APPENDIX II

SYSTEMS OF PHONETIC AND TONETIC

NOTATION AND SOME PROBLEMS

OF CONNECTED SPEECH

I. Phonetic Symbols

1. Explanation of Narrow, Broad, and Extra-Broad Notation

Study of a language may be helped a great deal if sounds to be used in a given word or sentence are represented in some unambiguous say, so that there can never be any doubt which sound is meant. English, however, is a very badly spelled language. In it the same letter often stands for a number of distinctly different sounds as, for example, the letter a in hat, hate, hall, harm, etc. On the other hand, the same sound is represented by a variety of letters; for example, the vowel sound in bee is represented by ea in bean, eo in people, ie in piece, ei in receive, e in be, i in machine. An alphabetic system constructed on the basis of one symbol, and always the same symbol, foe each ‘distinctive’ sound, or and always the same symbol, for each ‘Distinctive’ sound, or ‘phoneme’¹ is said to be phonetic. By making use of a phonetic alphabet a learner may avoid all those mispronunciations which one from relying on the traditional spelling. A word or sentence respelled in a phonetic alphabet will not only enable a learner to avoid the distracting influence of the traditional spelling, but also assist his auditory memory, since it appeals to the visual memory in a more unambiguous way than the traditional spelling.

When an alphabet with a clear one-for-one correspondence between a single graphic symbol and a single phoneme is used in a transcription, it is understood that one graphic symbol stands for one or other sound included in the phoneme, the choice being determined by simple principles which may be stated once for all. A transcription of this, type, i. e., ‘one symbol per phoneme’ type is called a Broad Notation.

A transcription may also make use of different symbols to indicate two or more variant sounds included in a phoneme instead of leaving the choice of a particular sound to be determined by principles stated. For examples, the two l sounds in the word little, which are regularly differentiated by many speakers of English, are transcribed as [l] and , the principles for the choice of the two being that in the pronunciation of such speakers the former variety is used before a vowel and the latter before a consonant or finally. A transcription of this type, i. e., ‘more symbols than one per phoneme’ type is called a Narrow Notation.

There may be degrees of broadness or narrowness in either type of notation. Thus, a broad notation may make use of exactly the same number of different letters as that of the phonemes existing in a language; while the number of different letters may be minimized by making a letter do duty either with length-marks, with stress-marks, or by being combined with other letters, for two or more phonemes without violating the ‘one symbol per phoneme’ principle. A transcription that makes use of the minimum number of letters within the ‘one symbol per phoneme’ principle is called an Extra-Broad Notation, or Simplified (Broad) Notation.

The type of phonetic notation used in Japan most widely at present is the one formulated by Daniel Jones and used in many of his works including his English Pronouncing Dictionary. His transcription is not one that is the broadest possible. It may be made broader by eliminating some of the letters in his system, which are more or less inconvenient from certain points of view.

The importance of broad, and extra-broad or simplified (broad) transcription has come to be recognized in recent years as evidenced by preference given it over a narrow transcription for general purposes of teaching English as a foreign language. The merits of a narrow transcription have come to be doubted, except for showing fine distinctions, particularly dialectical, for, although an American or Britisher may use varieties of any phoneme, the designating of such delicate differences in transcription has not in general borne better results than teaching through the use of a phonemic transcription. In the Preface to a phonetic reader published recently Daniel Jones remarks, “Another commendable feature of the book is the use of ‘broad’, transcription, i. e. a system which employs the least possible number of special phonetic letters. It is the same as that used in Scott’s English Canversations.”² It is significant, too, that the British Council uses an extra-broad or simplified transcription in its periodical entitled, English Language Teaching.

In one form of extra-broad or simplified transcription the symbols in hair (h) and e in get (et) are levelled to e while the symbols in hum and in about are levelled to . The levelling of the last two vowels is a recent tendency and has provided grounds for controversy, pro and con, among teachers of English as a foreign language. From a purely practical point of view the levelling of and , it seems, can be fully supported, as well as that of and i or and e, since substitution of one for the other does not distinguish one word from another, and also since English and American speakers often substitute one for the other, both individually and dialectically.

Because of the importance of understanding fully the nature and purpose of phonemes the following quotation, almost in full, is taken from the Preface by Daniel Jones, to a phonetic reader.

“I believe this to be the first book ever published in which a correct pronunciation of English is recorded accurately by the simplest possible type of phonetic transcription, i. e. a system employing the minimum number of symbols which will keep all words distinct from each other. Such a system involves using only 29 letters, a length mark, a stress mark and (very occasionally) a syllabic mark. This is entirely adequate for teaching foreign learners how to use the English speech-sounds properly, after they have learnt how to make them.

“The idea of transcribing languages by a phonetic system which is accurate and at the same time as simple as possible is by no means new. The advisability of writing in this way was pointed out by the great pioneer of practical phonetics, Henry Sweet, in his Handbook of Phonetics (1877), and the “Broad Romic” used by him in his Elementarbuch des Gesprochenen Englisch (first published in 1885) was actually a close approximation to the ideal of simplest transcription. That system contained in fact only one superfluous letter, namely ; if oo had been substituted for this, as it might have been in conformity with the rest of Sweet’s system, Mr. Scott’s style of transcription would have been anticipated 57 years ago.

“Most subsequent writers on phonetics, myself among them, have departed to a greater or less extent from the ideal of simplest transcription. However, after working for many years with more elaborate forms of phonetic writing, I have now reached the conclusion that the supposed advantages of introducing more than the bare minimum of signs are illusory, and that the simplest type of transcription is the best for almost all purposes, and particularly in the teaching of a spoken language to foreign learners.”⁸

Readers are asked to study the following comparative table.

Comparative Table

Narrow Transcription:

Explanation: accent; [l] clear l (in prevocalic position); , dark l (before consonant or final); [p^ｈ], aspirated p; ,half-length mark; quality of length mark: or when long; [ ], quality of relatively more lax short .

Broad Transcription:

pit [pit], bid [bid], hair [h], high [hi], how [hu], American: [hf], and hum [hm].

Explanation: [l] and are leveled to l, the distinction being obtained by the phonetic context. [i] in and [ ] are levelled to , the distinction being obtained by the phonetic context. [] in [h] has a different symbol than that for the “closer” [e] in get. [a] in diphthongs such as high and how, which is a front low vowel, has a different symbol from the back low [] in which is always long. [ ] in hum has a different symbol from the obscure vowel [] which [] in Received Standard (British) occurs only in unstressed positions.

Extra-Broad Transcription:

pit [pit], bid [bid], hair [he], high [hi], or [hai], how [hu] or [hau], American [hf], hum [hm].

Explanation: Based strictly on the principle of one symbol for one phoneme.

2. Comparison of So-called British and American Sounds

In connection with the problems of dialects or types of English, the following general observations may be found useful:-

(1) There are distinct variants of speech in every social class, and class variants in every district.

(2) Local variants become increasingly unlike one another in proportion to lack of education.

(3) They become more alike in proportion to increase in the amount of education acquired.

(4) The spread of education and the increased facility of oral communication tends to bring about the unification of the social variants in all districts.

(5) In England those speakers who use a type of English having more features in common with Southern English than with Northern English are considered educated speakers.

(6) In American those speakers who use a type of English having more features in common with Northern English than with Southern English seem to provoke least strangeness.

(7) The two types of English regarded as the speech of educated people in England and America respectively have few noticeable divergences, so that one is as well received as the other in all English speaking countries.

As a corollary of the above considerations, it is desirable for Japanese students to learn the type of speech used by educated people of either England or America, disregarding what is called the ‘sub-standard’ type of speech. It is inevitable that one who has mastered the special speech of one particular geographical area will meet with some difficulty of understanding when his contacts are with speakers from other areas who use more or less different types of speech. It would be preferable, however, to learn, if circumstances permit, the particular type of English spoken and considered normal by people with whom the learners will come in personal contact. Or they might be trained to speak any one type of English and to understand some other type when it is spoken to them.

Some of the chief differences between Southern British English and General American English are as follows:

Southern British	General American
(1) ‘r’ is sounded only before a vowel.	‘r’ is sounded wherever it is spelt, and the l is pronounced as a retroflex [r] in colonel, all British ’s having an [r] in G. A.
(2) The vowel in half, last, path, dance is that in father.	The vowel in half, last, path, dance is like that in bad.
(3) The vowel in stop, hot is more like in quality to the short of the one in talk.	The vowel in stop, hot sounds like the short of the one in father.
(4) The vowels of take, made, and of note, rode are more diphthongal than in G. A.	The vowels of take, made, and of note, rode are less diphthongal than in S. B.
(5) The [j] sound is more usually retained in words like new, tune, due, suit.	The [j] sound is more usually dropped in words like new, tune, due, suit.
(6) The [h] sound is less often heard in words like what, which, when.	The [h] sound is more often heard in words like what, which, when.
(7) The [t] sound is seldom heard after the [n] in words like once hence.	The [t] sound is often heard after the [n] in words like once, hence.
(8) Words like dictionary, cemetery, dormitory have no secondary stress.	Words like dictionary, cemetery, dormitory have a secondary stress.

It may be worth while to note that General American is a direct descendant of the standard English in the seventeenth century, when America was settled, and not a variant form of English derived directly from the present-day Southern British English. The divergences mentioned above mostly arose from the fact that the speech of England, particularly of London, which had changed a great deal from the standard form, came to be regarded as a new received standard form, the old standard form continuing to be spoken mostly in the North and America. This accounts for the many striking resemblances of Northern British and other British local dialects to general American English. These features of resemblances that are seen today have disappeared from Southern British but have remained in General American.

3. Comparative Table of Types of Phonetic Symbols

The following tables contain only a few examples of those symbols that are based on the principle of using a district symbol, whether comprising a single letter or combination of letters, for each phoneme. This does not, however, imply that other systems of phonetic notation do not deserve our attention. Each system may have its special merit in respect of the purposes it is used to serve. For instance, Webster’s symbols that show the pronunciation by adding diacritical marks to the orthographic form may help to show the relation between the sounds and the spelling, though the system may result in encouraging the so-called spelling pronunciation at the expense of more natural and recommendable pronunciation of spoken English. Even the use of kona-gaki notation may be justified in thee very elementary stage as a feasible memory cure, if proper precautions are taken against distorting the pronunciation into a Japanaized one. It is a common sense to say that any letter can be used for any sounds, if only clear definition is given and a consistent use is made of it.

Generally speaking, however, the use of the International Phonetic Alphabet from an early stage of learning a foreign language has the approval of phoneticians and many progressive teachers. As early as in 1921 Dr. Harold E. Palmer said, “It has been ascertained experimentally that those who have been taught to read and to write a language phonetically become quite as efficient spellers as those not so trained. In many cases the phonetically trained student becomes the better speller.”⁴ Twenty-six years later, in 1947, P. A. D. MacCarthy says in an article in a professional journal, after discussing the use of phonetic transcription in the classroom in the initial stage, “At the next stage, the transition to ordinary spelling is made, systematically, going over again the language material that is already familiar. This serves as a valuable period of recapitulation. Experience has shown, moreover, that those who have been introduced to English spelling in this say, (i. e., after having worked with tile sounds only, for a preliminary period), so far from being confused by the change-over, actually make better spellers in the end (quite apart from making better speakers) than those who have been plunged into the intricacies of the current orthography from the start. This is doubtless due to the clear and orderly habits of thinking instilled by the systematic method of a well-managed transition.”⁵

If some teachers feel that phonetic spelling may interfere with the learning of orthographic spelling at some stage, such interference may be due to (1) teaching resulting from a lack of confidence and skill in the use of phonetic script and mishandling of the change-over, (2) lack of understanding or interest in the phonetic approach, (3) and consequent lack of understanding and interest on the part of the pupils, or (4) some cause quite divorced from the introduction of phonetic script. But even if there were interference with the learning of orthographic spelling, the pains of conquering the difficulty would be well compensated by the mastery of a good pronunciation. The difficulty, moreover, of learning new phonetic symbols will be minimized if an Extra-Broad transcription is used.

Table I Consonant Symbols

Key Words	Type I		Type II		Type III		Type IV	Type V	Type VI
pip	[p]		[p]		[p]		[p]	[p]	[p]
bob	[b]		[b]		[b]		[b]	[b]	[b]
tut	[t]		[t]		[t]		[t]	[t]	[t]
did	[d]		[d]		[d]		[d]	[d]	[d]
kick	[k]		[k]		[k]		[k]	[k]	[k]
gog		[]		[g]		[]	[g]	[]		[g]
fife		[f]		[f]		[f]	[f]	[f]		[f]
valve		[v]		[v]		[v]	[v]	[v]		[v]
thin		[θ]		[θ]		[θ]	[θ]	[th]		[θ]
than
say		[s]		[s]		[s]	[s]	[s]		[s]
zoo		[z]		[z]		[z]	[z]	[z]		[z]
show		[]		[]		[]	[]	[sh]
azure		[]		[]		[]	[]	[zh]		[]
how		[h]		[h]		[h]	[h]	[h]		[h]
church		[t]		[t]		[t]	[t]	[ch]		[]
judge		[d]		[d]		[d]	[d]	[j]		[j]
mum		[m]		[m]		[m]	[m]	[m]		[m]
nun		[n]		[n]		[n]	[n]	[n]		[n]
king		[]		[]		[]	[]	[ng]		[]
lull		[l]		[l]		[l]	[l]	[l]		[l]
way		[w]		[w]		[w]	[w]	[w]		[w]
whale6		[hw]		[hw]		[hw]	[hw]	[hw]		[hw]
yet		[j]		[j]		[j]	[j]	[y]		[j]
rate		[r]		[r]		[r]	[r]	[r]		[r]

Table Ⅱ Vowel Symbols

bee

[ii]

[i]

[]

[ij]

pity

[i]

[i]

[i]

[]

[i]

[i]

bed

[e]

[e]

[e]

[]

[e]

[e]

bad

[]

[a]

[]

[]

[a]

[]

palm

[aa]

[]

[]

watch (S. B.)

[o]

[o]

[]

[]

[o]

[]

-(G.A.)

[]

[]

[]

[o]

[]

paw

[oo]

[]

[]

lord (S. B.)

[oo]

[]

[]

-(G.A.)

[o]

[]

[r]

[r]

[r]

full

[u]

[u]

[u]

[]

[u]

[u]

fool

[uu]

[u]

[]

[uw]

bird (S. B.)

-(G.A.)

[-]

[r]

[r]

mother(S. B.)

[]

[]

[]

[]

[]

[]

-(G.A.)

[]

[-]

[]

[r]

[r]

cut

[]

[]

[]

[]

[u]

[], [o]

( also for G.A.)

pay

[ei]

[ei]

[ei]

[e]

[]

[ej]

by

[i]

[ai]

[ai]

[a]

[]

[aj]

boy

[oi]

[oi]

[i]

[i]

[oi]

[j]

now

[u]

[au]

[au]

[a]

[ou]

[aw]

go

[ou]

[ou]

[ou]

[o]

[]

[ow]

ear (S. B.)

[i]

[i]

[i]

[]

[i]

[i]

-(G.A.)

[i]

[-]

[r]

[r]

[ijr]

air (S. B.)

[e]

[e]

[ ]

[ ]

[ ]

-(G.A.)

[e]

[-]

[r]

[ar]

[ejr]]

four (S. B.)

[o]

[o]

[ ]

[ ]

[ ]

[ ]

-(G.A.)

[o]

[-]

[r]

[r]

[owr]

poor (S. B.)

[u]

[u]

[u]

[ ]

[ ]

[u]

-(G.A.)

[u]

[-]

[r]

[r]

[uwr]

art (S. B.)

[aa]

[]

-(G.A.)

[ ]

[-]

[r]

[r]

[r]

Notes on the Terms “Southern British” and

“General American”

The term “Southern British” (abbreviated S. B.) is used to designate that form of English spoken most commonly be educated people in Southern England including the capital. It is the form of English which is considered in England at least to be freest from Provincial features and which other speakers of English in Great Britain may endeavor to acquire. The geographical area of southern British could be roughly shown on the map by drawing a line from Warwick running due east and another running due south, according to Fuhrken.⁷

Most dictionaries published in England, including Daniel Jones’ English Pronouncing Dictionary, record the pronunciation of this form of English. Jones believes the type of pronunciation he has

recorded in his dictionary to be the type “most usually heard in everyday speech in the families of Southern English people who have been educated at the public schools,” the term ‘public school’ being used in the English sense. The pronunciation, however, is not confined to this class of people but is used also by those outside of this category.

The term “General American” (abbreviated G. A.) is used to designate that form of English spoken most commonly by people who live in about four-fifths of the area of the United States and who consist of two-thirds of the total population. Geographically it is described by A. C. Baugh as covering the Middle Atlantic States, i. e., New Jersey, Pennsylvania, and all of New York west of the Hudson; the states of the Old Northwest Territory (Ohio, Indiana, Illinois, Michigan, and Wisconsin), most of the region west of the Mississippi River, and a few other states.⁸

Most dictionaries published recently in the United States of America, including Webster’s International Dictionary of the English Language (2nd edition), record the pronunciation of this form of American English. Kenyon and Knott’s Pronouncing Dictionary of American English (G. & C. Merriam Co., Springfield, Mass. 1944) records G. A. pronunciation (termed North) in the first place and other variant forms if there are any. R. H. Gerhard, referring to G. A. pronunciation, says:

“In view of the vastly greater geographical extent of American English, it is understandable that there can be no single standard so conveniently formalized as we find for British usage. The closest approach to this, however, is what is commonly designated as General American pronunciation.”⁹

Remarks

The Symbols listed above are copied from the following sources:

Type I: R. H. Gerhard, A Handbook of English and American Sounds, Shimizu Shoin, 1949. Extra-Broad Transcription. Designed, to transcribe both G. A. and S. B.

Type II: N. C. Scott, University of London, known as Simplified Phonetic Transcription, and having the approval of Daniel Jones.

Type III: D. Jones. An English Pronouncing Dictionary, Dent, 1937. Broad Transcription.

Type IV: J. S. Kenyon & T. A. Knott, A Pronouncing Dictionary of American English, Merriam, 1944. Narrow Transcription.

Type V: E. L. Thorndike, Thorndike Century Beginning Dictionary, Scott, Foresman, 1945.

Type VI: Leonard Bloomfield, Language Henry Holt, New York, 1933.

In preparing these tables we have had no intention to recommend any one particular type, though it may be regretted that so far no standard system of notation has been set up either at home or abroad. It is hoped that some unified form should come to be used by all dictionary compilers and textbook writers. In order to avoid needless complications, and in view of the various cries for simplified notation, we have used tentatively Type I symbols in the following pages.

II. Description of English Sounds¹⁰

1. Vowels

Vowel sounds are characterized by the resonances formed in the mouth cavity, and these resonances depend primarily on the shape assumed by the mouth cavity in pronouncing a sound. In diagrams to show the vowel sounds, the “location” for each vowel represents the approximate position in the mouth cavity of the highest part of the tongue in the pronunciation of that sound. All English vowels are voiced, and care must be exercised against devoicing them ( as is often done in a Japanese word, such as shikashi) when they occur between breathed consonants.

(a) Short Vowels: [i] [e] [] [] [] [] [o] [u]

[i] is similar to short Japanese イ, but laxer and more central, so that it sounds somewhat intermediate between Japanese イ and エ.

[e] is similar to a short Japanese エ, but laxer and more open.

[] is between Japanese エand ア, with a “flat” or “shallow” quality produced by a slight raising of the sides of the tongue. It is a help, too, to draw back the corners of the mouth during the pronunciation of this sound.

[] is a little like Japanese ア, but the central part of the tongue is higher and the sound is more obscure and deeper. When unstressed, the sound is very short and obscure; when stressed, the tongue is somewhat lower and more retracted, and the jaws are wider apart.

[] is entirely different from any Japanese vowel. One way of forming this vowel is to raise the tip of the tongue and curl it more or less backward toward the roof of the mouth, without actual contact of the point. Another way of forming it is to raise the center of the tongue toward the roof of he mouth. In both ways, there is considerable tensing of the muscles in the pharynx. This sound does not occur in Southern British pronunciation, which generally replaces it with [].

[] is much like Japanese ア, but with the tongue more retracted and with considerable distension in the pharynx. The jaws, too, are usually more open than for any other English vowel. Short [] does not normally occur in Southern British pronunciation.

[o] is somewhat similar to Japanese オ, but more open. This sound is not greatly different from a variety of [] pronounced with rounded lips.

[u] is very similar to Japanese ウ, but with the lips rounded and drawn in at the sides.

(b) Long Vowels:

is similar to Japanese イー, but a little higher and more tensed.

resembles [] but is longer and tenser, with the central part of the tongue raised closer to the roof of the mouth and the jaws very close together. This sound does not normally occur in General American pronunciation.

is similar to [] but a little higher, longer, and more tensed. The jaws are very close together. This sound does not occur in Southern British pronunciation.

is very similar to [ ] but longer. The tongue is more retracted than for Japanese アー, with greater distention in the pharynx.

is similar to Japanese オー, but a little more open. The back of the tongue is higher than for short [o]-a little less so in American usage than in British-and the lips are more closely rounded. There must be no movement of the tongue or jaws during the pronunciation of this sound.

is similar to [u], but tenser and longer, with the lips more closely rounded and drawn in at the sides.

N. B. The longer varieties of [], sometimes transcribed as , and [e], sometimes transcribed as , are not listed above as they are used only as variant forms.

(c) Diphthongs: [ei] [i] [oi]; [i] [e] [o] [u]; [u] [ou]; [i] [e] [] [o] [u]

The English diphthongs, though written with two vowel symbols, are really monosyllabic glides between two vowel positions, with greater prominence at the beginning of the glide. These digraphs, therefore, are to be interpreted as composite symbols, the first element being given full vowel quality in pronunciation, while the second element serves only to indicate the direction and approximate termination of tongue movement in the glide. With this general explanation, and with the accompanying diagrams to show relative tongue positions, the diphthongs should require little additional description.

[ei] and [ou] both start from considerably higher tongue positions than for those for short [e] and [o], respectively, and the starting position fro [ou] is normally more central and a little lower in British usage than in American.

For both [i] and [u], also, the starting position of the tongue is commonly more advanced than for short [].

The diphthongs moving toward [] do not occur in General American pronunciation, though the combination [i] occasionally occurs in that form of speech as a dissyllabic juncture of two vowels, e.g., as in idea .

The diphthongs moving toward [] do not occur in Southern British pronunciation. When followed by a vowel, a diphthong moving toward []− i.e. [i], [e], [], [o], or [u]− and the corresponding combination of a vowel and [r] −i.e.,[ir],[er],[r],[or], or [ur] −are pronounced in practically the same way in General American speech; e.g., as in hairy [hi] and very [vri]. The former transcritption, however, appears to be preferable when the diphthong with [] is at the end of the root-word and the vowel that follows is party of the suffix added to the root-word, e.g., poorer [p], but rural [rrl]; bearing [bi], but burial [bril].

2. Consonants

Consonant sounds are characterized by differences in place and manner of articulation. Again, some are “breathed” (i.e., pronounced with breath only), and some are “voiced” (i. e., pronounced with breath to which tone has been added by vibration of the vocal cords). In all cases, however, it is important to dissociate consonants from any accompanying vowel sound and also to learn to join consonants together according to the English speech habits.

(a) Plosives: [p], [b]; [t], [d]; [k], [].

[p], [t], and [k] are breathed, while [b], [d], and [] are voiced. These sounds are practically the same as in Japanese but care must be exercised to avoid adding any vowel sound when they occur at the end of English words, and also to avoid making a plosive [] a nasal []when [] is really needed in a word.

The lips must be firmly closed for [p] and [b], the tips of the tongue placed against the upper teethridge for [t] and [d], and the back of the tongue raised to the soft palate for [k], and []. For all of these sounds the nasal passage is closed by the raising of the uvula.

(b) Nasals: [m], [n], [].

These are voiced sounds pronounced through the nose. For [m] the lips are closed, for [n] the tip of the tongue is placed against the upper teethridge, and for [] the back of the tongue is raised to the soft palate. When a vowel sound follows, [m] and [n] are similar to the Japanese consonants in ma-gy and na-gy respectively, while [] is similar to the consonant sound in ga as pronounced by many persons in the middle of a word-e. g., as in hagaki haaki. In final position all three are slightly lengthened but otherwise unchanged, and so are radically different from the Japanese [] (=ン), as in hon [ho]. The sound [] never occurs at the beginning of an English word.

(c) Lateral: [l].

This is a voiced sound pronounced around the sides of the narrowed tongue, with the tongue-tip touching the upper teethridge. In articulating this sound the back of the tongue may be either raised toward the roof of the mouth or lowered down away from it. Acoustically the former articulation gives a “dark” sound, and the latter a “light” or “clear” sound somewhat similar to Japanese ウ. Most people use the “light” l before vowels and the “dark” l before consonants or in final position.

(d) Fricatives: [f], [v]; [θ],; [s], [z]; [], []; [h].

[f], [θ], [s], [], and [h] are breathed sounds, while [v], , [z], and [] are voiced. All fricative consonants are produced by expelling the air-stream through a very narrow passage formed in the mouth or, in the case of [h], the glottis. They are sometimes called “continuants” because the can be prolonged as long as the breath lasts.

[f] and [v] are characterized by audible friction between the upper front teeth and the lower lip.

[θ] and are characterized by audible friction between the tip of the tongue and the upper front teeth.

[s] is the same as the breathed consonant sound in Japanese サ, ス, セ, ソ, with the breath directed along a narrow channel over the center of the tongue and the characteristic friction produced between the blade of the tongue and the upper teethridge. [z], which is voiced, is identical with [s] in every other way.

[] is similar to the breathed consonant sound in Japanese シ, but characteristically with stronger friction, produced between the front of the tongue and the fore part of the hard palate.

[] is voiced, but is identical with [] in every other way.

[t] and [d ], sometimes classified as affricate consonants, are formed in the same way as [] and [] respectively, with a preceding plosive stop made by bringing the tip of the tongue in contact with the teethridge before the fricative consonants.

[h] is a breathed glottal fricative, articulated by a partial closing of the vocal cords, in which neither the tongue nor the lips should have any part. It is, therefore, practically the same as the breathed consonantal part of Japanese ハ, ヘ, or ホ, but different from that of ヒ (where the friction occurs between the raised tongue and the roof of the mouth) or in フ(where the friction occurs between the two rounded lips). This sound never occurs at the end of an English word, though the letter h sometimes does in traditional spelling.

(e) Semivowels: [j]; [w]; [r].

An English semivowel is a voiced gliding sound in which the speech organs start from a vowel position and immediately change to that of whatever vowel it precedes. It is always short and nonsyllabic, and is used only in prevocalic position.

[j] is a semivowel starting from the position of a high, tensed [i]. When it precedes the vowel [i], it starts from a higher and more tensed position.

[w] is a semivowel starting from the position of a high, tensed [u]. When it precedes the vowel [u], it starts from a higher and more tensed position, with closely rounded lips.

[r] is a semivowel starting from the position of a high, tensed []. When it precedes the vowel [], it starts from a higher and more tensed position. In Southern British pronunciation, [r] is also sometimes fricative in initial position, with the blade of the tongue then raised closer to the teethridge.

III. Stress Marks and Tonetic Symbols

1. Methods of Marking Stress

“Stress is defined as the degree of force with which a sound or syllable is uttered.”¹¹ In polysyllabic words it is usually possible to distinguish many degrees of stress. If we take the word ability, we can hear that the strongest stress fails on the second syllable, the weakest on the third, and the next weakest on the first, or, to show this in figures we could say that the word ability has [3-1-4-2] stressing, 1 standing for the strongest, 2 for the second strongest, etc. But for practical purposes such fine-cut accuracy is not needed. The main thing a student wants to know in learning a word is which syllable in it carries the chief stress. Words that have two stressed syllables are said to have ‘double stress.’ Occasionally we distinguish three degrees of stress: ‘primary stress,’ ‘secondary stress,’ and ‘no stress.’ For example, the word examination contains syllables of ‘no stress,’ ‘secondary stress,’ ‘no stress,’ ‘ primary stress,’ and ‘no stress.’

One way of indicating stress in a phonetic transcription is to put a stress mark for primary stress and for secondary stress; or for primary stress and for secondary stress. Thus, ability is marked or and examination is marked or The use of vertical marks before the stressed syllables is the practice observed by most recent phoneticians in Europe and America.

In Japan there has been for many years another way of indicating stress, namely, by putting the stress mark for primary stress and for secondary stress right above the vowel symbol of the stressed syllable.¹² This may be justified by the fact that the center of the stress is always on the vowel, and also by the fact that the stress may actually begin midway in the articulation of a consonant, thus making it difficult to show exactly where the syllable is to be divided¹³. It may be added that in the case of diphthongs, also known as kinetic vowels, the mark is placed over the first element, because in English there is an intra-syllabic accent on the first component. For instance,

There are other methods of showing stress. Webster’s system has the stress mark for primary stress or for secondary stress placed after the stressed syllable instead of before it. It is unfortunate that there exist two entirely opposite positions for stress marks. To underline or encircle a stressed vowel or syllable may be mentioned as an optional way of marking stress in an unambiguous way.

2. Word Stress and Sentence Stress

Every English word whether mono- or polysyllabic, has at least one stressed syllable when it is pronounced separately. In most dictionaries the stress mark is not given in the case of monosyllabic words, but this does not mean that such words have no stress. Stress we hear in words pronounced separately is called ‘word-stress.’

Stress in English words is a very important feature of the language. It cannot be learned by means of simple rules. In many cases there is no rule as to the position of stress, and when rules can be formulated, they are generally subject to numerous exceptions. It is therefore necessary for us to learn the stress of each word individually.

When words are in a group, they are not pronounced in the same way as when they are pronounced separately. Some words lose their stress altogether and others retain theirs more or less. In general, the relative stress of the words in groups depends on their relative importance in meaning. The more important words are, the stronger is their stress. Usually words that convey important meanings are what are called content words: namely

1. nouns,

2. principal verbs (verbs which do not stand in an auxiliary position before other verbs),

3. adjectives,

4. adverbs of time, place, and manner,

5. demonstratives,

6. interrogatives, and,

7. indefinite pronouns.

Words that do not usually carry important meanings are what are called function words; namely,

1. pronouns (personal and reflexive),

2. auxiliary verbs (used before other verbs),

3. prepositions,

4. connectives (conjunctions, relative pronouns, etc.),

5. articles, and

6. adverbs of degree.

Sentence stress is by no means fixed in all contexts. As the center of the speaker’s attention is shifted according to the context, function words may also have sentence stress.

Examples of Sentence Stress Marks

What do you think of the weather?

I think it will be quite warm tomorrow.

This express train often arrives late.

He told me that that that that that boy used was right.

3. Nature of Intonation

Intonation is defined as “the variations which take place in the pitch of the voice in connected speech.”¹⁵

Every language has its characteristic tunes, its own typical types of tones, or in terms of modern phonetics intonation patterns or intonation contours.¹⁶ When one language is spoken with the intonation of another language, it is often difficult to understand it or at least gives the effect of strangeness or unnaturalness. In many languages, notably in English, intonation has an important role of adding various shades of meaning, showing the speaker’s attitude to the lexical or dictionary meaning of the words used.

In tone languages, of which Japanese is one, the intonation is fixed on words. Thus, in standard Japanese, the word hashi, bridge, is always pronounced with a lower tune on the first syllable in any kind of context, while hashi, chopsticks, is always pronounced with a high tune on the first syllable. On the contrary, in stress languages, of which English is one, intonation is never fixed on words when standing in context; it is changed freely in various ways according to context. Thus, the word pencil in a sentence like “I have a pencil” (and ordinary statement) is, as is the case with the word apart from its context, pronounced with a higher pitch (accompanying the stress) on the first syllable, while the same word in a sentence like “Have you a pencil?” (an ordinary question), is pronounced with a lower pitch (accompanying the stress) on the first syllable. Sometimes the variations of pitch (from high to low or vice versa) are great, sometimes slight, and sometimes more complicated than a mere rise or fall.

4. Methods of Marking Intonation

A problem that has troubled phoneticians very considerable is how to devise a practical system of recording intonation. It should ideally be easy to write, print, and read, but so far we have not arrived at an ideal method.

The following are some of the well-known methods devised by various phoneticians. They are listed approximately in a chronological order, but neither exhaustiveness nor exact arrangement is claimed.

Henry Sweet¹⁷ devised a system to indicate a rising tone , heard in questions, such as What?; a falling , in answers to questions such as No!; a falling-rising , as in Takecare! ; and a rising-falling , as in Oh!: as an exclamation of sarcasm. The tone marks were put before the word they modified; if they modified a whole sentence, they were put at the end of it. If no tone-mark was written, a comma or a question mark implied a rising tone, a colon or semicolon a falling tone.

Subsequently Daniel Jones¹⁸ demonstrated by drawing speech curves on a musical staff that intonation was in almost constant movement. He obtained these curves by picking up the needle of a gramophone at various points during the playing of each syllable, and charting the pitches which he heard. Thus, the sentence, “Then he’ll be here by twelve, and we’ll have the service at a quarter past,” was recorded like this:¹⁹

H.O. Coleman, taking a hint from Jone’s curves, devised what he called numerical notaion.²⁰ He divided the pitch variations into nine degrees, giving each syllable one or more of the figures 1 to 9. He hoped his method of notation would be found clear and sufficientlyt exact, while involving no typographical difficulties. To cite a few examples adapted from his article:

6-1　　 1　　 3　4　　 9　　 1　　 1　 1-8

Why, you never had one before.

2　　　 3　　　　 5　　　2　　2　　2　　 4-6　　 8-1

They answered with a prolonged “Oh”

4-2-3　3　 3 　　　4　　4　　　5　　　　 6　 　　 1

Eight is the number she SPOKE of.

(The capitals indicate sentence stress.)

In 1922 Harold E. Palmer, in his English Intonation²¹ showed that the sentence could be broken into several parts of one or more syllables each, and that each part might have its own intonation contributing to the whole. These parts he called the nucleus, head, and tall of the intonation. By means of a horizontal or slanting bar for the head and about half a dozen kinds of arrows for the nucleus he managed to describe many small variations of English intonation, the marks leading themselves to facility of printing. Later,²² in 1933, the same author selected six of the most useful intonation patterns giving them somewhat fanciful terms, Cascade, Dive, Ski-jump, Wave, Snake, and Swan. Examples of his tonetic notation from his latter work are given below.

Note that in Palmer’s notation the initial syllables or words left unmarked have a medial or a low tone, that the falling tone indicated by or occurs in the syllable that follows the mark, the remaining syllables keeping a low tone all through, and that the modulations shown by the head-marks and , and the rising arrows and are distributed all over the following syllables.

In the same year that Palmer’s English Intonation appeared, Walter Ripman published some material with intonation marks.²³ These symbols were similar to H. O. Coleman’s except that they were mostly of three figures 1 to 3; very occasionally figures 4 and 5 being used. This notation of Ripman’s might well give an indication that English intonation is subject to an analysis in terms of four levels of tone. Examples from his passages analyzed are as follows:

1　　 2　 2　 1 　1 　3　 2　1

The labour of the day is done.

　1　 2　 3　 3　　 3 3 　1 　2　　 2 　2 　 2　2 2　 2-1

The mysterious spirit has gone to take its airy rounds.

　4 　　4 　2 　3 　　3 　3 　　 3 　3　-　　1

Write me as one that loves his fellow-men.

　2　 5　 4　 5　 5　　　 4　　　 3 2　 -　 1

And lo! Ben Adhem’s name led all the rest.

In 1926, Lilias E. Armstrong and Ida C. Ward published an entirely different analysis which made an influential contribution to the field.²⁴ Their system was later adopted by Daniel Jones, in his Outline of English Phonetics, 3rd edition, 1933, and by Maria Schubiger, in her Role of Intonation in Spoken English, Cambridge, 1935, and other writers. They utilized a stream of dots and curves corresponding to syllables between two (or three) lines representing high and low (and medium) pitch, and limited the basic structure of English intonation to two tunes and modifications of them. The following examples are taken from Jones’s An Outline of English Phontetics,²⁵ Chap. XXXI.

Note that in this marking the bigger dots indicate that the corresponding syllables are stressed.

Last on our list comes that most recent and important work done in the field of American English by Kenneth L. Pike.²⁶ He approached the problem from a phonemic point of view, so to speak, and selected a limited number of intonation types which would serve the foreign student most practically and simply. He holds that a frame of the relative pitch levels that are significant in American English consist of four steps:

No. 1 = extra high

No. 2 = high

No. 3 = medium (Basic voice level)

No. 4 = low

In natural speech there well always be fluctuations, but these fluctuations are not significant; that is, they do not matter linguistically. It is the points where the pitch levels change that are signifficant.²⁷

Pike transcribed intonation in two ways; either by means of lines, as:

or by numerical figures, as:

The sign shows that the ‘primary contour,’ or as Palmer and others put it, the nucleus of the intonation pattern, begins there. The hyphen after or before a figure shows that the particular pitch is part of an intonation contour to be complete with the following or preceding pitch level or levels.

Under the pressure of teaching English as foreign language Pike selected the following four primary contours:

Each of these four contours may also appear with the precontour 3-. The following examples are given to show these selected contours.

Note: A slanting line shows that the pitch rises or falls within the syllable marked. In numerical notation a single slanting line shows the division of possible breath-groups and a double one the end of an utterance.

5. Comparative Tables of Types of Tonetic Notation

Partly by way of summary of the preceding section and partly for purposes of comparing different types of notation given to the same kind of pitch modulations we give below in tabular form two illustrative sentences with tonetic symbols.

Note: As it is impossible to find sentences treated commonly by the individual authors mentioned above, we have marked the sentences as nearly as we could imagine the authors would have done.

6. Comparison of So-called British and American Intonation

There are doubtless considerable differences between British and American habits of intonation. Very little has been done, however, in comparative studies of the two. Nor is it easy to compare them, because of the scantiness of literature, especially of American intonation now available.

It has been pointed out that the intonational patterns of British and American English are essentially the same.²⁸ This means that none of the differences is ‘significant’ or important enough to change the meaning. In other words, the two habits of intonation are mutually understandable to both types of speakers.

One of the differences pointed out by many writers is that American speech normally consists of a smaller range of tonal rise and fall, particularly in regard to the higher pitches. This is obviously one of the reasons why American English sounds to our ears more monotonous than British English.

Another of the differences often pointed out is that in American speech the ‘precontour’, or the unstressed head of the phrase, is normally mid or low pitched and very seldom high pitched as is often the case with British speech. It may safely be stated that the patterns which H. E. Palmer indicated as containing a high head (‘Cascade’ and ‘Wave’) are normally replaced in American speech by those starting on a much lower-and often less stress-tone.²⁹

For example:

British: I don’t think it’s true.

British: Are you there?

Similarly, it seems to us that Palmer’s ‘Ski-jump’ and his ‘Snake’ pattern with a high-pitched head have no place in normal American speech. For example:

British: How strange it is!

A third point of difference we may note is that in British speed changes of tone more normally proceed by glides, or continuous rise or fall, especially in the primary contour, while in American speech they more normally proceed by leaps or steps when the primary contour falls on a word of more than one syllable or a phrase.

Thus:

The above is a description of some of the features more or less characteristic of the so-called British and American English. Warning is however needed against hasty conclusions that the two types of speech are clearly separated by the Atlantic Ocean. Both forms are used within the respective countries.

7. Specimen Passages with Stress and Intonation Marks

A. After K. L. Pike

B. The Same Passage with Tone Marks after H. E. Palmer

(Transcribed in Phonetic Alphabet)

 

IV. Other Problems of Phonetic Sequence

1. Length of Sounds

The length of a sound, or the length of time during which a sound is held on continuously in utterance, usually varies according to its phonetic context. A trained ear can distinguish many degrees of length, but for practical purposes it is sufficient to distinguish two degrees, long and short. When it is desired to distinguish an intermediate degree, this intermediate degree may be termed ‘half-long.’³⁰

A. Length of English Vowels

The main facts concerning length in English sounds are given in the following rules.³¹

Rule 1. Those vowels called ‘long vowels’ and written in some types of notation with the length-mark and also the ‘diphthongs’ are longer than the other English bowels in similar phonetic context.Thus the vowels in bead, bard, board, food, bird, bowed are longer than the vowels in bid, bed, lad, rod, good, bud.

In General American the four vowels in pit, pet, put, nut are shorter than the rest. Other vowels, such as those in glass [ls], box [bks], dog [do], are generally pronounced fairly long even if they are not followed in written symbols by the length-mark.³²

Rule 2. All English vowels, whether long, diphthongal, or short, are shorter when followed by a voiceless consonant than when they are final or followed by a voiced consonant. Thus the vowels in seat, pierce, hit are shorter than the corresponding vowels in see or seed, pier or beard, hid.

Rule 3. All English vowels, whether long, diphthongal, or short, in strongly stressed syllables are shorter when a non-stressed syllable immediately follows in the same word. Thus the first vowel in reader is shorter than that in read, and the first vowel in running is shorter than that in run.

Rule 4. All English vowels, whether long, diphthongal, or short, are shorter in non-stressed syllables than in strongly stressed syllables. Thus the vowels in the unstressed syllables in although, partake, idea until are shorter than the corresponding vowels in also, party, island, under. The so-called ‘schwa’ [], when representing an obscure vowel as distinguished from the sound represented in some types of notation by the symbol [ ], is especially short and even drops out altogether in an unstressed syllable, as in of course [()vks or ()fks], history [hst()ri]. The same vowel which occurs in a stressed syllable is regarded as normal in the north of England and the United States of America,³³ this obscure vowel never occurring in a stressed position in British Received Standard.

B. Length of English Consonants

Rule 5. Final consonants are longer when preceded by a short vowel than when preceded by a long vowel or diphthong. Thus the [n] is longer in pen than in pain or seen, and the [l] is longer in hill than in heel or hole.

Rule 6. The consonants [m], [n], [1], are longer when followed by voiced consonants than when followed by voiceless consonants. Thus the [m] in number is longer than that in jumper, the [n] in wind is longer than that in wink, and the [l] in hold is longer than that in bolt.

2. Sound-Junction and Word-Linking

Sounds are usually affected more or less by other sounds near to them in the phonetic sequence. The chief factor in the occurrence of variant sounds in a phoneme is the nature of the adjacent sounds. This will be illustrated in the following paragraphs. Sounds are affected by their adjacent sounds whenever they occur in a continuous sequence, both within a word and in neighboring words. Sounds are also affected by their position in a word, by the degree of stress with which they are pronounced in a word, and so forth. When sounds are unstressed the tongue is more lax, and hence undergo phonetic changes.

A. Use of Incomplete Plosive consonants³⁴

When a plosive consonant is followed by another identical or different plosive or by [t] or [d], the articulating organs take up their position for the second consonant before the articulation of the first is completed. Thus the plosion of the first consonant is either extremely weak, or completely inaudible, or, in the junction of two identical or similarly articulated plosives, the stop is held on from the beginning of the first to the release of the second. To show this graphically,

No Separation of the vocal organs takes place between the two plosives [pp], [pb], [bp], [tt], [td], [dt], [tt], [dd], [dt], [kk], [k], [], [k].

For example:

lamp-post [lmppoust] ([p] lengthened)

went down [wnt dun] (Vocal cords begin to vibrate in the middle of the stop.)

bed-time [bdtim] (Vocal cords stop vibrating in the middle of the stop.)

B. Nasal and Lateral Plosion³⁵

When a plosive consonant is immediately followed by a nasal, as in button written [rtn], sudden shopman [pmn], oatmeal [utml], the plosive is not pronounced in the normal may. The explosion in such plosives is produced by the air suddenly escaping through the nose, instead of the mouth. Many Japanese are apt to pronounce them in the normal way and to insert erroneously the vowel [] between the plosive and the nasal.

When [t] or [d] is immediately followed by the lateral consonant [l], as in little [ltl], middle [mdl], the plosive is not pronounced in normal way. The explosion of the plosive is not pronounced in the normal way. The explosion of the plosive in such groups is produced through the sides of the tongue, with the tongue-tip pressed against the teethridge. If the plosive is pronounced normally, as is often the case with many Japanese, the effect tends to be an erroneous insertation of the vowel [].

C. Partial Devoicing of Voiced Consonants

In English, most voiced consonants, more especially the voiced plosive [b], [d], [], and voiced fricatives [v], [z], [], , are fully voiced only when they come between two other voiced sounds. In initial and final position they are regularly ‘devoiced’ to a greater or lesser extent. This may be shown for the word birds [brdz] by a diagram, thus:³⁶

Note: Devoiced articulation is shown by —— in contrast with the voiced articulation shown by .

Many Japanese are apt to use fully voiced consonants where partially devoiced ones are needed, so that in their pronunciation of a final voiced consonant a slight additional [] is hard. It may be noted in passing that the degree of devoicing varies according to individual speakers, the style of speaking, the phonetic context, etc. With some speakers voiced consonants, particularly voiced fricatives, often sound as if they were voiceless.

Partial devoicing of English voiced consonants takes place also in the articulation of [l] or [r] before a strongly stressed vowel, when it is preceded by [p], [t], or [k]. Thus play, try, and cry might be transcribed and ³⁷ respectively, apart from the length of the consonants. In a diagram these words may be shown thus:

D. Juxtapositional elision and Assimilation

(1) Elision is “the disappearance of a sound.”³⁸ It is important to know in what kind of phonetic sequence a sound which exists in a word said by itself is dropped when a word is linked with another.

There are many kinds of such elisionl, which often escape the notice of foreign students of English. The degree and frequency of elision depends chiefly on the speed of speaking; the more rapid the speaking is the greater the frequency of elisions. The following are examples of juxtapsidonal elisions commonly made in ordinary (not rapid) speech:

kind man (elision of [d])

sit down (by some people; elision of [t])

take care [tik] (elision of [k])

In deliberate speech these juxtapositional elisions are less frequently heard.

(2) Assimilation is “the process of replacing a sound by another sound under the influence of a third sound which is near to it in the word or sentence.”³⁹ It is important to know how sounds are assimilated by their neighbors in a particular sound-sequence.

There are many types of assimilation. The most important are (1) assimilations of breathed consonants to voiced ones, and voiced consonants to breathed ones, (2) assimilations affecting the position of the organs in pronouncing consonants.

Examples of assimilation:

(1) (a) The reduced forms of is and has.

Tom is here. [tm z h.]

Tom has been here. [tm z bin h.]

Jack is here. [dk s h.]

Jack has been here. [dk s bin h.]

(b) The ending ‘-s’

(Plurals) boys [boiz], pens [penz], knives [nivz].

books [buks], caps [kps], hats [hts]

cf. faces [fisiz], roses [ruziz], houses

(Posessives) girl’s [lz], girls’ [lz], Bob’s [bobz], Mary's [mriz].

cook’s [koks], cooks’ [koks], Jack’s [dks], Bett’s [bets].

cf. horse’s , horses’ , fox’s [fksiz], foxes’ [fksiz], Jones’s [dunziz], Jones’ [dunz].

Note: The ending -s of the Third Person Singular Present Verb is treated in the same way as that of the noun.

(c) The ending ‘-ed’.

showed [oud], robbed [robd], begged [bed], looked [lukt], dropped [dropt], missed [mist], wished [wit].

cf. wanted [wntid], banded [bndid].

(d) The word ‘used’.

 

Compare		The knife was used to cut it.
		I used to cut with a knife.

(e) The word ‘have’.

I have to ask him.

I have to do it.

Note: In ‘have to’ the pronunciation [hf tu], [hf t] is common in ordinary speech in both the United States and England.⁴⁰

(f) Miscellaneous.

Compare		the news in the paper
		the newspaper41

It cost fivepence.

(2) Replacement of [s] by [], or of [z] by [].

horse-shoe

this shop

Yes, she is. [j iz.]

Does she?

butcher’s shop [bt p]

E. Some Cautions concerning Word-Linking

In English all words and syllables beginning with a vowel should in connected speech be normally linked with the final sound of a preceding word or syllable. This simple rule is not always observed in the pronunciation of many Japanese, the effect of which is an insertion of a glottal stop ⁴² where no such sound should occur. Thus It is on an armchair is in their pronunciation whereas it should be or more naturally,

When a word ending with the letter r (e) is immediately followed by a word beginning with a vowel sound, then an [r] sound is generally inserted in the pronunciation of Southern British. Thus far by itself is pronounced , but when it is followed by away the two words are linked by an [r] sound, as

There are, however, special circumstances in which a final r(e) is not pronounced in the linking. The principal cases are (1) when the vowel in question is preceded by [r], for example, the Emperor of Japan nearer and nearer [nr nd nr]; (2) when a pause is possible between the two words (even if no pause is actually made), for example, he shut the door and went away

It is to be noted, however, that in Southern British the insertion of a linking r is not essential, there being many people who do not insert it where others do.⁴³ In General American no linking r as such is necessary, as all final r’s in the spelling are sounded.

3. Relation between Rhythm and Sentence-Stress

There is noticed a strong tendency towards rhythmic regularity of strong stressed in connected English speech. This gives rise to the following important variations in the normal stressing of words and syllables.

(1) “Double-stressed” words, or words having two strong stresses, may lose the first stress when immediately preceded by another stressed syllable, or the y may lose the second when immediately followed by another stressed syllable. Thus the word fifteen may lose its first stress in a ten fifteen(10.15) train , and may lose its second in fifteen children .⁴⁴ The reason is that concurrence of stresses tends to make the rhythm incongenial to the ear of English-speaking people.

(2) When one word qualifies another, both words normally have a strong stress. Examples: Bank Holiday , New York , north wind . However, groups of words such as these may lose one of their normal stresses: example: last Bank Holiday he went to New York City , the strong North Wind said .

(3) Attention should also be paid to the very common combination of verb+adverb. Both the verb and the adverb normally bear strong stress, as in come in sit down [st dun], but in special rhythmic context one of the stresses often disappears, as in Do come in! , I can’t sit down He put his hat on , but He put it on .

4. Strong and Weak Forms

A number of very common English words of mostly one syllable have more than one pronunciation, the type of pronunciation in actual speech depending on the position in a sentence, the rhythmic context, and other factors. These words have what is called a ‘strong form’ and one or more 'weak forms.’ Of these about fifty most important ones are give in the list.

The strong form of a word is the pronunciation it usually has when said by itself, as an isolated word. This form is used in connected speech when the word is stressed. It is also used sometimes at the beginning or end of a breath-group, or a group of words to be spoken in one breath, even when the word bears no stress. For example:

Who is coming? I am.

Can anyone speak English?

Yes, i can.

The weak form or forms of a word are used only in unstressed positions. In natural speech the weak forms of these common words are for far more frequent occurrence than their strong form. Even in a reading style, which is much slower than a normal speaking style, the use of weak forms generally sounds more normal than the consistent use of strong forms.

Many of these words, in their weak forms, have the unstressed vowel [] in S.B. and [] (or [] if spelt with r)in G.A. The following are the chief ones that belong to this category: a, an, and, are, as, at, can, do (before consonants), for, from, had, has, have, of, sir, than, that, the (before consonants), them, to, (before consonants), was, were. (For full list see next section.)

A few words with in the strong form have a weak form with [i]; for example, the has a weak form used before a vowel, as in the others The following are the chief words in which the weak forms have [i]:-be, been, he, me, she, the, (before vowels and occasionally before [j]), we. The word been in the strong form is normally pronounced in General American with short [i].

A few words with in the strong form have a weak form with [u]; for example, the word to is pronounced [tu] in something to eat Chief words of this group are: -do (before vowels and before [w]), to (before vowels and in final position), you.

Some words have weak forms with no vowel at all; for example, the first do in How do you do? Chief words of this group are:-am, and, do, had, has, have, is, not, Saint, shall, should, us, will, would.

5. List of Strong and Weak Forms

Note: (S) stands for strong form and (W) for weak form.

(1) Articles

(2) Verbs

Note: The weak forms are not used after [s], [z], [], [], [t], and [d].

Note: The strong form [we] is occasionally used in S.B., but this is rarely used in G.A.

(3) Pronouns

　　

Note: In initial positions this form is always normal.

Note: “Let’s...” is the form used in proposing to do something.

Note: ’em is the weakened form of an old pronoun hem, meaning ‘them,’ though it is popularly considered as the abbreviation of ‘them.’

Note: A demonstrative ‘that’ has no weak form.

Note: ‘that’ as a relative noun and a conjunction is normally pronounced in the weak form in a sentence.

(4) Prepositions

(5) Conjunctions

Note: The conjunction ‘that’ is normally pronounced in the weak form in a sentence.

(6) Miscellaneous

Note: The strong form is used only in slow deliberate speech.

Note: ‘There’ as an adverb of place has no weak form. Compare the above example with “There is the sea!”

1 A ‘distinctive’ sound or ‘phoneme’ is a sound which is capable of distinguishing between different words a language. A sound in a language may vary a good deal according to different individuals and, even with one individual, according to its different phonetic surroundings or conditions. But if the particular phonetic differences do not interfere with an easy and spontaneous reception on the part of the hearer, they are usually ignored. A group of sounds that are phonetically different but psychologically the same and never distinctive of meaning is called a phoneme. Thus, when we speak of the “l phoneme,” we refer to all the varieties of l sound in the language of a single speaker, or in a uniform dialect, that occur under different phonetic surroundings or conditions but are regarded as the “same sound.” It may be noted that a phoneme as an abstract entity as described above allows for different interpretations. See W. F. Twaddell, On defining the Phoneme, Language Monographs, Vol. 16, Baltimore, 1935

2 E. Leonard Tibbitts, A Phonetic Readerfor Foreighn Learners of English, Heffer & Sons Ltd., Cambridge, 1946

3 N. C. Scott, English Conversations, Preface, Heffer & Sons LTD., Cambridge, 1942

4 Harold E. Palmer, The principles of Language Study, George G. Harrap & Co. Ltd., London, 1921, p. 153

5 P.A.D. MacCarthy, “Phonetic Transcription and the Teaching of Pronunciation,” in English Language Teaching, Vol, II, No. 1, The British Council, 3 Hanover Street, London W 1, September 1947, p.18

6 In the pronunciation of those who aspirate the sound; otherwise pronounced [w].

7 G. E. Fuhrken, Standard English Speech, Cambridge University Press, 1932, p.15

8 Albert C. Baugh, History of the English Language, D. Appleton-Century Company, 1935, p. 448. Quoted by permission of the publishers.

9 Robert H. Gerhard, A Handbook of English and American Sounds, Shimizu, Shoin, Tokyo, Copyright, 1949, p.86

10 This is largely based on the material prepared by R. H. Gerhard with his kind permission.

11 D. Jones, An Outline of English Phonetics, W. Heffer and Sons Ltd. Cambridge, 1936, p. 227

12 See Sanki Ichikawa, An English Pronouncing Dictionary, Kenkysha, Tokyo, 1926; A. S. Hornby, E. V. Gatenby, and A. H. Wakefield, Idiomatic and Syntactic English Dictionary, Kaitakusha, Tokyo, 1942

13 See G. E. Fuhrken, Standard English Speech, Cambridge, 1932, pp. 86-87

14 Be and Have used as principal verbs are exceptional. They are often unstressed.

15 D. Jones, An Outline of English Phonetics, W. Heffer and Sons Ltd., Cambridge, 1936, p.255

16 Daniel Jones, Harold E. Palmer and others have used the term ‘pattern’ probably borrowing it from psychology. Kenneth L. Pike, in his Intonation of American English, Michigan, 1947, uses the term ‘contour.’

17 See H. Sweet, A primer of Spoken English, Clarendon Press, Oxford, 1890, pp. 32-33

18 See D. Jones, Intonation Curves, Leibzing and Berlin, 1909

19 “Intonation Curves”, in D. Jones, The pronunciation of English, Cambridge, 1919 (2nd Ed.), p.96

20 See H. O. Coleman, “Intonation and Emphasis,” in Miscellanea Phonetica, Association Phontique Internationale, 1914, pp.6-26

21 See H. E. Palmer, English Intonation with Systematic Exercises, Cambridge, 1922

22 Harold E. Palmer, A New Classification of English Tones, The Institute for Research in English Teaching, Tokyo, 1933, pp. 6, 11, 14, 18. 25, and 30

23 Walter Ripman, Good Speech, Dent & Sons, London, 1922, pp. 59-73

24 See Armstrong & Ward, A Handbook of English Intonation, Teubner, Leipzig, and Heffer, Cambridge, 1st Edition, 1926, 2nd Edition, 1931

25 D. Jones, Ibid, pp. 259,262

26 See Kenneth L. Pike, The Intonation of American English, University of Michigan Press, 1945

27 See Charles C. Fries, Teaching and Learning English as a Foreign Language, University of Michigan Press, 1947, p.21

28 See Robert H. Gerhard, “Brief Notes on ‘A New Classification of English Tones,’ (by H. E. Palmer) with special reference to American Intonation,” The English Review (Tokyo University of Literature and Science), October 1945, p. 357

29 See R. H. Gerhard, op. cit., p. 358

30 See D. Jones, An Outline of English Phonetics, p. 216

31 See D. Jones, op. cit. pp. 216-220

32 See Leonard Bloomfield, “The Stressed Vowels of American English,” Language, Vo1. XI, No. 2, June, 1935, p. 101

33 See Robert H. Gerhard, A Handbook of English and American Sounds, Shimizu Shoin, 1949, p. 118,

Daniel Jones, Differences between Spoken and Written Language, Association Phontique Internationale, 1948, p.7,

and Charles C. Fries, Teaching and Learning English as a Foreign Language, University of Michigan Press, 1947, p. 12

34 See D. Jones, An Outline of English Phonetics, W. Heffer and Sons Ltd., Cambridge, 1936, pp. 134-4; P. A. D. MacCarthy, English Pronunciation, Heffer, 1947, pp. 114-5

35 See D. Jones, An Outline of English Phonetics, W. Heffer and Sons Ltd., Cambridge, 1936, pp. 144-5

36 See P. A. D. MacCarthy, op, cit., pp. 99-100

37 The Symbol underneath a phonetic letter signifies a devoiced variety of a sound.

38 D. Jones, An Outline of English Phonetics, W. Heffer and Sons Ltd., Cambridge, 1936, p. 214

39 D. Jones, op. cit., p. 202

40 See Kenyon & Knott, A Pronouncing Dictionary of American English, Merriam, 1944, p. 198

41 In the United States [n(j)zpeip] is also used.

42 See under “Glottal stop” in The Kenkysha Dictionary of English Philology, Kenkysha, Tokyo, Copyright 1940, 1949. Also see under Daniel Jones, An Outline of English Phonetics, W. Heffer & Sons Ltd., Cambridge, 1936, p. 138

43 See D. Jones, An Outline of English Phonetics, 1936, p. 182

44 For other examples see Jones, An Outline of English Phonetics, 1936, pp. 240 et seq.