity of the mute, producing a sonant. The organs should pass rapidly from the mute to the vowel position, giving a well marked glide, from which no part is curtailed, and of which no part is unmusical. This is what the singer must acquire by very careful and assiduous practice. Final Glide on to a Mute. Medial and Double Mutes. — The " final glide " from a vowel on to a mute is slightly different. Englishmen have no tendency to the insertion of any but the clear or check glottid at the moment that the mute posi- tion is reached, accompanied by an instant cessation of voice. If the voice is continued so that resonance of the voice takes place in the cavity of the mute, a sonant is generated. But, if after the mute position is reached, all breath and voice remain stopped, the speaker is not comfortable. Another vowel may follow, as paapaa, and then we have a recurrence of the first case, the analysis being p-{-aa-\-p-\-aa, so that we have a vowel between two glides -\-aa-\-, and a mute between two glides -{-p-\-, with absolutely no interval between them, just as at the turning point of a pendulum, motion in one direction ceases and in the opposite direction commences, at the same moment. This gives the " medial " effect of a mute. If however we pause between the two glides as in p-\-aa-\-p ...p+aa (see the break on p. 26a), or in the word hop-pole hop...poal, the effect of "doubling" is pro- duced. Children in syllabising their words, regularly double their consonants, when they are written double, as happy hap-. ..pi, but not where they are written single as rapid rap-. ..id. But in English speech, the consonant is not doubled in either case. Both consonants in hapi, rapid are medial, the sound api being common to each. In Italian, when mutes are written double, they are always so pronounced, as Bacco Baak-...koa, not Bak-oa. In ordinary glossic the break is not written, as the double or single consonant, and mark of strength, are sufficient Thus hap-i, rap-id, Baak-hoa, toabak-oa. Glides between Words. — If the following vowel belong to a new word, there is great diversity of usage. In English the mute 88 SPEECH IN SONG. IStCT. XII. sometimes runs on, but most frequently does not, as : pack it up pak...it...up, not pak-i...tup, or pak-itup, with a medial t. In German, it is, I think, generally separate as : pack ein paah...;aayn. In Italian the case cannot occur, because no word ends in a mute. In French, when the final mute affects the preceding vowel it usually affects the following in virtue of the ' union ' or liaison leeaizoan', as : grand homme gr'ahn'taom, with medial t, and this is one reason why English people find it so difficult to discover where a word ends in French. In EngHsh also when a hissed or vocal consonant follows, the final mute does not become medial. If a mute follows of course there can be no question of its being medial. Glide from Mute to Recoil. — If the mute however ends a clause of a sentence, as in : he's mute heez mewt, mayhap maihap-, some cake sum kavk, there is nothing to use up the store of wind and the speaker necessarily desires to get rid of it in the usual way of breathing, and also to release the contact. He commonly effects both at once by opening his glottis, and letting flatus escape by the mouth ; this is called the (flated) "recoil." But in doing so he necessarily makes a glide consisting entirely of flatus, as in ■meu-\-t-\-°h, ha-\-p-\-°h, kai-\-k-\-°h which is marked and distinct in each case, and thus the mute is really made " medial," not between two vowels but between vowel and flatus. If the sensitive back of the hand be directed to the mouth, and p°h, t°h, k°h be said forcibly, it will be found that not only the character of the sound, but the direction and force of the wind and even its temperature, differ in each case. But this "flated recoil" should not be used by the singer unless he finds it absolutely necessary for intelligibility. He can relieve the pressure of air by advancing the uvula, so that the breath passes quietly through the nose, and he can release the contact noise- lessly or at most by a very gentle click, written hap°, meut°, kaik°, which can be effected even while the breath is held. This '■ silent " or " clicked recoil" is the only one which the singer should in general -use. Occasionally the error is made of allowing the voice to resound in the mute position and then to cut off the voice and end by flated recoil, as habp°h for hap. This however rather more frequently occurs in final sonant glides, as madt°h for mad. It must be avoided in any case. In French the final mute never has the English recoil. When it is pronounced at all, a very slight sound of voice h', or a faint eo is generally inserted in speech. In verse this eo is counted as a distinct syllable, and in music it receives a note as long and accented as any other vowel. Glide from and to Sonants, and Recoil from Sonants. — For sonants there is much less difficulty to an Englishman, Sect. XII.] CONSONANTS AND THEIR GLIDES IN GENERAL. 89 Italian or Frenchman, for the initial glide is entirely vocal, the voice beginning before contact is broken. The Germans dis- tinguish the letters they write p, b oy t, d in a. different way, when they do distinguish them at all, which is not always. The p is either very tight as pp and the b a very loose p, which may be indicated by the single letter ; or else the p is ' markedly post-aspirated by using the strong gradual glottid jh, while b is either simply loose p, or a very exaggerated resonance, puffing out the throat like a turkey-cock's, and hence called a blahlaut hlae-laawt or " puffing sound." Occasionally a different effect is produced which some German orthoepists attribute to " whisper " and others to what they call an " implosion " (p. 14c). We need not trouble ourselves with these peculiarities. The English p and b are perfectly intelligible and may be used when- ever these letters are written in German. Similarly for t, d. In the final glide on to a sonant in " the pause," that is at the end of a sentence or a clause, even an Englishman in declama- tion has a tendency to add a final h', as mad-h'. If this is made very brief it is unobjectionable for the purpose of distinguishing a sonant from a mute termination. It is at least better than adding a recoil to the former. That is, when mat mad canno*' well be distinguished by the true glide ending abruptly in the first, and in a very brief sonant resonance sensibly lengthen- ing the syllable in the second, it is better to distinguish them as mat mad-h' than as mat°h mad. In German the difficulty is got over by never ending in a sonant or a buzz, except a vowel or buzz follows, so that if they were German words both mat and mad would become m,at (which a German hears as maet.) Glides from and to Hisses. — The glides from hisses do not present so much difficulty. The chief point to be attended to is not to run the voice into the hiss by beginning the voice too soon, as seal szeel for seel, or ending it too late, as piece peezs for pees. The buzz is so much more powerful than the hiss and the habit of using 5Z- initial and -zs final for z (p. 17c, d) so strong that the ear would feel that szeel peezs were meant for zeal peas zeel peez rather than seal piece seel pees. Glides from and to Vocals. — The glides of vocals to and from vowels are so like those of vowels to vowels, already treated at length (pp. 68-80), that they do not require further notice. Initial and Final Concurrence of two Mutes.— It is evident that one mute cannot glide on to another. Hence some flatus or click must be interposed when two mutes come together and cannot be made to glide off into adjacent vowels or hisses. Initially the difficulty is cut in English by omitting the first con- sonant, as ptarmigan taa-rmigu'n. But at the end of a word there is no escape. 'Act, kept,' must be akt°h, kept°h, or else ak-t-h' hep-t-h' (the last is quite un-English), and in French go SPEECH IN SONG. [Sect. XU either h' or eo is always used, as acta aalrteo. As soon as a vowel follows the difficulty disappears, as actor akter, acteur aakteor'. Glide from Mute to Htss and Hiss to Mute.— In English a difficulty is felt in pronouncing a mute before a hiss at the begin- ning of words, and the mute is generally lost as in psalm saa-m, pshaw shau, but no difficulty is felt when the mute is preceded by a vowel as apse aps = a+p-{-s, not a+p...s, depth depth = de+p + th not dep...th, axe aks = a-\-k+s, not ak...s, so that it is easy to practise initial glides of the kind and say pthee, pfaa, psai, ksoo, kshoo. In German pfau pfaaw actually occurs, though pf- or p'f- was probably an older form. Initial English Consonant Glides. In the following list of such glides from consonant to con- sonant as are used initially in English, the letters are filled up to words in glossic spelling, but ordinary spelling is not usually added. The words are arranged in the alphabetical order of the initial consonants, printed in capitals. A few remarks are added occasionally. BLak, BR'oun, BWoi (an error for boi or booy buoy,) DR'au, DWaurf, FLoa, FR'og, GLuv (often pronounced dluv, as a much easier combination,) GR'aas, GWaa-noa, KLoaz (often called tloaz, seldom kloadhz clothes), KR'um, KWilt (or kw'ilt), PLai, PR'eim, SFee-r, SKauld, SKR'ip-teur (or skr'ip-cher,) SKWee-z (or skw'ee-z,) SLoa, SMau-l, SNoa, SPee-k, SPLash, SPR'ed, STand, STR'ai, SWai-r, SHRee-k (many have a difficulty in this combination and say sree-k,) TR'out, TWain (or tw'ain,) THR'ee, THWak. Final English Consonant Glides. In the following principal final glides of consonant on to con- sonant after a vowel the letters are again filled up into words in glossic, not ordinary spelling. The arrangement is in the alphabetical order of the consonants printed in capitals, and generally the first consonant is medial, gliding finally on to the preceding vowel and initially on to the following consonant. A few remarks are added occasionally. r'uBD (here r'ubd-h' in the pause is not absolutely necessary, but would be heard at the end of clauses ; in : he rubbed him hir'ub-dini, the omission of the h renders the glide easy, but in : he rubbed him, hir'ubd-him- the sonant glide must suffice, no one would endure hir'ub'd-h'hinr.) ruBZ (in the pause regularly r'ubzs), r'eeCHT, br'eDTH (observe neither br'eddh nor br'etth, there is no reality in the rule that two voiceless or two voiced elements must come together, considered as a hard and Se=t. XII.] CONSONANTS AND THEIR GLIDES IN GENERAL. Ql fast regvilation,) br'eDTHS, paDZ, br'eeDHD, br'eeDHZ, kuFS, theFT, iheFTS, fiFTH (formerly fift, much difficulty is often felt in saying fifth) beGD (same remarks as for r'-ubd,) beGZ, juy (judge, the J is really two consonants dy'zh\) aKS, fiKST, siKSTH, siKSTHS, aKT (where as before explained akt"li. is inevitable), aKTS (no °h required after t, on account of follow- ing s, s&y a+k...t+s,) buLB, buLBZ, fiLCH {ths.t\s filty'sh') fiLCHT, weiLD, weiLDZ, eLF, eLFS, (as: in the elf's power in dhi elfs pou-r), enguLFT-, tweLFTH, tweLFTHS, buLJ (really buldy'zh', but apt to be pronounced buldy'sh',) buLJD, eLK, eLKS, suLKT (in : he sulked to his heart's content hisulk-t... toots haa-fs ku'nten-t, the t in sulkt is not heard, but the tongue is brought up to its proper position for t immediately after the k silence is reached, and then "held" as it were for some time, before it glides on to the oo ; the pronunciation marked is highly colloquial ;) fiLM (one syllable, but I is held as the lips close for m, by which means the effect of a vowel before m is avoided) fiLMZjfauLN (one syllable, the point of the tongue retained motionless, but the sides of the tongue brought up to the D position, and at the moment they attain it, the uvula advanced and the nasal passages opened,) heLP (compare with bulb, and observe how much shorter help is, because in it the voice is cut off by p, whereas in bulb the voice continues into the sonant,) heLPS, heLPT (in the pause, add °k,) eLS, (else) hiLT (observe how short the t cuts the word, compare twilled twild,'^ hiLTS, heLTH, heLTHS, sheLV, sheLVZ, eLZ (ells,) heMD, laMP (observe the shortness of tn occasioned hy p,) laMPS, ateMP-T (here °h is added in the pause, some speakers endeavour to supp^^ss /> and say atem't,) ateMP'TS (here the s supplies the place of a previous °h, many speakers suppress the t and say atemp'S, but those who suppress p retain t and try to say atem-ts;) haMZ, fliNCH (some try to say flinsh, but they really say flinsh'), fiiNCHT, haND, haNDZ, thou-zeNDTH, thou-- zeNDTHS, chaiNJ (some try to say chai-nzh, but at most say chain-zh', and chain-dy'sk' is common, which would be considered as chavnjsh,) chavNjfD, heNS (hence), miNST (minced,) MNT (observe how short the n is cut, com- pare its length in hand,) hiNTS, teNTH, teNTHS, heNZ (hens), wiNGD, thiNGK, thiNGKS, bUNGKT, leNGKTH, leNGKTHS (observe the k in the last five words, as well as the ng, and note that the action of the k is to cut short the sound of ng ; observe in the two last words, that a very large number of speakers use n for ngk and say lenth, strenth, a habit not to be imitated,) wiNGZ, taPS, wePT (in the pause add °h,)u^dePT-S (adepts; but of course, sing a'^dep-ts, the -^ is inserted to shew that d is not medial as in ad-ipoas, but purely initial,) dePTH, dePTHS, baa-R'B (in this and the following words the R' is 92 SPEECH IN SONG. [Sect. XII. always written in, in order that the singer may exercise himself in pronouncing it,) baa-R'BD, baa-R'BZ, aa-R'CH, aa-R'CHT, haa-R'D, berR'DZ (in this and following words r is also printed small, for then er is merely long w, so that berr'dz is bu-rdz,) whau-R'F, wkau-R'FS (wharfs), erR'^f (urge), erR'K (irk), sherR'KS (shirks) erR'L (earl), wherR'LZ, aa-R'M, aa-R'MD, aa-R'MZ, lerR'N, erR'ND (earned), lerR'NT (learnt, compare these last two in respect to length) lerR'N Z, cherR'P, cherR'JrS, cherR'PT (in the pause add °h), feerR'S (fierce, r small, because eer is now a murmur diphthong, with R' added), werR'ST (worst), erR'TH (earth) berR'THS (births), staa-R'V (starve), serR'VD, staa-R'VZ, feerR'Z (fears, compare fierce feerr's,) kaaSK (no °h is necessary, though very usual in the pause, and perhaps should be added for distinctness, but kaa'sk° is enough ; the vowel varies to a' and a both long and short), kaa-SKS, haa-SP, haa-SPS, kuSPT (add °h in the pause), r'iST (wrist), r'iSTS (wrists, observe a common provincial error r'is-tes, which avoids the harshness of the combination -sts, necessarily a long unsingable sound, which is most easily and lightly taken by pro- longing the 5 gently, and suddenly striking the tip of the tongue to the palate, but withdrawing it instantly, so that the s, which continues, is only momentarily interrupted ;) huSHT, wiTS, ai-TTH (eighth), ai-TTHS (eighths), baa-THS (bath's, the vowel varies as a', a long and short,) shee-THT (often shee-dhd, sheathed), haa-VD, haa'VZ, plee-ZD. Exercises on Consonant Glides. — The singer should exercise himself in these combinations by singing various musical phrases on them, such as any of the preceding exercises, taking each note to the same word, slowly at first, but afterwards as quickly as may be, till he can take them rapidly and without harshness or violent interruption to melody. Thus Ex. 30. 5r^ V ^ M3 ai-tth avtth ai'tth artth ai'tth ai'tth ai'tth ai'tth ai'tth artth ai'tth Or he may take a whole air and sing it with a succession of these words for each syllable of the original words, using both lists. An excellent and difficult exercise is to select any initial and any final combination of consonants from these lists, and interposing in order the vowels ee, ai, aa, au, oa, 00, i, e, a, o, u, uo, and diphthongs ei, oi, ou, eu sing the result to the first part of God save the Queen. Thus if pi- and -nz are selected, sing smoothly, and without adding on an s after the z, the following bars first very slowly and then increasing the time till they are sung very rapidly. Sect. XII. XIII.] Ex. 31. SYLLABLES. 93 » i g ^ — — ?~ w^=^. ^ ^ pUe-nz plai-nz plaa'nz plairiiz plownz ploo'nz plinz plenz planz m qSi S /)/onz /i/Mnx pluonz pleinz ploinz plounz pleunz (not ploonz) XIII.— SYLLABLES. Syllables without Consonants, separated by Silences. — Closely connected with glides to and from consonants is the division of words into "syllables." This division is so natural that it forms the basis of versification, and hence of vocal music, in all countries. A single vowel begun or ended by any glottid forms by itself a syllable. When the final glottid is "clear" the distinction is very marked. A succession of simple vowels may thus form syllables, as in the Scotch au ai 00 (meaning ' all one wool '). Usually the voice is not quite removed, but becomes practically inaudible, as oolite oa--T-ba-i-leit. A diphthong or triphthong is considered as a single syllable, the glide or slur (stronger than in the last case) binding all the sounds together, as : I owe you ei oa-w eu, your ear eu-r eer = wo-u i-u, where however naturally r' would be interposed. In these cases the clear glottid, weakened slur, or silence separates the units of speech, and no difficulty is felt, except perhaps in the last murmur diphthongs, whirh might be sepa- rated into eu-^er, v-^er = loo-i-u, i--i-u producing in each case the effect of two groups, on account of the less cementing power of the slur (-=-) in comparison with the glide (+). Syllables separated by Vowels. — But reverse the order, and instead of piling an ee+ or 00+ diphthong on a +ee or +00 diphthong as in i+aa+fto, pile the second on the first as in aa+ee+aa (ayah, a familiar Indian word for a nurse) or as in the French faience faa+ee+ahn's, paienne paa+ee+aen, vaillant vaa+ee+ahn', (usually wntten faayj/ahn's, paayyaen, vaayyahn') and the ear immediately recognises two syllables. The French consider that the central ee is " doubled " (hence the usual" Glossic writing), but on listening attentively it will be found to be only slightly lengthened, without even a central slur, that is to hefaa+ee-\-ahn's rather than even faa+ee-hee+ahn's. Now in this case where does thfj first syllable end and the second 9+ SPEECH IN SONG. [Sect. XIII. begin ? Clearly the ear being accustomed to a short sound of ee in both the first and second diphthong, naturally distributes the length of the ee between the two syllables, and gives half to each. The ear makes the syllables divide in the middle of the uniting ee, which has a measurable length. Practical Division of such Syllables. — But two practical questions arise, (i) how is the last case to be indicated in writing ? and (2) how is it to be indicated in speaking and singing 1 Now (i) as to writing there is no occasion to indicate it at all. Paayaen should shew that aa glides to ee and ee to aen. But many might read paa...yaen, omitting the first glide. Few perhaps would read /iaa)'...aew, omitting the second. Hence as a merely practical scheme paayyaen is written, and the division paay-yaen follows immediately. But (2) in speaking We do not indicate sylla- bles at all except to the ear, and the ear is quite satisfied with paa-\-ee-\-aen, ee being as "medial" as /> is in happy hap-i, because both lie between two glides. It is only in the entirely artificial dictation of syllable by syllable, that we should neces- sarily say paay-yaen, preferring the separation of the ee to mis- sing the indication of a glide. We in fact dictate two words paay...yaen,vi\ih. a break or silence between them. Now this necessity often occurs in singing. If we sing Ex. 32. w -R=^ -^-jj— ^— *--^-»i-^- ^E3=rt: 7) — r Ces saez a yahs la sont zaay yaa laa soan' pai en - nes paay yaen neo we are obliged by the silences in the music, in order not to miss glides, to double the z, the y (that is ee) and the n, and say saezzaayyaa paayyaenneo in place of saezaayaa paayaeneo, where these sounds are not doubled in speech, but merely have a double glide, as sae-{-z-\-aa-\-ee-\-aa, paa-\-ee-'rae-\-n-\-eo. If we were dictating syllables to write in glossic, we should have to do the same thing, but then, in order to avoid writing the double letter, we must have a silent sign, as extension of the arm, to shew that the vowel " runs on " and is not taken twice with a pause between, or dictate with null, to shew that there is no pause in speech, thus aay (null) yaa, or, as we might write, laez Q zaay Q yaa. It would be understood in this case that the letter following the is annulled. Syllables separated by Buzzes and Vocals. — The buzzes and vocals have been partly considered in these examples, but not entirely. We have no cases of doubled ee in English, but we have cases in which buzzes and vocals are doubled, by lengthen- ssci. xn:.i SYLLABLES. 95 ing them and making a sensible diminution of loudness in their middle. Compare the phrase : an unowned unknown punisher u'n...uir...oa-nd...U!r-i-noa-n...ptc)f+ish+er. The n in weak u'n (for an in common speech) is perfectly distinct from the following strong Ufv, in un-...oa-Hd, as also is the n of that un- from the following oa-, and the d of un\..oa-nd from the following j', 'fp if well understood, will render the very important distinction between wh, /', / and w, v', v, more easy to seize. There is no need to practise 'fp, 'vb. Group II. Tongue Point Contacts— <" d", t' d', T D, f fL. T, mute, i. 6. D, SONANT, ii. 6. The lips are open. The upper surface of the point of the tongue is pressed firmly against the hard palate, just behind the upper gums, but not reaching up to them or touching them at all. The outer margin of the tongue is pressed against the palate and upper side and back teeth, so as to completely prevent the passage of the air and yet to leave a considerable cavitj' between the palate and the front and back of the tongue, which is wedge shaped, with its thick end turned towards the throat. Glides of T and D on to Vowels. — The glide cannot be made, as in case ol p,b by having the tongue placed ready in the proper position, and hence the glide is longer and more marked. The glide t+aa is very easily made because the point of the tongue has merely to be dropped from the t to the aa posi- tion ; and similarly for aa + t where the point is raised. For t-{-ee there is more difficulty because not only has the point to be dropped but the front raised ; ee-\-t is easier. For t-\-oo there is not much difficulty because the lips can be previously adjusted, and hence we have only to drop the point and raise the back of the tongue slightly. Similar remarks apply to other vowels. For t-\-ue we have the changes involved in t+ee and t+00 united, but the lips being prepared, the real difficulty is the same as for t+ee. These important glides must be studied in the same way as p, b, with all the vowels, as aa taa, aa aat, aa taat, and then with d, and afterwards with an alternation of t and d, as taat taad daat daad, taa daa taa, daa taa daa, and then with an alternation 1&2 SPEECH IN SONG. [Sect. XIV. of p, b, t, d as paa baa taa daa, baa daa taa paa in various orders,' and they should be checked, at least at first, by silent symbols as explained under ^, b. p. 100&. Glides of T on to Consonants. — In the beginning of Eng- lish words t apparently glides on to r' and w only. It really also glides very frequently on to I, where k-\-l is the theoretical pro- nunciation, as clay tlai for klai, because tlai is much easier, the tip of the tongue remaining fixed while merely the sides are loosened for I; whereas in klai there is a violent change from contact with the back of the tongue to contact with the tip. In case of tree tr'ee, the tongue sinks instantly to the proper form for r\ and nothing like tu-r'ee- should be endured. In case of tw, it is really more common to adjust the, lips for w before the t- contact is released, than to do so and raise the tongue at the same time. The former action is represented by tw', the latter by tw. Thus 'twain' is rather tw'ai-n than t+w+ai-n. In fact in saying the latter it is difficult to avoid an inelegant insertion of an obscure voice sound, as tu+wai-n. But at the end of words we have +^+s (or at least what is so considered, though, as we shall see, the sound is rather -\-t+s'), in cats kats, and t-\-th in eighth ai-tth. This last is very curious. In saying slowly (sustaining the final consonant, or at least keeping the tongue point against the palate), the words: bread that is cut thin,, bred dhat iz hut thin, it will be found that the tip of the tongue has to slide along the palate, over the gums and down to the teeth, and it is during this slide that the glide takes place. So in ai-tth, the tongue point slides from the t to the th position. In words beginning with ' ch ' or terminating with 'ch, tch'as: chest such match chest such match, it is usually thought that we have the glides t-[-sh and -\-t-\-sh, but very close observation has convinced me that this is not the case, and that the real glides are ty'+sh' and +ty'-\-sh' which will be considered further on. Glides of D on to Consonants. — The d-\-r\ d+w follow the rule for t+r' and t+w, such words as 'dwell' being rather dw'el than dwel. The glide +d+z (or rather +d+z') also occurs, but not +d+dh. The termination +dth in breadth width br'edth width is not a' glide unless a t, as sometimes happens, is interposed, as br'edtth, and even then there is not a glide from t. In br'edth, the d voice simply ceases and the th hiss begins, at least I cannot hear a glide in the transition. T' MUTE i. 5. D' SONANT ii. 5. The position of the tongue point for English t, d, is, so far as I know, peculiar to English. In other languages the point of the tongue comes much more forward, at least on to the gums, as for t', d' , and may advance as far as the teeth for t", d". The t', d' are universal in German, Italian Spanish, and French, but are heard in English only in Sect. XIV.] CONSONANTS AND THEIR GLIDES IN PARTICULAR. T03 the glide on to r' or rather r" (vi. 5), in some dialects, as sVr"ong for strong, which has a peculiarly thick effect. It is not at all worth while for an Englishman to trouble himself to use t', d' in place of his native t, d, when speaking or singing foreign languages. Most Englishmen would be entirely unable to distinguish t, t' and d, d' when they heard them. Hence they need not be practised. T" MUTE i. 4. D" SONANT ii. 4. These differ from t\ d\ only in having the tongue point fully advanced against the teeth quite as far as for th. This very dental form is occasionally heard in German, Italian and French, and possibly Spanish. It is regular in modern Greek and the Indian languages. The Indians recognise French t\ d' as their own " dental " t", d", and distinguish them sharply from the English t, d, . which they identify (wrongly) with their own ,i, ,rf, called cerebral ser'-ibr'el by English grammarians. These consonants need not be practised. ,T MUTE i. 7. ,D SONANT ii. 7. The tongue point is brought against the hard palate considerably behind the gums, and is " reverted " that is the under part of the tip, and not the upper part, presses against the palate, so that the tongue is bent back, with the point directed towards the throat, and on looking into the mouth the under part of the tongue only is seen. By this means the front of the tongue is hollowed on the side nearest the throat and the result gives a peculiar dull character to the. preceding or following vowel. The margin of the tongue rests on the teeth much in the same way as before, and completes the contact, which leaves rather a square than a wedge shaped resonance cavity. This is the true Indian cerebral form with which Indians confuse English t, d. It seems to occur in the Southwest of England, where the derived ,r (vi. 7) and jL (viii. 7) are certainly found, and was probably West Saxon. It seems to be the foundation of our untrilled „r (iv. 7'), and of our common sh, zh (iii. 7 and iv. 7), and hence is worth attention, although it need not be practised. Group III. Tongue Front — ty' dy\ ch' j\ ky' gy'. CH' MUTE i. 9. J' SONANT ii. 9. The tongue is first put into the position for ee, and then the point being pressed against the lower gums, and the front raised to press against the very, back of the hard palate, the contact is completed by the margins of the tongue and the teeth. The shape of the mouth, therefore, as seen from the open lips is wedge shaped and the thick end of the wedge is in front. The resonance cavity behind is extremely smajl and narrow. Does this consonant occur in English ? Mr. Goodwin thinks that our initial ' ch, j ' in chest jest, are really. M j', and it is certain that clr^st, j'est, if especial care is taken: 104 SPEECH IN SONG. [Sect. XiV not to introduce flatus after ch', are scarcely if at all to be distin- guished from chest jest, except by the greater neatness and absence of hiss and much buzz. Now this is important for the singer, whose especial object it is to avoid these unmusical inter- ruptions as much as possible. It may therefore be worth a singer's while to practise ch'aa j'aa &c. with all the vowels as initial glides, for the purpose of using them as much as possible. But final glides on to ch',j' do not produce the effect of ch,j at all, just because they have no flatus, and hence it becomes necessary to add on sh' (iii. 6) and zh' (iv. 6). It is certain that such'sh' cannot be distinguished from such, or r'ifdzh' from r'ij, but it is more than doubtful whether the gain in using ch',j' initially at all counterbalances the loss of acquiring entirely new habits of speech. Though it was necessary to mention ch' j' , it does not seem at all necessary to practise them. The Indian sounds which all Englishmen confuse with ch,j, are really these ch',j' but they never occur at the end of words, being always replaced by other sounds which are easy to pronounce in that position. TY', MUTE i. 8. DY', MUTE ii. 8. Palatalisation and Labia- lisation. The consonants ty' , dy' are generated in English but are not acknowledged, so far as I know, in any language but Hungarian. The position for t or V being assumed, the whole front of the tongue is afterwards brought along the palate, so that the resonance cavity behind is almost the same as for cW , but the front aspect as viewed from the lips is very different ; for cK the whole upper tongue from the point to the further part of the front can be well seen, but for ty\ only the under part of the tongue is visible. Hence ty' may be described as an attempt to take a t and y or rather a ch' at the same time. In the same way dy' is an attempt to take d and y or rather j' at the same time. When a weak ee or t, the first part of an ee-\- diphthong, p. 72c, follows a t or d there is always a tendency to take it at the same time as the consonant, which changes tl- or ty into ty' , and d^- or dy into dy' , just as tw, dw pass into tw', dw'. The first change is called " palatalisation " the second "labialisation," most conveniently pronounced pal-utal-izai-shen lai-bial'izavshen. Now the glide from ty', dy' to a vowel, is difficult to perform without in- troducing some flatus or buzz. Hence in place of ty'+aa, dy'+aa, it is found esLsiei to sa.y ty' -j-sh' +aa, dy' +zh' +aa, where sh', zh' (iii. 6 and iv. 6) will be subsequently considered. Herein we have a complete explanation of some common English changes, as ver-teu = vevtioo into ver-ty'oo and that into ver-ty'sh'oo, that is ver-choo as it is commonly pronounced, and ver-deur=ver-dioor into ver-dy'oor ver-dy'zh'oor=ver-joor and then into ver-jer. As even dy'zh' is apt to become dy'sh', an unpleasant hiss is generated in both cases. Hence the singer is strongly recommended to avoid this very unnecessary change and Sect. XIV.] CONSONANTS AND THEIR GLIDES IN PARTICULAR. I05 sing ver'tioo, ver-dioor with pure i, and d, keeping the vowel i distinct, and not suffering the t and d to be palatalised. Hence ty', dy', so far from being practised, except as parts oi ch,j when otherwise generated, should be strictly avoided. KY', MUTE, i. 10. GY', SONANT, ii. 10. These are palatalisations of A, g, generated by a following i or y && ty', dy' are generated. To obtain their positions, first assume the position for k, in which the back of the tongue plugs up the passage to the throat, and bring the front of the tongue along the hard palate, but turn the point down. We have then the position for ch' with this excep- tion that the back of the tongue is high, so that in fact there is no resonance cavity behind it. In present English there is still a trace of ky' gy', which singers and actors are apt to exaggerate disagreeably. Thus, there was a fashion in England a hundred years ago, to introduce a ky', gy' for k,g before any sound of aa, a, ei, as cart ky'aa-rt, garment gy'aa-rment, sky sky'ei. This is often exaggerated nowadays into keeaa'rt keeyaa-rt, skeeyer, which of course must be carefully avoided. The only word in which gy' remains in great force is girl gy'erl, vulgarly gal gy'al, but even here gerl is considered better, provided that it does not degenerate into guurl. Generation of CH, J from KY', GY'. — The glide from ky',gy' to a following vowel is difficult to execute neatly v/ithout introduc- ing a flatus or buzz which necessarily take the form of sh', zh' (iii. 5 and iv. 6), so that in place of ky'+aa, gy'+aa, we easily say ky'+sh'+aa,gy'+zh'+aa OTgy'+sh'+aa, and the gHde ky'+sh' is necessarily indistinguishable from ch'+sh' or ty'+sh', so that ch, j result from ky', gy' as well as from ty' , dy'. All our words which came from old English, as choose chooz, got their c/z in this way, and such double forms as kirk church kerk cherch, work werch werk werch, brig bridge br'ig hr'ij, are remnants of the unchanged and changed forms of which there are numerous examples in our language. Most of our words that came from old French with ch and j (which have sh and zh in modern French) were derived in this way from Latin k and g or else from Latin jv worked up io f and so to j'zh' orj, a.sjtist from old Latin Justus yoos-toos, of which the first stage was the Latin ;'oo5-foos, whence came the Italian yooi-ioa a.nd¥rQuch. joostjuest now shuesi. It is therefore seen that the three forms ty' , ch' , ky' are so related, that whereas ch' only uses the front of the tongue, ty' uses both front and point, and ky' both front and back, and that the natural transition from all of them is to sh' before the vowel which follows. Hence it is that all of them generate ch, for which various orthoepists have mistaken them all. Of course there is no occasion to practise these sounds, as the singer can use ti-\-, ch, ki+ with much greater ease to himself, and, except as regards ch, with much greater musical effect. I06 SPEECH IN SONG. [Sect. XIV. Group IV. Tongue back — K G, kw' gw', K, MUTE, i. II. G, SONANT, ii. II. The back of the tongue raised as for oo, but brought close to the arches of the palate, so as entirely to block the entrance into the mouth. It is almost-im- possible to release this contact without making a click or smack, from the separation of the wet tongue and the soft palate. Glides of K and G from and on to Vowels. — The glides on to all vowels are quite distinct, and the nearest related vowel is oo. Practise this pair of consonants in the same way as was recom- mended for/>, b, trying aa kaa, aa aak, aa kaak with all vowels, up and down the compass, and varying with aa gaa, aa aag, aa gaag, and kaak kaag, gaak gang, kaa gaa kaa, gaa kaa gaa, and so on. It is also very advisable to try such phrases as paa taa kaa, daa paa gaa, kaa daa paa, gaa paa taa, and so on varying k, t, p, and g, d, b and making sure that the sounds are heard. 1 heard lately of a great tragedian who used to practise declama- tion on Hampstead Heath and send on a friend to determine the distance at which each word was distinctly audible. It is similarly advisable for a singer to determine the distance at which his glides to and from mutes can be readily seized. The final glide will be found the most difficult because the voice loses its energy and dies away before the glide is reached, and audibility depends on the smartness and firmness of the glide. Glides of K and G on to Consonants. — Initially k-{-l, k-\-r', k + w' and g+l,g+r\g->rW occur. But k+l is difficult on account of the great change of position of the tongue, and the comparative want of mobility in the back of the tongue. Hence t-\-l-{-, d+l+ are constantly substituted at the beginning of words, as cleaned gloves tleend dluvz. Sometimes k-\-r'-\-, g+r'-\- are also founa difficult. The forms k-\-w-\-, g-{-w-\- will be considered imme- diately. At the end of words -\-k-\-s is common, as axe aks, makes mai-ks. Yet initial k-\-s-\- though common in Greek is never used in English, and 'Xerxes xylography' become Zerkseez zeilog'ru-h-fi. Of course -{-kt cannot form a glide, and hence when no vowel follows the recoil must be used. There is an exception when a t follows. In phrases like, " he will act too," we do not say hee wil akfli -too, but apparently merely hee wil ak...too, but the speaker is generally aware that he passes from ^ to i immediately, (silently of course), and then holds the t position, as if he had a doubled t, like : at two at -too, and comes down at last energetically on to the following -too. Hence he feels that he pronounces t in both words and says hee wil akt too, but his doing so remains his own secret, and never reaches his audience, which can infer it only from a relatively prolonged (though of course absolutely very short) silence. Sbot.xiv.] consonants and their glides in particular. 107 KW, MUTE, i. 12. GW SONANT, ii, 12. The mouth being disposed for k, the lips are gathered as for 00, and then the glide is made at once from this compound position to the vowel. The, glide kw'aa is therefore quite different from kaa or waa, but is very liable to pass into either. These forms seem to be the proper sounds of ' qu gu ' in queen kw'ee-n, guano gw'aa-no. In German they do not occur at all, kw' being replaced hyk + v' + ,. in Italian they are practically replaced by koo-, goo-, as quanti guanti koo+aan-tee goo+aan-tee, and similarly in Spanish. In French, kw' at least occurs in quoi kw'aa. As the consonant hw'^ is frequent in English it should be well practised, as aa kw'aa, ee kw'ee &c. It never occurs finally. Division Straits — there is a narrow passage for the breath, owing to a slight relaxation of a contact, producing Con- tinuants, which can be sustained indefinitely. Subdivision Central Steady— the strait passes along the central, line from the mouth to the lips, and the sides of the passage are steady. Class iii. Flated or Hisses — unvocalised breath passes through the central passage and produces a hiss, full of rapid beats. Class iv. Voiced or Buzzes — ^voice is driven through the central passage with considerable difficulty and rather an unpleasant effect, seeming like a mixture of hiss and voice. Group I. Lip Straits— W^// W, f v', F V. WH, HISS, iii. I. W, BUZZ iv. i. The lips being placed for p, the contact is slightly loosened, and a small central aperture is formed, less than for 00, while the contact on each side scarcely extends beyond the outer edges of the lips, for the flatus gets in between them, and between the teeth and the lips, puffing out both upper and under lip, when the flatus is strong. The back of the tongue is raised to the 00 position. Flatus gives wh, voice gives w. The w is very common in English, but, so far as I know, does not occur in any other European language. It is really a buzz, and on prolonging it gives a disagreeable sound and not 00. The singer is recommended to use the Italian substitute, 6d-\- (see p. 74a), which gives no buzz, and would not be heard as any thing else but w, except before 00, uo, where w must be used, as : we would have wooed once ooee wuod ha'v woo'd ooun's, for the spoken wee wuod'uv woo-d wun'S. Then wh will become hdd-\-, and even the voiced aspirate (p. 231^) may be Employed; But the singer and speaker should never omit to distinguish between wh and w, as is so frequently the case, thus: while wheil, in singing hdd-{reil, wile weil in singing 66-\-eil, and not both weil, as is common in London among the most highly educated speakers. I08 SPEECH IN SONG. [Shot. XIV. Glides from W and WH. — Practise initial aa waa whaa, &c. for all vowels with great care, for the combination is frequent. As w does not now occur as a final, it has been used in glossic as a substitute for +00, p. 71a, and therefore will be practised with the diphthongs. W also does not occur medial, hence if it comes between two vowels in glossic, it is best to indicate by a hyphen or accent mark whether it forms a diphthong with the first, or a buzz gliding on to the second, as cowing kuwing, away u-wav, as ku-wing- uwai would be irrecognisably different. In approximate glossic, where ou is used for all forms of the diphthong, no error could arise from omitting the hyphen, as uwai' kou-ing. No glide now occurs from w to any consonant as r' and I, and although we retain the written letter in wright write r'eit, we do not retain it in (w)lisp lisp. It seems probable that the old wr-, wl-, were labialised r'w'-{-. lw'-\-, and in that form they occur in French roi, loi, r'w'aa, Iw'aa, which are usually con- sidered to be r'waa, Iwaa and which the singer may take as ■K'6d-{-aa, ld6-\-aa. There are traces of w in 'wr,' to be found in our dialects, as : write vr'eet in Aberdeenshire, and w-h'-r'eit w'-h'-lisp have been heard in the North by persons still living. F, Hiss, iii. 3. V, BUZZ, iv. 3. The lips are arranged as for 'fp, (p. 1016, introduced for this reason,) but the lower lip does not come windtight against the upper teeth, so that the flatus for /escapes between the soft lip and the hard sharp edge of the upper teeth, in a thin, much interrupted stream, which is driven up against the inside of the upper lip and is then reflected down- wards. The first escape of the air gives a very peculiar sharp thin hiss, to which the upper lip adds a deeper element, giving a hollow character to the hiss. It is necessary to preserve some of this hiss for identification, because, if reduced too much, f-\-aa sounds like 'fp-\-aa, the glide for which is identical, and this may be readily confused with p-\-aa. Glides of F from and on to Vowels. — In the glide /+ there is not much danger of prolonging the hiss, but there is great danger of so doing in the final glide +/, and hence the singer should practise shortening this /. The shortening of a hiss is best and most quickly effected by stopping the pressure on the lungs, so that there is no air to pass. The hiss does not stop immediately, for the air already in motion has to pass, and the time taken up thus suffices to give sufficient duration to the hiss. The singer is recommended to practise aa paa faa, aa aap aaf, with all vowels and at all parts of his compass. In the initial glide /+, take care that the voice is neither put on too soon nor too late, so that neither fvaa nor f°aa-aa occurs. In the final glide +/, take care that the voice is not carried on to the /, producing aavf, for example. In medial +f+ as laughing Sect. XIV.] CONSONANTS AND THEIR GLIDES IN PARTICULAR. IO9 laa+f+ing, both errors must be guarded against. In double +ff+, there is a little relaxation of force in the middle as: stiff feet stif-^feet, half friend, half foe, haaf-i-fr'end, haaf-=rfoa. Glides from F to Consonants. — All hisses can be added at pleasure before or after mutes or other hisses, but in English we find only /+;+ flow _^oa, /+r'+ fry /r'«, 5+/+ sphere 5/eer, +/+S laughs laa-fs, +/+< laughed laa-ft, +f-\-th Mfh fifth, (for which the older form, still sometimes heard, is fift.) In Germany we have p +f+ as pfaff pfaaf, pfropf pfraopf, but in order to say this, the lower lip is generally brought into the position for / immediately, and hence ^fp is generated, which relaxes perfectly on / as 'fpfaaf, 'fpfrao'Jpf. This is a very modern German sound, the genuine being//' or p'f. No Vin German or Spanish. — The hiss/ is well represented in English, German, Italian and French, though absent from many languages. The buzz v, which is quite easy to English, Itahans, and French, does not exist in Spanish (except on paper) and is almost impossible to a German ; at any rate I have not heard it from Germans who have spoken English for many years, and who are equally at a loss for w and for v. German orthoepists generally assert that the German ' w ' is w, at least in the North of Germany. It is however contrary to my own experience of a three years' residence in Dresden, where I had opportunities of hearing Germans from many parts of the country, and contrary to my experience of Germans in attempting to speak English. However, Englishmen may, if they find the real consonant v' too difficult, always use v for the German ' w.' Glides from and to V. — In the initial glide, v + , there is no difficulty, except in not putting on the voice soon enough, which would give fvaa, a result the singer should specially avoid. The final glide +v, as a rule falls into +v/ in the pause, because the voice dislikes to prolong a buzz (pp. lyd, 89c), and this the singer has to fight against strongly, as he should never introduce an unnecessary hiss. V never glides on to a consonant at the beginning of English words, and only on to d, z at the end of words, and that rarely, as: halved haa-vd, halves haa-vz. F', hiss, iii. 2. V, BUZZ, iv. 2. — Dispose the hps as for p% (p. lood), relax the contact and emit flatus or voice, coming out through the lips in a thin stream, without any interference with the teeth, or any reflection from the upper lip. The action is the same as in blowing liquid to cool. There is no puffing out of the lips, and the position of the tongue is indifferent. The sound is altogether softer and pleasanter than/, v both in hiss and buzz. Neither/' nor w' is acknowledged in English, Italian or French, and/' is not acknowledged in German. But in the initial ' pf ' of the Germans, although pf is professed to be said, either//' or p'f was certainly the oldest form, and is I believe still extensively no SPEECH IN SONG. [Sect. XIV. used. In Hungarian/', v' seem to be always used for/, v, as I found by the extreme difificulty which an Hungarian whom I knew, felt in acquiring the sound of v, which he had not noticed to be different from his own v'. And the English singer may without fear of detection use the much pleasanter/', v', at any rate initially, in place of/, v. They make the initial glide / + , v'+ quite as well as/, v, but not the final glide. German and Spanish use of V for V. — The Germans so far as I have observed (and I have paid especial attention to the point for many years) invariably use v' for their ' w,' which however occurs only as an initial, so that we have v' + but not +v'. Thus : wie v'ee, was v'das, schwelle shv'ael'u, quelle kv'ael'u. The Spaniards invariably use v' for both their b and i), which are distinguished in writing by dint of many rules, but are never distinguished in speaking. This beautiful consonant, far superior to either w or v, is worth acquiring, although v will be always intelligible in either German or Spanish. The speaker should however guard against w, or even vw, as some recom. mend, for the German v'. Group H. Tongue Point Straits— Ti7 DH, t'h d'h, S Z, s' z\ sh' zh', SH ZH, ,.»-. The hisses and buzzes which can be made with the tongue point are nearly endless ; the few here selected have all special values to the speaker or smger. TH, hiss, iii. 4. DH, buzz, iv. 4.. The teeth are near together, the pomt of the tongue is brought against the upper tieth (as for t", i. 4) so that the upper side is concealed, and the thickness of the tongue can be seen between the teeth. The tongue is therefore not between the teeth, though an efficacious way to instruct a German, Italian or Frenchman to pronounce something approaching to this wondrously difficult consonant, as they consider it, is " put the tongue between the teeth, and try to say 5." The flatus and voice force their way between the tongue and the edge of the teeth (as the side passage is stopped as for t"), and then strike the upper lip, much as for/, v. In fact the distinction between/, v and th, dh Is very slight, and/, th are constantly confused by children, and interchange in dialects. When uttered as simple hisses without gliding to a vowel, I have found the mere hisses, even when pronounced by 300 persons at once, difficult to distinguish. The glide however makes a considerable difference, as th-\-aa closelj' resembles t"+aa (p. 1036), and hence t+aa nearly, but f+aa resembles 'fpaa almost precisely, and hence />+aa nearly. It is the hiss Sect. XIVJ CONSONANTS AND THEIR GLIDES IN PARTICULAR. HI superadded which makes the difference. In case of the buzz and sonant, v+aa and b+aa are very difficult to distinguish, and dh+aa becomes confused with d'+aa or z-\-aa by foreigners. It is singular that in the Shetlands- and Orkneys, and also in Kent and East Sussex, the words 'the this that them there these those ' &c. are pronounced with initial d instead of dh. , When to use TH and when to use DH. — Singularly enough as the two sounds tK, dh are not distinguished in ordinary English orthography I have found even highly educated English people who were unaware that two sounds existed. Hence it is necessary that the singer should carefully distinguish them. ' Though ' is dhoa in England and thao in Scotland, and the word becomes a perfect shibboleth, even for the most highly educated, yet very few of either nation are aware of this national difference. The "pronominal words" than dhan, that dhat, the dhi dhu, their dhair, them dhem, then dhen, thence dhens, there dhair, these dheez, they dhai, thine dhein, this dhis, thither dhidher, those dhoaz, thou dhou, though dhoa, thus dhus, thy dhei, as is here seen, begin with dh, all other words having initial ' th ' in writing are pronounced with th. It is well to contrast words by interchanging th, dh, as to call thy thigh thei dhei in place of dhei thei, or to pronounce: they breathe thin breath, thai br'eeth dhin br'edh in place of dhai br'eedh thin br'eth; this interchange is marked enough to render the words unintelligible. In the middle of words dh is most usual, as : ieather fedh-er, leather ledh-er, nether nedh-er, mother mudh-er, fsither faa-dher, hither hidh-er, and to sa.y feth-er, leth-er, neth-er, muth-er, faa-ther, has a remarkable effect. Of course nothing nuthing=noa thing is an exception. At the end of words both th and dh occur. Bathe baidh, sheathe sheedh (sometimes sheeth), meathe tneedh (usually meed), breathe br'eedh, wreathe r'eedh', bequeath bikw'ee-dh, lathe laidh, loathe loadh, swathe swaidh, hithe heidh (as in Rotherhithe Rodh-er-heidh), blithe bleidh (often bleith), writhe r'eidh, scythe seidh, tithe teidh, withe widh (some- times with), c\othe kloadh (the substantive 'clothes' generally drops the dh and becomes kloa-z often tloa-z), to soothe soodh (often spelled 'to sooth'), and all that end in '-the' have dh. Seeth seedh, with widh (some say with, and in composition it is always with as herewith heerwith' &c.), booth boodh, smooth smoodh, to mouth moudh, are all the words, not names, ending in ' -th ' which have dh ; the rest, which are numerous have th. But both boa-th is called boa-dh by Scotchmen, and north naurth, south south, take dh in northern nau-dhern, southern sudh-ern. Some names, as county Meath Mee-dh, Bishop Lowth Loudh, are exceptions. In the west of England the initial th often becomes dh, and before r it becomes d, as three dree (more properly cL;ree). The initial glides th + ,dh + , on vowels are easily 112 SPEECH IN SONG. [Sect. XIV distinguished, but in the pause -\-dk is apt to become +dhth, by dropping the voice (p. ijd), which the singer must avoid, but then he must not suppress the buzz too much, or br'eedh will sound like hr'ee, just as in Spanish ' ciudad,' which is meant to be theudhaadh' (or at least t'heeoo'd'haad'h') is apt to sound as theudhaa-, the final dh, as a Spaniard expressed himself to me, being " rather felt by the speaker than heard by the listener." This arises from the Spaniard having no tendency to use -\-dhth for +dh. Glides from TH and DH to Consonants. — The initial glides are th-\-r\ th-\-w as : three thwacks thr'ee thwaks, but as thw is not very easy, thw' is generally said as thw'aks, that is, the lips are closed for w at the same time that th is formed. We also find +th-\-s in path's, bath's paa-ths haa-ths (in the posses- sive case, the plural 'paths baths' are paa-dhz baa-dhz). Th glides from i in eighth avt+th, but after d, it seems to be independent as wid...th, though some perhaps say widtth, and some actually say witth. S, HISS, iii. 5. Z, Bvzz, iv. 5. These are the typical hiss and buzz. The tongue holds the side teeth and sides of the hard palate quite tight by its margin, as far as the division between front and point, and the point is lowered slightly, so as to leave a narrow flat passage between the arched torigue and the hard palate, which can be much diminished. If, without altering the position of the front and back of the tongue, we raise the point till it touches the palate we shall find that it strikes the gum? precisely, and hence the S strait must be regarded as an opened t' contact. By forcing voice through, we obtain the typical buzz Z, and it is important for subsequent use to force the buzz very strongly and for a long time, by way of an experiment, when it will be found that the stiff but unsupported tongue point will vibrate and tingle as the vibration is resisted muscularly, and that even the inside of the lips share in this unpleasant vibration. One consequence is that it is decidedly disagreeable to sustain the buzz Z, while it is pleasant and easy to sustain the hiss S, so that there is a constant tendency to slide off from Z to S. It should be observed that Z is a generated sound in most Indo- European languages and certainly in four of the five here con- sidered, although in the Semitic languages, as Hebrew and Arabic, it was original. It does not occur in Spanish. It is con- sequently generally written ' s ' in English, German, Italian and French, except in words taken, or presumed to be taken, from the Greek (which now has a z, although the old sound of the letter now so pronounced, has been much disputed), and it is necessary to know when s and when z should be said, which must be learned from a pronouncing dictionary in cases of doubt. Scotchmen should observe especially not to use z for 5, as in : pre- SecT.XIV.] CONSONANTS AND THEIR GLIDES IN PARTICULAR. II3 cisely Tiecemher prisei-zli Dizem-her, which has a strange effect to English ears. In the southwest of England initial z for s is a very characteristic usage. Glides from and to S. — The glides s+aa, aa-\-s, aa-\-s-i-s-\-aa, aa-\-s...s-\-aa present no difficulty. They are not so perfect as those from t', but the hiss is so characteristic that no ambiguity can arise. At the same time the hiss is so piercing that it forms one of the principal difficulties of a singer. It is generally pos- sible to find one's place in a psalm-book by noticing where the s falls, as this is the one sound which comes most certainly to the ear. The singer therefore has to strive to lessen this serpentlike effect as much as possible. Hence in s+aa, he should practise making the s as short as possible for audibility, testing by silent signals from a distant friend, and using all the vowels in turn. This is obtained by hurrying on to the vowel as rapidly as is con- sistent with actually sounding the s. It is of course possible to omit the hiss altogether, by raising the tongue to the 5 position and dropping it for the vowel before any hiss or buzz escapes. The effect is that of an imperfect dental t' , having nearly the same glide, but by the imperfection of its construction, being neither a proper mute nor a proper continuant, but only the latter portion of the glide. A similar glide is heard at the end. Indicating this im- perfect s by ,s, we find ' Cease, blust'ring Boreas,' converted into ,see,s blUySt'ring Bao-rr'ia,s, which is very difficult to distinguish from ee blwtring Bao'rr'ia — except by the speaker himgelf who is initiated into the mystery by feeling his tongue move. No one who has the least wish to be intelligible will murder words in this way. They had better sing a simple laa throughout. In the medial cases the same thing is quite possible, and even +j^s+ may be got over by taking +5. ..5+ as missent mi+s...s+en-t, for mi+s-i-s+en-t. It is in finals that the great trouble is felt. In such words as : ice eis, grace gr'ai-s, cease see-s, hiss his, toss tos, there is a great tendency to lengthen the hiss. Now we ought not to shorten it by dimin- ishing force as we come to it, (the common expedient), because then we lose both the glide and the characteristic hiss, except for ears a foot or two distant. We can, however, cease to expire, as recommended for +/ (P- loSrf), or we can drop the tongue suddenly. The former is certain, the latter is apt to produce some other hisses. Hence the singer must diligently practise aa+s &c. with all vowels at a distance. Glides to and from Z. — The glide z+aa initial is easy, but not frequent in England, as in zeal zee-l, zephyr zef-er. Initial z for s is the rule in German, but; when the preceding word does not end with a vowel, the initial z is usually taken as sz-\-. In English the difficulty lies in guarding against +zs final. On both points, see p. 18, with the exercise, and in addition practise II 114 SPEECH IN SONG. [Sect. XIV. ^he following " nonsense verse," till the serpent s is charmed without entirely losing its hiss. Ex. 36. i feEE S W J 1 ^ those dhoa-z those dhoa-i seas ! see-z ! those dhoa-z i ^ S -^ -Nr- - w boist boi-s reus trus seas see-z those dhoa-z rest - less sur - ges rest- - les ser- - jez i ^ ^ S bil-lows vast, those bil--oaz vaa-st, dhoa-z sur - ges cease-less ser- - jez see- s -les rise r'eiz Glides from S and Z to Consonants. — The independent character of the s hiss allows it to be prefixed and affixed easily anywhere, see the lists on pp. 90-gi. This is not so much the case for other Consonants. See the same lists. SH, HISS, iii. 7. ZH, buzz, iii. 7. The tongue is retracted as much as for J:, and, if not fully reverted as for that letter (p. 103c), at any rate, has its front part hollowed much in the same way as it is for ,<, from which contact this strait is evidently formed, as has been always recognised by the Indian gram- mariaais. Put the tongue as for f, and release the contact by darting the point only of the tongue very slightly forward, bringing it just away from the palate and leaving the front and back in its former position. The hollow effect of this cavity at the back of the tongue is sometimes augmented by pouting the lips, and this is so well recognised, that the mere assumption of that position, without any hiss, is understood to mean hush!, but it is not usual in speech and should be avoided in singing. SH though now well developed in English, German, Italian and French, did not exist in Anglo-Saxon, old High German and Latin whence they came. It is still not found in Dutch, and Spanish, and also Welsh. In English, German and Italian it came through sky'-\-, and in French through ky\ The ZH buzz is found in English and French only, sparingly and only medial in the former, frequently in all places in the latter, as English division divizh-enj pleasure plezh-er, French je zheo, kge ahzhai, age ahzh. Those who find a difficulty in producing it initially must try beginning with sh to get the position of the tongue, and then putting on Sect. XIV.] CONSONANTS AND THEIR GLIDES IN PARTICULAR. II3 the voice as j'ai shzhae, but he must rapidly shorten and finally extirpate sh, because any semblance of that sound initially is disagreeable to French ears. Glides from and to SH. — It will be well for the singer to practise the contrast between the glides of s and sh, as s+aa, sh-\-aa, aa-\-s, aa+sh, with all vowels, and to ascertain, by silent signs, that the distinction is heard at a distance and that the hisses, though audible, are not too prominent. There is some- times a little difficulty in passing from s or « to sh, as in sun- shine sun-shein, suns shine sunz shein, 'sashes sash-ez. The word shee unemphatic, often becomes the mere hiss sh, and is then added on freely to other consonants, as : she told me so shtoa-lmi sou. But sh does not occur regularly before con- sonants initially except in shr' + , as shrine shr'ein, which many people find a difficulty in saying, and reduce to sr'ein, a fault to be carefully avoided. In G,erman however sh is regular before r' I, m, n, v' as : schreien shr'aayn, schlaf shlaa-f, Schmidt shmeet, schnee shnai, schwur schv'oor'. In the initial 'st-, sp-' it also regularly occurs, at least in the form sh', over most of Germany, even Hanover, which long kept to st-{-, sp + , having adopted the full sht+, shp-\- as in : stehen shtavn, spielen shpee-len. No public singer or actor would venture to say stai-n spee-len, and hence Englishmen should carefully avoid this sound. But the termination '-st' is very frequently pro- nounced -sht by the German vulgar as : August , ist stark Aawguosht isht shtaar'-k and this has to be very carefully avoided, as it is as bad a vulgarism as " dropping the h " in English. Say Aawguost eest shtaar''k. ,,R, BUZZ, iv. 7'. — The tongue being placed as for f, and the voice set on, the tip is slightly removed from the hard palate, so that flatus or voice may pass over it and produce a slight imperfect hiss or buzz by the way, but only the buzz is in use. The mode in which the position is really reached seems to be from the vowel u, .by raising slightly the point of the tongue, hence I term ,,r " the point rise." This consonant is largely used by Englishmen and Americans in place of r\ The received pronunciation prefixes u to r', after long vowels, and permits r' to be omitted if no vowel follows, (see p. 63c and PP' 75'77)' ^^^ after a short vowel r' must be used if a vowel follows (Americans however generally say U -i- me ^^r-i-i-ku), a.nd r' must be used initially. When ,,r is used initially, the same sort of imperfect glide occurs as for ,5 (p. ii3e), but there is a -slight buzz, as. voice passes. The singer should especially •cultivate r', which is absolutely necessary in Italian, always introduce it slightly when either necessary or permissible in English, and never condescend to the abortive ,/. Observations , on the Spanish ' soft r ' tend to shew that it is at least very like , r, Il6 SPEECH IN SONG. [Sect. XIV. and it may be written r simply as in English. It is always used except at the beginning of syllables, even when medial, as para paa^jaa, amar aamaa\/, breve v'^j'e-ve. In other cases, and when "rr" is written in Spanish, r' is used as onra aon-r'aa, hieno ye-r'ao. The other hisses may be referred rapidly to these three principal forms TH, S, SH. T'H, Hiss, iii. 4'. D'H sonant, iv. 4'. For fk the tongue is pressed against the teeth, but for t'h it just not touches them. This is said to be the Spanish ' c ' before ' e, i,' and Spanish ' z ' in all places, for which however th may be unhesitatingly said, as the difference is slight and diificult to seize. This t'h is also the defective s in lisps lifhpt'h, where it is generally confused with th. This t'h is also believed to be heard in the best Tuscan pronunciation of certain Italian words in place of ts or ts', as in vizio vee-t'hyoa for vee-ts'yoa, but the Englishman may certainly take the ts' with confidence. The buzz d'h is the Spanish 'd' in deidad d'heyd'haa-d'h or d'heyd'haa- (see p. 112a), and the z of the lispers lit'h-perd'ht'h. It need not be practised. S', Hiss, iii. 5'. Z', Bvzz, iv. 5'. If in depressing the tongue point from the t' position it be not brought so low as for s (for which it may even touch the lower gums,) but be kept very near the gum indeed, the- hiss s' is produced. It may be heard in the natural pronunciation of final ' -ts ' in English, as cats kats', wits, wits' &c., and then if the hiss be prolonged it may be contrasted with s. Similarly pads padz' gives dz'. In German the 'z' is probably always called t's' as in zuzuziehen t's'oo't's'do-t's'ee-n, but Englishmen need only try after ts thus tsoo-tsoo-tsee-n, and they will be sufficiently right. Considerable difficulty is usually experienced by Englishmen with this initial; Let them put the tongue in the position for t' and force the contact into a strait by flatus ; the violent result thus obtained may soon be tempered down, but for German and Italian singing the sound should be rendered quite familiar. In Italian almost all words beginning with ' z ' have t's', and medial ' zz' is pronounced t't's' as zio t's'ee-oa, stuzzichi st'oot'-t's'eekee, for which an Englishman's tsee-oa stuot-tsikee are quite near enough in singing. But in about 100 Italian words (which must be learned from the dictionary) ' z ' is d'z' and ' zz' is d'd'z'. The following are the most frequent of these : manzo maan-d'z'oa, garzone gaar'd'z'oa-nai, amazzone aa-.maad'-d'z'oa-nai, azzurro aad'-d'z'oor'-r'oa, brezza hr'aed'-d'z'aa, bizzarro beed'-d'z'aar'-r'oa, bozzo baod''d'z'oa, caprezzo kaapr'aid''d'z'oa, dozzina doad'- d'z'ee-naa, gazza gaad'-d'z'aa, gazzetta gaad-d'z'aW't'aa, lazza< rone, load' -d'z'aar'ownai, mezzo maed''d'z'oa, pozzo paod'-d'z'oa (meaning ' a hill,' but meaning ' a well ' it is poat'-t's'oa, two words well to contrast for ao and oa, as well as for d'z', t's'), sbct.xiv.] consonants and their glides in PAETICULAR. il7 razzo r'aad'-d'z'od, rozzo r'oad'-d'z'oa. In all these words the Englishman need not trouble himself to distinguish d'., d'z' from d, dz. When the sound of ' z ' is not " energised " in Italian, the t', d' are omitted, and s', z' simply are said, at least in Tuscany. But as an Englishman would be sure to fall into s, z, which would create confusion, he is recomiriended to keep to ts, dz. SH', HISS, iii. 6. ZH', buzz, iv. 6. This is derived from t, as sh is from f, and 5 from V. The true position of English t being assumed, with the tongue therefore retracted, although convex to the palate (and not concave as for sh), the point is very slightly rejeased and the result is sh\ which has the effect of a very fine sh. Just as v' may be defined as v without the teeth, so sh' may be defined as sh without the hollow at the back of the tongue. The sound is consequently much neater and finer than sh, and it is therefore worth cultivation by the singer to replace sh. It is really the sound used in initial st-, sp-, in mid Germany, as stehen sh'tai-n, spielen sh'pee-len, for which the true sh, especially when the lips are pouted, sounds too coarse to an educated Saxon ear. CH, J. — The principal peirt which sh', zh' play in English and Italian is in the formation of the common sounds of CH, J, of which the real analysis seems to be ty'sh', dy'zh', or (when final) frequently dy'sh' (pp. i8a, lo^rf). In Italian when these are not energised, they sink to sh', zh', which, from a Tuscan or Roman mouth are usually taken by Englishmen to be simple sh, zh< Thus faoce faaty'ty'sh'ai, fasce faash-ai, face faa-sh'ai, for which an Englishman may be satisfied to say faaPchai, faash'ai, faa'chai respectively, which will always pass muster. Glides from and to CH and J. — The glides ch + ,j+ should be well practised with all vowels, especially the Italian doubles, faccio faat-chyoa, raggio r'aad'jyoa. The finals -\-ch, -\-j also require much attention in English (they do not occur in Italian) and +ch in German, as the sh' is apt to be painfully obtrusive. Observe that in English ' ch ' and ' tch ' final both mean ch as such, match mach, and that ' dge ' final represents / as bridge hr'ij. This final -\-j is very commonly +dy'sh', held to be +jsh or -\-dsh, as distinguished from -\-dzh, the usual analysis of -\-j. Group III. Tongue Front Straits — sy' zy', YH Y, ky'h gy'h. YH, HISS, iii. g. Y buzz, iv. g. The y is the well known English initial y as yacht yot, and German initial j. as jagd yaakht. The tongue in the position of ch', (p. lojii), is released in its middle, where a flat groove, finer than for ee, is formed throughout its length. The voice gets through this groove with Il8 SPEECH IN SONG. [Sect. XIV. much difficulty, and the buzz of y is very disagreeable when forced. Hence singers are recommended never to use it, except before ee in ye yield yee yeeld, year yeer, where its use is unavoidable, but always to employ the musical and easy i-\- in its place (p. 72c), thus : yet i+et, yacht i-\-ot, yawn i+au-n, yolk i-\-oa-k. The flated form yh occurs only in a few words, as : hew hue yhioo, huge yhiooj, human yhloo'tnu'n, humour yhwo'mer (often eu-mer simply), and may be always repre- sented by h as hioo, hiooj, hioo-mu-n, Moo-mer, generally written heu, heuj, heu-men, hewmer. Compare wh, p. 10yd. Nevertheless the sound yh should be well understood because as Germans themselves confuse it with ky'h, it becomes an easy transition to that otherwise difficult sound. KY'H, HISS, iii. 10. GY'H, buzz, iv. 10. The tongue is placed in the position for ky' (i. 10) and is released into almost the same position as that for jA, but the back of the tongue being much more nearly in the position of k, the consonant ky'h has more of a ky' effect. The buzz of this is more difficult, but varies from y in the gy' direction. It is essential for any one who would sing German to master the sound ky'h, which is of very frequent occurrence after the sounds of ee, ai, ae, aay, aoy, r', I, n and is represented by ' ch ' in ordinary spelling, as ich eeky'h, siech zee-ky'h, echt aeky'ht, eiche aayky'hu, euch aoyky'h, durch door'ky'h, milch meelky'h, manch maanky'h. Where a short vowel does not glide on to it, yh might be used in its place. Care has to be taken that no sound of sh' or sh escapes ; nicht neeky'ht, not nish't, or nisht; welch v'aelky'h not v'aelsh' or v'aelsh. English people persist in believing that sh, zh (the latter of which is an impossibility to a German), are said and sung for ky'h, gy'h, calling fechten fesh-tn, and berger ber-zher, in place of faeky'h-ten, baer'gy'h-ur'. This must be guarded against. However much they persist in believing that they hear sh, zh, I can assure them after years of observation in Germany, and long discussions with Germans, that it is their own ears and not the German singers who are in fault. Ky'h does not begin any syllable in German, except the diminutive termination '-chen,' as madchen mardky'hen. Gy'h occurs initially before ee and ai, ae, as regierung ru-gy'hee-r'uong, general gy'hainu-r'aa-l, but only in adaptations from foreign words. In other cases, initial 'g' is full g as gegeben gu-gai-ben. The use of y as a degenerated gy'h, is a Berlin vulgarism of the lowest type. But at the end of words there is a difference of practice. The termination -ig, is regularly -eeky'h over all Germany, and in inflexion becomes -eegy'h or -eeyh as konig keo-neeky'h, konige keo-neegy'hu or keo-neeyhu. Otherwise in the North of Germany the letter 'g' remains always g. But in the middle and south final ' g ' becomes ky'h Sect.xiv.] consonants and their glides in particular. 119 where it would be so pronounced if ' ch ' were written, as sieg zee-ky'h, siege ze&-gy'hu, talg taalky'h. SY', HISS, iii. 8. ZY', buzz, iii. 8. The tongue in the position ty' is released by turning the point down and leaving a flat central channel. The resulting sounds sy' zy' are acknow- ledged only in Polish, but they have been mainly instrumental in the change of sy-\- through sy'+ and perhaps through sh' + into sh. Thus in English, no longer ago than in the days of Shakspere, all auch words as 'mo-tion, compas-sion, pa-tient, o-cean' and the like had as distinct a see-r- as they still have in French, form- ing two syllables {see-i-) as the rule, and one (sy+ = see+) as the exception. All these have now pure sh in English. It is by means of this change, also, that a few words have passed from see-T- in Dutch and Welsh to sh-\-, although as a rule sh is unknown in these languages. Many orthoepists have indeed analysed sh into sy, which should at least be sjih, which again is related to sky'h, and Briicke has actually proposed that sh is an attempt to pronounce s and ky'h at the same time, an hypo- thesis he seems to have founded on the German use of ' s, ch, sch ' to represent the sounds of s, ky'h, and sh. See Ex. 28, p. 85. Group IV. Tongue Back Straits — kh gh, "r, kw'h, gw'h. KH, HISS, iii. II. GH, buzz, iv. 11. The tongue being placed in the position for k, is released so as to form a narrow passage between the back of the tongue and the uvula and soft palate. This produces a hoarse hiss, approaching to a rasping of the throat, and as there is frequently an accumulation of moisture about the place the hiss is often accompanied by a rattle. The hiss form is common in Scotch, and should be studied therefore for Scotch songs, it was also common in Eng- land in the sixteenth century and may still be heard occasionally from elderly peasants in the. North West, but it does not occur in received English. Both hiss and buzz are common in Gerjnan, but never occur at the beginning of a syllable. The kh occurs after the sounds of aa, oa, ao, 00, uo, only, though after the labialised vowels kw'h seems to be usual. Thus ach aakh, machen maakh'en, sache zaakh-u, sprache sh'pr'aa'khu ; dach daakh-, tag taa-kh, tage taa-ghu ; doch daokh, auch aawkh, brauchen braawkh'en. In Spanish however kh occurs initial, as in : junta khoon-taa, gefe khe'fe, Ba.da}oz Baadaakhaot'h-, Xerez or Jerez Khe-ret'h, Mexico or Mejico Me-kheekao, Quixote or Quijote Keekhao'te. Mr. Tito Pagliardini however considers the Spanish sound to be gh as ghe'fe not khe-fe. Care must bp taken not to confuse this sound with h. In Modern Greek ky'h-, k h &nd gh all occur initial, but gj'^ is changed intojy; the ky'h is the Greek letter X (which we call kei and Greeks call ky'hee) before the sounds of ee, ae, and kh otherwise ; and r (which I20 SPEECH IN SONG. Sbct XIV. we call gam'u and they call ghaam-maa), is pronounced as y before the sounds of ee, ae, but is gh otherwise. "R, BUZZ iv. ii'. This is a " uvula rise," that is an abortive putting forth of the stiff point of the uvula, while the tongue for kh is further advanced. It is a degradation of the uvula trill 'r, (p. 123a) similar to what „r is of the reverted .r (p. 115c). It has been observed between vowels in Northumberland, replacing r' as very vaa"ri, which sounds very like the Italian vaaee. It need not be studied. KW'H, HISS, iii. 12. GW'H, buzz, iv. 12. These are merely kh, gh, with lips rounded as for 00, and are apt to occur after the sounds of ao, 00 in German, as noch naokw'h, buch buukw'h, rather than naokh, boo-kh, but German orthoepists acknowledge the last only. In Southern Scotch however kw'h is a frequent initial, spelled ' quh,' as quhat kw'haht. It need not be studied. Division Straits p. 1076. Subdivision, Central Trilled — the central passage of flatus and voice is periodically interrupted by some vibrating partion, as lips, tongue, or uvula, through, over, or under which the breath passes, setting it in vibration without muscular action in the vibrating parts, and producing the effect of a more or less perfect beat in music. Class V. Fluted Trills or Chirps — unvocalised breath sets the vibrating parts in action. Class vi. Voice Trills, or Trills proper — voice sets the vibrating parts in action. Group I. Lip Trills — ^pr 'br, 'wr. 'PR, FLATED CHIRP, V. I. 'BR, VOICED TRILL, vi. I. The HpS being quite loosely brought together in the position for P, flatus is driven through them, forcing them open, and they then close by their own imperfect elasticity and open again. This is a very common pastime of the youngest babies. The sluggishness of the action of the lip makes the vibration slow, and almost noise- less, but by somewhat stiffening the lip, or, better still, confining the lip with a ring, as in the mouthpiece of a trumpet, the lipa can be made to move fast enough to produce a musical tone. The action deserves attention because it is of precisely the same nature as the action of the vocal chords in the larynx in producing voice, and the tube of the trumpet acts as a resonance cavity in precisely the same way as the mouth and nose in speech. Also' observe that in both cases, and in trills generally, the action is not muscular (except as regards tightening) but purely elastic. When the voice is used, the modification produced by the trill, is that of a rapid interruption and restoration of voice ; and it can De made very powerful, when the lips move rapidly. The sound Sect.XIV.] consonants AND THEIR GI,IDES IN PARTICULAR. 121 '6r is used by coachmen in Germany as a signal to horses to make them stop. 'WR, VOICED TRILL, vi. 3. Many persons are unable to make the true point trill r', and take refuge in this lip trill 'wr, which is constantly confounded with w by hearers, as in "very right," commonly written " vewy wight," in place of ve'wri 'wreit. The lips are disposed as for p', but slightly separated, so as to allow air. to pass, the lips are much stiffened, so that they cannot flap, as for 'pr, but the soft inner part, just beyond the part of the lip usually seen, vibrates with the passing voice, producing the necessary interruptions, which make it supply the place of r'. It is important to recognise the real nature of this sound in order to remedy the error. The first step is to keep the mouth wide open, which renders 'wr impossible ; the next is to learn the true r'. Group II. Tongue Point Trills — r"h r", r'h R', /. R'H, flated chirp, v. 6. R' voiced trill, vi. 6. So many persons have a difficulty in pronouncing r' even before a vowel, and so many more in pronouncing it when not before a vowel, and it is so necessary for the proper pronunciation of any foreign language, especially Italian, that it should be pronounced easily in any situation, and it so much conduces to the intelligibility of English singing to be able to touch an r' finally and lightly, that to gain a perfect mastery over it is worth any amount of practice to a singer. The method I have found efficacious with quite young children is this. Pronounce z strongly and prolong it, till tingling in the tongue is felt (p. ii2c). Then observe that this tingling arises from resisting the air which continually endeavours to make the tongue move. Cease to resist it, by relaxing the stiffness of the tongue, and the tongue begins to shake at once, at first imperfectly, but when the mode of action is thoroughly understood, the speaker will let the tongue alone, to be moved by the air only and not the muscles. The desire to introduce muscular action is the greatest difficulty to be overcome. It would be impossible by voluntary muscular action to make the tongue move fast enough, and the most it can do is to alter the elasticity of the tongue. The old method of getting r' from a repeated d, d, d errs in two ways, first by inducing muscular action, secondly by bringing the tongue against the palate, to the position of a complete contact. The tongue never reaches the palate in r', so that there is merely more or less interruption, never an extinction, of sound. The distance that the tongue moves in each vibration, the rapidity of vibration, and the length of time that it lasts, are all important elements for consideration, and distinguish different forms of r'; they are all much less in English than Italian. A " faint trill," such as may be used after a " murmur diphthong " (p. 75c) is one made with very few 122 SPEECH IN SONG. [Sect. XIV. vibrations, and generally of small extent, though the rapidity may be very great. There are great varieties in the amount of the tongue which is allowed to vibrate, and these distinguish several kinds of R, but as these have no interest for the singer they are passed over. The r' is here derived from z, but strictly speaking, the point of the tongue is more retracted than for z, although the point is kept further from the palate than in case of zh'. The distinction need not be anxiously insisted on. The flated form, r%, a peculiar hiss, does not occur in English and need not be studied. Exercises on r' are very important. First r'-\-aa, r'-\-ai, &c. with all vowels and at all pitches, next aa+r', ai+r' &c. also with all vowels, with a careful distinction between English ee-rr', ai-rr', oa-rr', oorr' where the trill is necessarily preceded by " murmur diphthongs " which modify the vowels (pp. 75-6), and the foreign ee-V, ai-r', ae-r\ owr' ao-r', oo-r' in which there is no murmur diphthong and no change of vowel. Com- pare German hier hee-r', with English here heer, German heer hai-r', and herr haer', with English hair, German rohr r'oa-r', with English r'oar, German uhr oo-r', with English poor; Italian dir dee-r', with English deer, Italian aver aavai-r' and meatier maesteeae-r' , with English air, Italian cuor kwaor' and amor aamoa-r' with English oar, ItaHan pur poor' with English /loor. Also practiseaa+r' &c. with shortvowels, as Italian parte paar'-tai, terra taer'-r'aa, corre koar'-r'ai, torre taor'-r'ai, German harz haar'ts, herz hder'ts, herr haer'. It is quite necessary that these should be brought out neatly at all pitches. English people constantly say (to foreign ears) paa-tai, ter'-aa, kor''ai, tor''ai, haa'ts, herts, hair. These errors must be avoided, R"H, FLATED CHIRP, V. 5. R" VOICED TRILL, vi. 5. R" differs from r' as t' from t. For r" the tongue is really quite in the posi- tion for z, so that if the point were raised d' would be generated. The sound occurs probably always after i', d', abroad, so that we have English tr'ee, Italian t'r"ee, and certainly t'r"-i- occurs in English dialects. The singer, however, need not trouble himself with these varieties, but may simply sing his usual tr'ee. The flated r"h is not known in practice. ,R, VOICED TRILL, vi. J. The tongue being placed in the posi- tion for ^t (p. 103c), the point is released and allowed to vibrate by the action of the breath. This trill is universal in the Southwest of England, and indeed has been observed from Devonshire to Tunbridge Wells in Kent. It may be an original Saxon sound. But it is here only necessary ±o know it in order to avoid it. It has an interest however in the fact, that the common "point rise" „r (iv. 7, p. 1 15c) so much used about London, is merely a variety of this reverted ,r, caused by not allowing the point to trill. Skct.XIV.] consonants AND THEIR GLIDES IN PARTICULAR. 123 Group IV. (there is no Group III.) Tongue Back or Uvula Trills — 'rh 'r, 'kh 'gh, 'rw'. 'RH, flated rattle, v. ii. 'R voiced trill, vi. ii. The tongue being a little more advanced than for kh, the uvula is allowed to vibrate by the action of the air, care being taken not to stiffen it in any way. The voice form is very common. It is the usual form of trill throughout North Germany, and the northern half of France. In England it is used in the county of Northumberland, where it is known as the " burr." But in all places, wherever it is, it is recognised as a defect. The Comedie- Fran9aise proscribes it. No singer should ever employ it. The " beats of interruption " produced by r' are bad enough for sing- ing, but the hollow, rough, mucous rattle of 'r is still worse. There are, however, many varieties among those to whom it is familiar, and some young Parisians can produce a very metallic rattle with their uvula. The flated form 'rh occurs unrecognised, as a final in those parts of Germany where 'r is used, as hier, hee'rh, and it is difficult to distinguish from 'kh or even kh. But the singer's business is steadily to avoid both 'r and 'rh, and invariably to use the Italian r'. 'KH, flated rattle; v. ii'. 'GH, voiced trill, vi. ii'. These are simply kh, gh, with a vibrating uvula, and, often, a mucous rattle. They are used in Swiss German, and they may be heard, occasionally at least, for kh, gh, in Holland. The 'rh, 'r insensibly fade into 'kh, 'gh, and though a distinction is gener- ally insisted on, it is seldom made. The English singer has simply to avoid these sounds, not merely in German, but generally. 'RW, voiced trill, vi. 12. In the County of Northumber- land, not only is 'r used, but a labial letter following or preceding causes the lips to close somewhat, thus producing a labialised consonant 'rw'; this happens also at the end of words, as letter laet'o'rw' , which sounds like a bad let-au to a Londoner. Roth- bury 'Rw'oth'bo'rw'i, sounds most like a bad woth-boi to a Londoner. It is needless to say that the singer must carefully avoid this sound. Division, Straits. Subdivision, Lateral Tremulous — there is a central obstacle and the air escapes on both sides, or at least on one side of it, causing the edge which it passes to tremble, but far less than for /. Class vii. Flated Lateral, — flatus driven through the side passages. Class viii. Voiced Lateral — voice passed through the side passages. 124 > SPEECH IN SONG. [Sect. XIV Group II. (there is no Group I.) Tongue Point Laterals —I'h V, Ih L, Hh 'I, ;. LH, FLATED LATERAL, vii. 5. L, VOICED LATERAL, viii. 5. The typical consonant of this subdivision is the common Eng- lish L. The tongue is placed in the position of t, and then the sides of the tongue are released, while the point remains against the palate. Observe the difference of position between r' and I. Both are derived from t, but for r' the sides remain fixed while the point is released and allowed to vibrate, and for I, the point remains fixed, while the sides are released, and their margins allowed to tremble under the influence of passing breath. The apertures being comparatively large, there is less impediment to the passage of the voice than for any other consonant, so that I can be sung on (though it should not be, pp. 28c, 606) , and frequently forms the vocal part of a syllable, as in apples ap-lz. The vocal character of I has led to its universal adoption, when joined to the most musical vowel aa, in the syllable laa, as the sound on which an air should be sung without words. But o should be practised with all the other vowels, as lee, lai, laa, lau, loa, loo. The final glides il, el, al, ol, ul, uol also require diligent practice. The glides of mutes and sonants, and even hisses and buzzes on to / and from I are very numerous, as in plough plou, blow bloa, and Alfi help gulp, and great care must be taken not to give the semblance of the introduction of another vowel, as pu-^lou, bu-r-loa-, he+ulp which have a very unrefined effect. Lh is not used intentionally in English, but may be heard in place of leo final in French, when a voiceless consonant follows. Thus peuple poep-lh for poep-leo. The French sound is never poep-l, ending like pee-pl in English. L'H FLATED LATERAL, vii. 5. , L' VOICED LATERAL, viii. 5. These are derived from f as lh, I were derived from t. They need not be acquired, as foreigners are satisfied with our lh, I. 'LH FLATED UNILATERAL, vii. 6'. 'L VOICED UNILATERAL, viii. 6'. These differ from lh, I only in releasing the right side of the tongue from the t position, instead of both sides. The tongue is then in nearly the same position as for clicking to make a horse go on. The mouth is also frequently, but not necessarily closed on the left side. The flated form is the Welsh II, the voiced form does not occur in Welsh. The sound of 'lh is peculiarly breathy, and is reckoned as a guttural aspirate by the Welsh. It should be acquired, because Welsh names con- taining this flated lateral, frequently occur in English songs, as in Llangollen 'Lha'n-gao'lh-en, where Thlangoth-len should be avoided. 'Lh need not however be much practised. ,L, VOICED LATERAL, viii. 7. This is formed from Jt (p. 103c) Sect. XIV.] CONSONANTS AND THEIR GLIDES IN PARTICULAR. I25 in the same way as / from t. It occurs in southwestern dialects, but has to be learned only to be avoided. Group III. (Group IV. does not occur) — ly', LY', voiced lateral, viii. 8. This is derived from ty' (p. I04£) as / is from t. It is not an easy or pleasant buzz to sustain, and hence must be taken quietly and briefly. It may be con- sidered as an attempt to say / and y at the same time. If we say million mil-yen, prolonging the /, and gliding it on to thejy, the lateral ly' will probably be formed by the way, as mil- -ly' -yen; but naturally it does not occur in English. This sound used to exist in French as well as in Italian and in Spanish. It iS unknown in German. It is however quite lost in modern French pronunciation, a simple y being substituted, thus ' mouille, veille, vaillant, email ' which were formerly mooly'ai, vaely', vaaly'ahn', aimaaly' but are now simply mooyyai, vaey, vaay- yahn', aimaay. But there is no tendency to make such a change in Italian, where Zjv'is written ' gl ' and is always followed by an * i.' Hence it must be carefully practised as ly'ee, ly'ai, ly'ae, ly'aa, ly'ao, ly'oa, ly'oo, in : gli ly'ee, dateglielo daa-taily'ailoa, paglia paa-ly'aa, voglio vao'ly'oa. This lateral is also quite distinct in Spanish where it is written '11,' as: Have ly'aa've, lleno ly'e-nao, and in Portuguese, where it is written ' Ih.' Family NASALS — nasal passages open, the uvula advanced ; no passage of air through the lips. Class ix. Flated Snorts or Snuffles — flatus forced through the nose, but resounding also in mouth, generally scarcely audible. Class X. Voiced Hums — voice resounding in portion of mouth and in the nasal passages at the same time. Group I. Lip Nasals — mh M. MH, flated snort, ix. i. M, voiced hum, x. i. The mouth is disposed as for p, and the uvula is advanced so as to allow flatus and voice to pass through the nose. For mh there is usually a scarcely audible sound, like that in quiet breathing with the mouth shut, when the tongue is however generally raised to the ng position, and the action is therefore rather ngh. In attempting to clear the nose, or in blowing the nose, when the outer nostrils are compressed, there is a decided snort. . Snoring is orinasal, requiring a vibration of the uvula, and cannot be heard if the mouth is closed — a cure of value, but difficult to apply. The hum m is Wfell known, and its musical effect is enhanced when the tongue is kept quite low in the au position, so that the whole cavity of the mouth is occupied. In running up the scale it will be found, however, that the tongue rises for the uppei, 126 SPEECH IN SONG. ISect.XIV. notes. M, however, should not be sung upon, and in chasm kaz-m it should be sung u'm (p. 6i) and the same sound should be sung for those obscure weak syllables written -em in Glossic as bosom buoz'em, sung buoz-u'm. Glides from and to M, — In the glides great care should be taken not to carry on the nasality from m to the vowel, or nasalise the vowel by anticipation. Sing maa not nifla, and aam not flam where prefixed ( , ) indicates that there is a slight orinasality (p. 37^). Practise with all the vowels. Be careful also to avoid mpaa. MP. — When m is final it is apt to be prolonged. It is short- ened by cutting off nasality and allowing a sonant oral resonance as lamb, but the b is not now recognised. It is made very brief by cutting off the nasal passage and assuming the p position as in lamp, and, when there is no recoil as in lamp°h, and no subse- quent hiss, as in lamps, or vowel, as in stamping, the effect of p is solely to shorten the m.. In lamplighter lamp •... lei- ter the p acts in this way only, and does not glide on to the I, there is therefore a complete break or silence between m and I. Com- pare a lamb lifted u lam lif-ted and a lamp lifted u lamp lif-ted, and observe that we should never say u lamp°h lif-ted, though we generally say hee lif-ted u lamp°h. Before a subsequent t, shortening by assuming the p position is regular in English as attempting u-^ temp... ting. Group II. Tongue Point Nasals — n"h n", nh N, ,m. NH, FLATED snort, ix. 6. N VOICED HUM, X. 6. The n is the usual hum, with the tongue in the t position. The mouth may be open or shut, it does not affect the result. The oral resonance being much less, the musical effect is worse than for m, and the tendency to nasalise vowels which glide on and off it, greater. A following t shortens the nasal, as in' tent-maker tent-... marker , where no recoil would be heard, as in hee hiai-d u tent°h. For the general nature of the glides see mj p. 126a. Although nh is barely audible, it is more so than mh, and it is still used occasionally in Cumberland for ' kn ' in know nhao. This use of nh was regularly recommended by orthoepists; from one to two hundred years ago. N"H, FLATED SNORT, ix. 5. N" VOICED HUM X. 5. These differ from «/f, n solely by using the t' position in place of the t position. The difference, which is scarcely audible, may be entirely neglected. ,N VOICED HUM, X. 7. Here the tongue is in the ^t position. The difference between this ,« and the usual n, though very per- ceptible in humming, may be neglected. It is only in the south of England that it is likely to occur. S«CT.XIV.] CbMSONANTS AND THEIR GLIDES IN PARTICULAR. 127 Group III. Tongue Front Nasals — ny'. NY', VOICED HUM, X. 8. The tongue is in the position for ty' in place of t. The nasal resonance is not so good as for n, and when there is no glide from ny', it is chiefly by this deficient resonance that ny' is distinguished from n, as French peigne paeny', Boulogne Boolaony', Cologne Kaolaony'. The last two words are constantly used by Englishmen and very seldom pronounced aright. . Beware of Buolon-, Buoloa-n, Buolong\ Buolovn, and Ku-lon-, Ku-loa-n, Ku-long-, Ku-lovn. When a vowel follows ny', the attempt to pronounce n and y at the same time produces at least an intelligible effect, and the real sound may be generated in onion un-yen, by prolonging the n, in gliding on toy a.s-un--\-ny'-\-yen compare 'million' p. 125a. This sound occurs initial in (colloquial) French and In (literary) Spanish, and final in (literarj') French. It occurs between vowels only, and never at the beginning of a word in Italian. It is un- known in German and English proper. Italian signorstf««y'oa-r', campagna kaampaa-ny'aa; Spanish iiaque ny'aa-kai, iioiii? ny'ao'ny'ao ; French peignant paeny'ahn'. The sound should be very carefully mastered. Group IV. Tongue Back Nasals — ngh, NG. NGH, elated snort, ix. ii. NG voiced hum x. ii. The ng is formed with the tongue in the position of k, arid hence without any oral resonance proper, although that of the pear shaped upper pharynx is of course appreciable. The resonance is therefore the worst possible for singers, and should never be dwelled on. The snort here is the most natural one for blowing' the nose, as, the passage into the mouth being completely blocked, more force can be exerted through the nose. The sound is common in English and German, even when final, .and in Italian before k, g. It is said not to occur in Spanish even before k, g. It is quite unknown in French, where it is replaced by the orinasalisation of vowels. Glides to and from NG. — Ng never occurs at th*^ beginning; of words or syllables in English, German, or Italian, but is frequently medial in German, and occasionally so in English, as sing-er rather si+ng+er than si-{-ng...er, German Sanger zaeng-er. But the common practice in English (not followed^ in German) is to introduce g as long longger longgest, but to long, a longer too long, u long-er. In some words usage hesitates, as England Ing-glend, Ing-lend, English Ing-lish, Ing-glish, German England Aeng'laant, Italian Inghilterra Eeng'geeltaer'-r'aa. In ItaHan ' nk, ng' are always +ngk + , 4-ngg-\-' as banco baang-koa, lungo loong-goa. The resonance 'of ng is shortened by k, as thank thangk, not thank, for this would imply the preservation of n before k in 128 SPEECH IN SONG. [Sect. XIV . XV the same syllable, which is difficult for English, Germans or Italians. We preserve it however by a break as income in\..kum, ingoing in-...goa'ing. At the end of words many Germans say ngk, whenever there was not an omitted 'e,' as gesang gu-zaangk' ; but not otherwise as lang' laang for ' lange.' Many however do not make the distinction, and singers are recommended to use the simple ng. In English the only vowels which occur before ng are short i, a, 0, u as sing, sang, song sung. In German short ee, ae, ao, uo are common. But no long vowels occur before ng in either language, although only long vowels occur before ng in Icelandic. XV.— WORDS, ACCENT AND EMPHASIS. Principal Syllable of a Word. — One syllable or a collection of several syllables maj' have a separate significance given to it or them, and then form a word. But the habit of collecting syllables into words is modern and artificial. In real speech we gather syllables into groups which do not by any means always agree with that division to which we are accustomed in dictionaries. These groups are more or less determined by some principal syllable which forms the nucleus round which the others are arranged. This pre-eminence of one syllable over the rest, is effected in several ways — principally by " length," by " pitch, "- and by " force," — and different nations differ considerably in their habits in this respect. Syllabic Length or Quantity. — This is determined by length of vowel, and length of consonant, and is fixed in each English and German word, or nearly so, as it was in ancient Latin and Greek, but is very variable in Italian, Spanish and French. In English and German also very numerous consonants can come together without producing the feeling of length. In Italian and French two consonants at the end of a syllable, or between two syllables decidedly lengthen the first. But ques- tions of length are decided for the singer by the composer of the music. Long vowels must be sung to short notes and short vowels to long notes, and as the consonants and glides have to be made as brief as is consistent with intelligibility, singers must diligently practise this extension and diminution of the natural length of their vowels, and learn to make a distinction between naturally long and naturally short vowels in closed syllables, as pee-t pit, pai-t pet, by the nature of the glide, recollecting that for long vowels the glide is weak and sluggish, and for short vowels it is strong and smart. Syllabic Pitch or Tone. — The musical pitch of the vowel, whether it was high, or low, or descended from high to low while the vowel was held, was the favourite mark of distinction Sect. XV.] WORDS, ACCENT AND EMPHASIS. I2g in older nations, who founded their rhythm on length only, independent of pitch, and grouped their syllables round that of the highest in pitch. This habit, which still clings (with varia- tions) to Norway and Sweden, is lost to the rest of Europe, in the sense of being fixed to particular syllables in particular words. But everywhere there is a " song of speech," often profanely termed a " sing-song," which varies extremely from locality to locality, and may be studied in the educated pronunciation of Edinburgh, London and Paris, and the uneducated pronunciation everywhere. But this is of no consequence whatever to the singer to whom a series of definite pitches has been assigned by the composer, which most frequently do not agree with the natural song of speech of any country ; indeed we know that Greek composers paid no attention to it 2,000 years ago. Syllabic Force or Stress — Strong and Weak Syllables. — The third mode of distinction is by variation of " force." This is the one now prevalent in English, German, Italian and Spanish, but not in French. In the former languages syllables are strong and weak in words of more than one syllable by a fixed rule which serves to distinguish words and shew what syllables belong to them. Hence we divide syllables into "strong" and "weak," according as they have much or little force. In English and German the weak syllables are usually spoken " obscurely," not so in Italian and Spanish, and not so also in French, except as regards final ' e,' which is generally lost in speech, but is retained in singing. In French however the distribution of strength and weakness is not uniform, but seems to depend on grammatical construction, and the differences of strength are not so marked as in the other languages. In speaking English the weak syllables, and their peculiar obscurations are of great importance, but in singing the only thing to be considered is how to indicate that obscuration. It may be said generally that the weak terminations with any written vowel before r, I, m, n, obscure that vowel to an u or u' which does not glide tightly on to the following conso- nant, and is written e in Glossic, as already explained (pp. 61, 63). When the weak vowel is an open ' a,' that is, one in which the sound does not glide on to a following consonant, it is regularly u in speech, but in singing becomes a' (pp. 48^, 59^). When the weak vowel is 'i, y,' or an 'e' in an open syllable, it is generally i as pity fit'i, possibility pos-i-^bil'i-r-ti but may be i' as pit-i'. When written ' o, ow, ough ' it is usually oa, as window win-doa, not witrdu or winder (the last of which would imply permission to add an r' , which could not be granted), promote pr'oamoa-t, but promontory pr'om-entu-^r'i or pr'om-7intur'-i. Similar rules apply to weak words, as they are esteemed, although they are usually grouped as mere adjuncts to other words, as: to go to bed tu-i-goa' tu-^bed. The native English singer will I 130 SPEECH IN SONG. [Sect. XV know what syllables are weak, and sing them accordingly. The weak syllables are often sung to notes of considerable length, and strong syllables are sung to short notes. Thus " scatter her enemies, and make them fall " would be spoken skat-er herr' en-imlz n mai-kdhem /awl, but are sung Ex. 37. tr skat--ur' herr' 5^ en-T-ni-r-nnz ^ :3z and maik dhem \faul where the chief force of a bar falls on ' and ' and hence its sound is restored to and. In glossic, force and length are rnarked by the same sign in English and German. The turned period (•) placed immediately after a vowel indicates that it is long and strong, and after a consonant it indicates that the preceding vowel is short and strong. In the case of the indistinct weak el, em, en, when it is intended to shew that they have a distinct sound, a (•) is annexed as novel- in contradistinction to the usual novel, jewel jeu-eV, as distinguished from either jewel (which is vulgar) or jewil which is common. In Italian and Spanish the mark of force (') is placed after a vowel to indicate force without length, in cases where an Englishman would natur- ally add length, and after a consonant when the syllable sepa- rates after it, so that the vowel is also strong. But the length of vowels is not fixed in Italian and Spanish ; it is generally of middle length, shorter than for our long and longer than for our short vowels, but may be Ifingthened or shortened even in speech almost at pleasure. It is only force which is fixed, and this deter- mines the rhythm of verse. In French no mark of force is put, because it is never fixed in any word, and even among French- men there are great differences of opinion. Many still believe that they always put most force on the last syllable of a word (not being 'e'), as would have been natural when the language was growing out of Latin. But this was disputed by Frenchmen 300 years ago. The consequence is that while music (especially of simple airs) generally gives the strong notes of a bar to the strong syllables, in English, German, Italian and Spanish, no rule whatever but the composer's fancy, seems to be followed in French, and the final ' e ' which is lost altogether or hurried over obscurely in speech, rises to all the dignity of a long and strong vowel in verse, thus the spoken words aalaajleor duehelahzh become Ex. 38. ^ ^ HfeROLD. T fc iP~?t ^' -^-P-^ a aa la fleur du bel laaf fleor dueb bael a - ge ah -\- zheo Sect. XV.] WORDS, ACCENT AND EMPHASIS. I3I Accent. — The principal syllable pointed out by " the higher pitch " was said to be " accented," being that to which the song fell (ad cantum). When the feeling for pitch was superseded by that for force, at the time that the feeling for length was lost (close of the third century, a.d.), the " strong " syllables were said to be "accented," and alternations of strong and weak syllables were held to represent the former rhythmical alternations of long and short syllables, so that the names of such groups were transferred, as " iambus " which meant a short syllable followed by a long one, was held to mean a weak syllable followed by a strong one. The Latin names for poetical feet were thus transferred to all lan- guages but French, which versified by the number of syllables (including unspoken final ' e ') and by the division of words. In this Primer " accent " means a " fixed " custom in placing stress on a certain syllable of a word. Where that strong syllable lies must be learned from a dictionary, as no rule can be given, but for singing this is of no importance as the music points out its position. In French there is no accent at all in this sense, and the various stress implied by the music is entirely an arti- ficial creation of the composer. Emphasis. — But words are grouped into clauses, and in each clause there is one word more conspicuous than the rest, which is termed " emphatic." The unemphatic words (as usually divided) are often merely weak syllables, and are then grouped with the neighbouring words. But what has occasioned their separation in writing is the fact that they are sometimes em- phatic, and then are not so grouped. This emphasis is written in Glossic by the same turned period placed before the whole word, and corresponds to placing a word in Italics, but is used mora frequently, as : he never said so hee -never sed soa. It also distinguishes different senses of words which are distinguished by force in speech, as : John gave his two cakes to two boys, and Tom gave his two, too, to two, too, yon gai-v 'hiz 'too kai'ks too •too boiz, n Tom gai'v 'hiz 'too, 'too, too -too, -too. The above are all the principal phenomena of speech, syste- matically explained with great brevity, but at sufficient length, it is hoped, to be useful to singers of the five languages con- templated. Annexed are lists of the elementary sounds of those languages with a reference to the pages where they are explained, Bsrving as an index to the details of the book 132 SPEECH IN SONG. [Sect XVI. XVI.— INDEX OF THE ELEMENTARY SOUNDS IN (RECEIVED) ENGLISH, GERMAN, ITALIAN, SPANISH, AND FRENCH. *^* The figures refer to the page, and the letters a, 6, t, and d to the first, second, third and fourth quarter of the page. I.— ENGLISH. Glottids. I4(J. f eel pe-l. i8c. (Not written.) Aspirate. 22c. h =: hj or hf, hay hai. 22c. 2^d. Strong and Long Vowels. iz8c. 129*. ee. beet beat bee't. 40c. at (without aftersound of ee, see ai). bait bate bai't. 44a. a', ask a'sk. /^yd. 4g». aa. baa Jaa. 49a. au. maul mawl. 516. cy. fur fiT fer. 63,c. oa (without aftersound of 00, see od). coal Aofl'i. SS'^- 00. cool koo'l. 55(Z. Strong and Short Vowels in Closed Syllables. 128c. i2g&. i. knit nit\ 41c. e. net nei'. 446. ». gnat ««*•. 47J 0. knot not'. 52a. «o. iootfuot. 55c i^. nut »!«i. 62c. u\ 59c. Sung in /, m, n, and spoken for indistinct e in e/, «m, en. 6ob,c. 61. Weak AND Short Vowels in Open Syllables. 128c. i2gt. ft', trochee troa'kee. 420. a'. China Cheitia'. 48^. i. witty wJJ-j. 41C. i'. witty wit'V. 41C. «j. railway ravlwai. 44a. fl«. august atigus't. 516. « and «»'. China Cheinu, finer feiner. 48^. 59^. 646,c. 00. influence in-flooens. 55^. Recoil. j!'.°/j, r/i, A°A, flated and /i", i?, k° clicked. 88c. Diphthongs. 67. aay. aye H'JJ'. 67. ei = uy or ay. height Mi. 68c. 04 = oj/. foil foil. 6gd. ou ■=. uw ox a'w. {owl foul, yja. eu = 100 feud fewd. 72c. eey = i-iir'. hear Aee'y. 75(f. air =: <•«/. there dhai'r. 76a. oay = «07(r'. oar oa'r. . 766. oor = twur\ poor poo'r. ybc. aar = aa'ur' or flfl'ii or aa or aa*/ at pleasure, partyjaa'rt. ybd. aur^au'Hr' or atrii or an OT airr' a.t pleasure, short shatrrt. yya'. ai =: aiy, that is, ai with vanish or aftersound of £d. pay pai. 44c. 78«. od r= ofl'ie', that is, Off with vanish or aftersound of 00. know nod. 53c. 786. Triphthongs. 79^^. eir. fire fei-r. yyc. our. sour sou'r. yyc. Consonants. gSrf. p. pea. pee', ggc. b. bee bee', ggc. t. toe ioa'. loic. d. doe doa'. loic. iy' in ch. cheese chee'zzzty'sh'ee'z, 104c. 117c. dy' in > jest fest = dy'zh'est, 104c. 117c. k. cave kai'v. 106a. g. gave gai'v.^ loba. ky'. cart ky'aart, antiquated. 1056. /r>''- girl^-j/'cW. 1056. Sect. XVI.l INDEX OF ELEMENTARY SOUNDS. 133 kw\ queen kw'ecn, written kwee-n. 1070. gw'. guano gw'aa'noa, written gwaa-noa. loya. wli. wheel wheel. 107c. w. weal wee'l. 107c. /. feel/se-/. 108c. V. veal vee-l. 108c. th. thin tJiin. iioe. in. dh. ihen dhen. iioc. nr. s. seal ses'l. 112c. z. zeal zee-l. 112c. sh. mesh mesh, 114c. zh. mea.sure mezh'er. 114c. ,,»". occasional defective ' r,' merry me^yi. 115c. sV in ch::=ty'sh''. cheese=:chee'z = ty'sh'ee'z. 117c. zA' in j = dy'zh'. jes-t:=.dy'zh'esi. 117c. jz/j. hew j//ie«. Ii7(f. J/, you yoo:=eu. iijd. r. vowel u diphthongising with previous vowel, and fol- lowed at pleasure by r'. See above under the diphthongs eer, air, oar, oor, aar, aur. r\ trilled ' r.' rut r'ut. izic. I. lamb lam. 124a. apple ap-l. 28c. 606. m. me mee. 135c. chasm kaz-m. 28c. 606. n. knee »««. 126c. open oapjt. 28c, 606. ng. sing sitig. 28c. 60b. 127c. II.— GERMAN. Glottids. i4if. ( ; ). erinnern :er';itfer^ii. igc. Aspirate. 22c. h. hand hdant. 22c. Long and Strong Vowels. 1281;. 129c. ee. liebe lee'bu. 40c. ai (without aftersound of ee). gegen gargy'hen. 44a. ae. spTSichesh'pr'ae'ky'hu, usually sh'pr'ai-ky'hu. 446. aa. wahr v'aa'r\ 49a. oa (without aftersound of 00). schoos shoas. 53c. 00. schuh shoo. 551^. ue or ui. giite gue-tu. 57b,c. eo. Goethe Geo-tu. 57^. Short AND Strong Vowels. i2Sc. 1296. ee or i. nicht neeky'ht. 426. a« or e. netz K«is. 446. aa. kalt AaaW. 49a. ao. holtz haolts. 52c. N. Ger- man /lorts. 526. Ko. hund hiwnt. 556. ««. fiille /««/■«. 57c. oe. kopfe koep'fu. 58a> Short and Weak Vowel. 128c. 129&. u. rache r'aa'khu, also pro- nounced r'aa'khai. ^gd. u'. gabel gaa-bu'l, lieben lee-bu'n, pronounced gaa-bel lee'ben with English indistinct, el, en. sgrf. Diphthongs. 67. aay. eile aaylu, hain haayn. 67. oy. leute laute Zoyto. .70a. 71a. 3ai£'. haus haaws. yid. Consonants. 98^. pjorp. pack pjaak, or paak. ggc. 86d. b. band baant. ggc. Often merely a faint />. 893. t'j or t. tadel t'jaa'd'u'l, 01 taa-del. lozd. d' ord. darstd'twr'st, or duor'sf 102d. ty' in c7j = ty'sh'. Deutsch DoicK IIJC. k-h of k. komm k-haom or kaom 106a. 86d. g. gonnen goeiven. 106a, /'. pfau pf'aaw or pfaaw. logd. v'. wie v'ee. nob. 134 SPEECH IN SONG. [Sect. XVI. /. ieind faaynt. io8c. (,there is no V. logi.) s. reissen raaysen, fusz foo's, II2C. zn tsoo = t's'oo. ii6c. z. sie zee. II2C. s/j'. stab s/t'tea*/. 1176. Also in ch^ty'sh', punzsch puonch. 117c. s/t. schiessen shee'sen. 115c. >>. jaj/aa. 117^. ky'h. ich eeky'h, euch oyky'h. 118b. gy'h. io\ge faol-gy'hu. 1186. M. sache zaakh'U. iigc. gh, tage taa-ghu. iigc. r'. reich r'aayky'h. 121c. 'r is also common. 123a. /. lamm lamm. 124a. Ischl Eesh-l. zSc. 60b. m. Marz Maer'ts. 125c. ». nun noo'n. 126c. ng. finger feeng'ur'. izyc. III.— ITALIAN. Glottid. I4(?. ?. anno taan'noa. iSc. (Not written.) 128c. Vowels, medial length. ee. spiri sperr'ee. 40c. ai (without aftersound of ee). fedefai'dai. 44a. ae. straniere str'aaneeae'rai. 446. aa. (aXo faa-toa. 49a. (with a tendency towards «'. 48c.) ao. poco pao'koa. 52c. o« (without aftersound of 00). amore aamowr'ai. 54c. 00. scudo skoo'doa. ^^d. Diphthongs, mostly slurred. 26a. 68«. aa^. vai vaay. 6gc. aaw. causa kaawzaa. yjd. aew. neutro nae'wtr'oa. "jid. yae. jeri yacr'ee. 73c. eu. piu peu. "j^b. wao. uomo wao'moa. 74^* Triphthongs, ."jgd. yaey. miei myaey. waoy. vuoi vwaoy. Consonants. gSrf. p. parte paar''toa, b. bardo baar''doa, t' or t. ta.ntotaan'toa. 80a. 99c. 99c. loic.i02a. d' or d. detto dait-toa. loic. i02d. k. caro kaa-r'oa. io6a. g. augusto aawgoos'toa. io6«. /. iasti faas'tee. 108c. V. vasti vaas'tee. io8c. s. sano saa'noa. ii2c. pezzo paet'tsoa = paef't's'oa. i i6c. vizio vee-t's'yoa, sometimes vee'fhyoa. ii6c. 2. smorto zmaor''toa, rosa r'ao-zaa. 112c. mezzo maed'dzoa = maed''d'z'oa. ii6c. sh' in cA. facce faat'chai ^ faaty''ty'sh'ai ; sometimes s^' is used alone as face faa'sh'ai. 117c. zh' in y. giudice joo'deechee =: dy'zh'oo'deety'sh'ee. 117c. s/f. sciolto shaol'toa. 114c. »■'. terra taer'-r'aa. 121c. 122c. /. la /ad. 124^. ly'. gli /[yVs. 125a. m. mano maa-noa. T2^d. n. no «flo, non »oa«. 126c. ny'. ognano oany'oo'noa. i2ya. ng. lungo laong-goa. 127c. IV.— SPANISH. Glottid. i4(i. t ha.y )aay. i8c. (Not written.) Aspirate (none). Strong or Weak vowels, length medial. 12SC. ee. y ee, aqui aakee', dificil deefee'theel. 40c. Sect. XVI.] INDEX OF ELEMENTARY SOUNDS. 135 *. e e, sobre sao'bre. 446. (there is no at.) aa. la laa, hacer aaths-r. 49a. ao. no ftao, castellano kaaste- ly'aa-nao. 52c. (there is no oa.) 00. su soo, minuscula meenoos- koolaa. 55d. (There are only five vowels.) Diphthongs, rather slurred. 26a. 68a. aay. dabais daa'baays, bayle baay-le. 67. ey. veisveys,peynepeyne. 6ga. aoy. sois saoys, voy vaoy. "job. aaw. ^axisa. paaw sua. yid. ew. deuda dhewdhaa. jid. yaa. gracia graa-thyaa, 73c. ye. cielo thye'lao. 73c. yao. precio pre-thyao. 73c. yoo or eu. ciudad theudhaa'dh.'jic. ti/aa. fragua fraa-gwaa, Juan Khwaan. 74^. we. dueno dwe-ny'ao. 746. wee. ruido r'wee'dhao, buytre bwee'tre. 746. wao. arduo aafdwao. y^d. Spaniards also reckon the following slurred vowels as diphthongs. eaa. linea lee'tieaa, Boreas Bao'reaas. eao. virgineo veerkhee'neao. 6e. h6roe e-ro-e. Triphthongs, also slurred, ygd. yaay. precinis pre'thyaays. 80a. yey. vacieis v'aa-thyeys. 80a. waay. santiguaissaantefgwaays. 80a. wey. averigueis aaveree'gweys, hue}' bwey. 80a. Consonants. 98. p. puede pwe'dhe. ggc. b. The Spaniards endeavour to separate a b from v', as baron baarao'n (a baron), varon v'aarao'n (a man of worth), but do not in general succeed, " la B se confunde por lo comun en castellano con la V conso- nante en quanto a su pro- nunciacion." (B is com- monly confused in pronun- ciation with V in Spanish). Ortografia de la lengua Castellana compuesta por la Real Academia Espanola, Madrid, 1792. 99c. t' or t. todo tao'dhao. lozd. d' or d, seems not to exist, it either falls into dh, or is elided, as in the termination -ado, as pecado pekaa'ao. lozd. k. como kao'mao. xo&a. g. gana gaa'naa, guerra ge'r'aa (written 'gu' before 'e, i, y'). io6a. v'. voluntad v'aoloontaa'dh, " su sonido el mismo que el de la B " (its sound is the same asthatofB).Ort.Cast. 1106. /. fuerte fwe'rte, (there is no v.) io8c. t'h or th. censo the'nsao, ciento thye'niao, zumo thoo-mao. 112a. I 166. d'h or dh. toda. tao-dhaa. 112a. 1 166. s. su soo, los loos (there is no z). I12C. sh' in ck, chapin chaapee'n := ty'sh'aapee'n, (zh' in j is not found). 117c. kh. justo khoo'stao, gefe khe'fe, Quixote orQuijote Keekhao-te (some consider it to be gh). ijgd. ..v (or r, never r'). " suave para la qual nunca se duplica la R " (soft, for which R is never doubled). Ort. Cast. p. 70, as arado aaraa'oo, breve v're-v'e, amar aamaa-r. 115c. 1160. f'. " R fuerte," used only at beginning of syllables, or when written 'rr,' razOn r'aathao'n, malrotar maal- r'aotaa-r, barra baa'r'aa. 121C. /. leve le-ve. 124a, 136 SPEECH IN SONG. [Sect. XVI, ly\ Have ly''aa've. misma mee'smaa. tiene tye-ne. 126c . nifiez neeny'e-th. nunca noo'ngkaa. 125a. 123^. m. n. ny'. nifiez neeny'e-th. 127a. ng. nunca noo'ngkaa. 127c. Note. — The only letters doubled in Spanish spelling are ' a,e,i, o,c,n,' as in Saavedra Saa-aav'e'dhraa, preemi- nencia pre-emeene-nthyaa, piisimo pee-ee-seemao, loor lao-ao-r, accidente aaktheede'nte (in which cc represents two different sounds) ennoblecer en-naov'lethe'T [in which there isatrue doubling p. g^d). In '11, rr,' the duplication implies a single new sound, as ly', r', quite different from U, rW. v.— FRENCH. Glottids. T^d. }. auneoa'w. (Not written). 18c. Aspirate (none, at most a break, as : le onze leo...oan'z. 25a.) Oral Vowels. Strong and either long or short, the tendency is to make them short, except for the purposes of expression. 128c. ee. titre teetreo, partie paar'tee, synonyme seendoneem- 40c. _ ai (without aftersound of ee). et6 ditiii, esprit aispree. 44a. ae. proems pr'aosde, complftte koan'pldet. 446. aa. {sitfaat, femmefaam. ^ga. (now usually replaced by a' p. 48c in Paris). ah. gras gr'dh, pas pah, casser kdhsai, always long, gener- ally used when 'a' is cir- cumflexed or comes before ' s.' 49c. ao. raotii mdoteef, hotit dot. 52c. oa (without aftersound of 00). mots maux moa, beau hoa, always long. 53c. 00. fou foo, toute toot, voiate voot. 55^. ue. muse muez, hutte iiet. 57c. eo, je zheo (or zhoe), deux deo peu i>eo (some do not dis- tinguish «o from oe). 57iJ.58». oe. peur poer, neuf noef, peuple poepleo. 58a. Very Short Vowel, (a substitute for the recoil). eo. possible pdoseebleo, peuple poepleo. S8d. Orinasal Vowels. G5. aen'. pin pain paen'. 66b. ahn'. dans dahn'. 66c. oan'. non noan\ 66c. oen'. brun br'oen'. 66c. Diphthongs, rather slurred. 26a. aay. 6mail aimaay. 6gc. aey. reveil r'aivaey. 69c. ooy. oui ooy. yob. wee. y^d. uey oTwy'ee. lui luey Iwy'ee. yoc. oey. oeiloey,Tecu.eilreokoey. yod. wy'aen'. Juin Zhwy'aen', quin- quag6sime kwy' aen kwaa- zhaizeem. yid. waa, oie waa. 74c. Consonants. g8d. p. pas pa,h, nappe naap. ggc. b. biton bdktoan\ ggc. t' or t. th6 tai, un grand homme oen' gr'ahn't aom. loic. lozd. d' or d. donner daonai, raide raed. loic. i02e. k. carte kaar'i. 106a. g. gar9on gaar'soan'. 106a. f. carafe kaar'auf. io8c. V. vie vee, veuve voev. loSc. s. son saon'. fa9on fdasoaft', ambition ahn'beesyoan'. 112c. z. zh\e zdel, rose rdoz. 112c. sh. cheval sheovdal, chercher shaer'shai. 114c. (The French represent the sound of ch in foreign words by ' tch,' meaning tsh.) zh. jardin zhaar'daen', dge ahzh, 114c. (The French repre- sent the sound ofj in foreign words by ' dj,' meaning dzh.) Sei.1. CVI.XVIIJ INDEX TO MUSICAL EXAMPLES. 137 »■' or aor'. art aar\ inc. (The greater number of French- men in the north " gras- seyent " g'raasaey, or use the uvular 'r in place of the trilled r', p. 1236.) /. le leo, €lhvt ailaev. (^44. ly'. is heard only from old people, as M. Littr6, and is replaced by ^. 125a. m. mfeme maem. 125^. «. ennemi aenmee, bonne baon 126c. ny\ agneau aany'oa. 127a. ng (is not found). XVII.— INDEX TO MUSICAL EXAMPLES AND EXERCISES. 1. Harmonics. 7. 2. Compasses and Registers. 16. 3. Effect of 2 final. hiz foa-z umavzing, &c. 18. 4. Voiced or Singer's Aspirate. 24. 5. Ee aa 00, I m ng r\ z v, b g sung to ' See the conquering hero comes.' 27. 6. Oo, oa, aa, at, ee resonances, (Koenig.) 31. 7. Pitch Favourable to Vowels, (Seller). 32. 8. Pitch Favourable to Vowels, (Helmholtz). 33. g. Effect of Resonators on Vowels. 33- 10 and II. Helmholtz's Artificial Vowels. 34. 12 and 13. Effect of Pitch on ee, i. 43- 14 and 15. Effect of Pitch on ai, e, ae. 46 and 47. 16, 17! and 18. Effect of Pitch on aa. 50. 19 and 20. Effect of Pitch on ao,oa. 55- 21. Effect of Pitch on 00. 56. 22. Comparative Effect of Pitch on 00, ao, aa. 56. 23. Effect of Pitch on eo, oe. 58. 24. Effect of H. Philips's oap-n see. 6a. 25. Exercise on vernu'-l, simbu'-l, &c. 61. 26. Generation of ej. 67. 27. Generation of OM. 71. 28. Seiler'sConsonant Resonances. 85- 29. Seller's Vowel and Consonant Resonances compared. 86. 30. Exercise on Consonant Glides, as ai'tth. 92. 31. Plee-nz, plarnz, plaa-nz. Sec, sung to ' God save the Queen.' 93- 32. ' Ces ayahs-li sont pa'iennes.' Sung to "Twas in that garden beautiful.' (Balfe.) 94. 33. 'An unowned unknown pu- nisher.' (Same air as Ex. 32) 95- 34. ' Happy man.' Sung to march in overture to William Tell. (Rossini.) 95. 35. ' Dites, ma jeune belle.' (Gounod.) 98. 36. ' Those seas, those seas, those boisterous seas,' sung to 'The sea! the sea!' (Neukomm, 114. 37. Skat-ur' herr' e7t-i-ni-T-miz. ('God save the Queen ') 130. 38. 'A'lafleurdubelige.' (Herold). 130. CONTENTS. I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII. XIII. XIV. XV. XVI. XVII. Introduction .... Short Key to Glossic The Nature of Musical Sound Speech and Song Production of Speech Sounds Glottids Physems Voice, Flatus, Silence, Glide, Slur, Break Table of Transition from Voice to Silence Table of Singableness of Speech Sounds. Ex. 5 Vowels in General . Vowels systematically arranged The Speaker's Vowel Trigram The Singer's Vowel Trigram , Vowels in Particular Vowel Glides and Slurs, Diphthongs and Triphthong Consonants and their Glides in General , Systematic arrangement of Consonants Syllables Consonants and their Glides in Particular Words, Accent and Emphasis .... Index of the Elementary Sounds in (Received^ English, German, Italian, Spanish and French Index to Musical Examples and Exercises 3 4 5 9 II 14 19 25 26 27 30 38 39 40 40 67 80 84 93 98 123 132 137 NOVELLO, EWER AND CO.'S MUSIC PRIMERS. Edited by Dr. STAINER. A DICTIONARY- OF MUSICAL TERMS, BY DR. STAINER & W. A. BARRETT. (Compressed from the Imperial 8vo Edition by K. M. ROSS.) One SHILUN3; Paper Boards, Is. 6d. OPINIONS OF THE PRESS. " A careful and judicious abridgment of the larger work, and will be found useful to those by whom the original is inaccessible." — Athenaum. "A most serviceable abridgment of Stainer and Barrett's well-known Dic- tionary. Few, if any, words that the student can wish explained are absent from this little volume, which should be in every amateur's library." — Daily Telegraph. " Of even more general utility is ' Musical Terms.' As a dictionary of re- ference this little volume should be on the table or shelf of every musician." — Graphic. " Most of the special information contained in the large volume has been re- produced in this sWUing pamphlet, the compression having been judiciously made by Mr. K. M. Ross." — Illuitrated London News. " In its condensed form the work contains a mass of information which is rarely to be met with in books of the class. It is particularly copious in German terms, wliich are now becoming of nearly as much use as the Italian." — British Mail, " The musical learner will find much useful information on the principles ol music and on many technicalities of the art only to be found in high-priced productions." — Perthshire A dvertiser. "No individual with any pretensions to a musical taste should be without this excellent book of reference." — Perth Citixen. " An important addition is made to Messrs. Novello's series of ' Music Primert' in the shape of an abridgment, or rather compression, of Messrs. Stainer and Barrett's well-known 'Dictionary of Musical Terms.' That work has ranked as a standard one ever since its publication, and Mr. K. M. Ross, who has performed the task of reducing it to the slim dimensions of its ' primer ' form, has executed the business with evident care and intelligence. For its dimensions, this little book is unquestionably the best dictionary of musical terms in our language." — The Scotsman. " The issue of this abridgment of Messrs. Stainer and Barrett's dictionary was a decidedly ' happy thought.' Many amateurs cannot well afford the price of the complete work, while many others find its varied contents, if not super- fluous in themselves, an embarras des richesses as part of a book of reference for words in common use. The present issue exactly meets the wants ol the many. Its pages are crowded with succinct definitions of musical termmology, hardly a word being omitted No other work of the kind conveys so much accurate information within the same compass. The little book is the buit shilling's worth in all musical literature." — The Musical Times. LONDON & NEW YORK: NOVELLO. EWEK & CO. Crown Octavo, cloth, 2s. 6d. THE MUSIC OF THE BIBLE WITH AN ACCOUNT OF THE DEVELOPMENT OF MODERN MUSICAt, INSTRUMENTS FROM ANCIENT TYPES By JOHN STAINER, M.A,, Mus. Doc. " Th9 veiy excellent illustrations scattered throughout the book materially increase itj value ; and we sincerely hope that Dr. Staiaer's labours may be rewarded as they deserve, for not only has he accumulated a large mass of reliable information upon a subject of the deepest interest, but he has thrown additional light upon much of this evidence by giving us the result of his own practical experience." — Musical Times. ~~ " THE Life & Labours of Vincent Novello BY HIS DAUGHTER MARY COWDEN CLARKE. " It is my father's music To speak ' his ' deeds." — Shakespeare. Demy Octavo, cloth, gilt edges, price 3s. 6d. ; with Portrait, from a painting by his Son, E. P. Novello, engraved by W. HuMPHRYS. Crown Octavo, price is. LETTERS FROM BAYREUTH DESCRIPTIVE AND CRITICAL OF WAGNER'S DER RING DES NIBELUNGEN WITH AN APPENDIX By JOSEPH BENNETT Special Correspondent of the Daily Telegraph. " Mr. Bennett is well-known to English musicians as one of our ablest cHtici. A musician himself, he a Athaliah 3 o Esther 3 o Sosanna 3 o Th4odora 3 ° Belshazzar 3 o The Messiah, edited by V. Novello (Sol-fa, is.) 2 o The Messiah, ditto. Pocket Edition ... i o The Messiah, edited by W, T' Best ... 2 o Israel in Egypt, edited by Mendelssohn 2 o Israel in Egypt, edited by V, Novello. Pocket Edition i o Judas Maccabeus (Sol-fa, IS.) 2 o Judas MaccaB/Eus. Pocket Edition ... i o Samson (Sol-fa, is.) 2 o Solomon 2 o iEPHTHA 2 O OSHUA 2 O lEBORAH 2 O Saul 2 o Chandos Te Dedu I o Dettingen Te Deum i o Utrecht Jubilate i o O PRAISE the Lord with one consent (Sixth Chandos Anthem) i o Coronation and Funeral Anthems. Cloth 5 o Or, singly : The King shall rejoice o 8 Zadok the Priest o 3 My heart is inditing o 8 Let thy hand be strengthened ... o 6 The ways of Zion i o Ode on St. Cecilia's Day i L'Allegro 2 o HAYDN. The Creation (Sol-fa, is.) 2 o The Creation. Pocket Edition i o The Seasons (Each Season, singly, is.) ... 3 o First Mass in B flat (Latin) i o Ditto (Latin and English) i o Second Mass in C (Latin) i o Third Mass (Imperial) (Latin and English) i Ditto (Lat;n) i o Sixteenth Mass ^Latin) i 6 The Passion ; or, Seven Last Words ... 2 Te Deum (English and Latin) i o Insan.£ et Van* Cur/E (Ditto) 4 BATTISON HAYNES. The Fairies' Isle (Female Voices) 2 6 H. HEALE. Jubilee Ode i 6 C. SWINNERTON HEAP. Fair Rosamond (Sol-fa, 2s.) 3 6 EDWARD HECHT. Eric the Dane 3 O MAY I join the Choir Invisible i GEORGE HENSCHEL. Out of Darkness (Psalm 130) 2 6 HENRY HILES. Fayre Pastorel S ^ The Crusaders 2 FERDINAND HILLER. NALA AND DaMAYANTI 4 " A Song o? Victory ^ ° HEINRICH HOFMANN. Fair Melusiha ' ° Cinderella ^, — ^ f ° Song of the Norns (Female Voices) ... i o HUMMEL. s. d. First Mass in B flat i o Communion Service, ditto 2 o Second Mass in E flat i Communion Service, ditto 2 Third Mass IN D i o Communion Service, ditto 2 o Alma Virgo (Latin and English) o 4 Quod in Orbk (ditto) o 4 W. H. HUNT. Stabat Mater 3 H. H. HUSS. Ave Maria (Female Voices) z F. ILIFFE. St. John THE Divine i JOHN WILLIAM JACKSON. I cried unto God i 6 W. JACKSON. The Year 2 o D. JENKINS. David and Saul (Sol-fa, 2s.) 3 A. JENSEN. The Feast of Adonis x o W. JOHNSON. EccE Homo z o C. WARWICK JORDAN. Blow ye the trumpet in Zion i 6 N. KILBURN. The Silver Star (Female Voices] x 6 ALFRED KING. The Epiphany 3 o OLIVER KING. By the waters of Babylon (Psalm 137) ... i 6 The Naiads (Female Voices) 2 6 J. KINROSS. Songs in a Vineyard (Female Voices) ... 2 6 Ditto, Sol-pa o 6 H. LAHEE. The Sleeping Beauty (Female Voices) ... 2 6 Ditto, Sol-fa o 6 LEONARDO LEO. Dixit DoMiNus i H. LESLIE. The First Christmas Morn 2 6 F. LISZT, The Legend of St. Elizabeth 3 o Thirteenth Psalm 2 o C. H. LLOYD. Alcestis 3 Andromeda 3 Hero and Leander i 6 The Song of Balder i o The LoNGBEARDs' Saga (Male Voices) ... i 6 The Gleaners' Harvest (Female Voices) ... 2 6 A Song of Judgment 2 6 W. H. LONGHURST. The Village Fair 2 o HAMISH MACCUNN. Lay of the Last Minstrel (Sol-fa, is.6d.)... 2 6 Lord Ullin's Daughter (Sol-pa, 8d.) i G. A. MACFARREN. Songs in a Cornfield (Female Voices) ... 2 6 May Day (Sol-fa, 6d.) i •The Soldier's Legacy (Operetta) 6 Outward Bound i ORATORIOS, &c.— Continued. A. C. MACKENZIE. The Dream OF JuBAL The Story of Sayid Jason The Bride (Sol-fa, 8d.) The Rose of Sharon (Sol-fa, 2s.) Jubilee Ode - The Cotter's Saturday Night The New Covenant Veni, Creator Spiritus F. W. MARKULL. Roland's Horn F. E. MARSHALL. Prince Sprite (Female Voices) J. B. McEWEN. The Vision OF Jacob J. H. MEE. HoRATius (Male Voices) MENDELSSOHN. Elijah (Sol-fa, IS.) Elijah (Pocket Edition) As the Hart pants (Psalm 42) Come, let us sing (Psalm 95) When Israel out of Egypt came (Sol fa, gd.) Not unto us, O Lord (Psalm 115) St. Paul (Sol-fa, is.) St. Paul (Pocket Edition) Hymn of Praise (Lobgesang) (Sol-fa, is.) ... Lord, how long wilt Thou forget me Ditto, Sol-fa Hear my prayer (s. solo and chorus) Ditto ditto Ditto, Sol-fa Lauda Sion (Praise Jehovah) (Sol-fa, gd.) ... The First Walpurgis Night (Sol-fa, is.) ... Midsummer Night's Dream (Female Voices) Athalie (Sol-fa, is.) Antigone (Male Voices) (Sol-fa, is.) Man is Mortal (Eight Voices) Festgesang (Hymns op Praise) Ditto (Male Voices) Christus (Sol-fa, 6d.) Three Motets for Female Voices Son and Stranger (Operetta) LoRELEY (Sol-fa, 6d.) CEdipus at Colonos (Male Voices) To the Sons of Art (Ditto) Ditto, Sol-fa Judge me, O God (Psalm 43) (Sol-fa, ijd.) Why rage fiercely the Heathen My God, why, O why hast Thou forsaken ME (Psalm 22) Sing to the Lord (Psalm g8) Six Anthems for the Cathedral at Berlin. For 8 voices, arranged in 4 parts Ave Maria (Saviour of Sinners). 8 voices MEYERBEER. Ninety-first Psalm (Latin) Ditto (English) B. MOLIQUE. Abraham MOZART. King Thamos First Mass (Latin and English) SfivENTH Mass in B flat Communion Service in B flat, Ditto Twelfth Mass (Latin) Ditto (Latin and English) (Sol-fa, gd.) Requiem Mass Ditto (Latin and English) Ditto Ditto, Sol-fa ... LiTANIA DE VeNERABILI ALTARIS (IN E PLAT) LiTANIA DE VeNERABILI SaCRAMENTO {iN B flat) Splendente te, Deus. First Motet O God, when Thou appearest. Ditto Have mercy, O Lord. Second Motet Glory, Honour, Praise. Third Motet s. d. 2 6 3 o 2 6 1 o 5 o 2 6 2 o I 6 2 6 2 6 2 o 4 E. MUNDELLA. Victory of Song (Female Voices) DR. JOHN NAYLOR. Jeremiah J. NESVERA. De Profundis HERBERT OAKELEY, Selection from A Jubilee Lyric REV. SIR FREDK. OUSELEY.. The Martyrdom of St, Polvcarp R. P. PAINE. The Lord Reigneth (Psalm 93) The Prodigal Son Great is th^ Lord PALESTRINA. Missa Assumpta est Maria Missa Pap£ Marcelli Missa Brevis Missa *'0 Admirabile Commercium" H. W. PARKER. The Kobolds C. H. H. PARRY. De Profundis (Psalm 130) Ode on St. Cecilia's Day (Sol-fa, is.) Blest Pair of Sirens (Sol-fa, 8d.) ... AjAX and Ulysses ,. Prometheus Unbound Judith L'Allegro (Sol-fa, is. 6d ) Eton DR. JOSEPH PARRY. Nebuchadnezzar (Sol-fa, is. 6d.} B. PARSONS. The Crusader T. M. PATTISON. May Day (Sol-fa, 6d.) The Miracles of Christ iSol-fa, gd.) The Ancient Mariner The Lay of the Last Minstrel A. L. PEACE. St. John the Baptist PERGOLESI. Stabat Mater (Female Voices) (Sol-fa, 6d.) CIRO PINSUTI. Phantoms — Fantasmi nell* ombra A. H. D. PRENDERGAST. The Second Advent E. PROUT. Damon and Phintias (Male Voices) The Red Cross Knight (Sol-fa, 2s.) The Hundredth Psalm Freedom , Hereward Queen Aim6e (Female Voices) PURCELL. Dido and ^neas Te Deum and Jubilate in D J. F. H. READ. Harold Bartimeus Caractacus The Consecration of the Banner ... In the Forest (Male Voices) Psyche .„ J. V. ROBERTS. Jonah W. S. ROCKSTRO, The Good Shepherd ROLAND ROGERS. Prayer and Praise s. d. I 2 6 3 5 2 6 2 3 3 6 1 6 2 O 2 6 2 6 2 6 X 6 2 6 4 1 6 2 6 I 6 a 6 4 ORATORIOS, Sec—Continued. ROMBERG. The Lay of the Bell (New Edition, trans- lated BY THE Rev. J. Troutbeck, D.D.) Ditto, Sol-fa The Transient and the Eternal (Sol-fa, 4d.) ROSSINI. Stabat Mater (Sol-fa, is.) Moses in Egypt CHARLES B. RUTENBER. Divine Love C. SAINTON-DOLBY. Florimel (Female Voices) CAMILLE SAINT-SAENS. The Heavens declare — Cceli enarrant (Psalm ig) FRANK J. SAWYER. The Star in the East SCHUBERT. Mass in A flat Communion Service, ditto Mass in E flat Communion Service, ditto Mass in B plat Communion Service, ditto Mass in C Communion Service, ditto Mass in G Communion Service, ditto Mass in F Communion Service, ditto Song of Miriam (Sol-fa 6d.) SCHUMANN. The Minstrel's Curse The King's Son MiGNON's Requiem Paradise and the Peri (Sol-pa, is. fid.) Pilgrimage of the Rose Manfred Faust Advent Hymn, "In Lowly Guise" New Year's Song (Sol-fa, fid.) H. SCHUTZ. The Passion of our Lord BERTRAM LUARD SELBY. Choruses and Incidental Music to "Helena IN Troas" J. SHORT. Mass (S. George) Mass (S. Joseph) E. SILAS. Mass in C JOASH R. SLOMAN. Supplication and Praise HENRY SMART. King Rent's Daughter (Female Voices) .. The Bride of Dunkerron (Sol-pa, is. fid.) J. M. SMIETON. King Arthur (Sol-fa, is.) Ariadne (Sol-fa, gd.) ALICE MARY SMITH. The Red King (Men's Voices) The Song of the Little Baltung (ditto) Ditto, Sol-fa Ode to the North-East Wind Ode to the Passions A. SOMERVELL. Mass in C minor CHARLTON T. SPEER. The Day Dream SPOHR. Mass (Five Solo Voices and Double Choir) Hymn to St. Cecilia s. d. 6 2 6 3 6 3 6 2 I fi SPOHR. — continued. s d. Calvary z 6 Fall of Babylon 3 o Last Judgment (Sol-fa, IS.) i The Christian's Prayer i o God, Thou art great (Sol-fa, 6d.) i o How lovely are Thy dwellings fair ... o 8 Jehovah, Lord OF Hosts o 4 JOHN STAINER. The Crucifixion (Sol-fa, gd.) i 6 St. Mary Magdalen (Sol-fa, is.) 2 o The Daughter of Jairus (Sol-fa, gd.) ... i fi C. VILLIERS STANFORD. Eden 5 o The Voyage of Maeldune 2 fi Carmen Saculare i 6 The Revenge (Sol-fa, gd.) l 6 God IS our Hope (Psalm 46) 2 o CEdipus Rex (Male Voices) The Battle of the Baltic H. W. STEWARDSON. Gideon J. STORER. The Tournament E. C. SUCH. Narcissus and Echo God is our Refuge (Psalm 46) ARTHUR SULLIVAN. The Golden Legend (Sol-fa, 2s.) Ode for the Colonial and Indian Exhibi- tion Festival Te Deum W. TAYLOR. St. John the Baptist A. GORING THOMAS. The Sun-Worshippers E. H. THORNE. Be merciful unto me VAN BREE. St. Cecilia's Day (Sol-fa, gd.) CHARLES VINCENT. The Village Queen (Female Voices) The Little Mermaid (ditto) W. M. WAIT. The Good Samaritan R. H. WALKER. Jerusalem WEBER. In Constant Order (Hymn) I 6 Mass in G (Latin and English) i o Mass in E plat (ditto) i o Communion Service in E flat i 6 Jubilee Cantata i o Preciosa I ° Three Seasons ' ° S. WESLEY. In exjtu Israel " 4 Dixit Dominus i ° S. S. WESLEY. O Lord, Thou art my God i o C. LEE WILLIAMS. The Last Night at Bethany (Sol-fa, is.) ... 2 o THOMAS WINGHAM. Mass in D 3 o Te Deum (Latin) i CHAS. WOOD. Ode to the West Wind 1 o J. M. W. YOUNG. The Return of Israel to Palestine ... 2 fi 3 Most of the above Works may be had in paper boards at 6d. each extra, or handsomely bound°nclolh, with red or gilt edges, at is., is. 6d.. or 2S. each extra. VOCAL ALBUMS. ARNE, THOMAS A.— Twenty Songs i BACH, J. S.— Twenty Sacred Songs i BEETHOVEN— *Twenty-six Songs (Vol. I.) • ..I 'Seventeen Songs (\ ol. II.) . . i »Twenty-two Songs (Vol. III.) . . i BENDL, KAREL— *Gipsy Songs. First Series . . . . . . 2 Ditto. Second Series. (English and Bohemian) . . . . . . 2 Twelve Songs ("Loving Hearts") 2 BENNETT, STERNDALE—* Twelve Songs . . . . . . . . I 'Ditto .. .. Cloth gilt 2 BENNETT, GEO. J.— Ten Songs (Robert Burns) . . . . . . 2 Twelve Songs (Shelley & Rosetti) 2 BERLIOZ— Summer Nights (English and French) . . . . . . i BISHOP, SIR HENRY R.— Twenty Songs . . . . . . . . I BRAHMS, J.— Twenty-two Songs . . i Selected Songs (English, French, and German Words). For high or low voice. Six Books each 3 BRIGHT, DORA— Twelve Songs . . 2 DANNREUTHER, E.— Six Songs (D. G. Rosetti) 2 Five Songs (W. Morris) . . 2 DIBDIN, CHARLES — Twenty-one Songs . . . . . . . . I DVORXk, ANTONIN — *Sixteen Songs (Op. 2, 5, 17, and 31) . . 2 Eight Love Songs (English, Ger- man, and Bohemian Words), Op. 83 3 ELLIOTT, J. W.— National Nursery Songs and Rhymes. With sixty- five Illustrations Cloth gilt 7 FRANZ, R.— Thirty Songs . . . . i Fourteen Songs (Robert Burns) 2 GOETZ, HERMANN - "Eighteen Songs (Op. 4, 12, 19) . . . . 2 HANDEL — Twelve Songs for Soprano 2 Twelve Songs for Contralto . . 2 Twelve Songs for Tenor . . 2 Twelve Songs for Baritone or Bass -i. HAYNES, BATTISON— Elizabethan Lyrics . . . . . . . . 2 HAYDN— Ten Canzonets . . . . i HILL, LADY ARTHUR— Holiday Songs . . . . . . . . 2 HOOK, JAMES— Twenty Songs . . i KING, OLIVER— Six Songs (Baritone) 2 LISZT— Twenty Songs .. .. i * These Songs have -Twenty-one Songs. . — Cycle of six Love LODER. E. J.- MACCUNN, H. Lyrics . . . . . . . . 2 MACKENZIE, A. C. — Eighteen Songs. Three Books. . . each 2 Eighteen Songs. One Vol. Cloth gilt 7 Spring Songs . . . . . . 2 MARIANI— Twenty-two Songs (Ital.) 2 MENDELSSOHN — Songs. .(With Portrait.) Folio. Cloth gilt 21 *Songs .. .. .. ..4 *Ditto .. .. Cloth gilt 6 *Songs. (Deep Voice) .. ..6 'Ditto ditto Cloth gilt 8 Solo music in "Elijah," Soprano, Contralto, and Tenor . . each i Ditto, Bass 2 MOORE— Irish Melodies . . . . 2 Irish Melodies . . Cloth gilt 4 Irish Melodies. Folio. Cloth gilt 21 PURCELL, HENRY— Twelve Songs 2 RANDEGGER— Sacred Songs for Little Singers. (Illustrated) . . 2 Ditto . . . . Cloth gilt 5 RUBINSTEIN, A. — Twenty-five Songs . . . . . . . . I SCHUBERT— Twenty Songs (Mezzo- Sop.) I Twenty Songs (Contralto) . . i . Twenty Songs (Sop. or Tenor) . . i 'Schwanengesang (Swan Songs) I *Die Schone Miillerin (The Fair Maid of the Mill) .. .. i 'Winterreise (The Winter Jour- ney) , Op. 89 . . . . . . I SCHUMANN— 'Songs. Folio. Cloth gilt 10 'Myrthen (Twenty-six Songs), Op, 25 I ■*Liederkreis(TwelveSongs), Op. 39 i ■ 'Vocal Album . . . . . . 2 'Ditto . . . . Cloth gilt 4 'Woman's Love and Life (Eight Songs), Op. 42 . . . . . . I -'Twelve Songs (Op. 35) . . . . i Dichterliebe (A Poet's Love). Sixteen Songs (Op. 48) . . i VARIOUS COMPOSERS- English Folk-Songs . . . . 2 Old Ireland (Irish Melodies) . . 2 The Sunlight of Song. Forty-six Illustrations . . Cloth gilt 3 'Volkslieder Album (Forty Songs) 2 'Ditto .. .. Cloth gilt 4 German and English Words, LONDON & NEW YORK: NOVELLO, EWER AND CO. NOVELLO'S PIANOFORTE ALBUMS Edited by BERTHOLD TOURS. In Numbers, each One Shilling; or, Cloth Volumes, each Four Shillings. Ho. 1. — Bach. Twenty Pieces from Petits Preludes. Suites Anglaises. Partita Nos. i to 3, &c. No. 2. — Bach. Twenty Pieces from Suites Anglaises, Suites Franpaises. Partita Nos. 4 and 5, &c. No. 3.— Bach. Twenty Pieces from Petits Preludes, Concertos, Suites Anglaises, &c. Nos. 1, 2, and 3, in One Vol., cloth, 4s. No. 4. — Handel. Twenty-four Pieces from Suites i to 7, Sonatas, *' Harmonious Blacksmith," &c. No. S. — Handel. Twenty-four Pieces from Suites 8 to 12, Gavottes, &c. No. 6.— Handel. Twenty-four Pieces &om Suites 13 to 16, Water Music, &c. Nos. i, 5, and 6, in One Vol., cloth, 4s. No. No. 7. — Marches. Fifteen Pieces, including Mendels- sohn's " Hero's March," Rakoczy's March, Bridal March, " Lohengrin," &c. 8. — Marches. Fifteen Pieces, including Meyer- beer's Coronation March, Cornelius March, Pilgrim's March, &c. No. 9. — Marches. Fifteen Pieces, including Mendels- sohn's Wedding March, March " Tannhauser," &c. Nos. 7, 8, and 9, in One Vol., cloth, 4s. No. 10.— Gavottes, Hinnets, &c. Sixteen Pieces, including Rameau's Gavotte, 2immermann's Gavotte, Calkin's Minuet, Silas's BourrSe, &c. No. 11.— Gavottes, Minuets, &c. Sixteen Pieces, including Gluck's Gavotte, W. Macfarren's Bourree, Bach's Bourree, &c. No. 12.— Gavottes, Hinnets, &c. Sixteen Pieces, including W. Macfarren's 2nd Gavotte, Sir J. Benedict's Gavotte, Mozart's Minuet, &c. Nos. 10, 11, and 12, in One Vol., cloth, 4s. No. 13.— Wollenhaupt, J. Ten Pieces, including Marche Hongroise, Scherzo Brilliant, &c. Ho. IJ.— Wollenhaupt, J. Ten Pieces, including " La Gazelle," " Mazeppa Galop," &c. Ho. IS.— Wollenhaupt, J. Ten Pieces, including " Les Clochettes," " Feu FoUet," &c. Nos. 13, \i, and IS, in One Vol., cloth, 4s. Ho. 16.— Schweizer, Otto. Eight Scottish Airs (arranged for four hands). No. 17.— Spindler, Fritz. Nine Pieces, including "Murmuring Rivulet," The Pilgrims' Chorus (" Tannhauser "), &c. No. 18.— Bplndler, Fritz. Nine Pieces, including " L'Oisillon," "Le Carillon," "The Evening Star " (" Tannhauser "), &c. Ho. 19.— Spindler, Fritz. Ten Pieces, including " Jeu des Ondes," Valse Melodieuse, Spmmng Song (" Flying butchman"), &c. Nos. 17, 18, and 19, in One Vol., cloth, 4s. Ho. 20.— Goetz, Hermann. Five Compositions. No. 21.-Goetz, Hermann. Four Compositions. Ho. 22.-Goetz, Hermann. Six Compositions. HOS. 20, 21, and 22, in One Vol., cloth, 4s- No. 23.— Rheinberger, Josef. Seven Compositions. No. 2{.— Rheinberger, Josef. Eleven Compositions. No. 33. — Rheinberger, Josef. Seven Compositions. Nos. 23, 'ii, and 29, in One Vol., cloth, 4s. No. 26 Tours, Berthoid. Juvenile Album. Eight Characteristic Pieces (L)uets), 2S. Nos. 27 and 28.— Moscheles, J. Domestic Life. Twelve Characteristic Duets, Two Books, each 2s. The Two Boolts, in One Vol., cloth, 4s. No. 29.— Kjerulf, Halfdan. Nine Pieces. Op. 4, Nos. I to 3 ; Op. 12, Nos. i to 6. Ho. 30.— B3eiuir, Halfdan. Ten Pieces, Op. 24, Nos. I to 4 ; Op. 27, Nos. I and 2 ; Op. 28, Nos. I to 4. No. 31.— Kjerulf. Halfdan. Op. 28, Nos. 5 and 6 ; Op. 29; an I Twenty Songs arranged for the Pianoforte by the Composer. Nos. 29, 30, and 31, in One Vol., cloth, 4s. Nos. 32, 33, and 34. The National Dance Music of Scotland. Arranged for the Pianoforte by Alexander Mackenzie ; with additions by his son, Dr. A. C. Mackenzie. Nos. 32, 33, and 31, in One Vol., cloth, 4s. No. 33.— Mackenzie, Dr. A. C. Eight Pieces, Op. 13, Nos. I to 5; Op. 15, Nos. I to 3. No. 36.— Maclienzie, Dr. A. 0. Nine Pieces. Op. 20, Nos. I to 6 ; Op. 23, Nos. I to 3. No. 37.— Maclienzie, Dr. A. C. Six Songs. Tran- scribed for the Pianoforte by Guiseppe Buonamici. Nos. 33, 38, and 37, in One Vol., cloth, 4s. No. 38.— Altschul, Rudolf. Fifty Hungarian National Songs. No. 41.- Liadotr, Anatole. Twenly-one Pieces. Op. 2, Nos. I to 14; Op. 3, Nos. I, 3, 4, and 6; Op. 6 ; Op. 15, Nos. I and 2. No. 42.— Liadoff, Anatole. Seven Pieces. Op. 4, Nos. 1104; Op. 7, Nos. 1 and 2 ; Op. 11. No. 43.— Liadoff, Anatole. Ten Pieces. Op. 8; Op. 9, Nos. I and 2 ; Op. 10, Nos. i, 2, and 3 ; Op. 13, Nos. I to 4. Nos. 41, 42, and 43, in One Vol., cloth, 4s. No. 44.— Cui, C^sar. Thirteen Pieces. Op. 20, Nos. 1 to 12 ; Op. 21, No. 3. No. 49.— Cui, CSsar. Eleven Pieces. Op. 21, No. 41 Op. 22, Nos. I, 2, and 3 ; Op. 31, No. 2 ; Op. 39, Nos. I to 6. No. 46.— Cui, C^sar. Seven Pieces. Op. 22, No. 4 ; Op. 29, No. 1 ; Op. 30, No. I ; Op. 35i "OS. i and 2 ; Op. 40, Nos. 2 and 4. Nos. 44, 49, and 46, in One Vol., cloth, 4s. No. 47.— Schubert, Franz. Four Impromptus. Op. 90. No. 48.— Schubert, Franz. Four Impromptus. Op. 142. No. 49.— Schubert, Franz. Moments Musicales (Op. 94), and Adagio and Rondo (Op. 145)- Nos. 47, 43, and 49, in One Vol., cloth, 48. T r.wr.nN & NEW YORK: NOVELLO, EWER AND CO. VOCAL DUETS. F. ABT. II u TWENTY - FOUR DUETS (Sop. and Cont.). Book i . . 2/6 — TWENTY - FOUR DUETS (Sop. and Cont.). Book 2 . . 2]6 — STERNDALE BENNETT. FOUR SACRED DUETS .. 1/0 — F. H. COWEN. SIX DUETS (Sop. and Cont.) 2/6 — E. DANNREUTHER. FIVE TWO-PART SONGS.. 2/6 — E. C. FORD. SIX TWO-PART SONGS .. 2/6 — MYLES B. FOSTER. SIX TWO-PART SONGS .. 1/0 — SIX TWO-PART ANTHEMS 1/0 — Singly, Threepence each. BATTISON HAYNES. SIX DUETS (Sop. and Cont.) 2/6 — SIX TWO-PART SONGS .. 2/6 — OLIVER KING. SIX DUETS (Sop. and Cont.) 2/6 — JOHN KINROSS. SONGS OF THE FOREST. Six Two-part Songs . . . . 1/0 — Singly, Threepence each. MENDELSSOHN, H THIRTEEN TWO - PART SONGS 1/0 Tonic Sol-fa . . . . 0/8 THIRTEEN TWO - PART SONGS .. .. Folio 2/6 THIRTEEN TWO - PART SONGS (German and English) 2/0 MOORE. IRISH MELODIES. Duets.. 1/0 GIRO PINSUTI. SIX TWO-PART SONGS .. 2/6 SIX TWO - PART SONGS. Second Set 2/6 CARL REINECKE. TWELVE CANONS (for Two- part Female Chorus or Two Solo Voices) 1/6 RUBINSTEIN. EIGHTEEN TWO - PART SONGS H. SMART. NINE SACRED DUETS (Sop. and Cont.) 2/6 SCHUMANN. THIRTY FIVE VOCAL DUETS (German and Eng- lish words) 2/6 CHARLES WOOD. SIX TWO-PART SONGS. For Solo Voices (or Female Chorus) 2/6 2/0 4/0 2/6 416 LONDON & NEW YORK : NOVELLO, EWER AND CO. ORATORIOS, 8cc.— Continued, HAN DEL — continued. Alexander Balus Hercules Athaliah Esther Susanna Theodora Belshazzar The Messiah, edited by V. Novello (Sol-fa, is.) The Messiah, ditto. Pocket Edition The Messiah, edited by W. T Best Israel in Egypt, edited by Mendelssohn Israel in Egypt, edited by V. Nuvello. Pocket Edition Judas Maccab;eus (Sol-fa, is.) Judas Maccabeus. Pocket Edition Samson (Sol-fa, is.) Solomon Jephtha Joshua '. Deborah Saul £handos Tb Deum Dettingen Te Deum Utrecht Jubilate O praise the Lord with one consent (Sixth Chandos Anthem) Coronation and Funeral Anthems. Cloth Or, singly : The King shall rejoice Zadok the Priest My heart is inditing Let thy hand be strengthened The ways op Zion Ode on St. Cecilia's Day L'Allegro HAYDN, The Creation (Sol-fa, is.) The Creation. Pocket Edition The Seasons (Each Season, singly, x&.) First Mass in B flat (Latin) Ditto (Latin and English) Second Mass in C (Latin) Third Mass (Imperial) (Latin and English) Ditto (Latin) Sixteenth Mass (Latin) The Passion; or, Seven Last Words Te Deum (English and Latin) iNSANiG £T VANiE CURM (DiTTO) BATTISON HAYNES. The Fairies* Isle (Female Voices) ... H. HEALE. Jubilee Ode C. SWINNERTON HEAP. Fair Rosamond (Sol-fa, 2s.) EDWARD HECHT. Eric the Dane O MAY I join THE ChOIR INVISIBLE GEORGE HENSCHEL. Out of Darkness (Psalm 130) HENRY HILES. Fayre Pastorel The Crusaders FERDINAND HILLER. Nala and Damayanti A Song of Victory ... HEINRICH HOFMANN. Fair Melusina Cinderella Song of the Norns (Female Voices) s. d. 3 o 3 o 3 3 o 3 o 3 o 3 o 1 6 2 O X o 4 2 6 1 6 3 6 3 I 2 6 6 6 2 6 4 I 2 o 4 o I o HUMMEL. s. d. First Mass in B flat i o Communion Service, ditto 2 o Second Mass in E flat i Communion Service, ditto 2 Third Mass in D i o Communion Service, ditto 2 o Alma Virgo (Latin AND English) o 4 Quod IN Orbe (ditto) .... o 4 W. H. HUNT. Stabat Mater 3 o H. H. HUSS. Ave Maria (Female Voices) i F. ILIFFE. St. John THE Divine i o JOHN WILLIAM JACKSON. I cried unto God i 6 W. JACKSON. The Year z o D. JENKINS. David AND Saul (Sol-fa, 2s.) 3 o A. JENSEN. The Feast of Adonis i W. JOHNSON. EccE Homo 2 o C. WARWICK JORDAN. Blow YE THE trumpet IN .Zion i 6 N. KILBURN. The Silver Star (Female Voices) i 6 ALFRED KING. The Epiphany 3 o OLIVER KING. By the waters of Babylon (Psalm 137) ... 1 6 The Naiads (Female Voices) 2 6 J. KINROSS. Songs in a Vineyard (Female Voices) ... 2 6 Ditto, Sol-fa o 6 H. LAHEE. The Sleeping Beauty (Female Voices) ... 2 S Ditto, Sol-fa 6 LEONARDO LEO. Dixit Dominus i o H. LESLIE. The First Christmas Morn 2 6 F. LISZT. The Legend of St. Elizabeth 3 o Thirteenth Psalm 2 o C. H. LLOYD. Alcestis 3 Andromeda ... 3 Hero and Leander i 6 The Song OF Balder i o The Longbeards' Saga (Male Voices) ... i 6 The Gleaners' Harvest (Female Voices) ... 2 6 A Song OF Judgment 2 6 W. H. LONGHURST. The Village Fair 2 HAMISH MACCUNN. Lay of the Last Minstrel ( Sol-fa, is. 6d.)... 2 6 Lord Ullin's Daughter (Sol-fa, 8d.) i o G. A. MACFARREN.- Songs in a Cornfield (Female Voices) ... 2 6 May Day (Sol-fa, 6d.) i o The Soldier's Legacy (Operetta) 6 o Outward Bound i ORATORIOS, &c,~ConHnued. A. C. MACKENZIE. , s. d. The Dream of Jubal 2 6 The Story of Savid 3 o Jason ... 2 6 The Bride (Sol-fa, 8d.) ... i o The Rose OF Sharon (Sol-fa, 2S.) 5 o Jubilee Ode 2 6 The Cotter's Saturday Night 2 o The Neik Covenant i 6 Veni, Creator Spiritus S o F. W. MARKULL. Roland's Horn a 6 F. E. MARSHALL. Prince Sprite (Female Voices) 2 6 J. H. MEE. HoRATius (Male Voices) z MENDELSSOHN. Elijah (Sol-fa, IS.) 2 o Elijah (Pocket Edition) i o As THE Hart pants (Psalm 42) i o Come, LET us SING (Psalm 95) i o When Israel out of Egypt came (Sol-fa, gd.) i o Not unto us, O Lord (Psalm 115) i o St. Paul (Sol-fa, is.) 2 o St. Paul (Pocket Edition) i o Hymn of Praise (Lobgesang) (Sol-fa, ib.) ... i Lord, how long wilt Thou forget me ... i o Ditto, Sol-fa o 4 Hear my prayer (s. solo and chorus) ... i o Ditto ditto 4 Ditto, Sol-fa 3 Lauda Sion (Praise J ehovah) ( Sol-fa, gd.) ,„ 2 o The First Walpurgis Night (Sol-fa, IS.) ... i o Midsummer Night's Dream (Female Voices) i o Athalie (Sol-fa, is.) 2 o Antigone (Male Voices) (Sol-fa, IS.) ... 4 o Man is Mortal (Eight Voices) 1 o Festgesang (Hymns of Praise) i o Ditto (Male Voices) ... i o Christus (Sol-fa, 6d.) i o Three Motets for Female Voices i o Son and Stranger (Operetta) 4 o LoRELEV (Sol-fa, 6d.) ... i o CEdipus at Colonos (Male Voices 3 o To the Sons op Art (Ditto) i Ditto, Sol-fa "o 3 Judge me, O God (Psalm 43) (Sol-pa, ijd.) o 4 Why rage fiercely the Heathen o 6 My God, why, O why hast Thou forsaken me (Psalm 22) o 6 Sing to the Lord (Psalm 98) o 8 Six Anthems for the Cathedral at Berlin. For 8 voices, arranged in 4 parts ... o 8 Ave Maria (Saviour of Sinners). 8 voices 1 MEYERBEER. Ninety-first Psalm (Latin) i o Ditto (English) i o B. MOLIQUE, Abraham 3 o MOZART. King Thamos 1 o First Mass (Latin and English) i Seventh Mass in B flat i o Communion Service in B flat, Ditto ... i 6 Twelfth Mass (Latin) i o Ditto (Latin and English) (Sol-fa, gd.) i o Requiem Mass 1 o Ditto (Latin and English) i Ditto Ditto, Sol-pa ... i o LiTANIA DE VeNERABILI ALTARIS (IN E FLAT) I 6 Litania DE Venerabili Sacramento (in B FLAT) I 6 Splendente te Deus. First Motet ... o 3 O God, when Thou appearest. Ditto ... o 3 Have mercy, O Lord. Second Motet ... o 3 Glory, Honour, Praise. Third Motet ... o 3 E. MUNDELLA. s. d. Victory of Song (Female Voices) z DR. JOHN NAYLOR. Jeremiah ... ... v ■ 3 J. NESVERA. De Profundis a 6 HERBERT OAKELEY. Selection from A Jubilee Lyric ... ... x REV. SIR FREDK. OUSELEY. The Martyrdom of St. Polycarp 25 R. P. PAINE. The Lord Reignetk (Psalm 95) z The Prodigal Son z 6 Great is the Lord i PALESTRINA. MissA Assumfta est Maria 2 6 MissA Pafjb Marcellx 2 MissA Brevis 2 6 H. W. PARKER. The Kobolds i o C. H. H. PARRY. De Profundis (Psalm 130) s Ode ON St, Cecilia's Day (Sol-pa, IS.) ... 2 Blest Pair OF Sirens (Soi^fa, 8d.) i o AjAx AND Ulysses ... i i o Prometheus Unbound ... ... 30 Judith ,>..,.-. 5 L'Allegro . .,. >^ 2 6 Eton > 2 DR. JOSEPH PARRY. Nebuchadnezzar (Sol-fa, IS. 6d.) 3 B. PARSONS. The Crusader 3 6 T. M. PATTISON. May Day (Sol-fa, 6d,) ... i 6 The Miracles of Christ (Sol-fa, ga.j ... 2 The Ancient Martnttr ... ,_ 2 6 The Lay of THE i-..oi .Minstrel ... 2 6 A. L. PEACE. St. John the Baptist 2 6 PERGOLESI. Stabat Mater (Female Voices) (Sol-fa, 6d.) z GIRO PINSUTI. Phantoms— Fantasmi nell' ombra z A. H. D. PRENDERGAST. The Second Advent i 6 E. PRO ax. Damon and Phintias (Male Voices) 2 6 The Red Cross Knight (Sol-fa, 2s.) ... 4 The Hundredth Psalm z Freedom , • ... 1 o Hereward 4 Queen Aim^e (Female Voices) 2 6 PURCELL. Dido and ^neas 2 6 Te Deum and Jubilate in D i J. F. H. READ. Harold 4 o Bartimeus ... ... ' ... I 6 Caractacus 26 The Consecration of the Banner i 6 In the Forest (Male Voices) i Psyche 5 J. V. ROBERTS. Jonah : ... 3 a W. S. ROCKSTRO. The Good Shepherd 2 6 ROLAND ROGERS. Prayer and Praise 4 ORATORIOS, &c,~Continued. ROMBERG. The Lay of the Bell (New Edition, trans- lated BY THE Rev. J. Troutbeck, D.D.) Ditto, Sol-fa The Transient and the Eternal (Sol-fa, 4d.) ROSSINI. Stabat Mater (Sol-fa, is.) MobEs IN Egypt CHARLES B. RUTENBER. Divine Love C. SAINTON-DOLBY. Florimel (Female Voices) CAMILLE SAINT-SAENS. The Heavens declare — Cceli enarrant (Psalm ig) SCHUBERT. Mass in A flat Communion Service, ditto Mass in E flat Communion Service, ditto Mass in B flat Communion Service, ditto Mass in C Communion Service, ditto Mass in G Communion Service, ditto Mass in F ' ... Communion Service, ditto Song of Miriam (Sol-fa ,6d.) SCHUMANN. The Minstrel's Curse ... The King's Son Mignon's Requiem Paradise and the Peri (Sol-fa, is,.6d.) Pilgrimage of the Rose Manfred Faust ... ' Advent Hymn, "In Lowly Guise" New Yf*^«?'s *= no (Sol-fa, 6d.).-. 7^- H. SCHUTZ. The Passion of our Lord^',, BKP'^r.AM LU'AiM>l SELBY. Chorul. :D ' ■ Incidental Music to " Helena IN Troas ". . J. SHORT. Mass (S. George) Mass (S. Joseph) E. SILAS, Mass in C JOASH R. SLOMAN. Supplication and Praise ... " HENRY SMART. King Rent's Daughter (Female Voices) .. The Bride of Dunkerron (Sol-fa, is. 6d.) J. M. SMIETON. King Arthur , Ariadne (Sol-fa, gd.J ALICE MARY SMITH. The Red King (Men's Voices) The Song op the Little Baltunq (ditto) Ditto, Sol-fa ... Ode to the North-East Wind Ode to the Passions A. SOMERVELL. Mass in C minor CHARLTON T. SPEER. The Day Dream SPOHR. Mass (Five Solo Voices and Double Choir) Hymn to St. Cecilia s. d. z o 8 1 Z 6 o 2 6 2 6 I 6 3 6 3 6 2 O z o 4 2 6 2 O 2 6 2 O I O I o 8 1 o a a fi 2 O 2 I O SPOHR. — continued. s d. Calvary 2 6 Fall op Babylon 3 Last Judgment (Sol-fa, is.) i o The Christian's Prayer z God, Thou art great (Sol-fa, 6d.) i How lovely are Thy dwellings fair ... o 8 Jehovah, Lord OF Hosts o 4 JOHN STAINER. The Crucifixion (Sol-fa, gd.) z 6 St. Mary Magdalen (Sol-fa, is.) 2 The Daughter op Jairus (Sol-fa, gd.) ... i 6 C. VILLIERS STANFORD. Eden 5 o The Voyage OF Maeldune 2 6 Carmen SiCCULARE z 6 The Revenge (SoL-FA,9d.) i 6 God IS OUR Hope (Psalm 46) 2 o CEdipus Rex (Male Voices) 3 o The Battle of the Baltic i 6 H. W. STEWARDSON. Gideon 4 J. STORER. The Tournament 2 E. C. SUCH. Narcissus and Echo ... 3 o God is our Refuge (Psalm 46) i o ARTHUR SULLIVAN. The Golden Legend (Sol-fa, 2s.) ... 3 6 Ode for the Colonial and Indian Exhibi- tion I Festival Te Deum i W. TAYLOR. St. John the Baptist 4 o A. GORING THOMAS. The Son-Worshippers , i E. H. THORNE. Be merciful unto me z VAN BREE. St. Cecilia's Day (Sol-fa, gd.) i CHARLES VINCENT. The Village Queen (Female Voices) ... 2 6 The Little Mermaid (ditto) ... 2 6 R. H. WALKER. Jerusalem 3 WEBER. In Constant Order (Hymn) z 6 Mass in G (Latin and English) z o Mass in E flat (ditto) z o Communion Service IN E FLAT i 6 Jubilee Cantata i o Preciosa Z Three Seasons ... z o S. WESLEY. In exitu Israel 4 Dixit Dominus i S. S. WESLEY. O Lord, Thou art my God i C. LEE WILLIAMS. The Last Night at Bethany (Sol-fa, is.) ... 2 o THOMAS WINGHAM. Mass in D 3 o Te Deum (Latin) i 5 chas. wood. Ode to the West Wind i o J. M. W. YOUNG. The Return of Israel to Palestine ... 2 6 Most of the above Works may be had in paper boards at 6d. each extra, or handsomely bound in cloth, with red or gilt edges, at is., is. 6d., or 25. each extra. NOVELLO, EWER & CO.'S MUSIC PRIMERS EDITED BY Sir JOHN STAI NER. 1. The Pianoforte (Price 2.s.) - - - E. Pauer. 2. The Rudiments of Music {Price is.) W. H. Cummings. 3. The Organ (Price 2s.) - - - - J- Stainer. 4. The Harmonium {Price 2s.) - - - King Hall. 5. Singing {Price 4s. Paper Boards, 54.) A. Randegger. 6. Speech in Song (Singer's Pronouncing Primer) {Price 2s.) - - - A. J. Ellis, f.r.s. 7. Musical Forms {Price 2s.) - - - E. Paukr. 8. Harmony {Price 2s.) J. Stainer. 9. Counterpoint {Price 2s.) - - - - Dr. Bridge. ro. Fugue {Price 2s.) James Higgs. r.i. SciENTiiTC Basis of Music {Price is.) ■ Dr. Stone. X2. Double Counterpoint {Price 2s.) - - Dr. Bridge. 13. Church Choir Training {Price is.) Rev. J. Troutbeck. 14. Plain Song {Price 2s.) - - - Rev. T. Helmore. 15. Instrumentation {Price 2s.) - - - E. Prout. 16. The Elements of the Beautiful in Music {Price is.) ----- E. Pauer. 17. The Violin {Price 2s.) - - - Berthold ToURS. 18. Tonic Sol-fa {Price is.) - - - - J. Curwen. 19. Lancashire Sol-fa {Price is.) - James Greenwood. 20. Composition {Price 2s.) ----]■ Stainer. 21. Musical Terms {Price is.) - Stainer and Barrett. ■£2. 1"he Violoncello {Price 2s.) - - Jules de Swert. 23. Two-part Exei-xises (396) {Price is.) James Greenwood. 24. Double Scales {Price is.) - - Franklin Taylor. .^5. Musical Expression {Price 3s.) - Mathis Lussy. 26. Solfeggi {Price 4s. Paper boards, 5s.) F. A. Marshall. Or, in Three Parts, is. 6(1. each. 27. Organ Accompaniment {Price 2s.) - - Dr. Bridge. 28. The Cornet {Price 2s.) - - - - H. Brett. 29. Musical Dictation, Part i {Price is.) - Dr. Ritter. 30. Musical Dictation, Part 2 {Price 2s.) ■ Dr. Ritter. 31. Modulation {Price 2s.) . -• . . James Higgs. 32. Double Bass {Price 2s.) - - - - A. C. White. 33. Extemporization {Price 2s.) - - - Dr. Sawyer. 34. Analysis of Form {Price 2s.) - - H. A. Harding. 35. 500 Fugue Subjects {Price ^s.) - Arthur W. Marchant. 36. Hand Gymnastics {Price is. 6d.) T. Ridley Prentice. 37. Musical Ornamentation (i«i'/;e Press) Ed. Dannreuther. (to be continued.) Any of Ihe above may be had strongly bonnet in boards, price 6d. each extra. LONDON & NEW YORK: NOVELLO, EWER AND CO.