Open Access

JSON Export

{"created":"2022-01-31T13:18:12.029347+00:00","id":"lit28753","links":{},"metadata":{"alternative":"Studies from the Yale Psychological Laboratory","contributors":[{"name":"Tuttle, E.H.","role":"author"}],"detailsRefDisplay":"Studies from the Yale Psychological Laboratory 10: 96-117","fulltext":[{"file":"p0096.txt","language":"en","ocr_en":"PHONETIC NOTATION\nBY\nE. H. Tuttle.\nI. General principles.\nIn order to discuss speech sounds in a concise and intelligible manner, we must represent them by signs of fairly definite meaning. Ordinary orthography will not do for this purpose, since in nearly every language there is more or less of irregularity in the relation of spoken words to their written forms. Thus, in English, groups of letters are often used for simple sounds, and conversely ;\t\u2018 rough \u2019 = \u2018 ruff, \u2019\t\u2018 sword \u2019 =\n\u2018soared,\u2019 \u2018phlox\u2019 = \u2018flocks.\u2019 In German \u2018wird\u2019 = \u2018 wirt(h)/\n\u2018 stadt \u2019 = \u2018 statt,\u2019 \u2018 viel \u2019 = \u2018 fiel,\u2019 \u2018 feld \u2019 = \u2018 f\u00e4llt,\u2019 \u2018 so(o)le \u2019 = \u2018 sohle,\u2019 \u2014z rimes with\u2014ds, \u2014ts, etc. In French \u2018 car\u2019 = \u2018 quart, \u2019 \u2018sansr = \u2018 sang \u2019 = \u2018 sens \u2019 = \u2018 sent \u2019 = \u2018 cent,\u2019 \u2018 saoul \u2019 = \u2018 sous \u2019 = \u2018 soue,\u2019\n\u2018 ceins \u2019 = \u2018 ceint \u2019 = \u2018 cinq \u2019 = \u2018 sain \u2019 = \u2018 saint \u2019 = \u2018 sein \u2019 = \u2018 seing,r \u2018souhait\u2019 = \u2018soi.\u2019 In Italian half the letters of the alphabet are used in two or more different manners.\nIt is evident that we need for scientific purposes a sign system free from such serious faults as these. It should, however, be kept in mind that to construct and employ a mathematically exact system would be practically impossible, because of the linguistic and physiological differences between individuals.1 Many of the phonetic symbols used here must therefore be understood as representing groups of slightly different articulations, sounds or properties.2\nOne of the first points to be considered is how to distinguish phonetic letters from orthographic ones. Many writers, overlooking the importance of this consideration, make no distinction between orthographic and phonetic spelling, with the result that their work is sometimes unintelligible, or, worse yet, misleading. Thus, when a French writer speaks-\n1\tThe formula system of representing sounds is evidently intended as an approach to mathematical accuracy, but can hardly be called a phonetic transcription in the proper sense of the term ; it is rather a description in a sort of physiological shorthand. (JES-persen, Fonetik, K\u00f6benhavn 1897-1899.)\n2\tThe fact that speech does not consist of a series of adjacent sounds independent of one another, as implied by an alphabetic representation, will be left out of account here.\n96","page":96},{"file":"p0097.txt","language":"en","ocr_en":"Phonetic flotation.\n97\nof \u201cu allemand \u201d or \u201c y anglais,\u201d 1 we are more or less in the dark as to what he is talking about.\nIn some cases it may be found convenient to use phonetic and orthographic signs in combination ; thus for instance in a discussion of stress or pitch, much space can be saved by writing \u201c 1Ja\u00b0pa2nese \u201d instead of \u2018 \u2018 \u2018 Japanese \u2019 'djie'fifniis, \u2019 \u2019 where the particular vowels and consonants used are of no especial importance. But this freedom should not be extended too far; such mixtures as \u2018\u2018mamzel, via, dja, syisi, membr?,\u201d2 are highly objectionable, as they might seriously mislead a reader not perfectly familiar with the language represented.\nAs the scientific notation employed in this article contains roman3 letters, it will be enclosed in brackets, to prevent confusion with orthography. The English textual system is in italics, which will permit dispensing with the [ ].\nAnother important point in phonetic work is consistency. This also has received none too much attention, even from the best writers. Passy uses two different modifiers on the same page4 5 to indicate syllabicness. Rousselot uses \u201c h\u201d for the two quite distinct fricatives of German \u2018 hage. \u20195 Ellis employs the sign ( \u2019 ) in two apparently quite different ways : (\u2019) = [/], and (\u2019n) = syllabic or long [\u00ab] ;6 7 another phonetician coming upon a text containing numerous (\u2019)\u2019s, mistranscribes it\u2019 into an unpronounceable form containing syllabic v, z, etc. Sweet transcribes the Arabic dental fricatives sometimes by \u201c$ ]> \u201d'and sometimes by \u201cS <V\u2019 without apparent reason for this distinction.8 Storm uses \u201c \u00e6, e, e\u201d sometimes for the same vowel (that of \u2018air\u2019), sometimes for different sounds.9 Hempl seems to use \u201cf\u201d and \u201cv\u201d with the same value.10 11 Sie vers employs \u201c r' \u201d to mean /-modified r, \u201c e\u2018\u201d to mean e followed by nonsyllabic/, and to indicate a half-sonant o.n There is a similar inconsistency in the use of superior letters in the system of Jespersen,12 and in that of Murray.13 We find one of the vowels\n\u25a0Rousselot, Principes de Phon\u00e9tique Exp\u00e9rimentale, 157, 181, Paris 1897.\n2 Storm, Englische Philologie, 2te auflage, 187, Leipzig 1892.\ns I use \u201c roman \u201d in contrast to italic ; \u201c Roman \u201d in contrast to Greek, Russian, etc.\n\u2018Passy, Sons du Fran\u00e7ais, 5e \u00e9dicion, 58, Paris 1899.\n5 Rousselot, Principes de Phon\u00e9tique Exp\u00e9rimentale, 410, 554, Paris 1901.\ne Ellis, Early English Pronunciation, 10, London 1869-1889.\n7\tSievers, Grundz\u00fcge der Phonetik, 4te auflage, 41, Leipzig 1893.\n8\tSweet, Practical Study of Languages, 31, 233, New York 1900.\n9\tStorm, as before, ix, 56.\n10\tHempl, German Orthography and Phonology, 156, xxx, Boston 1898.\n11\tSievers, Grundz\u00fcge der Phonetik, Ste auflage, 185, 180, no, Leipzig 1901.\n'\u201cJespersen, Fonetik, 229, 515, K\u00f4benh\u00e0vn 1897-1899.\n13 Murray, New English Dictionary, vol. I, p. xxv, Oxford 1888.","page":97},{"file":"p0098.txt","language":"en","ocr_en":"98\nE. H. Tuttle,\n(a small \u201c 0 \u201d) that is defined as \u201c short \u201d in Hamilton\u2019s notation used with the length-modifier, and a vowel ( \u201c q \u201d ) that is defined as \u2018 \u2018 long \u2019 \u2019 used for the short vowel of \u2018not.\u20191 In Passy\u2019s system the character \u201cj\u201d is used in three entirely different ways : to indicate a consonant ; to indicate qualitative modification of the preceding vowel ; and to indicate both qualitative and quantitative modification.2\nOf equal importance with typographical consistency is linguistic consistency. Sweet transcribes English \u201c long e \u201d as vowel + consonant, and then gives \u2018 sea \u2019 as an illustration of a final long vowel.3 After criticising Sweet for calling English \u00e4 and \u00f6 vowel diphthongs, but \u00eb and oo vowel-consonant groups, Sievers transcribes \u2018 yE \u2019 with a diphthong, and < wound \u2019 with a simple long vowel.4 Hempl calls w in \u2018 way \u2019 a consonant, but y in \u2018 yet \u2019 a vowel,5 although these sounds are really of the same general character. Some of the dictionaries6'transcribe \u2018allow\u2019 with two V s, although only one has been pronounced during the last ten or twelve centuries. Until very recently the American Dialect Society has used \u201c<7as in not\" along with \u201c\u00f4, ou as in no \u201d and \u201coi as in coin,\" 7 notwithstanding the confusion that is liable to arise from such a system.8 9\nConsistency in the physiological basis of a transcription is also very desirable. When we find a fricative defined as being similar to the Arabic and German glottal catch, and the sounds [ot] and [\u00ab] called fricatives, while [yz] and [75] are classed as occlusives,8 we feel tempted to doubt the accuracy of certain other statements made by the same writer. The two Italian z-sounds are often described as simultaneous t and s, d and 2,10 although such a combination of stoppage and nonstoppage of the air-current is physically impossible. One well-known sound system divides the vowels according to the assumed tongue-positions into \u2018 front,\u2019 \u2018 mixed \u2019 and \u2018 back,\u2019 and \u2018 high,\u2019 \u2018 mid \u2019 and \u2018 low \u2019 ;\n1\t[Hamilton,] With the linguists, Herald, April 1902 (Toronto), p. 90.\n2\tExpos\u00e9 des principes de !association phon\u00e9tique internationale, Suppl\u00e9ment au Ma\u00eetre phon\u00e9tique de novembre 1900 (Passy), p. 8, 12, 15, 18.\n3\tSweet, Primer of Phonetics, 21, 43, Oxford 1890.\n\u2018Sievers, as before, 163.\n5 Hempl, as before, xxx, xxvii.\n\u00ab [March,] Standard Dictionary of the English Language (Funk), New York 1895.\n[Barnum,] Webster\u2019s International Dictionary of the English Language (Porter), Springfield 1901.\nI Dialect Notes, vol. i, p. 27, vol. ii, p. 190.\n8\tDialect Notes, vol. i, p. 233, 452.\n9\tAraujo, Fon\u00e9tika Kastellana, 24, 36, 37, Madrid 1894.\n10\tD\u2019Ovinio und Meyer, Die Italienische Sprache, Grundriss der Romanischen Philologie (Gr\u00f6ber), Iterband, 491, Strassburg 1888.\nEllis, as before, 800.","page":98},{"file":"p0099.txt","language":"en","ocr_en":"Phonetic notation.\n99\nthe last class is defined as having the lowest possible position of the tongue,1 the results of such a definition being apparently overlooked. In pronouncing a \u2018 low-front \u2019 vowel, where is the back of the tongue ? It cannot be above \u2018 high \u2019 nor below \u2018 low \u2019 ; it cannot be \u2018 high \u2019 nor \u2018 mid\u2019 (for either of these positions would produce \u2018back \u2019 vowel), and consequently must be \u2018low.\u2019 In other words \u2018low-front\u2019 is identical with \u2018low-back.\u2019 Further discussion is hardly needed to demonstrate the frailty of the whole system.\nThe legibility of phonetic characters is an important consideration. In one system we find \u201cs\u201d and \u201cs\u201d used for different sounds.2 Sweet uses for [y] and for [j>] two very similar signs3 \u2014 so similar, in fact, as to be misprinted in one of his own books.4 * Frequently \u201cce\u201d and \u201cce\u201d are used in the same alphabet.6 Vietor transcribes two German sounds by\tand \u201cy.\u201d6 Passy employs \u201cg\u201d and \u201cg\u201d for\ntwo sounds as nearly distinct as ^ and t in his so-called \u201cinternational alphabet \u201d ;7 this, of course, makes difficult reading in texts where both signs are used.8 Lepsius usesa \u201c t\u201d with a microscopic \u201c s\u201d beneath it in transcribing Slavic languages, because these happen to represent the sound group ts by a single letter.9 Rousselot often employs minute superscript letters as modifiers;10 in the case of nasalized \u201ca\u201d the tilde is so small that it might easily be mistaken for a macron or grave accent.11 The more new types an alphabet has, the less easy will it be to read and to remember ; this reason, as well as the one mentioned below, explains to a large extent the failure of new-type alphabets to come into common use. The illegibility of his organic alphabet in its script form has been admitted by Sweet himself ;12 for the ordinary educated person the printed characters are extremely hard to read.\n1 Sweet, as before, 14.\nI\tCentury Cyclopedia of Names (Smith), vii, New York 1899.\n\u2019Sweet, as before, 97, 98.\n* Sweet, Practical Study of Languages, 17, New York 1900.\n\u2019Vietor, Elemente der Phonetik, 4te auflage, Leipzig 1898.\nVietor, German Pronunciation, 2d edition, Leipzig 1890.\nSievers, Grundz\u00fcge der Phonetik, 5te auflage, Leipzig 1901.\n6\tVietor, as before.\n7\tMa\u00eetre Phon\u00e9tique (Passy), since 1894.\ns Vietor, Aussprache des Schriftdeutschen, 4te auflage, Leipzig 1898.\nDjelali, Contes et Chants Arm\u00e9niens, Paris 1899.\n9\tLepsius, Standard Alphabet, 2d edition, London 1863.\n10\tRousselot, Principes de Phon\u00e9tique Exp\u00e9rimentale, 332, 614, Paris 1897-1901.\nII\tRousselot, as before, 224.\n12 \u201cAs I had some difficulty in deciphering your postcard\u2014which isn\u2019t your fault but the fault of visible speech which I\u2019m afraid is quite unsuited for practical work\u2014I write in Roman letters.\u201d Sweet, quoted by Spieser, lautfrift, Ma\u00eetre Phon\u00e9tique, 1895 X 139.","page":99},{"file":"p0100.txt","language":"en","ocr_en":"IOO\nE. H. Tuttle,\nIt is not sufficient that a system should not be difficult to read ; it must not be unpleasant to read, if it is to be at all widely adopted.\nIn order to be pleasant to read, a textual alphabet should be as uniform as possible; a mixture like \u201c iygli/'t/\u00e0ild \u201d is highly objectionable, where the italics serve no phonetic purpose.1 An exception to this principle may be made, however, where it is desirable to call attention to certain sounds. Thus in an italic text for English students it might be well to transcribe the Russian palatalized consonants or the French orinasal vowels by roman letters, or by heavy-face ones.\nCompactness is also an important feature of an alphabet. The clumsiness of using several characters for a single sound becomes a serious waste of space in cases like \u201c \u0153oe \u201d 2 for long [3], and \u201c p[h] \u201d 3 4 for English p. Practical experience has shown that it is best to indicate quantity, stress and pitch by separate modifiers ; these signs should be as small as is compatible with legibility. In a scientific notation the use of modifiers may be considerably extended.\nThe employment of many diacritics should be avoided ; a transcription with a mark over or under nearly every other letter, or with two and three marks on some, gives to the printed page a repellant aspect.* A single word like \u201cz\u00fc-z\u00e9\u201d (Vietor\u2019s notation) = French \u2018jugeait\u2019 [jyje] will illustrate this fairly well. In the system of Lepsius, the second vowel of French \u2018emprunte\u2019 is written \u201c 0\u201d with a straight line under it, another over it, a tilde over this, and an acute accent over the tilde; in Rousselot\u2019s alphabet the corresponding sign is \u201cce\" with a vertical line under it, a grave accent, a macron, and a tilde over it.\nA fourth requirement is that, so far as possible, familiar letters should not be used in unfamiliar ways ; a transcription like \u201c psje \u201d 5 for French \u2018monsieur\u2019 is very disagreeable to a person familiar with Greek. In English, where every letter is used in more than one way, any fixed use must be to some extent unfamiliar. For the vowels a, \u00e6, e, i, o, u, y, we can hardly do better than return to the Old English values ; the consonants should, of course, be used with their commoner or older sounds, except in the case of the superfluous ones (c,j, q, x).\nA phonetic notation, to be suitable for general use, should be easy to print with types now in common use. Although Passy has recently\n1\tSweet, Practical Study of Languages, 17, New York 1900.\n2\tSweet, as before, 41.\n4 Passy, Sons du Fran\u00e7ais, 5e \u00e9dicion, 120, Paris 1899.\n4 Vianna, Exposi\u00e7\u00e2o da Pronuncia Normal Portuguesa, 99, Lisboa 1892. ViETOR, German Pronunciation, 2d edition, 113, Leipzig 1890.\n5Sweet, History of Language, 28, New York 1900.","page":100},{"file":"p0101.txt","language":"en","ocr_en":"Phonetic notation.\nIOI\ndenied the validity of this proposition,1 he practically admits it by using in one of his most important books2 his easily printed sonant-velar-fricative sign instead of the proper new character to represent the occlusive [,\u00a7\u2022].\nThe casting of new types requires a considerable outlay of time and money, and consequently many writers, publishers and printers would refuse to have anything to do with a phonetic notation containing them. On this account the various new-type systems that have been constructed, including those now employed by Bell3 * and by Sweet,1 seem to be total failures so far as general adoption is concerned. The numerous Roman-basis alphabets now used all seem to require a larger or smaller number of new types, and are therefore all equally unsuited for practical use ; for the employment of a single specially cast character makes a system almost as difficult to use as would the employment of a hundred.\nWhile it is highly desirable that new types should not be used, it is perhaps even more important not to waste old types that are readily available and suitable for phonetic purposes. In the systems of Ellis and of Lepsius capitals are used as in orthography, without any phonetic significance ; this greatly increases the number of new types in one, and of digraphs in the other, besides making the alphabets much more difficult to use. In one of the best of the dictionaries5 we find the same sound represented in different ways : \u2018 come cum, \u201d \u2018 came k\u00eam. \u201d Such inconsistencies are not permissible in a phonetic alphabet ; a system containing them is really only a reformed spelling, not a phonetic one.\nII. Details. a. Types used.\nThe most readily available types are of course those of the Roman alphabet : a b c d e fghijklmno pqrstuvwxyz\\ the corresponding capitals may also be employed in a scientific notation -.ABC D E F G HIJ K L MN O P QR S T U V WX FZ. As small capitals are made only in roman, and as many of them are more or less similar to lower-case letters (c i j k o s u v w x z), no use of them will be made here.\nOf the \u2018 \u2018 accents \u2019 \u2019 commonly made for English fonts, the following are used : \u00e0 i etc., \u00e2 \u00ea etc., \u00eb \u00ef \u00f6 \u00fc ( ; of the ligatures, \u00e6 fi ; of the\n1\tPassy, lynite hnetist, Ma\u00eetre Phon\u00e9tique, 1901 XVI 106.\n2\tMichaelis et Passy, Dictionnaire Phon\u00e9tique de la Langue Fran\u00e7aise, Hannover 1897.\n3\tBell, Sounds and their Relations, Salem 1881.\n1 Sweet, Primer of Phonetics, Oxford 1890.\n5 [March,] Standard, as before.","page":101},{"file":"p0102.txt","language":"en","ocr_en":"102\nE. H. Tuttle,\nfall-sized numerals, 2 j; of the punctuation marks,\t-j- J ||.\nThe Danish alphabet furnishes two very serviceable types, \u00e0 and 0, and Greek a large number : \u00df y Se\u00c7 y ft iX\u00c7p<jT<px<pu>'These foreign types are in a few cases chosen arbitrarily; in most instances they are used with approximately their native values, or else are to be considered as modifications of similar Roman letters.\nThe number of available characters can be increased by means of different styles of type \u2014 as roman, heavy-face, superior, inferior \u2014 and by means of cut and turned letters. It should be noted that some types are not suitable for use in this last manner ; turned b, d, n, p, q, s, u, x, z are rather hard to read. Turned o is indistinguishable from direct o, unless the alinement is very uneven ; the use of the two together1 makes a text practically illegible.\nCut types are of two sorts, face-cut and body-cut. The process of body-cutting is sometimes undertaken by printers in preference to ordering special characters from a foundry, such as s, accented \u00e6, capital \u00e7, etc. It is very objectionable not only by reason of the considerable time and expense involved, but because the cutting and adjusting are often poorly done, lying as they do outside of the typesetter\u2019s regular line of work. Face-cutting is a comparatively simple matter ; it should not, however, be employed for any great number of signs. In the textual notation (English) of this article, only one face-cut type is needed, the length-sign ', made from an exclamation point. Certain kinds of face-cutting should not be undertaken without proper consideration of the other principles involved. Thus the use of dotless i or j may seem a very simple expedient; but to use\u201ci\u201d and \u201ci\u201d2or and \u201cj\u201d in the same alphabet is a serious mistake, because in printing the dots are sometimes broken off, so that whenever the reader sees one of these types he is uncertain whether the mutilation is accidental or intentional.\nb. Scientific notation.\nA scientific notation should be so constructed as to be capable of providing a suitable transcription for any speech sound. An attempt will be made here to outline such a sign-system, due regard being paid to the principles considered in the preceding pages, and to the present usages of philologists.\nOne of the first things to be symbolized is the motion of the air-current. While the exact method of indicating this is not very important, it is\n1\tAraujo, as before, 143.\n2\tExpos\u00e9 etc., as before, p. 14.","page":102},{"file":"p0103.txt","language":"en","ocr_en":"Phonetic notation.\n103\nimportant to have some method, in order to avoid the confusion that is to be found in some writers.1\n[||] respirate (current due to lung-action).\n[J] irrespirate (current not due to lung-action).\n[f] ingressive (current passing into vocal organs from external air).\n[).] egressive (current passing from vocal organs to external air). Irrespirate and inspirate sounds are often used in interjections, as [ttTw]> [||f/4j] ; otherwise ordinary speech is generally expirate in European languages. The modifiers [||] and [i] will therefore be usually understood, where the contrary is not indicated.\ni. Consonants.2\nLABIAL\nBILABIAL\nocclusive, [b ; p] F. b ; p.\ncorresponding orinasal, [m] E. m.\nvibrant, [m] in [A\u00bb1] expressing a feeling of coldness.\nfricative, [/? ; Dutch w ; Japanese f. The sign similar to u used\nfor [/9] by Passy is quite objectionable, because it is a new type ;\nalso because in italics it must be printed as \u201c v,\u201d3 * the distinction\nbetween \\v\\ and [/3] thus being lost, or else the relation of \u201cu\u201d\nand \u201cv\u201d must be just reversed, in \u201c v\u201d and \u201c v,\u201d* in order not\nto disturb the universal association of v with v.\nstrictive, [<\u00ab] a rare substitute for E. r.\nBILABIAL-MARGINAL\nocclusive orinasal, [m\\ 5 a common respiration-articulation with sonancy added ; [mf] = \u201cyes.\u201d\n\u25a0Passy, Sons du Fran\u00e7ais, 5e \u00e9dicion, 116, Paris 1899.\nRousselot, as before, 488.\n2 The terms \u201cconsonant\u201d and \u201cvowel\u201d will be used here in the ordinary manner, without discussion of the hitherto undetermined distinction between the two classes. It is noteworthy that the consonants can in nearly all cases be referred to fairly definite genetic conditions, or articulations ; while this is not easily done for the vowels, as is shown by the widely differing vowel-theories upheld by various writers.\nWhere two consonant-signs in the list are separated by a semicolon, the first is sonant, the second surd ; where only one is given, it is sonant unless the contrary is stated. Articulations are oral unless stated to be orinasal.\nThe following language names are abbreviated to their initial letters : American English, British English, English, French, German, Hungarian, Italian.\n3Ma\u00eetre Phon\u00e9tique (Passy), 1897 XII 80, 81.\n* YTetor, Aussprache des Schriftdeutschen, 4te auflage, 5, 17, Leipzig 1898.\n5 To be associated with m.","page":103},{"file":"p0104.txt","language":"en","ocr_en":"io4\nE. H. Tuttle,\nBILABIAL-DORSAL\nstrictive, [w] E. consonant w ; approximately a combination of the labial [\u00ab\u00ab] -position and the lingual [\u00ab] -position.\nLABIOBIDENTAL\nsurd fricative, \\fi] in [\u00eef/tfi1] sometimes used to urge on horses.\nLABIODENTAL\nocclusive, [b ; p]. corr. orinasal, [m]. fricative, [v ; f] F., I. v; f.\nDENTAL\nfricative [j ; /,.] ; inspirate [/,*'] is sometimes used as an interjection of pain.\nLINGUAL1\nmarginal\nINTERDENTAL\nfricative, [5 ; #] E. sonant th ; surd th. The signs \u201c 8 j> \u201d for these sounds are in general use among Germanic philologists ; the chief objection to them is that italic ]> is very similar to certain styles of p, so that one is liable to be misprinted for the other.2 Another important consideration is the fact that many printers, and even type-founders, have difficulty in furnishing italic 8 J>.3 roman 5 $ are needed, 9 might be used for the latter, while the former could be made by cutting 8 ; or 5 might be considered the roman letter corresponding to italic S. The German printers\u2019 crossed d, evidently intended as an improvement on 8, is unsuitable for practical use, being much inferior to it in legibility.4\n1\tNames of lateral articulations are in italics.\n2\tModern Language Notes, 1901 XVI 154 (c. 307).\nEmerson, History of the English Language, 235, New York 1897.\n3\tSweet, Practical Study of Languages, 135, 233, New York 1900.\nSweet, History of Language, 28, 125, 131, New York 1900.\nSievers, as before, 147.\nEmerson, as before, 241.\nHempl, as before, xxvi, xxviii, xxtx, 128.\nAmerican Journal of Philology (Gildersleeve), 1901 XXII 427, 430.\n4Compare Lloyd\u2019s Northern English (Leipzig 1899) with Sweet\u2019s Elementarbuch des Gesprochenen Englisch (Oxford 1895).","page":104},{"file":"p0105.txt","language":"en","ocr_en":"Phonetic notation.\n105\nPOSTDENTAL\nfricative, [o ; <r] sometimes distinguished in English, though perhaps never distinctive : |[\u00ab\u00a3*'] = [sikshs] \u2018 sixths \u2019 ; [hmzs>3tn\\ = [h\u00e6zdoni] \u2018has them.\u2019\nALVEOLAR\nocclusive, \\d\\ /] E. d; t. corr. orinasal, [\u00ab] E. n.\nocclusive, [d; r]. Independent lateral occlusives are rare ; [J+t1] or [+tri] *s use(^ t0 urSe on horses. In words like \u2018 wildly \u2019 and \u2018 faultless \u2019 the occlusives are necessarily lateral, and do not need to be marked as such in an ordinary transcription, vibrant, [f] Scotch r.\nfricative, [z ; j] E., F. z ; E., F., G. \u201chard\u201d s. fricative, [\u00cf ; f\\ . Welsh 11 = [j] ? strictive, [\u00ff] Londonese consonant r ? strictive, [/] G. 1.\nALVEOLAR-DORSAL\nstrictive, [/] E. 1 ; a [/] with approximately [\u00ab] -modification of the dorsal tongue position.\nCEREBRAL\n[d t n r z s j 1] differ from [d tnrzsu /] in having the tongue point much farther back.\nPAGINAL\nfricative, Qy ; f~\\ F. j, E. si in \u2018vision ; E. sh, F. ch, G. sch. These signs, now coming into common use, are preferable to the German philologists\u2019 z s by reason of their greater legibility and ease of writing, and also because the Bohemian accented letters are difficult to obtain in this country. The chief objection to f is its similarity to ordinary f\\ this may be avoided by using an italic f with no kern at the bottom. If a roman character is needed, the letter f might be used ; the resemblance could be increased by removing one serif. The articulation of the paginais seems to be quite variable; the assumption that [/] is necessarily formed farther back than [r] 1 is as incorrect as the older theory that [/ ] is simply [s] .2\n1\tSie vers, as before, 131.\n2\tHenry, Comparative Grammar of English and German, 27, London 1894.","page":105},{"file":"p0106.txt","language":"en","ocr_en":"io6\nE. H. Tuttie,\ndorsal\nPREPALATAL\nocclusive, [y ; c]1 H. gy ; ty.\ncorr. orinasal, [yz] Italian gn, Spanish n.\nfricative, [j; \u00e7] G. j g in \u2018j\u00e4ger\u2019; ch in \u2018machte, b\u00fbcher.\u2019\nfricative, ; \u00a3].\nstrictive, [/] E. y in \u2018year,\u2019 e in \u2018 ewe, few.\u2019 This sound is about as distinct from [y ] as [2] is from [5] ; it is a serious mistake to use a single sign, as some writers do, for both sounds. The relation of [/] to the strictive [z/] of F. \u2018lui\u2019 seems to be the same as that of the vowels [z] and [y].\nstrictive, [-1] I. gl(i). Passy\u2019s double use of italic X,2 sometimes for this sound, and sometimes for a /z-sound, is objectionable ; <( might be used instead of J., but the other form (\u00c6) should not.\nPALATOVELAR\nocclusive, [g; 0] E. \u201chard\u201d g; k.\ncorr. orinasal, [tj] E. ng in \u2018young,\u2019 n in \u2018younger, uncle.\u2019 fricative, [y ; y] G. g in \u2018l\u00e4ge\u2019 ; ch in \u2018macht, buch.\u2019 Passy\u2019s use of \u201cg\u201d for the sonant fricative3 is extremely objectionable, because it involves, in \u2018a roman notation, the use of a new type for the sound [y], and because the letter g is in most European languages commonly associated with the occlusive. The unpractical character of Passy\u2019s notation is shown by the fact that several books4 represented as using it employ \u201cg\u201d as an occlusive, while one5 uses \u201c g \u201d for [g] about twice as often as for [y].\nPOSTVELAR\nocclusive, [g ; k]. corr. orinasal, [9].\nvibrant, [>] a common form of German r (uvular trill), fricative, [q ; x]. A roman form more similar to y could be made by cutting off half the loop of q. strictive, [2].\nstrictive, [I] Russian \u201chard\u201d 1? The Polish crossed / (i) often used for this sound is objectionable because it is not easy to obtain, even from type-foundries, and because it is easily misprinted as z1.6\n1\tTo be associated with j \u00e7.\n2\tMa\u00eetre Phon\u00e9tique, 1897 XII 79, 1900 XV 66.\n3Passy, Sons du Fran\u00e7ais, 5e \u00e9dicion, III, Paris 1899.\n* Michaelis et Passy, as before.\nPassy, Le Fran\u00e7ais Parl\u00e9, Leipzig 1897.\nLloyd, Northern English, Leipzig 1899.\nSpiers, Senior French Reciter, London 1902.\n5Rippmann, Elements of Phonetics, 7, 14, 35, 40, 52, 57, 79, etc., London 1899.\n6 Ma\u00eetre Phon\u00e9tique (Passy), 1893 VIII 109, 1902 XVII I, 108.","page":106},{"file":"p0107.txt","language":"en","ocr_en":"Phonetic notation.\n107\nLARYNGEAL\nGLOTTAL\nsurd occlusive, [;] G. \u2018 \u2018 glottal catch. \u2019 \u2019 surd fricative, [/z] E. h, G. h.\nOther sounds of this class seem to be distinguished in oriental languages ; but until their articulations are better known, it is hardly necessary to provide signs for them.\nii. Vowels.* 1 * * * 5\nNORMAL\n[i] F. i, G. ie, I. i.\n[/] G. open i.\n[e] F. \u00e9, G. eh, I. close e.\n[\u00ab] G. e in \u2018 feld,\u2019 \u00e4 in \u2018 f\u00e4llt.\u2019\n[\u00e9] F. \u00e8, \u00ea, G. \u00e4h, I. open e.\n[\u00e6] E. a in \u2018 pats,\u2019 A. a in \u2018 past.\u2019\n[a] F. a in \u2018part, page, patte.\u2019\n[a\\ I. a.\n[q] F. \u00e2.\n[a] E. au, aw.\n[oY I. open o.\n[0] G. o in \u2018sollte, sonne.\u2019\n[o] F. eau, G. 00, oh, I. close o.\n[zz] G. open u.\n[u] F. ou, G. uh, I. u.\n1\tAs the relations of the vowels are a matter of much dispute, it may be well to give some reasons for the classification adopted here.\n(1) The normal series forms, when whispered, a fairly continuous pitch-scale ; (2) in pronouncing the normal vowels simultaneously with [V], the easiest is [ \u00ab] ; passing from\nthis in either direction, they become more and more difficult, [u] and [i] being about\nequally hard ; (3) in pronouncing the normal vowels, in the order given, the tongue-point is gradually retracted ; (4.) the vowels [a,], [\u00e0,], [a,], [\u00ab,] can be distinguished fairly well, while attempted nasal [as], [<?], [i] give [a,], their dorsal articulations being in front ofthat of [k]. This seems to show that the normal vowels lie in a single con-\ntinuous series, and that the 0- and \u00ab-articulations are behind the uvula.\n(5) The j-sounds can be formed without \u201c rounding,\u201d and the f-sounds with it, without any considerable change in quality ; (6) both these classes of sounds can be pronounced simultaneously with m ; (7) attempted simultaneous [/] and [a] gives a sound very similar to [/]. This apparently proves that the/-sounds are combinations of u- and z-positions (slightly reduced?), and are not (necessarily) \u201crounded\u201d z-sounds.\n2\tTo be associated with 0, as e is with e.","page":107},{"file":"p0108.txt","language":"en","ocr_en":"io8\nE. H. Tuttle,\nREDUCED\n[i] intermediate to [i] and [i\\. p] intermediate to p] and p].\n[\u00e8] intermediate to [e] and p], etc.1\nNEUTRAL\np] simple voice, the unaccented vowel of E. \u2018 fungus, wondrous. \u2019 NORMAL COMPOUND\n[y] combination of [u] and [i], F. u, G. iih, Danish close y.\n[y] com. of p] and pr], G. open ii.\n[0] com. of [o] and [e], G. \u00f6h, F. close eu.\n[0] com. of [0] and [<?], G. open \u00f6.\n[3] com. of p] and [e], F. open eu.\np]2 3 4 com. of [a] and pe], B. stressed u in \u2018 fungus \u2019?\nREDUCED COMPOUND\n[\u00ef] com. of [i] and [\u00f9], Russian bi. p] com. of p] and [ u\\.\n[\u00eb] com. of [e] and [o], etc.8\nThe use of most of the vowel signs requires little comment. The sign \u0153, often used for p], is avoided here in order to permit the use of \u00e6 instead of \u00e6, and because of its conflicting associations ; if \u0153 is to be used at all, it should be for the sound intermediate to p] and p>] .* If roman s is needed, Russian 3-like 2 might be used. The sign \u00e0 seems necessary for several reasons. The sounds \u00e6 and e are distinguished by Sweet 5 6 and by Passy 6 as here, but they use the single letter y for the vowels of B. \u2018 knot, nought,\u2019 as well as for the somewhat different one of F. \u2018 port, note. \u2019 The distinction is just as important in the latter case as in the former, and is therefore indicated here by \u00e2\u2014y. The character \u00e2 indicates the sound intermediate to the \u00ab-group and the \u00ab-group, just as \u0153 does the one between the \u00ab-group and the \u00ab-group ; and it is especially suited to American English, because commonly written with an a : \u2018 all, awl, haul, talk, watch,\u2019 etc.\n'Indicated by a grave accent over the corresponding normal vowel.\n2\tTo be associated with a.\n3\tIndicated by two dots over the corresponding normal vowel.\n4\tAtkinson, \u00e6lfybit, Ma\u00eetre Phon\u00e9tique, 1900 XV 61.\n5\tSweet, Practical Study of Languages, New York 1900.\n6\tPassy, Sons du Fran\u00e7ais, 5e \u00e9dicion, Paris 1899.","page":108},{"file":"p0109.txt","language":"en","ocr_en":"Phonetic notation.\n109\niii. Modifications.\nLENGTH\nIn many languages the time-element (duration, quantity, length) of sounds is an important factor of speech. In the Germanic tongues similar long and short vowels are often appreciably different in quality : compare English \u2018fairy\u2014harry, wren\u2014rein, pick\u2014pique, good\u2014food,\u2019 etc. ; from which fact there has arisen the common notion that a difference in quality necessarily accompanies difference of quantity.1 2 This idea however is readily proved false: in music the \u201cshort \u201d vowel of \u2018sat\u2019 may in one measure be sung to an eighth note, and in another to a half note, with any noticeable difference of quality. In some languages, as French and American English, words are distinguished solely by the duration of their vowels ; quantity signs are therefore often indispensable in a phonetic notation.*\nThe indication of length by doubling a letter is objectionable not only by reason of its sprawliness,3 4 but because it renders impossible, in a textual notation, a distinction like that of the diphthong ii in \u2018 key-ring \u2019 and the simple long i in * hearing.\u2019\nThe use of a colon (:) to indicate length has been adopted by many writers.* This may do well enough in the Romance languages, where final vowels are commonly short, but it is unsuitable in the Germanic ones, because at the end of a word it may easily be mistaken for and misprinted as a punctuation mark.5 A modified form with angular dots6 is rather illegible, resembling i with the dot broken off, so that in transcription \u2018call\u2019 looks like \u2018coil,\u2019 \u2018pause, paws\u2019 like \u2018poise,\u2019 etc.7\n1\tSoames, Introduction to Phonetics, 2d edition, 54> London 1899.\n2\tMuch of the otherwise excellent work published by the American Dialect Society is made more or less unintelligible by the lack of any systematic method of marking vowel-length.\ns Sweet, Primer of Phonetics, 100, Oxford 1890.\n4\tMichaelis et Passy, as before.\nPassy, Le Fran\u00e7ais Parl\u00e9, 4e \u00e9dition, Leipzig 1897.\nViETOR, Aussprache des Schriftdeutschen, 4te auflage, Leipzig 1898.\nRippmann, Elements of Phonetics, London 1899.\nSpiers, as before.\n5\tYet even in Romance speech interjections sometimes have final long vowels, so that confusion may occur. For actual examples, see Passy\u2019S Fran\u00e7ais Parl\u00e9 (4th edition) t\n\u2022 \u201c sokrr page 9, line 3, has a short vowel, while \u201co :\u201d,p. 11,1.8, is long.\n6\tPassy, \u00c9l\u00e9ments d\u2019Anglais Parl\u00e9, 3e \u00e9dition, Paris 1900.\n7\tSpiers, A moyen, Ma\u00eetre Phon\u00e9tique, 1902 XVII 87.","page":109},{"file":"p0110.txt","language":"en","ocr_en":"no\nE. H. Tuttle,\nThe popular method of marking quantity by the macron and breve, \u00e4, \u00e4, \u00eb, \u00eb, etc., is unsuitable for an extensive scientific notation, because it requires the casting of new types for long and short \u00e6, \u00e2, o, \u00f4, \u00fc, n, s, etc. To avoid typographical difficulty, it has been proposed that these marks should be put after the letter modified instead of over it.1 This plan is unsuitable for a textual notation, because if the marks used are large enough to be properly legible, they take up a great deal of space (half an em or more). As the length sign seems to be the only one needed in ordinary transcriptions, a mark like the macron, but in a different position ['J, is proposed here. If other signs are required in more accurate work, the following might be employed :\n[T] over-long.\n[\u2022*\u25a0] half-long.\n[\u201c] short.\n[\"] extra-short.\nThese are placed after the sound-sign modified.\nPITCH\nA common manner of indicating pitch, or intonation, is the use of slanting lines, angles and the like to mean rising, falling, high, etc.2 This method is (or might be made) a good one, as far as it goes ; but if it is used, care should be taken to provide for all necessary variations. In one system,3 where only level, rising and falling tones are indicated, we find that the \u201c tone-marks are put before the word they modify ; if they modify a whole sentence, they are put at the end of it \u201d ; using this it would be impossible to distinguish normal \u201c Is he ready ?\u201d with a rise throughout from \u201cIs he ready? \u201d with a rise through three syllables and a low tone for the fourth.\nTo indicate pitch with considerable accuracy, ordinary musical notation is sometimes employed. This system is hardly suitable for textual use, because of its bulkiness ; it might be replaced by numerals corresponding to the semitones : [\u00b0] for the lowest note of speech (which is usually a little higher than the lowest singable note), f1] for the next semitone, [2] for the next, etc. This provides for a variation extending\n1\tVi ANNA, kjresp\u00f6d\u00e4:s, Ma\u00eetre Phon\u00e9tique, 1890 V 105. Haldeman, Analytic Orthography, 83, Philadelphia i860.\n2\tSweet, Primer of Phonetics, 65, Oxford 1890.\nPassy, Sons du Fran\u00e7ais, 5e \u00e9dicion, 71, Paris 1899.\nSievers, Grundz\u00fcge der Phonetik, 5te auflage, 225, Leipzig 1901.\n3\tSweet, Primer of Spoken English, 3d edition, 3, Oxford 1900.","page":110},{"file":"p0111.txt","language":"en","ocr_en":"Phonetic notation.\nhi\nthrough nine semitones, a range that is not commonly exceeded in speech. If the complete octave were needed, superior io could be indicated by [x], and n by [Y], the groups [10], [01], [\u201c], etc. thus being left for glides within a vowel or syllable ; the second octave could be marked with [*], or some other arbitrary sign. Taking the lowest note as do, we should have these correspondences :\n[\u00b0] do\t[3]\tri,\tme\t[6]\tfi, se\t[9]\tla\t[*\u00b0]\tdo\n['] di, ra\t[4]\tmi\t[']\tsol\t[x]\tli, se\t[*']\tdi,\tra, etc.\n[2 *] re\t[5]\tfa\t[s]\tsi, le\t[Y]\tsi\nThe pitch-marks should be placed before the syllables modified. STRESS\nThe common method of indicating stress in English, by a mark ( ' ) after the accented vowel or syllable, is unsatisfactory because it affords no means of distinguishing ordinary \u2018without\u2019 from artificial or emphatic \u2018 with-out \u2019 ; the use of a mark on the vowel is objectionable for the same reason, and because of typographical difficulties with v, 3, a, \u00e2, etc. The use of an accent before the stressed syllable, as in one of Passy\u2019s systems, is entirely unsuitable for English texts because of conflicting associations with the ordinary use of the accent. The best system seems to be that employed by Sweet1: [\u2022] for strong stress, [:] for medium, and [-] for weak. If other distinctions are required, they might be indicated by combinations of these marks, as [ \u2022\u2022 ] for very strong, corresponding to orthographic italics. The stress signs are placed at the beginning of the syllable.\nSYLLABLES\nThe common philological notation for the nonsyllabic function is the mark ^ under the sound-sign modified. As this requires new types in most cases, a more practical means of indicating a nonsyllabic vowel would be to put the same sign after the vowel : [\u00ab],[<?], etc.\nOn the analogy of [\u201c_J we might use [,], the turned length sign, to indicate a syllabic consonant, without thereby implying any direct or fixed relation between quantity and syllabic function ; to use the same sign for these two things, as some writers do,2 is not advisable, because syllabic and nonsyllabic sounds alike can be either short or long.\n1\tSweet, Elementarbuch des Gesprochenen Englisch, 3te auflage, Oxford 1S95.\n2\tBell, Sounds and their Relations, Salem 1881.\nLloyd, Northern English, Leipzig 1899.","page":111},{"file":"p0112.txt","language":"en","ocr_en":"11 2\nE. H. Tuttle,\nIn the description of a textual notation given below these signs are not used, since in English the vowel-diphthongs have the second portion nonsyllabic, while syllabic consonants occur only before a consonant or a pause.\nSONATION\n[P 4> F 0 T~], etc. For capital f the letter f can be used. SURDATION\n[l m \" \u201d], etc. If it is necessary to distinguish whispered sounds from surds, they can be indicated by smaller letters printed on the line instead of superior.\nEXPLOSION\nAn occlusive before a pause or another occlusive has little audibility unless it is exploded ; in a scientific notation there is need of signs to indicate the explosion : ['] surd, [*] whispered, [\u2019] sonant. The last of these differs from the ordinary vowel 3 in having much less duration, and loudness, or sonancy, as may be seen by comparing \u2018 Rhoda \u2019 roud? with \u2018 road \u2019 roud. The sign ['] might be used in a very accurate textual notation to indicate aspiration of surds before any sound, as \u2018 stay \u2019 sfei,\n\u2018 try \u2019 frai, etc., although in such cases the ['] can hardly be considered a separate sound, but is rather a sign that the following vowel or strictive commences surd.\nWhere it is necessary to indicate the absence of explosion, this may be done by the use of a period; English \u2018actor, Bagdad\u2019 have [\u00a3.], [o\\], not [/&'], [^\u2019] like the similar French words.1\nNASALIZATION\nThe Polish and Portuguese signs for orinasal vowels have come into general use among phonetists ; unfortunately neither system is very practical, since only two letters are available in each (a, e; \u00e4, \u00f6). In a scientific notation we can, as Haldeman suggests,2 imitate the Polish sign by a turned apostrophe [,] after the letter modified: Fr. \u2018vin\u2019 [z>\u00e6(], \u2018vent\u2019 [wq,], \u2018vont\u2019 [wj, \u2018un\u2019 [sj. These vowels occur independently in English in some interjections, as [\u00e6^\u00e6j, [A?,], etc.; they may often be distinguished, but are not distinctive, in cases like [setnd~\\ \u2018send,\u2019 [witnd~\\ \u2018wind.\u2019\nComplete nasality, formed by closing the oral passage at the [k] -position, may be indicated by [,]. While the nasal vowels perhaps do not\n\u2019Passy, as before, 121.\n2 Haldeman, as before, 46.","page":112},{"file":"p0113.txt","language":"en","ocr_en":"Phonetic notation.\n\u201d3\noccur in any language, they are important for experimental purposes (p. 107, note 1).\nModification of one sound toward another can be represented by the use of inferior letters, as [7J in a common American pronunciation of \u2018winters,\u2019 [r.], [mj in Russian \u20187,\u2019 etc.\nAdvancement of articulation may be indicated by [,], as [/] postdental, [7 J interdental, [tt/J lingualabial ; retraction by [J, as [/] postalveolar, [t] postpalatal.\nAny modification desired might be indicated by inferior numerals ; for instance [rj, [r2], [rj, etc. to show the number of flaps.\nc. Textual notation.\nA notation used to transcribe a continuous text in a single language must differ somewhat froni'the scientific notation given above ;.especially in that digraphs like at, av a.v a3 for simple sounds are not permissible, except for length, stress and pitch modifiers; typographical distinctions like b, b, 6 should in general be avoided, except where it is desirable to call the student\u2019s attention to some unfamiliar modification or class of sounds in a foreign language.\nThe following list includes signs for the American English sounds that need to be distinguished in an ordinary literary text ; sounds that occur only or chiefly in exclamations are not taken into account.\nb \u2018 oBey \u2019 ; p1 2 3 \u2018 appear \u2019 ; m \u2018 OMit \u2019 ; syllabic m \u2018 cup and saucer \u2019 ; v \u2018 leaves \u2019 ; f \u2018 leaF \u2019 ; w \u2018 wight,\u2019 \u2018 white \u2019 hwaitd\nd \u2018 ri\u00fcer \u2019 ; t's \u2018 wriTer \u2019 ; n \u2018 hoNor \u2019 ; n] \u2018 teNDs \u2019 ;4 syllabic n \u2018 listEN \u2019 ; \u00f4 \u2018 eiTHer \u2019 ; >'/ \u2018 ef Her \u2019 ; z \u2018 raises razes \u2019 ; s \u2018 miss \u2019 ; s1 \u2018 misTS\u2019 ;5 6 j \u2018vision,\u2019 \u2018age\u2019 eid3 ; * f \u2018 fisH, \u2019 \u2018rich\u2019 ritfp \u2018nature\u2019 neitfo ; r= [1]\nSometimes pronounced where not written, as \u2018 warm( )th, dream( )t, some()thing.\u2019\n2\tA few speakers do not pronounce the h in this and similar words, so that \u2018 white \u2019 \u2014 \u2018 wight,\u2019 \u2018 where \u2019 = \u2018 wear,\u2019 \u2018 which ' = \u2018 witch,\u2019 etc.\n3\tOften inserted where not written, so that \u2018Welsh\u2019 =\u2018 Welch,\u2019 \u2018dense\u2019 =\u2018dents.\u2019 The consonants d t n l may become nearly interdental when adjacent to d or \u25a0&, as in\n\u2018 earthen,\u2019 and are slightly retracted after apical vowels, as in \u2018 bartered.\u2019\nf In careful or emphatic speech the d is often pronounced, so that the word is perfectly distinct from \u2018 tens.\u2019 In some pronunciations this long n does not seem to be distinguished ; likewise with long l.\n5\tIn artificial and emphatic speech the t is often pronounced : mists.\n6\tIn these consonant groups the occlusive and fricative belong the same class ; either may be slightly different from ordinary d 13 f. In my pronunciation the paginais and r have about the same lip-position as w.","page":113},{"file":"p0114.txt","language":"en","ocr_en":"E. H. Tuttie,\n114\n\u2018 meRRy \u2019 ;1 / = [/\u2022] \u2018 hoLy whoLLy \u2019 ; /' \u2018 hoLDs \u2019 ;2 3 4 syllabic / \u2018 handLEs \u2019 ; g \u2018 finGer \u2019 ; k% \u2018 CHoir Quire \u2019 ; rt \u2018 siNGer \u2019 ; syllabic rj \u2018 lookiNGglass \u2019 ; / \u2018 you Ewe,\u2019 \u2018 Hugh hue hew \u2019 hjtiu ; h \u2018 Hold.\u2019\nThe surd occlusives, p, t, k, when not followed by an occlusive are usually aspirated, though very weakly before an unstressed vowel. The importance of this aspiration is shown by the fact that English-speakers are liable to mishear unaspirated p t k as b d g.* The consonants k g -q vary somewhat according to the neighboring sound ; they do not however reach the full prepalatal position, as sometimes happens with French gu, qu: [\u00ab/] \u2018 quel,\u2019 [ri] \u2018qui,\u2019 etc. The sonant occlusives and fricatives are usually whispered before a surd ; the sonant occlusives become surd after a surd, but are kept distinct from p t k by the absence of aspiration ; h tends towards sonancy after a sonant. The glottal occlusive i seems to be often substituted for p, t or k between ori nasals, as in \u2018 bluntness. \u2019\nii \u2018 see sea, kEY-ring \u2019 ; [z'i] ; with some speakers nearly [i '].\nft \u2018spEAring, pEriod\u2019; [z'] ; some replace this by a diphthong, h ; similarly with long e, 0, u.\n\u00d9 1 SpEARS, pIERS pEERS \u2019 ; [z>J .\ni \u2018 spirit, busy, give \u2019 ; [z'J, nearer to [z] than to pure [z].\ni \u2018 studdEd studiEd, Accept Except, palAce, lettuce \u2019 ; unstressed and therefore rather variable, [z], \\_e] ; only preconsonantal.\nz \u2018 busiest, sixtieth, cerEal serial \u2019 ; unstressed ; approximately [e], varying towards [\u00eb] or [z] ; chiefly prevocalic.\nei \u2018 say, wEight wAit\u2019; approximately\tbut the second portion\nseems to vary towards [e] ; before r or p sometimes replaced by long [,?].\ne \u2018 says, mEn, mEAnt \u2019 ; [z] ; before g and -y this may become slightly diphthongal, ee, the second part being [e] : \u2018bEg, lEngth.\u2019\n1\tPublic speakers and singers often use ar=[r], with one or more flaps.\nSome Americans, like many Londoners, insert r as a hiatus-filler after <v and long d, a : \u2018 idea ( ) of, draw( )ing, Shah ( ) of,\u2019 etc. This r is probably not due, as has been supposed (Lloyd, Phonetische Studien, 1892 V 89), to apicality of the vowels, since it is used chiefly by persons who cannot (or at least do not) employ apical vowels in \u2018 hard, furthers \u2019 ; it is merely the result of analogy, like so many other speech changes (for example \u2018 nothing\u2014nothin \u2019 from a similar variation in present participles ; \u2018 we was\u2014he was \u2019 from the lack of inflection in all other preterits ; * different than \u2019 from \u2018 other than\u2019). Most words ending in ?, and many in long 4, a, are written with an r that is regularly pronounced before a vowel ; this prevailing duality of pronunciation has simply been extended to all words with these endings.\n2\tIn artificial or emphatic pronunciation sometimes houluz, perfectly distinct from \u2018 holes.\u2019\n3\tSometimes inserted where not written, as \u2018 leng( )th, streng) )th.\u2019\n4\tCuST, Report on Korean, Transactions of the Philological Society, 1877-8-9, p. 615.","page":114},{"file":"p0115.txt","language":"en","ocr_en":"Phonetic notation.\n\u201d5\ne1 \u2018 cAring, fAiry, bEArer \u2019 ; [e1].\neo \u2018 cARes, fAiRs, bEARs \u2019 ; [e^] ; sometimes replaced by \u00e9] [e*J. \u00e61 \u2018 bAd, pAth, pAst, hALve, smAsh, sAnd, nun \u2019 ; nearly the same in quality as short \u00e6, with perhaps a slight tendency toward [e1].1\nas'bAde, hAth, hAst, hAve, sAsh, plAnned, bAn\u2019; sometimes diphthongized to \u00e6e before g and rn as in \u2018 rag, rang. \u2019\na1 theoretical vowel of \u2018half, path, past,\u2019 etc., [a1] ; used chiefly by those who naturally employ long \u00e6 or a.\na1 \u2018 ALms, cALmer, fAther \u2019 ; usually on the [q]-side of [a] ; in eastern New England perhaps more commonly on the [a] -side.\na \u2018 yAcht, knot, comma, bother \u2019 ; identical in quality with the long vowel in most parts of the United States.\n\u2018 ARms, hEARt, fARther \u2019 ; [a'J or [a1,] ; the corresponding short vowel may be distinguished in weak syllables, as \u2018pARtake.\u2019\nai \u2018 Aisle isle, a ye EYe, high, hEight \u2019 ; [ai], the final element as in ei-, in the South sometimes nearly <z' (as} a\u00e6\u00ef).\nau \u2018 how, out, loud \u2019 ; [\u00e0\u00f9], with second element varying toward [o] ; in the South the first element is reduced \u00e6 or even e.\n&1 \u2018hAll, dAughter, sought, tAught, wALk.\u2019\n& \u2018hAlt, wAter, thought, cAught, wAsh, wAtch, squAnder,\u2019 identical in quality with the long sound ; not distinguished by some speakers, who use a or \u00e01 instead.\noi \u2018 boy, hoist \u2019 ; first portion [\u00ab], [/] or [<?], second as in ei. o' \u2018 sOAring, pouring, story\u2019; [;d] or [<?']; by some speakers replaced by \u00e01.\n03 \u2018 woRn wARn, soars, pours, foRward, foRm, foRce \u2019 ; [osj or [opJ ; by some not distinguished from long &. ou \u2018 sew so, soul, cOAt \u2019 ; [<3\u00f9], second portion as in au. \u00e0 used by a few speakers in \u2018whole, wholly, stone, coat, only,\u2019 etc.; [0] ; obsolescent.\nu the unstressed sound corresponding to stressed uu, as in \u2018 annual, gradual \u2019 ; one form of unstressed \u201c long o,\u201d as in \u2018 following \u2019 ; chiefly prevocalic ; [u] or [o].\nu \u2018 good, foot, book, bush, cushion, full \u2019 ; slightly reduced [u]. \u00f9 one form of unstressed \u201clong o\u201d before a consonant or pause, as in \u2018 follows, disobey \u2019 ; sometimes distinguishable, though never distinctive, in trisyllabic \u2018usually, gradually, actually\u2019; variable between [?i] and [o].\n1 In eastern New England this vowel seems to be rather uncommon, short \u00e6 or long a being used instead, as in southern England.","page":115},{"file":"p0116.txt","language":"en","ocr_en":"116\nE. H. Tuttle\nu1 \u2018mooring, enduring\u2019;1 [\u00ab'] .\nup \u2018 mooRS, enduRes \u2019 ;1 [upj.\nUU \u2018mood, boot, knEw\u2019;1 [wu], in some pronunciations nearly [u'].\n? \u2018AugustA, fungus, son sun\u2019; in most parts of the United States the stressed vowel is hardly to be distinguished from the unstressed one, [>]; often used byrestressing for \u201cshort o\u201d in \u2018what, was, because, from,\u2019 etc.; often inserted where not written, as \u2018chas()m, rhyth()m, fi()ie,\u2019 etc.\npi a rare form of unstressed \u201c long i \u201d ; pi] .\npu one form of unstressed \u201c long o,\u201d as in \u2018 follows, following.\u2019\n3 \u2018 altARs altERS, mannERS manORS \u2019 ; pj , might be transcribed as syllabic r, from which it is very slightly different.\nv\\ \u2018 fuRze fiRS, sERf surP; a point-modified vowel with dorsal articulation rather variable, from [a] nearly to [e], \\o], p], p] ; in New York and vicinity often pi, with same final element as in ei, at, oi. Long dorsal v may occur in \u2018stirring, purring,\u2019 but the short vowel p seems to be much more common in such cases.\nMonosyllabic vowel triphthongs, as aig, aug, etc., seem to sometimes be used ; they are not easy to distinguish from the dissyllables ai-g, ati-g, etc. Theoretically \u2018 flowers \u2019 has two syllables and \u2018 flours \u2019 one ; in ordinary pronunciation they are identical.\nMany persons do not use the point-modified vowels at all, but employ the dorsal ones instead, so that \u2018arms\u2019 = \u2018alms,\u2019 \u2018leader\u2019 = \u2018 Leda, etc., as in southern England. A serious fault of the alphabet used in Le Ma\u00eetre Phon\u00e9tique, and of that of Dialect Notes, is the failure to provide any suitable signs for these r-modified sounds.\nIn a similar way could be constructed textual notations for French and for German. In the former the chief difficulty would be the a of \u2018 patte \u2019 pat, \u2018 part \u2019 pair, as this italic form of a is made for only a few styles of type ; it might be replaced by roman a or A. Such notations as an, arh etc.,2 should not be used for the nasal vowels, since they encourage the foreigner\u2019s prejudice against learning or using the proper pronunciation ; the signs a s o 3 could be used for these sounds. The French sound-system, as reckoned by Passy,3 would then be written b pmvfw dtn\n'Stressed \u201clong u\u2019\u2019 following any lingual consonant other than k org, in the same word, has no j in most parts of the United States.\n2Soames, Introduction to Phonetics, 2d edition, London 1899.\nSweet, Primer of Phonetics, Oxford 1890.\n3 Passy, as before, 133.","page":116},{"file":"p0117.txt","language":"en","ocr_en":"Phonetic notation.\nTI7\ns s J J l 7- fi i i[ g k (\u00c4) i e z a a oous0yveao3, each of the vowels except v being capable of lengthening without difference of quality. German may be transcribed with the signs b p m v f d t n z .r (j) flrj\u00e7gk-t) (y) y i h i] i e1 s e1 a ai au o oy (or n, or si) o' u u' 3 0* y y< 3 ; if the quality of short i, u, \u00fc needs to be distinguished, t U y could be used for them.\nThe use of the stress-marks can be much simplified in texts where the nomic word-division is retained.\nWeak-stressed syllables do not generally need to be marked as such, except in the case of monosyllables.\nA large number of polysyllables in Germanic languages have the first syllable stressed ; in the Romance languages, except French, the penult is commonly accented ; the \u2022 can therefore be omitted before these syllables.","page":117}],"identifier":"lit28753","issued":"1902","language":"en","pages":"96-117","startpages":"96","title":"Phonetic notation","type":"Journal Article","volume":"10"},"revision":0,"updated":"2022-01-31T13:18:12.029353+00:00"}