1156 W4lh copy 2 A A = s: cz o Ai = en — o s X m 3D : I „, | m r% ^S ^^= CD 3 = ^^= .:_" 7 8 THE LIBRARY OF THE UNIVERSITY OF CALIFORNIA LOS ANGELES This ' SOUTHERN BRANCH, UNIVERSITY OF CALIFORNIA, LIBRARY, (LOS ANGfcLES, CALIF. THE N. E. A. PHONETIC ALPHABET WITH A REVIEW OF THE WHIPPLE EXPERIMENTS BY RAYMOND WEEKS JAMES W. BRIGHT CHARLES H. GRANDGENT \ i »' j THE NEW ERA PRINTING COMPANY LANCASTER, PA. 1912 1 I . I z, i PREFATORY NOTE. The experiments with the N. E. A. phonetic key alphabet, con- ducted by Assistant Professor Whipple, of the Cornell Psychological Laboratory, reported by him in a pamphlet recently issued, have had, with some people interested in the adoption of a scientific phonetic alphabet, an influence so out of proportion to their value and signifi- cance, that it seemed a possible service to the cause of education if one were to point out the inadequacy of Dr. Whipple's tests. The occasion has been made use of to suggest also some of the larger, more serious aspects of the problem. These may appeal to the schoolmen and educators who, as members of the N. E. A., are directly interested and concerned in the success of the alphabet, and aid them to recog- nize the advantages that may be secured to American schools by the adoption (A a scientific phonetic alphabet for educational purposes. Three appendixes offer information and some suggestions perti- nent to the discussion. January 22, 1912. THE N. E. A. PHONETIC ALPHABET In November, 1911, appeared a pamphlet entitled "Relative Efficiency of Phonetic Alphabets" (Baltimore: Warwick & York, Inc.), by Assistant Professor Guy Montrose Whipple, giving the results of a series of experiments in the Educational Laboratory of Cornell University, to test the relative merits of one of the key- alphabets in popular use (the Webster key) and the Proposed Key of the N. E. A. Committee of the Department of Superintendence, recommended by action of the Department of Superintendence at Mobile, February, 1911, to the National Education Association at its annual session last July. Dr. Whipple has "no hesitation in de- claring that the key proposed by the Committee of the National Education Association is inferior for pedagogical purposes to the Webster Key now in common use." Inasmuch as this pamphlet from an expert in psychology must be felt by many to offer scientific support to the objections that have appeared in a number of school journals, it may not be out of place (I) 1) to restate the opinions and aims of those who favor the general introduction of a simple, rational, and scientific phonetic alphabet, 2) to point out the deficiencies of keys of the Webster kind for really effective work, whether in primary or more advanced language study, and 3) to indicate what place phonetics, and the use of a phonetic alphabet, play in the remarkable advance of training in the living languages that has during the last generation been made throughout Europe; and then (II) to point out how far from conclusive and in fact how misleading at times are the results reached by Dr. Whipple. One would think that the overwhelming philological and phonetic authority behind the movement to adopt a scientific alphabet for general phonetic use would in the mind of a scientifically trained man of whatever other field of scholarship raise doubt of the sound- 5 6 THE N. E. A. PHONETIC ALPHABET ness of the ground upon which opposition to such a key took its stand. And it perhaps required some courage on the part of Dr. Whipple to put himself on record as aligned with various other in- fluences against the sense of all American and English philological and phonetic scholarship. There is not a reputable American or English philologist or phonetician that does not advocate the universal adoption of a scientific phonetic key, and declare a key of the Webster kind unfit for scientific uses.* There are no indications in his pamphlet that Dr. Whipple is conscious of this weight of expert opinion in favor of a scientific key. But a brief review of the movement will enlighten anyone on that point, and will reveal further the unanimity of aim and method in the actions of the associations that have joined in the reform. (1) Origin op the N. E. A. Alphabet The movement began in 1877, when the American Philological Association agreed upon an alphabet, the distinguishing features of which were the use of the vowel letters a e i o u in their Latin values, and the addition of three new forms a e u . There were thus eight vowel symbols, aaeioeuuf, which, marked (with a macron or otherwise), represented the long vowels heard in the words art air they eve note north rude burn ; unmarked, they represented corresponding short vowels. To make the alphabet easy of adoption for popular use, some consonant signs were retained that were later discarded, such as c as an alterna- tive to k , q and x as alternatives to kw and ks , and certain digraphs. After this first step the movement in this country for half a * The list of experts who, as members of the various committees, worked together to reach a common basis that should be at once rational, scientific, and practical, is given in Appendix I (page 83), and their professional standing may be learned from Who's Who. The list ought to impress every intelligent man with the presumptive good sense of their conclusions. t This symbol u , for the vowels heard in bum and but, is in some publications given the form of a small capital v. The original form, however, is the better. THE N. E. A. PHONETIC ALPHABET 7 generation appeared to make little progress; * but the leaven was working. With the further development of the science of phonetics, and a consequent significant change in the methods and aims of language study, a need was felt throughout a wider educational circle for a rational scientific alphabet precise enough for elemen- tary instruction in phonetics and simple enough for the widest pop- ular use. The Department of Superintendence of the N. E. A. took the initiative in calling a conference in Boston in 1903, in which representatives from the N. E. A., the American Philological Asso- ciation, and the Modern Language Association took part.f The conference at once accepted the 1877 alphabet as right in principle and as satisfactory in most of its details, and appointed a Joint Committee from the three associations to prepare a report that should embody the aim of the conference. That aim was "to agree upon and promote the general adoption of the best set of alphabetic symbols to be used in dictionaries and text-books for the accurate denotation of the sounds heard in English speech." The report of this Joint Committee, adopted after exhaustive study, and published in the fall of 1904, gave a careful review of the details of the problem, and submitted an alphabet for acceptance by the three associations. The revised alphabet was again con- sidered by a committee of the A. P. A. and the M. L. A., and later by a committee of the Department of Superintendence. The final revision by these committees took two directions. The former body changed the value of the sign a in order to give the symbol its original significance — the vowel sound in art, introduced the ligature se from old English manuscripts to represent the value of a heard in bat, and substituted j for the y-sound heard in you ; they chose the forms tj" and dg for ch as in church and j as in judge, allowing no alternatives ; they preferred ju to iu as the more exact representative of the sound of "long u." These changes were felt to be, in the main, in the direction of greater sim- plicity and greater historical accuracy. The N. E. A. committee, * A modification of the alphabet was used in the Funk & Wagnalls Standard Dictionary, 1890. t The American Dialect Society, and the U. S. Geographical Bureau, as well as various scientific and official bodies in England, had already acted upon the lines of the Philological Association's reform, and had adopted similar "scientific" alphabets. 8 THE N. E. A. PHONETIC ALPHABET on the other hand, retained the use of a as the symbol for the vowel sound in art, and a for the vowel of bat (in agreement here with international usage), used sh and dh for the consonants of she and church, instead of the M. L. and A. P. J and tj , fh and th for the M. L.-A. P. p and 5 . That is, the N. E. A. changes were in the direction of rendering the new alphabet no more strange to an eye accustomed to English spellings than was absolutely necessary. These two guiding principles, that the alphabet should be his- torically and scientifically justified, and that it should be also feasible for popular use, were the principles that the committees kept before themselves. The former necessarily was dominant; the latter, based upon expediency, was subordinate, but was given full consideration. The following table shows the detailed relations of the three alpha- bets, that of the Joint Committee, and those of the M. L.-A. P. and the N. E. A. committees on revision. Only the vowel and the divergent consonant symbols are here given; there is no difference of opinion about the other symbols: Key Word Joint Com. M. L.-A. P. N. E. A. art d artistic a aisle, find ai out, thou au ask a air, care a at a chew -c, t J prey e men e marine, eve i tin i mute iu jaw j, dg sing rj note 6 poetic o a a a a ai ai au au a a £e a ee a tj dh A e e e e A 1 i i i ju (iu) iu d3 i rj rj 6 o o o THE N. E. A. PHONETIC ALPHABET 9 Key Word Joint Com. M. L.-A. P. N. E. A. nor e e e not e e e oil ei ei ei ship J J sh thin p p fh that d 8 th mood u u u push u u u urge u u hut u u u yes, you y j y azure 3 3 3 about ) > 3 a a over j candid ") added f In this table, symbols separated by a comma (e. g., •€ , tj ) are sub- mitted as optional; either may be used. In the second column, iu is not an alternative of ju , but is intended to represent a slightly different pronunciation. In the case of ask path pass etc., the prevailing vowel in the south of England is that of art , in the greater part of the United States that of man . Some speakers in both countries avoid both sounds, and the intermediate value that they give the vowel is meant to be represented by the new symbol a of the Joint Committee. That symbol may remind some readers of the difference between American and English usage; but it is open to question whether it would not be better for our dic- tionaries to recognize the prevailing American usage and respell ask pass path etc. with a as in the case of at am bat etc., allowing as an alternative the symbol d (as in English usage). If that were done, the N. E. A. alphabet could drop the symbol a. The alphabets (M. L.-A. P. and N. E. A.) now offered for approval have undergone repeated scrutiny and most minute consideration by the foremost philological, phonetic, and educational scholarship of the country. It is true that they differ slightly from each other, 10 THE N. E. A. PHONETIC ALPHABET and a little more from that adopted by leading English philologists (as, for example, in the great Oxford Dictionary), and from the In- ternational Phonetic Alphabet. But the emphasis laid by oppo- nents of the reform upon the lack of uniformity among these various scientific bodies could come only from persons that have not been observant enough to recognize either the essential agreement of all these alphabets upon a common basis, a common method, and a common aim, or the ease with which any one who has an elementary knowledge of phonetics can pass from one to another. Within eight years the N. E. A., the M. L. A., and the A. P. A. have reached substantial agreement. (2) Objections to the Webster Key What led these three associations into a common effort to con- struct a scientific alphabet and secure its general adoption for phonetic work, including that in dictionaries and school books, was the utter unfitness for such work of the several phonetic keys then more or less used. The basis of those keys, namely, the English alphabet names and English pronunciation values of the letters, was unscientific, unstable, and misleading. On the ground of expe- diency they might have for a limited clientele and within a limited field some claim to consideration. It may be argued that a child in the mere process of learning to say the alphabet and to read and spell simple English words becomes accustomed to using some of the letters in the phonetic values assigned to them in such keys; that the eye of the child becomes more or less familiar with some combinations of these letters, and he can therefore later learn to use them as phonetic signs with somewhat less effort than he can learn signs wholly new in form. However, even if the like were true of all the characters in, for example, the Webster key, there would still remain a question of the economy of using that key. "If you were going to buy a watch to last you a lifetime, you would not make the choice turn on whether the dealer's shop were five blocks away or six." A phonetic alphabet should be consistent in its elements, and THE N. E. A. PHONETIC ALPHABET 11 precise enough for its purpose. For popular use it must be also simple and easy to learn. It may be worth while, for those who have not fully considered the matter, to point out why such a key as the Webster is felt to be "unscientific" and unadapted for use in serious phonetic work, even of an elementary kind. Its defects are the same in kind as those of the conventional spelling. To some extent in German, more so in French, most so in English, conventional spelling is a hindrance in the study of the sounds of these languages. To illustrate by English vowel spell- ings, a given letter does not stand uniformly for a given sound ( take tack tart talk culpable ) ; a given sound is not uni- formly represented by a given symbol ( ate bait bay gaol gauge great feign they ). Combinations of letters represent not compound but simple sounds, and sometimes one may say that neither of the letters forming the combination is a proper sign for the sound ( laugh meat great head conceive relieve sieve though ought fraught fool ). Other letter-groups stand for combined sounds made up of elements other than those represented by the letters involved ( thou now oil beauty ). Simple char- acters stand for diphthonged sounds ( mite mw mute ). No phonetician would for a moment think of teaching English sounds by means of English spellings. The inconsistencies and contradictions would utterly confuse and confound the learner. The like must be said of any key of the Webster kind. 1. The Webster key, even in its latest and best form (it has been considerably altered with each revision of the dictionary), uses a given letter (with varying diacritics, it is true) to represent radically different vowels (a as in ale, a as in at; e as in eve, e as in met; I as in bible, I as in bit). The words here are not rationally paired: the long of the vowel in at does not appear in ale but in care; the long of the vowel in met is not that of meet but is nearer that of mate; the vowel heard in bit does not, when lengthened, become the diphthong of bible, but approximates the vowel heard in bean. If the symbol a represents the vowel heard in ale, it should not elsewhere represent some other vowel. 2. The Webster key does not represent a given sound uniformly by a given symbol. The first element of the diphthongs heard in 12 THE N. E. A. PHONETIC ALPHABET ice and out is the vowel a of art; but in the Webster respellings that vowel is not at all represented in the one diphthong symbol ( I ), and in the other ( ou ) it is misrepresented as if it were the vowel of note. 3. In the Webster key certain double letters stand for simple vowels: oo and do for the vowels of pool and pull. The vowels here are approximately the same sound, long in the one word, short in the other (N. E. A. u and u , respectively); in neither case is o a proper symbol. 4. The Webster key uses simple characters for diphthonged sounds; for example, i for the diphthong in bible ( "ah-ee" or "ah-i," i e., ai ). If the word were respelled with the Webster characters for "ah" and "i" — as, to be consistent, it should be — the phonetic form would be baib'l .* Or take Webster u as in mute: if this word were respelled consistently, that is, in the Web- ster characters that represent its phonetic content, the form would be myoot ! 5. In English, all vowels in lightly stressed or unstressed sylla- bles have become altered in length or in quality or in both. That is a historical phenomenon of language. The result in present- day English is generally an "obscure" vowel which, whatever its original quality, approaches in quality the vowel of but, or, much more rarely, that in bit. These sounds are well defined vowels, and have as much independent existence as any other. Only in the most deliberate enunciation, as that of syllable by syllable, a pro- nunciation that would seem affected in even the most dignified and careful speech, is any of the original vowel heard. We are unac- quainted with any scientific alphabet of medium precision which recognizes more than two obscure vowels. For every practical purpose, then, two characters should be sufficient for the "obscure" vowels in English (as the symbols a and l in the N. E. A." alphabet). In the Webster key four wholly different devices ap- pear: (1) A dot above the phonetic symbol of the full vowel, as in prelate event anatomy formulate ; (2) a change to an italic form, with breve, as in infant recent control circumstance ; * Consistency is at times synonymous with honesty. What is to be thought of the man that cites this particular word to stir up the prejudices of his hearer against a consistent phonetic respelling and in favor of his own inconsistent form? THE N. E. A. PHONETIC ALPHABET 13 (3) italic form with dot over the letter, as in dbout ; (4) a tilde over the letter, as in inference . That is, the two "obscure" vowels are represented by ten symbols! In fact, on the Webster basis, there ought to be eleven symbols; one should be added, namely, I for the second vowel in limit. When one reflects upon the confusion in the Webster treatment of the "obscure" vowels — upon the use of ten symbols for two sounds, and the difficulty of learning them all; upon the use of some of these symbols now for the one ( a ) and now for the other ( l ) vowel, and the consequent impossibility of ever learning to asso- ciate unfailingly one specific sound with one specific symbol; and, on the other hand, upon the simple and adequate handling of the matter in the N. E. A. system, namely, two plain symbols for the two vowels — one can not help asking one's self what motives led the experimenter to exclude, as we shall see he did, this part of the two keys from a test supposed to demonstrate which is the better pho- netic key, which is "easier to learn, to remember, and to apply." In explanation, Dr. Whipple (footnote, p. 5) says: "It should also be borne in mind that in this experiment only those Webster symbols are used that correspond in power to the symbols of the Proposed Key. There are other symbols in the Webster Key used to indicate finer distinctions in sound, for which the committee has proposed no corresponding symbols. These other symbols were necessarily disregarded to insure comparable conditions." * The "other symbols" here referred to were either the symbols for certain sounds in foreign words (Webster ii k n ) or those for the obscure vowels. Dr. Whipple's words must refer to the latter. But no one who knows anything of English phonetics and of the pronunciation used by educated Englishmen and Americans of to-day, can explain Dr. Whipple's treatment of the matter except as that of a man not thoroughly instructed in what he was dealing with and imposed upon by conditions surrounding his task. In fact, the first part of the footnote just cited shows that the experi- menter was not thoroughly acquainted with his materials. He in- * In Missouri some years ago the state legislature passed a notorious partisan election law for St. Louis. The St. Louis Democratic "boss" of that time ex- plained that the law was meant to give his party only "a fair advantage" in the. city elections. 14 THE N. E. A. PHONETIC ALPHABET eludes among the Webster symbols used in the tests the Webster 6 as in maker and the ng in sing. He says: "In the Proposed Key, apparently the only symbol that can be used for the first of these sounds is 'v,' and the only symbol that can be used for the second is rj ." If Dr. Whipple read the N. E. A. key through to its end, he found in it the symbol "a for e in over." He prob- ably thinks that the Webster e is something different, and so would not recognize in the N. E. A. respelling mekar his own, and the current, pronunciation of the word maker. He evidently thinks — as do a good many people who have not considered it — that what is called the "obscure" vowel is heard only in careless or rapid speech. But the phoneticians — experts, who have trained themselves to recognize and classify speech-sounds, and who have made observations far more numerous and careful than any intelli- gent layman, or all of them, could make — these men say that the "obscure" vowel is the normal sound in the usage of the best speakers both English and American. The normal pronunciation (in N. E. A. symbols) of maker is mekar , of infant is infant , of anatomy is anatami , of culpable is kulpabl , of recent is risant , of propose is prapoz , of preface is prefis , of create is criet , of senator is senitar . Dr. Whipple may think that he does not pronounce these words in this way; that he makes cer- tain finer distinctions in sound which he supposes the Webster key to provide for. But — if he pronounces English as other educated Americans pronounce it — he does not use ten varieties of weakened or obscure vowels requiring transcription in an alphabet of medium precision. The truth is that the Webster key, in its treatment of the "obscure" vowels, evades and smothers the facts; and a reference to its "finer distinctions in sound" has something of the ring of im- position upon the ignorance of the reader. As for the ng in sing and the n in bank (Webster ng and g ), the difference — if there is a difference — is too minute to be considered in a dictionary for general use; one symbol is sufficient. None of the leading English phoneticians (Sweet, Soames, Jones, Rippmann) recognize more than the one sound, rj . For a simple statement of the matter Dr. Whipple may be referred to Professor Rippmann's elementary treatise, "English Sounds; a Book for English Boys and Girls" (Dent, 1911). In §§19, 20 the author THE N. E. A. PHONETIC ALPHABET 15 says: "In long the two letters stand for the nasal sound only; and we are going to use a special sign for this in future: rj , that is, an n with the tail of a g . In longer, rig has the value of "Sometimes we say rj when we write n , as in thank, anchor , tranquil ." But to return to noting the inconsistent and therefore "un- scientific " features of the Webster key. 6. The Webster key makes an extended and at the same time inconsistent, chaotic, use of diacritic marks. The use of diacritics in itself would not necessarily make the key unscientific if the marks were used consistently, that is, each always in its one significance. There is, however, a question as to the advantage of using diverse diacritic marks rather than diverse letter forms when the aim is to distinguish the qualities of sounds. The diacritic may very well denote length. In the words art note there are vowels heard also in the first syllables of artistic poetic respectively, but in the one case the vowels in question are longer than in the other. The macron put above the symbol will easily and clearly distinguish a long vowel. The symbol for the short vowel need not be marked. The N. E. A. alphabet, for ex- ample, distinguishes these four vowels with a minimum use of diacritics: d a , 5 o . But the question whether it is better to use many or few diacritics, in many or few senses, ought to be satisfactorily answered by the fact that competent teachers who have tried both declare against any further use of such marks than is absolutely necessary. The chief phoneticians of the world are explicit in their statements. At an annual meeting of the German Modern Language Association, Zurich, May, 1910, W. Vietor discussed the need of a consistent phonetic alphabet for modern language schoolbooks and dictionaries. He declared (1) that such an alphabet is most earnestly to be desired; (2) that it should be a genuine phonetic alphabet, not one made with diacritic marks; (3) that the alphabet of the Association Phonetique Internationale best answered the demand.* Sweet, the leading English phonetic authority, in his recent book, "The Sounds of English" (Oxford, 1908), says (§313): "The most objectionable * Die Neueren Sprachen, xviii, 549 (1910-11). 16 THE N. E. A. PHONETIC ALPHABET class of letters in a broad alphabet are diacritical ones." f Otto Jespersen, of Copenhagen, one of the keenest of phoneticians, in his "Phonetische Grundfragen" (Leipzig, 1904), §25, p. 19, says that to mark shades of sound variation for which there is no symbol in the conventional alphabet, the nearest recourse is to diacritic marks. These to a degree have long been in use in certain languages, the macron and the breve, for example, in the work of old Roman grammarians and metrists as marks of quantity. It is no wonder, therefore, that many phoneticians have taken refuge in such means, and have tried to make systematic use of them — some placing diacritics above the letters, some below, and some both above and below. But both principles lead to an impracticable alphabet. Lepsius in his so-called Standard Alphabet (1855) carried the practice to an extreme. "Since his time it is only dilettantists and beginners who, incapable of profiting by the warning of past experi- ence, think that they can construct a satisfactory phonetic alphabet by applying an array of diacritic marks over and under the letters of the Latin alphabet." The International Phonetic Association, whose headquarters are in France, explicitly avoids the use of diacritics wherever possible. All of the recent steps toward a uni- versal scientific alphabet in England or in this country — the alphabet of the Oxford Dictionary, that of various scientific bodies in England and America, that of the American Philological Association (1877) revised into the A. P.-M. L. and the N. E. A. alphabets now under discussion — make use of a minimum of diacritic marks. The opinions of these expert phoneticians, and the practice of the leading philologists, language associations and scientific bodies, furnish satisfactory guidance. Now, it must be noticed that the 1909 Webster key not only uses a multitude of diacritic marks, but uses them inconsistently. Leaving out those signs in the Webster key that are needed for certain foreign words, that is, counting those that furnish an English phonetic alphabet equivalent to the N. E. A. key, there are found in the Webster six vowel characters with the macron ( a e I 5 65 u ), four with the macron and a dot (a e 6 u ), ten with the breve (a&eeloo od u u), one with breve and a t By "broad alphabet" Sweet means one of medium precision, for general or national use; an exact alphabet specifically for phoneticians he calls " narrow." THE N. E. A. PHONETIC ALPHABET 17 dot — if one may trust one's unaided sight ( 6 as in s6ft dog g6d ), — three with circumflex (a 6 u ), one with two dots above ( a ), two with one dot above (a a ), one with tilde (e). Here are twenty-eight vowel signs, some of them with two dia- critics, all the rest with one, to represent vowel sounds that phoneti- cians recognize as about twenty-one in number! And there are two more, oi and ou , for two of the English diphthongs (those in oil and now ). Thus we find thirty Webster symbols for twenty- three vowels. The N. E. A. key, by way of contrast, serves the same purposes with twenty-three signs, just one for each sound : d a ai au a aaeeiiiuooeeeiuuuuai. Only eight of them are marked with a diacritic, the macron. The diph- thong-signs are consistently formed of their proper elements, so that they do not have to be learned for and by themselves. That is, one has to learn the form and phonetic value of only eleven characters (aaaeioeuuai). The macron, when applied, is always to be interpreted as a sign of length; it does not connote or accompany a significant change of quality. A scheme of symbols for English vowels could not be simpler. But set to work on the Webster diacritics. What does this or that diacritic mean? Length? or quality? or both? Compare the words care and can (Webster respelling, kar kan ). The quality of the vowels is, at least in ordinary American speech, the same. The vowel in care is long, in can is short. Is one, accordingly, to understand the circumflex as the sign of length, the breve of shortness ? Then how is one to interpret the Webster symbols for the vowels in old and all, namely, 6 and 6 ? The difference here is one of quality, not of length. The only conclusion that a rational mind can reach is that in a, the circumflex is a sign of length, and in 6 is a sign of quality, which is irrational. Again, the Webster symbols oo for the vowel in food and 66 for that in foot , represent sounds (not properly indicated by the symbol o ) that in an alphabet of medium precision may be regarded as one vowel. The macron and the breve are here evidently signs of quantity; but since only one of them is needed, it is foolish 18 THE N. E. A. PHONETIC ALPHABET to use both. Compare these symbols with those used for the vowels of ale and at , namely, a and a . Does this second pair of symbols, opposing macron to breve, represent a common vowel in the two words, a vowel long in the word ale and short in the word at ? This can not be, because the vowel heard in at does not, when lengthened, become the vowel heard in ale . One can not apply what one learns from the symbols oo 06 to the interpretation of the symbols a a , however spontaneously a simple mind would assume that a pair of symbols distinguished from each other by cer- tain diacritic marks must have the same relation to each other as obtains in another pair differentiated in the same way. Two cards marked each with a black spot are aces; but of two marked each with a red spot, one is an ace, and one is something else ! The learner fares no better who tries to reason from breve to macron in the Webster symbols I and i ; for the first stands for a simple vowel, that heard ( in ill , and the second for a diphthong whose main element is quite another vowel — namely, the diphthong heard in ice ( ai ). The Webster uses of diacritic marks are, in fact, tm-reasonable. Any attempt to classify the diacritics according to their uses will end with one's defying the world to do it. One must fix them in mind by sheer force of memory. The attempt to learn the English vowel sounds by means of a phonetic key so confused and contradictory must bewilder, even stupefy, a child's mind; and the teacher must bear a useless burden, perhaps not realizing it, who tries to induct children into the science of speech-sounds, with the help of such a system of "ocular phonetics run mad." As for the phoneticians, they will none of it! There is no living phoneticiaa who can regard a key of the Webster kind with anything but impatience. The Webster key, in comparison with such a key as the N. E. A., might be likened to the English system of pounds and shillings, in comparison with the American system of dollars and cents — except that this would be doing an injustice to the English system. It is because of this multitude of inconsistencies and contra- dictions in the Webster key, and in every key, which, like it, starts from the basis of English alphabetic and pronunciation values of the THE N. E. A. PHONETIC ALPHABET 19 vowels, that phoneticians and philologists and progressive teachers of the living languages (whether the mother tongue or a foreign speech) reject such a key in toto. As a guide to pronunciation, its only usable capacity, it is of questionable merit even for the unin- structed layman, for whom its leaning upon pronunciation values might be felt, at first thought, to be an advantage. In all other uses for which a phonetic alphabet is needed, it is worse than useless; it is execrable. (3) Uses of a Scientific Alphabet A simple, rational, and scientific alphabet is needed, as a means — (a) For teaching phonetics in schools of all grades: there should be elementary instruction for children learning to read, somewhat more thorough work for pupils taking up the study of a foreign language, and extended and thorough practice for language teachers. (6) For indicating pronunciation, in schoolbooks, dictionaries of the native and foreign languages, and all handbooks and general works of reference. The purposes of a phonetic alphabet are here named in inverse order of their importance as present public opinion would regard them; for there are at present among us few learners of English phonetics, relatively little study of the sounds that form English speech and little training in the correct and intelligent production of those sounds. But there is a vast army of users. From their point of view the main function of a phonetic alphabet is to indicate the pronunciation of words assembled in English dictionaries, ency- clopedias, atlases, handbooks, and all popular scientific treatises. For this purpose such an alphabet should be "easy to learn, easy to read, and easy to remember." As a matter of fact, the last requirement is, for the average uninstructed layman, not essential. He does not attempt to remember any key, but refers, when in doubt, to the top or bottom of the page, where he finds it. His chief need is that the symbols of the key be clear, explicit, precise. That established English spellings are quite unfit to indicate pronunciation, is universally recognized. Most dictionaries, and some encyclopedias, respell the words, and in many schoolbooks and handbooks words new to the reader are treated in like manner — a 20 THE N. E. A. PHONETIC ALPHABET more or less phonetic respelling is set alongside the conventional form. Manifestly, it would be of very great advantage if all these books used a common phonetic alphabet, if there were a national scientific alphabet. It will be worth while to review here some details of the problem.* It is generally agreed that a phonetic alphabet intended for wide use (1) should use one sign for one sound, (2) should, as far as possible, employ the letter forms already familiar, (3) should add few new forms, (4) should employ few diacritic signs, (5) should not mingle forms of differing styles in a way unusual in ordinary print (for example, italic with roman, capitals with small letters), (6) should design the necessary new symbols in harmony with existing letters. That is, the respelled word should offer as little as possible that is strange or startling to the eye. It would be well, also, to adopt an alphabet constructed in harmony with international usage. Nations to-day are too close to one another, intercourse between them is too constant and intimate, for each to go its own way as of old. The relation that a phonetic alphabet adopted in one country should have to those adopted in other countries, is briefly stated by Sweet, in §§8 and 308 of his "Sounds of English": "In dealing with the sounds of English it becomes necessary to adopt a phonetic notation. It is now generally agreed that the best way of constructing such a notation is to give the letters of the Roman alphabet the sounds they had in the later Latin pronuncia- tion, with, of course, such modifications as seem to be improvements or otherwise desirable, supplementing the defects of the Roman alphabet by adding new letters when required. This is the ' Romic ' or international basis. . . . "Whatever alphabet is adopted, it must be capable of modifica- tion so as to supply the want of (1) an international scientific 'nar- row' notation, in which all possible shades of sound can be expressed with minute accuracy by symbols of fixed value, and (2) an indefinite number of national 'wide' notations, each of which selects the minimum number of simplest letters required to express the prac- * See the brief but complete presentation of the matter in Section II, pp. 4-17 of the "Report of a Joint Committee on a Phonetic English Alphabet" (Publishers' Printing Co., New York, 1904) — the Report referred to on page 7 above. THE N. E. A. PHONETIC ALPHABET 21 tically necessary sound-distinctions of the language in question, ignoring those that are superfluous, so that all the national systems appear as modifications of a common basis, each diverging from it only so far as is made necessary by considerations of simplicity and ease of printing and writing both in long and short hand." That is exactly what is attempted in the N. E. A. alphabet — to effect such a modification of the international alphabet (of a thing, it is true, that does not yet have absolute existence, but which nevertheless is already fixed in all but its minor details), as shall adapt it to national use in this country. That the committee suc- ceeded in making an alphabet that is in substantial harmony with international usage, every American linguistic scholar will recognize. That in itself it is a good phonetic alphabet is evident upon con- sideration of what characteristics such an alphabet should have. The symbols used in respelling to indicate pronunciation should, above all, indicate the nature, or quality, of sounds. Less impor- tant, and applying mainly to vowel sounds, is the indication of length or quantity. A third essential thing is the indication of syllable stress or accent; but this last duty does not rest upon the letters of the alphabet. The manner of marking the quantity or length of vowel sounds is a matter for consideration. Shall quantity be indicated by the form of the letter, or by a diacritic mark? Now, expert opinion and practice are unanimously against a general use of diacritics. Quantity, however, is a thing affecting in the same way a whole series of vowels. If it is indicated uniformly by a given diacritic mark that has no other significance, one has to learn the meaning of this mark only once for all. And since in English fewer long vowels than short need notation, to employ a plain mark as the long- vowel symbol involves a minimum use of the diacritic* What diacritic should be chosen is to be decided by considerations of clearness to the eye, and durability in the printer's type. On these grounds the macron is the best. * In the construction of the N. E. A. and the M. L.-A. P. alphabets, as of that of the Joint Committee (given on pp. 8f.), the committees thought it better, in an alphabet of medium precision, to disregard the slightly diphthongal character of the long vowels (N. E. A.) e I o u heard to some extent in the United States, and generally in South England. 22 THE N. E. A. PHONETIC ALPHABET It is when one comes to choose the means of indicating the qualities of vowel sounds, for consonants give little trouble, that difficulty is met and controversy arises. The English vowel scheme appears, in its English dress, to be a monstrous thing, without form and void of meaning. There is no head or foot, end or side, up or down; absolutely no handle by which to take hold of it, nor point of view from which one can conceive of its having a form. One must, therefore, take the familiar vowel symbols, "a e i o u ," assign to each a phonetic value, and make additional symbols (either by adding diacritics to some of these letters or else by in- venting new forms) until there are enough letters to have one for each of the elemental vowel sounds in the language. For the purposes of a phonetic alphabet of medium precision, phoneticians recognize in English eleven vowels requiring tran- scription, namely, those of art ask at met. bit note not pool but (ov)er (add)ed. Indeed, the vowel of ask need not be included ; most of us pronounce it as we pronounce the vowel of at. The intermediate vowel, as far as present usage is concerned, is an invention of dictionary makers. We need, then, five (six, with a ) new vowel symbols. If we listen to what the experts say (pp. 15f . above), we shall not use diacritic marks to make them; for above each of the original letters we must at times put a diacritic sign of length, and to superimpose a second diacritic of quality would quadruple the ground of objec- tion to diacritics. We could make the one diacritic indicate at the same time both length and quality; in which case we must have three sets of diacritics, to mark a change (1) of length alone, (2) of quality alone, (3) of length and quality together. Which, again, multiplies the diacritic difficulty. Unquestionably, the best procedure is to make five or 6ix new vowel symbols, as was done for the N. E. A. alphabet. Then each symbol, in its unmarked form, has a recognized phonetic value, and never — whatever diacritic cap or tail is attached to it — stands for a radically different vowel. Horses may be of different colors, shapes and sizes, but you never mistake a horse for a cow. Whether the N. E. A. new vowel signs are satisfactory in design, is submitted to the intelligence and sense of harmony of the public. Diphthongs should of course be represented by the letters d&- THE N. E. A. PHONETIC ALPHABET 23 noting the simple sounds that combine to form them. No other way can be defended by logic of the schools or by common sense. One should be consistent. If one, for example, chose to use aw as the phonetic symbol for the first vowel in awful, and ie for the vowel in sieve, then one must, to be consistent, respell the word boij as bawie ; for the diphthong in boy is a union of those two vowels. Being provided with ten or eleven simple vowel signs, then, and the macron as a sign of length, we can represent all the vowels in English speech that the ordinary uses of an alphabet call for. And to translate into sound the twenty-three symbols thus made, we need to know the form and significance of just twelve things, namely, of eleven letters and one diacritic. By way of parenthesis, it may again be remarked that with the Webster key — since neither letters nor diacritics are uniform in value — not only each letter but also each combination of letters or of letter and diacritic must be learned by and for itself; in the which learning one's instinct for law and order and reason must first be bludgeoned into unconsciousness. One more question must be decided before our phonetic alphabet can be constructed: how shall we distribute the eleven phonetic values among the eleven signs? Phoneticians, philologists, our teachers of foreign languages, our government scientific bodies, and our learned associations, have chosen to use the vowel signs in their Latin values. To do so brings the notation into harmony with phonetic science, which, being a science, is international. It makes the notation a help instead of a hindrance in the study of foreign languages. It opens the literature of phonetics to teachers of English. It brings order and consistency into the alphabet, a consummation impossible when one symbol, as i , is used to denote such widely separated sounds as the vowel of bit ( i ) and the diphthong of bite ( ai ) . There is no phonetic kinship between the sounds; nor is there between the vowel of mate and the vowel of mar or mat, nor between the vowel of meet and the vowel of met. To use the letters a e i o u in their European or Latin values creates no difficulty for any one who knows even a little or wishes to learn a little about one of the foreign languages. On the contrary, it lightens materially the learner's task. But for one who knows only English, it looks like a different matter. There is the 24 THE N. E. A. PHONETIC ALPHABET heart of the difficulty, the gist of the opposition among laymen to the N. E. A. or any similar alphabet — so far as that opposition is not inspired by commercial influences.* To any one to whom the alphabet names are so ingrained that he can not separate a from ale, e from eve, i from ice, but must associate these letters with the sounds heard in the respective words, there comes a shock when he is for the first time told to write e for the sound in ale , i for the sound in eve , and ai for the sound in ice ; and it will take him a little time to regain his balance, to get his bearings. An alphabet like the N. E. A. is, how- ever, so much more simple and consistent in its elements, and in their application, that the mastery of it and its uses must for a rational mind be easier to attain than equal efficiency with a key more intricate, and inconsistent in many of its elements. There have been no investigations or experiments that show the contrary. The recognition of specific difficulty in learning several of the N. E. A. symbols applies only to the cases of persons to whom it has become a habit to associate with certain letters the phonetic value given by their alphabet names; and the difficulty itself is only temporary. If, on the other hand, an elementary study of English sounds were introduced where it ought to come, namely, accompanying the child's first attempts to read, and if a consistent phonetic alphabet were used in the instruction, the child would have less difficulty in learning the rational key, symbol by symbol, than in learning the Webster, would have fewer symbols to learn, would find in the proc- ess of learning a means of developing instead of stunting its reason- ing powers, would find pleasure in gratifying its instinct for order and system, would more easily attain a correct and confident pro- nunciation. * If the plea of the opponents of the N. E. A. alphabet to retain "our alphabet as it has been handed down to us" is felt to have any weight, one should consider that present pronunciations of the alphabet names, as of other words, are quite modern. For the greater part of their history the vowel signs a e i o u have had in English their continental values; and the names of these letters, until comparatively recent times, were simply the vowel sounds heard respectively in the words art they eve note rule. There is nothing sacred or immutable in our present alphabet-names; they could be changed without difficulty in one generation, and the change would be little more than a restoration of earlier usage. THE N. E. A. PHONETIC ALPHABET 25 All these results have been realized in teaching modern foreign languages in Germany, France, Denmark, and (more recently) in England. That the pupil must first learn accurately the sounds of the language, their physiological nature and relations, and must train his vocal organs to make them, is one of the cardinal principles of the "new method" of language teaching. In this elementary instruction an ever increasing number of teachers use at the be- ginning of the work a phonetic alphabet only. After the pupil, reading from a phonetic transcript, has learned thoroughly the sounds of the language, then he is gradually introduced to the con- ventional spelling. This principle of the reform, namely, that em- phasis should be laid upon correct pronunciation as the first thing to be attained in the study of a foreign language, and upon phonetic training as the means of attaining it, teachers as well of the native tongue in those countries are beginning to adopt and put into prac- tice. The extent, methods, significance and results of the "reform" in modern language teaching within the last generation should be considered in any decision upon a matter of such basic importance as the construction of a phonetic alphabet. Of that reform a leading English educational authority, J. J. Findlay, Professor of Education in Manchester University, said: "Quite deliberately the present writer ventures to assert that the 'reform' in modern language teaching now in progress is one of the most noteworthy events in the sphere of Teaching since the Renaissance, surpassing in impor- tance even the results of introducing Science into the Schools." The history and principles of the reform one may review in numerous educational monographs, and in articles in educational journals.* The first noteworthy public enunciation of the reform ideas was that of Vietor in his pamphlet of 1882, "Der Sprachunterricht muss umkehren!" In opposition to the grammatic-philological method then in full vogue, Vietor advanced the ideas championed before by * See especially the German publications like M. Finger's " Der fremdsprach- liche Unterricht" (Leipzig, 1907), and Victor's " Die Methodik des neusprachlichen TJnterrichts " (Leipzig, 1902); or Jespersen's excellent "How to Teach a Foreign Language" (English edition, Macmillan Co., N. Y., 1904). For the most recent account in English of the new method, see F. B. Kirkman: " The Teaching of Foreign Languages; Principles and Methods " (Clive, 1909). 26 THE N. E. A. PHONETIC ALPHABET such educators as Ratichius, Comenius, and Pestalozzi. His pam- phlet "demanded thoroughness of pronunciation, a more intensive study of reading, and inductive study of grammar based upon the reading. Above all, not the dead letter but the spoken word was to be put into the foreground of modern language teaching." * Of course some reformers went too far. They were accused of putting a smattering of conversational attainment in the place of solid mental training. But the essential principles of the move- ment were more seriously developed and built upon; and the prog- ress of the reform became so unmistakable and significant that it began (1901) to be reflected in slow-moving German officialdom. German official Lehrplane (Outlines of Instruction) in foreign languages in secondary schools, 1882, named correctness of pronun- ciation as the first requisite, but expressed doubt of attaining it. The Outlines for 1891, however, insisted upon it. The study of phonetics had in the meantime been introduced and applied to modern language instruction, and found to be a means of teaching right pronunciation. German officialdom recognized the results but not the means; for the official Outlines of 1891 forbade the use of phonetic script. (One should keep in mind that no satisfactory phonetic alphabet had then been agreed upon. In fact, at this time, Paul Passy in France was perfecting his alphabet, now known as the International Phonetic Alphabet, and used by the leading modern language teachers in Germany, as in practically the rest of Europe, including England.) Finally in 1901, Lehrplane for boys' upper grammar schools (filr die hoheren Knabenschulen) no longer forbade phonetic instruction and the use of phonetic script in beginning classes in foreign language work.f This official recognition of the usefulness of a phonetic alphabet is significant. The introduction into English schools of the continental "re- form" (including the use of phonetic script in beginning classes) is now taking place. The platform of progressive English language teachers is, in brief: Phonetics is "the science of speech-sounds and the art of pronunciation" (Enc. Brit.). "The advantages of be- ginning a foreign language in a phonetic notation are many and * Krause: " The Teaching of Modern Languages in German Secondary Schools," Monatshefte fur d. Sprache u. Pedagogik, X, 6 and 8. t Finger, p. 31. THE N. E. A. PHONETIC ALPHABET 27 obvious. A learner who has once mastered the notation and learned to pronounce the sounds the letters stand for, is able to read off at once any text that is presented to him without doubt or hesitation, and without having to burden his memory with rules of pronuncia- tion and spelling. ... If the learner begins with the phonetic nota- tion, and uses it exclusively till he has thoroughly mastered the spoken language, he will then be able to learn the ordinary spelling without fear of confusion, and quicker than he would otherwise have done." The experience of teachers who have adopted this platform may be given in a few representative statements: * "Every foreign language teacher finds that one of the most serious of the difficulties which confront him at the beginning of his teaching is the fact that he has to present to his pupils and make them acquire correctly a certain number of sounds. . . . The quick- est and surest way is by a systematic course of phonetics and regular practice in phonetic drill. . . . Once the phonetic alphabet is mas- tered, — this is not nearly so difficult as some think, — a regular sound drill of about five to ten minutes at the beginning of each lesson for about a week will work wonders. ... [A child finds] sat- isfaction in pinning sound down to a symbol that can represent only one thing. . . . [It] satisfies a child's sense of proportion and fitness of things. . . . Does the use of phonetics in writing spoil the pupil's ordinary spelling? ... It has invariably been my experience that when children have been taught phonetics from the beginning, they spell French words far more correctly than those who come to me with no phonetic training at all." — V. Partington, in Mod. Lang. Teaching, ii. (1905-6), 40-44. "True it is that the earlier it [phonetics] is made use of, the better for teacher and for taught. . . . No one who has tried the two methods, that of relying upon the pupil's memory, and that of giving him the phonetic script to help him remember, would care to give up the use of the script." — H. W. Atkinson, Rossall School, in Mod. Lang. Quarterly, hi., 54, 56. "Phonetics are used. . . . The idea that imitation is sufficient to teach the sounds of a foreign language is dying out." — Fr. Dorr, in Mod. Lang. Teaching, ii., 232. * These are all from leading English modern language teachers, and school inspectors — practical schoolmen. 28 THE N. E. A. PHONETIC ALPHABET "A beginner may hear a new sound a thousand times without reproducing it rightly. This is the 'imitation fallacy,' to suppose that mere hearing is sufficient. There must be knowledge and con- scious effort. ... Of all the classes I visited the pronunciation was worst in those held by foreigners teaching their own language to English children. ... A beginner should always hear and speak before he attempts to read and write." — E. R. Edwards, Mod. Lang. Quarterly, vii., 118. "At first we found it extremely difficult to combine the elements of good reading — articulation, accent, and rhythm. Our energies were almost entirely spent, in the lowest classes of all, on articula- tion, which, with the complexities of French spelling, is a serious difficulty to beginners of nine; so that when we came to deal with accent and rhythm, bad habits had already been formed. It became evident that all three must be combined from the very first. The solution of^the difficulty lay in the use of phonetic script in the initial stage. We introduced that of the Association phonetique internationale last January, and we have every reason to be satisfied with the result. It reduces to a minimum the difficulties of articu- lation. It trains the ear and the faculty of attention to a degree which, of itself, would give it a great educational value. . . . The stage of transition to ordinary spelling offers no difficulties whatever, if only it be carried out gradually and methodically. . . . Last, but not least, the use of phonetics keeps the teacher up to the mark in the matter of pronunciation, and helps him to correct his own defects." . . . "Unless the alphabet used is phonetic there is likely to be con- fusion just at the time when clear impressions and associations are all-important. In short, some sort of phonetic alphabet is the best 'mistake preventor' in the matter of pronunciation. . . . The one point I want to emphasize is that the first spelling used should be phonetic." — L. von Glehn, Merchant Taylors' School, in Mod. Lang. Teaching, v., 46, 154. So much for the elementary work in French and German as done in English secondary schools. Leading foreign language teachers in Europe and Great Britain agree in recognizing the necessity of basing the earliest stages of work in a foreign language upon thorough instruction and training in the elementary sounds of that language, and the great help that may be derived from the THE N. E. A. PHONETIC ALPHABET 29 use of a phonetic transcript. Jespersen's "How to Teach a For- eign Language," Chapter X (pp. 142-173), is an adequate pres- entation of the matter. What use is made or is to be made of phonetics in the elemen- tary teaching of English? That the method, now widely used and found to be so efficient with pupils beginning a foreign language, would bring about a like advance in the teaching of the mother-tongue, is felt by a very large number of linguistic scholars.* The consensus of opinion of the International Phonetic Association — a society numbering about one thousand members distributed throughout the world, drawn from the ranks of the foremost teachers and scholars — is expressed in "The Aims and Principles" of the Association (1904): "Phonetic writing can also be used with great profit in elementary school work for teaching to read the mother-tongue. Learning to read by the usual methods is at best a long and dreary task; but where phonetic texts are used it becomes short and easy. When once a child reads phonetic texts fluently, learning to read the common spelling is a mere trifle. This is a most important, though hitherto little known application of phonetic writing, which may yet revolutionize the systems of education all over the civilized world." Sweet long ago wrote: "If our present wretched system of study- ing modern languages is ever to be reformed, it must be on the basis of preliminary training in general phonetics, which at the same time will lay the foundation for a thoroughly practical study of the pro- nunciation and elocution of our own language — subjects which are totally ignored in our present scheme of education" (Preface to "Handbook of Phonetics," 1876). Later he said: "Phonetics, of course, should be begun in the nursery. The time will come when ignorance of phonetics will be held to disqualify a nurse as much as any other form of incapacity." In the same book from which the last sentence is cited, the author further says: " The reading-books in the native language should at first be mainly in simple prose, with only occasional pieces of simple poetry. They would, of * The new method applied to the teaching of English may be seen in part in the books of H. C. Wyld, " The Historical Study of the Mother Tongue " (1906), " The Place of the Mother Tongue in National Education " (1906), " The Teach- ing of Reading in Training Colleges " (1908), published by Murray, London. 30 THE N. E. A. PHONETIC ALPHABET course, be entirely in phonetic spelling on a Broad Romic basis, and with accurate marking of stress and intonation" (p. 245 of " The Practical Study of Languages," Dent, 1899). Sweet's words are beginning to bear fruit. W. Campbell Brown, in Mod. Lang. Quarterly, iii, 46, writes: "The use of a phonetic system by the teacher to organize the teach- ing of pronunciation is a comparatively simple matter. . . . Should phonetics be only and for the first time applied to the learning of foreign tongues? We should answer most decidedly in the negative. 1. Our own language has not a sufficiently fixed standard of pro- nunciation. The use of a Lauttafel [phonetic chart] might be made of great service in improving the pronunciation of the native tongue. 2. Phonetic signs and the use of the Lauttafel are much more rapidly learned through the medium of English than of a foreign language. 3. There is no reason why it should be the duty of the foreign language teacher to teach pupils the use of their organs of speech and the classification thereby of the different sounds of speech." It is only by the help of phonetics, "the art of pronunciation," that it is possible to deal effectively with vulgarisms and provincial- isms of pronunciation and secure uniformity of speech. On careful training of the ear and the organs of speech in the early years of the child's school life depends the acquirement of a clean and correct pronunciation, the overcoming of slovenly habits of speech. If bad habits are not got rid of before the child reaches the late grammar grades, he will in most cases be by that time confirmed in them. Such training is in itself an educational influence now beginning to be recognized. Pronunciation and reading are the first subjects the child studies. It is important that this earliest work should appeal to and satisfy the child's instinct for consistency. Contra- dictions in the material put before the child do more harm than to a more mature mind. Let a child feel from the beginning that there is system and reason in what he is doing; then his mind will take hold more readily, grow sensibly into a more efficient thinking machine. This fact explains the success of the experiments by Feline in France, Miss Soames in England, and Spieser in Germany, in teaching chil- dren to read first in a phonetic script, and only later in the conven- tional spelling. It is found that children taught in this way for two years have then no difficulty whatever in passing to ordinary spelling. THE N. E. A. PHONETIC ALPHABET 31 They make better readers, and better spellers, than children taught from the beginning with conventional spelling alone.* In our own country like experiments were made as long ago as 1866-70, in the public schools of St. Louis, under the administration of W. T. Harris, then Superintendent of Schools of that city. Dr. Harris's account of the experiments may be found in his annual reports. In 1866-67 a phonetic primer, printed in the phonetic alphabet constructed by Dr. Edwin Leigh, was used in the Seventh Grade of the Clay School (of which Harris was then Principal). Its introduc- tion "proved to be productive of the most satisfactory results. The class that finished it made very rapid progress in learning to spell in the common orthography after they were transferred to the ordinary type in the First Reader. But the best of all was the demonstration that the imperfections of articulation and the provincialisms of pronunciation current here can be completely eradicated by that thorough drill that is rendered necessary in teaching the Phonetic Primer. Besides, it was shown that all this can be done in less than the time required for completing the same textbook in the ordinary type!" {Thirteenth Annual Report, 1867, p. 56.) The St. Louis Board of Education adopted the innovation for the ensuing year, in the Seventh Grade, throughout the city. In ad- dition to the Primer, a Phonetic First Reader was introduced. The results Dr. Harris narrated and considered at length in his report for 1869 {Fifteenth Annual Report, pp. 95-98). He notes that "the phonetic system is found to produce the best spellers," and he gives a "summary of the advantages of the phonetic system, as intro- ductory to the art of reading: I. Gain in time — A given standard of good reading can be reached in about one-half the time. This has been thoroughly tested. II. Distinct articulation — clear-cut words, every element brought out in its purity — is secured. III. The logical inconsistency of the ordinary alphabet makes the old system a very injurious discipline for the young mind. The earliest studies * Such a result is to be expected from properly conducted experiments with the mother-tongue; for the method, when applied to beginning work in the mother- tongue, has to meet conditions essentially similar to those that surround the beginning work in a foreign language. In the latter, its success is established. See Professor Geddes' statement in Die Neueren Sprachen, xiii, 361. 32 THE N. E. A. PHONETIC ALPHABET should be the most logical and consistent. IV. The most important feature of the phonetic system is the substitution of the analytic drill, during the first year of training, for the loose word-method in vogue. . . . Pupils who are taught to read phonetically make better arithmetic and grammar scholars, and are more wideawake and attentive, have finer discrimination — in short, are more distinguished in those traits of mind that flow from analytic training." Again in the Sixteenth Annual Report (1870), p. 165, the Superin- tendent records gratifying results : "Gain in time, distinct articulation, better spelling — these are its merits. . . . The gain in time is nearly one-half, and the improvement in quality fully as great." By reference to the Schedules of Studies in the reports for 1866 and 1870, one may find that in 1866 pupils began the Second Reader in the third quarter of the Sixth Grade, in 1870 in the third quarter of the Seventh Grade, a saving of one year's time in less than two years' work as ordinarily done. Finally, in the Seventeenth Annual Report (1871), p. 133, Dr. Harris reviews the results of the four years' use of Phonetic Primer and Phonetic First Reader, and again states his conviction that the gains promised by earlier stages of bis experiment are remarkable and permanent. These early experiments were made with an alphabet much less simple and clear than that now proposed by the N. E. A. committee. The primary teachers of that time (1866-70) had no phonetic train- ing — a training now possible through the development of the science of phonetics. If the experiment were now tried, with the present understanding of the problem, on the scale that was possible to Dr. Harris, the results would inevitably overwhelm any opposition to the method, and to the means employed in carrying it out. Dr. Harris's measure of success, however, substantiates what modern language teachers are to-day finding out — the greater effectiveness of the "new method," with its reliance upon early and thorough training in phonetics and its employment of a scientific alphabet as an aid to the child's first steps in language. More recent primary English instruction based on phonetics is referred to in Modern Language Teaching, ii, 232: "I have seen ex- cellent results obtained from judicious and systematic use of pho- THE N. E. A. PHONETIC ALPHABET 33 netics in teaching children to speak and read their mother-tongue by Miss Nellie Dale, of Wimbledon." Miss Dale's work is based upon phonetics and involves thorough training of the children in speech-sounds. It is true she does not use a phonetic alphabet. She substitutes the device of printing letters in varying colors — vowel signs in red, voiced-consonant signs in black, voiceless in blue, silent in yellow. With what Professor Rippmann characterizes as nothing short of genius, she adapts her method to the child mind with excellent results. Nevertheless, even with her method, the inconsistencies of conventional spelling flare up in the face of the child, and must be overcome by memory alone; and Miss Dale's genius appears in the interesting devices by which she assists the memory. In the hands of teachers at large, a still more con- sistent means of representing sounds would lighten the task and re- lieve the child of part of the burden that her system in a measure still imposes upon the memory. In the kindergarten and primary, her device of colored letters and appropriate pictures would greatly increase the pleasure and inter- est of the child, whatever alphabet is used. It can not, however, have the full educational value of a strictly scientific and consistent scheme of symbols. A really scientific alphabet is needed from the beginning of the child's work with speech-sounds. The opponents of the N. E. A. and the M. L.-A. P. alphabets seem not to realize that the proposals of these educators and scholars are in fine with the general advance of linguistic science in the last two-thirds of the nineteenth century. That movement, of late quickened and directed by the special development of the science of phonetics, and the application of that science to language teaching, is older and broader than the modern language "reform." It affects all language work. The reform of Latin pronunciation in America and England was one of its minor currents, and had a history that the present reform in modern language teaching seems to be repeating. It is interesting now to review the prolonged controversy over the introduction of the Roman pronunciation of Latin into American and English schools. It offers many parallels to the present struggle over alphabets. Let us keep to the English pronunciation of Latin, said they of a 34 THE N. E. A. PHONETIC ALPHABET generation ago; it will help our students in their pronunciation and their use of the mother-tongue. The new-fangled way will be harder to learn, will put another bar in a stile already too high for many of our pupils in Latin to climb over. The "innovation" will throw additional burdens upon teachers of Latin, who must now not only re-instruct themselves in a strange and difficult subject, but must take up the task of teaching the subject to unwilling schoolboys. The English pronunciation of Latin is for us at once the "natural," the logical, and the helpful. We have already a mass of phrases in law, medicine and the arts, which we pronounce in the English way. Let us rally to the defense of our established, tried, and familiar ways against a change that threatens to subvert them. The reformers are themselves not agreed; they have only uncertainty and confusion to offer us. The ruin, moreover, will involve more than our Latin and our English scholarship. Our language itself is threatened! There is a foe in hiding. Spelling reform is on foot; and it and the reform of Latin pronunciation are alhes! That this is no exaggerated picture of the Latin pronunciation controversy of thirty years ago, and later than that in England, a glance into the literature of the struggle will show. A few specimen passages follow: "The whole question is whether it will tend to promote the study of Latin if we put another bar in the stile already too high for most boys in the form of a foreign pronunciation, to which they are unwill- ing to lend their tongues, and which deprives them to a great extent of the help in suggestion and to the memory of like sounding English words, derivative from the Latin." — A London Times correspondent, April 6, 1907. "But of this [i. e., of the value of using the Roman pronunciation in any serious study of Latin philology] boys can not easily be con- vinced; and accordingly they did not care for a change that seemed to them merely troublesome."— Headmaster of Westminster (1879). "I think 'reformed Latin pronunciation' a mere waste of time, and, if done on a fictitious, professor-made plan, absurd. . . . But cui bono 'Reform'? Not for any practical end."— Headmaster of Rugby (1876). "Philological skill is not dependent upon the accidents of utter- ance."— Principal of Phillips Academy, Andover (1879). THE N. E. A. PHONETIC ALPHABET 35 "The entire method is so unlike the English pronunciation of Latin as not to be understood by one who has not made a long and diligent study of it with careful and frequent practice. It makes the study of pronunciation a difficult, repulsive and unnecessary prelude to the study of Latin. . . . How easy and rational to pronounce the Latin word and its English derivative in the same way; vivid and vivid us, circumjacent and circumjaceo ! . . . Attempts to use other systems [than the English] are ineffectual, misguiding, and detri- mental to genuine scholarship, both Latin and English." — Undated pamphlet, reprinted from The North Carolina Teacher. "So far from the system resting upon a basis of truth, the want of harmony among its supporters discredits the evidence adduced in its behalf, and renders uniformity of practice impossible." — M. M. Fisher's "Three Pronunciations," 3d ed. (1884), p. 179. "Will you allow me to thank you for the satisfaction which your 'Three Pronunciations of Latin' has afforded me? Such a work was sadly needed, and I think it has come at the right time. My impres- sion is, that it has caught the ' Latin Pronunciation' on the down-grade, and has given it a blow which I sincerely hope will help to send it to perdition." — Id., p. 216. "I am a total stranger to you, but take the liberty of a brother professor to offer you my best thanks for your capital book on the pronunciation of Latin. . . . One thing I am fully persuaded of, and that is that our would-be classical reformers will not only murder Latin, but slaughter English in the bargain." — Id., p. 215. "There is another phase of this threatened revolution,* . . . the Spelling Reform. . . . That the Reform Association aim at revolution in English orthography is as certain as that two and two make four. That the reformed Latin pronunciation means revolu- tion in Latin is just as certain. . . . Reform in English spelling and reform in Latin pronunciation are natural allies. . . . The hard method must come into our vernacular as an inevitable necessity. For Professor March says, 'We shall pronounce [Latin], of course, as the Romans did.' Hence the English and the so-called Conti- nental modes of pronouncing Latin, according to Professor March, will no more be thought of; and, worst of all, and positively sure in process of time, will be the dragging of our noble English tongue, * The small capitals and the italics belong to the original. 36 THE N. E. A. PHONETIC ALPHABET with all its hallowed associations, back to the hard, harsh, and un- couth standard which Roman Latinists insist characterized the stately Latin two thousand years ago." — Id., p. 168f. Here are arguments much like those one meets nowadays against the proposed scientific alphabet — arguments from expediency, argu- ments from the quiver of an out-of-date scholarship, appeals to passion and prejudice. Vigorous expression was sometimes not wanting on the other side: "That [i. e., the English pronunciation of Latin] is merely a jargon produced by deafness, carelessness, confusion and isolation. It has no merit except a supposed utility to schoolmasters, which I think overestimated." — Another London Times correspondent. The reform of Latin pronunciation won its way in America and in England because it had the best linguistic scholarship behind it. The reform method of teaching living languages is winning its way to-day because present-day linguistic scholarship is behind it. The reform involves a study of phonetics as the basis of language in- struction, as the first step for the beginner, whether a child in the primary years just learning to read its mother-tongue or a boy in the higher grades taking up the study of a foreign tongue. Ad- vanced language work is still more dependent upon phonetics. Phonetics is the science of speech-sounds, the material of which language is composed. In all the Scotch Training Schools for teachers the study of phonetics is a required subject; and in England also, where the government Board of Education has recently lessened the requirement because of too great pressure upon the students, the Training Schools and the teachers are demanding that the subject be restored to the compulsory list.* Progressive teachers of English and of foreign languages in England are following close in the wake of their continental colleagues. It has seemed needful to point out (1) the weight of authority behind the movement to adopt a national phonetic alphabet in this country, and the substantial agreement of all our phoneticians, philologists, and leading educators in the results thus far attained, (2) the grounds of objection to a key of the Webster kind, and (3) * See Mod. Lang. Teaching, Feb., 1910, and a communication from F. M. Purdie on pp. 129-132 in Volume V (1909) of the same journal. THE N. E. A. PHONETIC ALPHABET 37 the great need — indeed, absolute necessity — of a national scientific alphabet to-day, as an instrument for the study and teaching of the native and foreign languages, and as a simple but accurate way to indicate the pronunciation of words in our multitude of school- books and reference works. These are important matters for con- sideration, and Dr. Whipple's pamphlet leaves most of them un- touched. The conclusions that he draws from his limited experiments, even if they were justified by his results, leave still a great deal to be said before one can intelligently and rationally decide against the adoption of the key proposed by the N. E. A. committee. And if it were shown that his experiments were materially unfair and quite in- adequate, and moreover that the results were at times wrongly in- terpreted, in some specific points actually reversed, and that his conclusions from them were therefore baseless, his work must, for every intelligent reader, offer little ground for opposition to the N. E. A. alphabet. That such, in fact, is the case, a review of his pamphlet will show. II It remains now to examine in some detail the experiments that Dr. Whipple made, the manner in which he handled his results, and the reasonableness of the conclusions that he drew from them. In his introductory paragraphs Dr. Whipple refers to the grounds upon which objection has been made to the N. E. A. alphabet: (1) "Practically every elementary textbook in spelling, reading, geog- raphy and history in use in this country must be revised"; (2) "Thousands of teachers will have to drop a familiar system and learn a new one"; (3) "It has also been pointed out that the Proposed Key is open to certain obvious criticisms from the standpoint of phil- ology." As to these objections, one must recognize that the first has been much magnified. There is not now, and has not been in the past, uniformity of usage in the mass of books referred to. The Webster key itself has been greatly altered with each revision of the diction- ary. Not all the textbooks assumed to use that key really do use it as it is or was. If Dr. Whipple will examine the spellers, readers, J 1 & 38 THE N. E. A. PHONETIC ALPHABET and other schoolbooks in current use, he will find a portion of them using a pronouncing key other than the Webster; and of those using the Webster, some modify it. Nor is it any more accurate to say, as some opponents of the N. E. A. alphabet urge, that the adoption of that alphabet will render useless the thousands of reference books in American libraries and homes. Besides the Webster dictionaries, only two such works published in this country use the Webster key, and only one of these, The New International Cyclopedia, is an important general reference work. On the other hand, the N. E. A. key has been adopted for future editions of the Standard dictionaries of the Funk & Wagnalls Company, and the International Alphabet in a new series of bi- lingual dictionaries for school use, namely, the International Uniform Dictionaries published by Hines, Noble & Eldridge, and a slight modification of the International Alphabet in the serviceable and inexpensive Dictionary of Hard Words, by R. M. Pierce, published by Dodd, Mead & Co., 1910. Other American dictionaries and cyclopedias either use a special key of their own (as The Century Dictionary and Worcester's), or else (as the majority of them) none at all. No work of English origin, of course, uses the Webster key. All such books will remain in use, as now, until they are worn out or superseded by better. Furthermore, to say that the adoption of the N. E. A. key will throw any current schoolbooks out of use is to mislead the public. Textbooks in our public schools are frequently changed.* Now, the N. E. A. alphabet could not at once be introduced everywhere, and it is not well that it should be; for teachers must first learn how to use it. t A manufacturer can not suddenly put improved machinery into his plant, however more efficient the new may be. His workmen must know how to use it. But that is the second objection that Dr. Whipple recalls: thou- sands of teachers will have the task thrust upon them of learning a new key. Wouldn't it be a good thing to happen? No one can ob- * In fact, publishers will be found that welcome a frequent change. The ethics of it is left to the reader's conscience. t Indeed, practically all of the objections that have been made in England to the introduction of the phonetic method in language teaching arise from abuse of the method in the hands of teachers not prepared to use it. THE N. E. A. PHONETIC ALPHABET 39 serve what a great advance has been and is being made in the teach- ing of living languages on the other side of the ocean, without recognizing that elementary as well as advanced language instruc- tion in this country will gain immensely if our thousands of teachers could have the opportunity of following their foreign colleagues in improved methods and improved means of language work. Give them the opportunity to do this better work. Few of them will not eagerly grasp at it. The task of learning so consistent and simple a key as the N. E. A. is not insurmountable; on the contrary, under proper guidance, it is negligible. By the time that new editions of texts are needed, teachers can be ready to use them, and the new, simple, rational alphabet proposed by the N. E. A. committee can be incorporated without financial loss to publisher or pupil. And the gain to American schools in efficiency will be beyond reckoning. The third objection that Dr. Whipple recalls — "certain obvious criticisms" of the N. E. A. key "from the standpoint of philology" — is nothing short of grotesque. Corral all the reputable English phil- ologists in England and in America; you will not find among them one champion of the Webster kind of key, nor one acquainted with conditions in this country who will not recognize the N. E. A. as a scientific alphabet well adapted to the situation which it is designed to meet.* No one of its supporters will say that it is really perfect, for it must in the nature of the case serve two masters, the necessity of having a scientific key for successful work in the study of speech- sounds in the primary room and higher, and, on the other hand, the need of making that scientific alphabet at the same time simple and easy enough for general use. The ideally perfect is unattainable. But the N. E. A. key is so immeasurably superior to the Webster — from the scientific philologist's point of view — that hesitation be- tween the two is for the philologist unthinkable. Dr. Whipple then puts a question as to a possible fourth objec- tion to the N.E. A. key: Is it not pedagogically inferior to the Web- ster? A minor bit of evidence against the key Dr. Whipple finds in the fact that "the only system of primers published in this country that incorporated a system of key notation at all like that now pro- posed has been a complete failure and has been withdrawn from the * In Appendix II the reader will find resolutions recently passed by the three leading philological societies of America endorsing the N. E. A. key. 40 THE N. E. A. PHONETIC ALPHABET market. Nevertheless," Dr. Whipple goes on to say, "it will not do to reject the Proposed Key on this evidence alone, since the failure of the primer in question may have been due to other defects than its phonetic system." The assumption that there were other defects in this primer is wholly gratuitous — at least, defects of a kind that Dr. Whipple's words suggest; for possibly the sole defect of the primer may have been in its being ahead of the times. In which case the remedy is to move forward! * Putting aside the minor bit of evidence that the fate of this primer may offer, Dr. Whipple tests the relative merits of the Webster key and the N. E. A. key by means of four experiments; and the conclu- sions that he draws from his results he summarizes as follows: "We have no hesitation in declaring that the key proposed by the Committee of the National Education Association is inferior for peda- gogical purposes to the Webster Key now in common use." One must first notice in Dr. Whipple's experiments a limitation of field, and certain underlying defects, which render his conclusions of less authority than many of his readers may think; and, in the second place, one must recognize that the results of certain of the tests were wrongly interpreted, so much so that at times the rational conclusion — if any conclusion at all is warranted from such hurried work — is the direct opposite of that which Dr. Whipple draws. (1) 1. Dr. Whipple's test of the two alphabets, especially in Experi- ment A, was limited to the use of them as pronunciation keys. A phonetic alphabet has other pedagogic uses. Dr. Whipple does not consider the relative merits of the two keys as, for instance, instru- ments for general phonetic work, that is, instruction in the nature and classification of speech-sounds and training in the reproduction of such sounds. The importance of such work, and the necessity of having a rational phonetic alphabet to do the work with, one may * To see ourselves as others see us, read the following from one of the leading En- glish educational journals: " We have very long wondered at the very unprogressive character of the American publisher of books for modern language teaching, and, in particular, at the absence of good schoolbooks that pay regard to the modern advance in phonetics" {Mod. Lang. Teaching, July, 1911, p. 143). THE N. E. A. PHONETIC ALPHABET 41 learn who considers only briefly what has been done and what is being done with phonetics in Europe and in England. Nor does Dr. Whipple undertake to show the relative merits of the two keys for beginning students in foreign languages. There is, indeed, no need to do this; for experience has already decided in favor of a scientific alphabet (see pp. 25ff. above). Dr. Whipple, again, takes no account of the absolute necessity of a scientific alphabet in all advanced study and training in the science of phonetics, a science that is the sole basis of all other scientific study of language. Furthermore, Dr. Whipple does not take notice of the fact that scientific bodies like the U. S. Geographical Bureau, the Bureau of American Ethnology, the Royal Geographical Society, have already adopted a scientific alphabet on the basis that the N. E. A. uses, and that their practice is in time sure to be followed by compilers of popular handbooks that use their material. So that, even if the Webster key were the easier to learn, it would not be universally serviceable. However, it was quite right for Dr. Whipple to put upon his investigations whatever limits his time and his material dictated; only, this must be emphasized: his conclusions, even if justified, are not the whole of the matter. 2. But the limitations marking his experiments are not all that lessens the force of his conclusions. He tests the two alphabets on conditions so unequal that his results were foregone; for (a) the per- sons submitting to the experiments (both grade pupils and college students) were distinctly, though perhaps in cases unconsciously, possessed of a knowledge of a greater portion of the Webster key than of the N. E. A.; (b) the time given to learning the keys was al- together too brief for the results to be permanent, the defect working, of course, to the disadvantage of the less familiar key ; and (c) those who conducted the experiments seem not to have been familiar with both the instruments they undertook to te3t, and not to have used the newer instrument to its best advantage. (a) The Webster key is based upon English alphabetic and pro- nunciation values.* In the mere process of learning to say the alphabet, a child begins to associate with the letters the sounds that are heard in their alphabet names. In spelling and reading his eye later meets time and again certain combinations of letters in a * The statement is of consequence in the present discussion chiefly when applied to vowel symbols. 42 THE N. E. A. PHONETIC ALPHABET given phonetic value, and grows more or less familiar with them, so that it should be easier for him to accept them as phonetic symbols. Further, if his textbooks provide respellings in the Webster or any like key, he will have, by the time he reaches the end of the Seventh Grade, a very definite sense or f eeling for certain symbols as embody- ing in themselves certain sounds. The Webster symbol 60 , for example, is striking enough, and the words in which it has the Web- ster value are frequent enough, to make the "double-o" an old ac- quaintance to a seventh grade pupil. Only one of the Webster pupils in Test 6 of Experiment A (Appendix B) failed on the "00"- part of the symbol; about one-third of them, however, confused the diacritics. It is impossible to avoid the conclusion, from a study of Appendix B (Whipple, pp. 29-30), that the chief source of error of those pupils using the N. E. A. key was the attempt on their part to give the alphabetic or some known pronunciation-value to the symbols. In fact, a theoretical consideration of the question is not needed here. The reader should merely consult Dr. Whipple's own figures, in Appendixes G and H, to see how overwhelmingly a child, after it has got fixed in mind the association of letter and sound forced upon it by conventional English spellings, interprets unknown or vaguely known phonetic symbols in the values suggested by con- ventional spellings— which, in the case of several Webster symbols, are identical with the Webster values. For a psychologist, Dr. Whipple underestimates astonishingly the degree in which the average seventh grade pupil must be as- sumed to be familiar with the Webster key. On page 9 he attempts to make some allowance for it. He says: "It may be objected that we have not allowed for the fact that a few pupils knew the long and short values of the vowels a and e in Key 1 [the Webster key]. This objection is partly met by the inclusion of one more character in our test. To obviate it entirely, we may exclude all reference to these four vowels. By consulting Appendix B it will be found that 19 errors were made with these four characters. When these are subtracted, we have left 20 characters and 18 pupils, a total of 360 cases, with a total of 91 errors, or 25.3 per cent wrong. Hence the elimination of the four Webster symbols known to a few pupils actually reduces the per cent of wrong pronunciations in this THE N. E. A. PHONETIC ALPHABET 43 key from 25.5 per cent to 25.3 per cent. Accordingly, the fact that our results in Test 6 indicate a decided superiority of the Webster Key can not be attributed to the slight knowledge of this key pos- sessed by a few pupils." Dr. Whipple here speaks apparently of the knowledge the pupils may be supposed to have had of the Webster key previous to their five lessons. Now, why does he assume that knowledge limited to these four symbols, a a e e ? Let the reader turn to page 22 of Dr. Whipple's report, where the author is stating the results of Experiment C with thirty-eight Ithaca grammar school children who (presumably) had had no previous instruction in either key. Dr. Whipple there says: "To be more specific: 'a' is pronounced according to Webster 132 times, according to the Proposed Key, 9 times; 'i' is pronounced according to the Webster Key 128 times, according to the Proposed Key, 4 times; 'e' is pronounced ac- cording to the Webster Key 121 times, according to the Proposed Key, once; 'u' is pronounced according to the Webster Key 45 times, according to the Proposed Key, 8 times." * On Dr. Whipple's own showing, then, he should have excluded also I and u ; for by Experiment C he proves that these, as well as a and e , are overwhelmingly pronounced in the alphabetic values of the letters (which are the Webster values of the symbols) by grammar pupils who (presumably) have had no previous instruction in the Webster key. If we " exclude all reference" to these symbols also, we shall have, for Test 6, on the Webster key, 18 symbols and 18 pupils, a total of 324 cases, with 85 errors, or 26.2 per cent wrong — an increase of .7 per cent instead of Dr. Whipple's decrease of .2 per cent. But to Dr. Whipple, as to Macbeth, came figures that "mar- shaled him the way that he was going." More than that, there are some things here pertinent that Dr. Whipple apparently did not see. He shows a blindness to certain facts that seems strange in a psychologist. Our present question, throughout Experiment A, is not only as to what Webster symbols the Cortland grammar pupils, before their instruction began, would be likely to pronounce in the Webster values, but also as to what help in learning other symbols they would have from their * For , the revised statement that Dr. Whipple makes in his footnote, p. 22, is incorporated above. 44 THE N. E. A. PHONETIC ALPHABET already fixed knowledge of alphabet-names and pronunciation-values of the letters appearing in the Webster symbols. Let us dismiss all recollection of Dr. Whipple's appendixes and tables, and ask ourselves the questions: What Webster symbols are, from the start, likely to be read by seventh grade grammar pupils in the value assigned them in that key? and what further symbols would be especially easy for them to learn, precisely by reason of the Webster leaning on alphabetic and pronunciation values? One can hardly be blind to the fact that the alphabet-names of the vowel signs would be the first sounds to suggest themselves, and in the case of the long vowel signs (a e i o u ) would at once fix themselves in the mind upon presentation of a familiar key-word. There is, further, no doubt that the Webster symbols oi ou , when presented with the key-words oil out beside them, would offer no difficulty. The combination oi every pupil had repeatedly met, in familiar words, and always with the value (N. E. A.) ei . So the letters ou , also, in one after another simple and familiar word, had for every pupil spelled (N. E. A.) au . For the Cortland grammar pupils, the acceptance of the Webster symbols oi ou meant simply the acceptance of the letters oi ou in their known pronunciation values. The Webster consonant symbol zh , too, after once being seen, would be remembered; it is easy to remember as a parallel to the perfectly familiar sh . Of the short-vowel symbols, e and I would offer little difficulty because the letters here involved appear (obscure vowels disregarded) in only two symbols, the so-called "long" and "short" (e e, i I), of which the former ( e i ) fix themselves at once with alphabet- name values, leaving the other two, without confusion and without difficulty, to be associated with familiar key-words end and ill. But the case would be different with a 6 u . These would be hard for the pupils to learn, because (in the first place) of the number of symbols using the same letter with varying diacritic marks: aaaaa;*oooo; uuu. Manifestly, * Confusion, it would seem, must inevitably arise among so many "a"- symbols. That it did arise, Appendix B (Whipple, p. 29) shows. Hence one wonders why Dr. Whipple chose & as one of the four symbols of which some of the pupils had "slight knowledge." If he had not included & in his little group of four "knowns," his figures would have been reversed; the result would have THE N. E. A. PHONETIC ALPHABET 45 the difficulty of remembering all the symbols in each of these series, and distinguishing each symbol from its comrades, is far greater than with a series of two, like e e or i I . A child is likely to re- member the first of each series here (a 5 u ), because its value is the alphabet-name of the letter; but on the others he would easily fall into confusion. Further, several of the vowels in question are muffled, back vowels, difficult to make with exactness, and to recognize with exactness when made. This is especially the case with the sounds whose symbols involve the letter u . Only by effective phonetic training can one bring ear and speech-organs to an intelligent and confident mastery of these sounds. To test un- trained or ill-trained children on them is to invite confusion. We hold it to be highly probable, then, that a grammar-school pupil would have little difficulty, if any, with the following Webster special symbols (of the 24 used in Test 6 of Experiment A) : a e I u e I oi ou zh . If Dr. Whipple cares to "exclude all refer- ence to these" nine symbols, well and good. But if he eliminates a a , e e without eliminating also I u I oi ou zh , he has his trouble for nothing. He can prove nothing unless he is con- sistent in his argument. It is not consistent to eliminate four symbols on the ground of their "slight" familiarity when there are at least six others equally or better "known" to the pupils at the start, in the midst, and at the end of their lessons. If we eliminate our nine symbols, we have left 15 symbols, or 270 cases, with 90 errors. That is, 33.3 per cent wrong, instead of Dr. Whipple's 25.5 per cent. Dr. Whipple might better have found an increase of 7.8 instead of a decrease of .2, in his percentage of errors made with the Webster key, in Table 2. Dr. Whipple may not be willing to go in the direction that these figures marshal him; but he must at least accept the figures of his own Experiment C. These point him the same way, if he will trust them for guidance as to the knowledge a grammar pupil is hkely to have of the Webster key — or, rather, is likely to appear to have; for at the beginning it is knowledge of the English alphabet and of English pronunciations. Indeed, no intelligent man can fail to recognize that, with pupils been an increase instead of decrease in the percentage of symbols wrongly pro- nounced. 46 THE N. E. A. PHONETIC ALPHABET of the Seventh and Eighth Grades, and with college students as well —that is, with persons for whom English conventional spellings and pronunciation values had become instinctive — the Webster key, because it is based upon alphabetic and pronunciation values, would start with a distinct advantage over a genuinely phonetic key like the N. E. A., even if the subjects of the test had had no previous specific instruction in the Webster. For the first few periods of learning, the Webster would be greatly helped by the fact that to every one who can read English a number of the Webster symbols would have become familiar in the Webster value along with other values, so that the learner has merely to fix in mind one of two or three already known values. These half-familiar symbols could be quickly learned. Throughout further study of the key the Webster would still find some help in the fact that in each case the key-word is a familiar English word containing in its ordinary spelling the symbol itself, and so helps the learner to remember it. These ad- vantages the Webster key would have (with pupils advanced beyond the primary) because of its leaning upon the English alphabet- names and English spelling values — the very thing about it that makes it — chiefly in its vowel symbols — impossible as a genuinely scientific phonetic key. On the other hand, consider the difficulty that some of the N. E. A. vowel signs must present at first sight to any one familiar with the English alphabet names and with English conventional spellings (cf. pp. 23f. above). Think of the shock it must bring when for the first time one who has for years been saying "a e i o u " , as in the English alphabet, is told that e is not e but a (Webster symbols are used because the matter is put from the alpha- bet-name point of view), I is not I but e, u is not u but 60. The N. E. A. alphabet begins the race with a heavy handicap. But Webster adherents (other than Dr. Whipple) will perhaps say: That is just what we assert. The Webster key is already in good part familiar; it is easier to learn because it can lean upon alpha- betic and pronunciation values already perfectly familiar to the pupil. The N. E. A. key is difficult because in the case of some symbols the pupil must turn topsy-turvy a set of associations as familiar to him as his own name. It is like asking him to write his name backwards. The Webster key is the "natural" key. THE N. E. A. PHONETIC ALPHABET 47 Dr. Whipple, indeed, is far wrong when he declares that the sub- jects tested in his experiments had only "slight knowledge of the Webster key," and worse than wrong when he attempts to prove it by eliminating only a part of the symbols that were certainly in appreciable measure familiar. (b) The assumption that the two keys were practically on equal footing is an error on Dr. Whipple's part that underlies all his work and destroys the force of his conclusions. Further than this, the time he gave to the experiments was inadequate for the portion of the Webster key that he used (the most puzzling portion of that key he excluded), and for the N. E. A. key, taught in the way that he taught it, the time was farcical. In Experiment A pupils of the Seventh and Eighth Grades met for study five times. Each time they pronounced once, following the director, each of a series of 44 or 46 sounds and key-words, at the same time looking at the phonetic symbol of the sound and at the key-word. The first time they did this, their attention was specific- ally called to the form of the symbol. That was the whole of their instruction. At the end of it the Websters were far from complete mastery of their key; the N. E. A.'s, having started pretty far behind their rivals, were also far from complete mastery of theirs. In the fourth and fifth lessons, however, they were catching up. Nor did the college students in Experiment B learn thoroughly either key; none of them reached complete mastery of even the Webster.* They started with a predisposition, ingrained, life-long, to assign to the symbols the English values. They may not have consciously learned any part of the Webster scheme; but in much of their reading and study, especially in the lower schools, they had here and there, with greater or less frequency, run across phonetic symbols of the Webster kind. Six of those symbols ( a e I 5 u oi ) were perfectly familiar to them, or at least learned with perfect ease; eight more ( a e I 6 u oo doou) should not have taken more than a second thought. The students' study of foreign languages was not of the importance in this test that Dr. Whipple assumes; for such knowledge is at best left-handed. English was * On page 19 Dr. Whipple, in comment on Teat 3, Experiment B, says that the considerable number of "common errors" "gives further evidence of the difficulty of this test." It gives also evidence that the keys were neither of them "learned." 48 THE N. E. A. PHONETIC ALPHABET their right-hand speech, in constant use and practice. It is the English value that unconsciously and instantly sprang up before them. The key-words were familiar English words in which the letter-portion of the symbols themselves stood in the Webster value. That the key-words were not shown to the students was of little con- sequence because persons of their maturity could instantly visualize the words. In these conditions it was only natural that to attain an errorless recitation would require more repetitions of the N. E. A. key than of the Webster; and that the knowledge of the N. E. A. symbols, having been on the average for a shorter time in the pos- session of the students, was not so stable. That this was the state of things with the college students, as with the grade pupils,every person who considers the matter will recognize; and the fact is, moreover, proved by Dr. Whipple's figures in Ap- pendixes G and H, where it is brought to the test with both school children and college students. The various tests demonstrated that English-speaking persons whose habits of speech have become fixed, require an appreciable time to memorize the symbols of either key; and that, when the time is inadequate, the less familiar key suffers more than the other from that lack of time. But the tests do not show which key is the hardest to fix completely in mind; the experiments were too brief, the learn- ing periods inadequate. Nor do the tests show how the N. E. A. alphabet would fare in primary instruction in reading, the place where the child should get its first knowledge of speech-sounds. Hundreds of trained men have experimented with phonetic alphabets, on a far larger scale than Dr. Whipple did. The experi- ments have covered a wide range of country, in hundreds of "labor- atories," and a wide range of time as well. The results of these multiplied thousands of experiments led the army of experimenters all to the same conclusion — a conclusion just the opposite of that which Dr. Whipple's brief five lessons with about seventy seventh and eighth grade pupils led him to publish. And these hundreds of experimenters who do not agree with Dr. Whipple are men who know the material that they are working in. The conclusions that the modern language teachers of all Europe and practically all the modern language teachers of Great Britain, and further that many teachers of the mother-tongue, too, in all those countries, have reached THE N. E. A. PHONETIC ALPHABET 49 — the conclusions of these hundreds of highly trained and experienced teachers, based upon an extended experience for each of them, is rather more trustworthy than those arrived at in one scientific labora- tory, on the basis of one week's work with some seventy school children and a few hours' work with three dozen or fewer college undergraduates. The matured opinion of these hundreds of teach- ers is that a rational phonetic alphabet — the N. E. A. kind — is the most efficient means by which to instruct pupils in the art of pro- nunciation, the easiest kind of alphabet to learn intelligently and to apply. (c) These teachers, it should be repeated, know something of phonetics, of the nature and relations of speech-sounds, of the diffi- culties inherent in their task, of the specific effects desired to be at- tained by a phonetic alphabet, of the specific qualities needed in such an alphabet to make it an effective instrument, and of the specific ways in which such an instrument is to be handled in order to make it efficient. It is not apparent that those who conducted the Cornell experi- ments were familiar with the material in their hands. Indeed, Dr. Whipple himself shows a lamentable uncertainty in the phonetic values of some N. E. A. symbols and the normal usages of educated Englishmen and Americans (cf. pp. 13ff. above); and he or his asso- ciates showed some unfamiliarity with the difficulty that every one who is not a trained phonetician has in recognizing and reproducing speech-sounds by imitation alone (cf. pp. 27f. above). In illustration of that difficulty, look at Dr. Whipple's Appendix B, page 29 of his pamphlet. In every case where a pupil translated an o-sign (6 6 6) into the sound a or a (Webster a. a ), a sound heard in the first syllable of artistic , it must be interpreted as an intention of the pupil to give the so-called "short o " of the key- words odd not . The great majority in this country pronounce these words not with the sound represented by the N. E. A. symbol e ( Webster 6 or 6 ), but with the clear vowel sound of artistic . To expect pupils who have this habit of pronunciation to recognize and correct their deviation from the value assumed in both keys to be the standard value of the o in not , by a process of merely hearing the sound five times, is to expect the impossible. Further evidence that the experimenter was not in position to 50 THE N. E. A. PHONETIC ALPHABET direct, properly, experiments in phonetics appears in his "Directions for Conducting Experiments with Phonetic Keys," Appendix I, p. 37. In the section (4a) under the heading Method (p. 38) one reads: " Now, the vowels, a , e , i , o , u , have different sounds, as you know, like 'ay' and 'ah' , 'ee' and 'eh' , and so forth." But a e i o u are letters, not vowels. A vowel has one measurably fixed sound — it is a sound. To say that "the vowel a has different sounds," when you mean that "the letter a repre- sents different sounds," is to use terms unscientific, and confusing to any one attempting to get clear notions about the relations between sound and sound, and between sound and letter. No experiments with phonetics, directed by persons who use so loosely the terms of the science, can be accepted as trustworthy. In several tests it appears that the experimenter does not realize that it makes any difference whether he gave the letter-name or the sound-value of the symbols he spoke of; which would not be surprising in an experimenter who is not at the same time a phonetician, but should be surprising in an experimenter who is a psychologist. What is necessary for the subjects of experiments with phonetic keys is some authoritative and competent instruction in speech- sounds, some training by practice in the recognition of the smaller differences in sounds and in the accurate reproduction of those sounds — some training of ear and speech-organs by an instructor who knows something of phonetics. Some of the Webster symbols would have fared better if the subjects under test had had proper instruction. But the N. E. A. key fared far worse because there was no at- tempt to present it on its most telling merits, namely, its inherent simplicity and logical consistency. These are qualities that make it incomparable with any key of the Webster kind (cf. p. 17 above). It is easy to learn because, when the eleven or twelve simple elements of the vowel-scheme are once mastered, the whole English vowel- scheme is in the learner's possession; for he needs only to combine the elements in a fixed and consistent way to represent the varied values of the twelve English vowel sounds not interpreted by the fundamental eleven signs. His instinctive, unconscious-acting rational sense does half the work. To throw away the help that reason gives, and to learn the N. E. A. vowel system in the same way that one must learn the Webster — that is, by sheer force of memory THE N. E. A. PHONETIC ALPHABET 51 — is like resorting to forge fires and sledge hammers when the subtle electric current would weld the metals with a fraction of the labor and time. That is what the conductors of the Whipple experiments did. They did not show their pupils how to learn and how to use the N. E. A. key. One may have in one's hand a delicate instrument, but that doesn't mean that one knows how to use it. A chemist, however competent in his own sphere, can not step into a watch factory and at once set to using the watch-maker's fine machinery without dis- astrous results; and when such results follow his attempt, they can not be interpreted as sufficient ground for condemning the machinery itself. Dr. Whipple recognizes that his investigations leave something to be desired in thoroughness and completeness. He does not, how- ever, appear to realize how inadequate and one-sided they were. Not five days but a year or two in time is necessary. The experi- ment should be made not with pupils or students who have already ingrained in them associations that years of use of the English con- ventional spellings have fixed, but with pupils who are beginning to learn the language — with children in the primary reading classes, or with American-born children of foreign parentage who have not yet fixed the one language in mind to the exclusion of the other. And, above all, the experiment must be in the hands of conductors whose training fits them for the specific task — of language teachers who know something of phonetics, who know how to teach it, and who have mastered the key that they attempt to use. In short, any experiments testing two or more specific phonetic keys, should be directed by expert phoneticians, and carried out by men who represent equal knowledge of and sympathy with the keys in question. It will take some time to make such an experiment; but it is better to walk six blocks for an accurate watch than five blocks for a poor one. But overtopping all these fundamental errors, there was yet a fourth error vitiating all of Dr. Whipple's experiments, which one finds it hard to excuse. Dr. Whipple used in his work the portion of the Webster vowel symbols which is least open to objection. He eliminated from both keys a portion, namely, the symbols for the 52 THE N. E. A. PHONETIC ALPHABET "obscure" vowels, in which the superiority of the N. E. A. key is most marked ,the confusion of the Webster most trying (cf .pp. 12-14 above). Whatever the grounds for the action, it was wholly un- warranted, and irretrievably biased the results. The Webster symbols for the obscure vowels (agouS ANG&LES, CALIF. UNIVERSITY OF CALIFORNIA LIBRARY Los Angeles This book is DUE on the last date stamped below. :j(i//( 7,'68(J1895s4) — C-120 yjowej L 007 114 701 '-■ uc SOUTHERN REGIONAL LIBRARY FACILITY AA 000 352 078 o