Introduction to the science of language, Volume 1 (of 2)

by A. H. Sayce

FOOTNOTES

[1] “Lectures on the Origin and Growth of Religion,” p. 65.

[2] See Jolly (translation of Whitney), “Die Sprachwissenschaft,” p. 640.

[3] 660-690.

[4] See the quotations in Steinthal: “Geschichte der Sprachwissenschaft bei den Griechen und Römern” (1863), pp. 181 sq.

[5] Proclus, p. 9.

[6] Anomalist as he was, moreover, Krates was not blind to the defects of language as it was commonly used, and it would appear that the ninth book of his Satires was devoted to the reform of orthography.

[7] He introduced the practice of writing r between two vowels instead of s, and banished the use of z “because its pronunciation resembles the sound that passes through the teeth of a dying man” (Pomp. Dig. i. 2, 2, 36, Mart. Cap. i. 3, § 261, ed. Kopp). Panætius had read his poetical “Maxims,” or “Sententiæ,” which Cicero calls “Pythagorean” (Tusc. iv. 2, 4).

[8] Max Müller: “Lectures on the Science of Language” (eighth edition, 1875), p. 111.

[9] It is given in Bekker’s “Anecdota,” pp. 629-643. Its authenticity is satisfactorily defended by Lersch, “Sprachphilosophie der Alten,” ii. pp. 64-103.

[10] Γραμματική ἔστιν ἐμπειρία τῶν παρὰ ποιηταῖς τε καὶ συγγραφεῦσιν ὡς ἐπὶ τὸ πολὺ λεγομένων. Μέρη δὲ αὐτῆς εἰσὶν ἕξ· πρῶτον ἀνάγνωσις ἐντριβὴς κατὰ προσῳδίαν, δεύτερον ἐξήγησις κατὰ τοὺς ἐνυπάρχοντας ποιητικούς τρόπους, τρίτον γλωσσῶν τε καὶ ἱστοριῶν πρόχειρος ἀπόδοσις, τέταρτον ἐτυμολογίας εὕρεσις, πέμπτον ἀναλογίας ἐκλογισμός, ἕκτον κρίσις ποιημάτων, ὅ δὴ καλλιστόν ἐστι πάντων τῶν ἐν τῇ τέχνῃ.

[11] A good idea of the character of his etymologizing may be gathered from the following quotation:—“Vestis nomen factum est per syncopen ex composito perestis, et mutato r in s (ut sæpe factum est), pesestis sive pesestas, a verbo per-edo, per-es, per-est; quo significatur, quidquid peredit et plane consumit et perdit materiam quamque, unde facta est, ut lues illa epidemica pestis appellationem obtinuerat.” Elsewhere he asserts that sin is derived from σίνειν, while so is merely ὥς reversed. But Junius is quite equalled by Scaliger, Voss, Wachter, and other philologists of the same school. Thus Scaliger says (“De Caus.” c. 35):—“Ordinis nomen Græcum est. Dicebant militibus tribuni—‘Hactenus tibi licet; hic consistes: eò progrediere, huc revertere; ὅρον δῶ,’ inde ordo;” and again (“De Caus.” c. 28) that quatuor is κατερα, i.e. και ετερα (the aspirate being dropped as among the Æolians), because when the Latins had counted “unum, alterum, tria; pro quarto dixere et alterum.” Scaliger, again, agrees with Voss in deriving “opacus ex Ope, hoc est, terrâ; nam umbræ et frigoris captandi causa in subterraneos se specus abdebant,” and pomum from πῶμα, because most fruits quench the thirst. Voss identifies the Latin rus with the Greek ἄρουρα, “præciso a,” and declares: “ab ἔπω, qua notat operor, venit Latinum opus.” Perhaps the various etymologies proposed for the word cause by Perottus will give the best illustration of what once passed for “a true account of the origin of words.” It is either (1) from chaos, as being the first cause of things, or (2) from καῦσις, because heat “kindles and inflames us” to action, or (3) “a cavendo,” because a cause forewarns (“cavet”) us that something should or should not be done, or, finally (4), “a casu, quia causa accidit.” To these Voss adds a fresh possibility, that causa comes from “caiso,” that is, “quærere seu petere.” Perottus, again, derives “locusta ex locus et ustus, quod tactu multa urat, morsu vero omnia erodat.” We cannot but be struck by the ingenuity of these old scholars. Wachter, however, offers us equally absurd etymologies in the field of modern High German. Thus he brings kämpfen (from campus) from kam, “the fist,” cat from ge-wachten, the French guêter (!), and agrees with Clauberg in making neigen the source of nacht.

[12] Where there is so much to choose from it is difficult to select; but perhaps the richest morsels of the book are the reference of the Latin suffix -or in words like sonorous, as well as the final syllable of Hebrew words like tabor, to the “Celtic” mhor, “great,” and the derivation of the Egyptian Rameses from the “Celtic” raromeireas, “gasconading.” The author, however, cannot claim to be facile princeps of the year in the matter of bad etymologizing. A certain Mr. Boult has printed two papers, read before the Literary and Philosophical Society of Liverpool, in which, among other novel statements, he informs us that city is derived from the “Celtic” sigh-tigh or “peace-house;” count from co-meas, “united assessment,” alderman from all-dor-meann, “chief of the great door,” and custom from cur-do-meas, “rent of assessment.” It is needless to observe that “Celtic” with both writers means the decayed forms of an Irish dictionary.

[13] “Whitney’s Sprachwissenschaft,” p. 660.

[14] “De Vulg. El.” I. xii. p. 46, cxvi. (ed. Fraticelli, 1833).

[15] Ed. Jebb (1733), pp. 44-56. We may notice that Bacon in this part of his book (p. 44) draws attention to the existence of the French (“Gallicorum”), Picard, Norman, and Burgundian dialects in France, which differ from one another in many idioms and uses of words.

[16] Edited by Graesse (1850).

[17] See Thurot: “Extraits de divers Manuscrits latins pour servir à l’histoire des Doctrines grammaticales au Moyen Age” (1869).

[18] “Histoire des Langues sémitiques,” p. 272.

[19] Thus Voss derives νεός from the Hebrew particle nâ, “now.”

[20] See second edition of first collection of “Fragmente zur deutschen Literatur” (1768).

[21] See Pott: “W. von Humboldt’s Verschiedenheit des menschlichen Sprachbaues” (1876), I. p. cciv.

[22] On this book of Bernhardi’s was founded Reinbeck’s “Handbuch der Sprachwissenschaft, mit besonderer Hinsicht auf die Deutsche Sprache” (1815)—intended for school use.

[23] “Lettere,” p. 415 sq. (Florence, 1855.)

[24] Max Müller: “Lectures,” p. 155.

[25] “Works, with Life,” by Lord Teignmouth (1807), iii. p. 34.

[26] The second edition of his work on Language, in six vols., “with large additions and corrections,” was published in 1774.

[27] Quoted by Max Müller: “Lectures,” i. p. 150.

[28] “Ueber die Verwandtschaft der malayisch-polynesischen Sprachen mit den indisch-europäischen” (1841).

[29] “Die kaukasischen Glieder des indo-europäischen Sprachstammes,” 1847.

[30] See the Edition of Pott, published in two volumes in 1876.

[31] Beginning in 1859.

[32] A new edition has just been brought out (1878), with a valuable appendix, by Scheler.

[33] “Observations sur la langue et la littérature provençales,” p. 14.

[34] “Charakteristik der hauptsächlichsten Typen des Sprachbaues,” pp. 20-75. Steinthal’s criticism is criticized in turn by Pott in his edition of Humboldt’s essay, “Ueber die Verschiedenheiten des menschlichen Sprachbaues” (1876).

[35] Edited by Steinthal (1856).

[36] Translated into French by F. de Wegmann (1859).

[37] “Die Darwinische Theorie und die Sprachwissenschaft,” 1863 (translated into French in the first part of the “Collection Philologique,” 1868, and into English by Bikkers, 1869); “Ueber die Bedeutung der Sprache für die Naturgeschichte des Menschen” (1865), and “Die Deutsche Sprache” (second edition, 1869).

[38] Translated into English by Keane for the “Library of Contemporary Science” (1877).

[39] “Essai sur la Langue Poul” in the “Revue de Linguistique” (Jan. et Avr. 1875).

[40] “Ursprung der Sprache,” with a preface by Häckel. Translated by T. Davidson (1869).

[41] “Ursprung und Entwickelung der menschlichen Sprache und Vernunft” (1869).

[42] See also Benfey’s article, “Einige Worte über den Ursprung der Sprache,” in the “Nachrichten von der k. Gesellschaft der Wissensch. zu Göttingen,” Jan. 30, 1878. Benfey here points out that just as we share a capacity of walking with the lower animals, so also do we share with them a capacity for communicating with one another by the help of a language of some sort. And he remarks pertinently that it was not harder for the first men to understand the meaning of what was said to them than it is for domestic animals nowadays to learn the meaning of the words and phrases we use in speaking to them or giving them orders.

[43] “Der Ursprung der Sprache” (1877).

[44] See his “Vergleichende Grammatik der indo-germanischen Sprachen” (1873), I. Appendix, pp. 56-98.

[45] “Agglutination oder Adaptation” (1873).

[46] “Mémoires de la Société de Linguistique de Paris,” ii. 5.

[47] “Anthropologie der Naturvölker,” i. p. 272.

[48] “Grundriss der Sprachwissenschaft,” I. i. p. 49 (1876).

[49] “City of the Saints,” p. 151.

[50] “Trans. Eth. Soc.” (1869), i. p. 283.

[51] “Expedition to the Rocky Mountains,” iii. p. 52.

[52] Sayce: “Principles of Comparative Philology” (second edition), p. 26.

[53] “De l’Esprit,” i. p. 2. Aristotle attacks Anaxagoras for holding διὰ τὸ χεῖρας ἔχειν, φρονιμώτατον εἶναι τῶν ζώων τὸν ἄνθρωπον (“De Part. Animal.,” iv. 10).

[54] “Festus,” p. 28.

[55] II. 405, ed. Burm.

[56] Max Müller: “Lectures,” ii. p. 179 (8th edition).

[57] Whitney: “Life and Growth of Language,” p. 120.

[58] See I. Disraeli: “Curiosities of Literature,” vol. iii. pp. 79, 80.

[59] Milligan: “Vocabulary of the Dialects of some of the Aboriginal Tribes of Tasmania,” p. 34.

[60] See Mr. Hartshorne’s Paper read before the British Association (1875).

[61] “Tropical South Africa,” p. 132.

[62] “Journal of the Anthropological Institute,” vi. 2 (Oct, 1876), p. 119.

[63] A “pig” is called poro-poro, and the act of “eating” nam-nam. We must remember, however, that just as a nurse will speak to a child in nursery-language, so a savage on being asked the name of an object may have recourse to onomatopœia, instead of giving the real native word.

[64] Plato termed it ἀπείκασμα (“Krat.,” 402 D, 420 C.).

[65] On the whole subject of the onomatopœic origin of words, see (but with caution) Wedgwood’s introduction to his “Dictionary of English Etymology” (first edition, 1859), and Farrar: “Chapters on Language,” and “Origin of Language” (1860), ch. iv. Compare Buschmann in the “Abhandlungen der k. Akademie der Wissenschaft zu Berlin” (1852).

[66] See page 82.

[67] Latin et, Greek ἔτι, Zend aiti, Sanskrit ati, are referred by Weber (“Indische Studien,” ii. p. 406) to the root at, “to go.”

[68] As in Chaucer: “Knight’s Tale,” 2488:—“But by the cause that they sholde ryse.”

[69] See “A Comparative Vocabulary of the Barma, Malayu, and T’hai Languages,” published at Serampore in 1810. A Siamese compound like lúk-mai, “fruit,” literally, “son of wood,” is an exact equivalent of the Hebrew “son of Belial” for “sinner,” or “master of hair” for “hairy.”

[70] Steinthal: “Charakteristik der hauptsächlichsten Typen des Sprachbaues,” pp. 165, 171, 173.

[71] See Du Ponceau: “Langues d’Amérique,” pp. 120, 200, 234, 236, 237.

[72] Pickering: “Indian Languages,” p. 26.

[73] Part iii. (1853), pp. 420-445.

[74] See Latham in the “Proceedings of the Philological Society,” vol. vi. (1852-3), p. 85; and in the “Transactions of the Philological Society” for 1856, pp. 40, 41.

[75] “Charakteristik,” &c., p. 113.

[76] This order of the pronouns was a later innovation in the language, and seems due to Semitic influence. In the older period of the speech the form was rû-n-mu or rû-mu-n.

[77] Suetonius: “De illustr. Gramm.,” 22. “M. Pomponius Marcellus ... quum ex oratione Tiberium reprehendisset, affirmante Ateio Capitone, ‘et esse illud Latinum, et, si non esset, futurum certe jam inde;’ ‘Mentitur,’ inquit, ‘Capito. Tu enim, Cæsar, civitatem dare potes hominibus, verbo non potes.’”

[78] Sir John Stoddart: “Universal Grammar, or the Pure Science of Language,” 2nd edition, 1852.

[79] By French writers.

[80] As by Schleicher.

[81] As by Ascoli.

[82] In A. J. Ellis: “Early English Pronunciation,” pp. 1293-1307, 1352-1357.

[83] Humboldt’s “Travels,” Engl. Tr., i. p. 322.

[84] Compare A. Erman: “Journey round the Earth through North Asia,” iii. § 1, p. 191.

[85] P. 159.

[86] v. p. 774.

[87] “Travels,” Engl. Tr., i. p. 310.

[88] So in Japan the learned class has introduced the Chinese characters under the name of Koyé or Won, and with them the Chinese pronunciation and order of words. In Koyé (that is to say in Chinese) the particles come first, then the verb, and, lastly, the case. The reverse is the case in Japanese, or when the characters are read as Yomi, that is, as ideographs standing for Japanese words. Thus, the Koyé “sed non videbo hodie illum” would have to be read in Yomi, “illum videbo hodie non sed.”

[89] Sayce: “Journal R. A. S.” x. 2 (1878).

[90] “Arische Studien,” i. 2, pp. 45-61.

[91] De Charencey in the “Revue de Linguistique” (1873), i. 1, p. 57.

[92] “Ein Problem der Homerischen Textkritik” (1876), p. 95. Pott (“Wilhelm v. Humboldt ueber die Verschiedenheit des menschlichen Sprachbaues,” i. p. 15) suggests that the change of the Latin demonstrative into the article of the Romance languages was due to Teutonic influences.

[93] Sayce: “Lectures upon the Assyrian Grammar and Syllabary,” pp. 61, 62, “Journal R. A. S.,” x. 2, pp. 251, 252 (1878).

[94] E.g. θῠω in Od. 260, Theok. iv. 21; Aristophanes and the Attic poets preserve the long vowel.

[95] Hackländer: “Ein Winter in Spanien” (Werke xxii.), ii. 78.

[96] N. H., iii. 20, 24.

[97] Schott: “Abh. der Berlin. Akad.” (1865), p. 440.

[98] Deffner: “Neogræca” in Curtius’ “Studien,” iv. 2, p. 307.

[99] Burton: “Etruscan Bologna,” p. 246.

[100] Karl Verner in Kuhn’s “Zeitschrift,” xxiii. (New Series, iii. 2.)

[101] For Greek synonyms, see Trench: “Synonyms of the New Testament” (1865).

[102] Apud Priscian. vii. 345.

[103] “Mémoires de la Société de Linguistique de Paris,” ii. 5.

[104] V. a 24, sve mestru karu fratru = “si major pars fratrum.”

[105] On the other hand, the Asiatic members of the family have certainly lost the distinction that existed in the parent speech between the vowels represented in the European members by ă, ĕ, and ŏ, so that the differentiation of the root ar may have been earlier than the period of Aryan separation. In the Finnic group roots are similarly differentiated by a modification of the vowel, kah-isen, koh-isen, and kuh-isen, for instance, being “to hit” or “stamp,” käh-isen and köh-isen, “to roar,” keh-isen and kih-isen, “to boil.”

[106] This happy term was the invention of Professor Max Müller.

[107] Mill: “System of Logic,” ii. p. 240.

[108] Grimm’s “Law: a Study” (1876).

[109] Böhtlingk: “Ueber die Sprache der Jakuten,” p. 168.

[110] “Introduction to the Study of the Chinese Characters” (1876).

[111] “De Div.” ii. 40, 84.

[112] Gallatin: “Synopsis of the Indian Tribes of North America,” in the “Archæologia Americana,” ii.

[113] “Die Melanesischen Sprachen” (1873), p. 4. According to Meyer (“Sitzungsberichte d. Oesterr. Akademie,” 1874, p. 301), “Riedel has made us acquainted with twenty-three dialects in some parts only of North Celebes, and the number of dialects in the whole island can only be estimated at hundreds.... But in New Guinea this dialectical variety is very much greater and more thorough-going, since there the very foundations of a state have not yet begun to be laid.”

[114] “Travels” (English translation), i. 298.

[115] Gibbon: “Decline and Fall of the Roman Empire” (ed. Milman and Smith), vii. p. 387. Dr. Deffner, however, asserts that there never was a tithe of that number of dialects in the country.

[116] Prince Lucien Bonaparte reckons that there are 660,000 Basques in Spain and 140,000 in France.

[117] Sayce: “Principles of Comparative Philology” (2nd edition), p. 87.

[118] The progress of cuneiform research has shown that a similar woman’s dialect existed among the Accadians, and “a woman’s language” is also said to exist in Bengal.

[119] “United States Exploring Expedition,” vii. 290.

[120] “Natal Colonist,” Sept. 3rd, 1875. Mr. Theal says (“The Cape Monthly Magazine,” June, 1877, p. 349): “A woman, who sang the song of ‘Tangalimlibo’ for me, used the word angoca, instead of amanzi, for ‘water,’ because this last contained the syllable nzi, which she would not on any account pronounce. She had, therefore, manufactured another word, the meaning of which had to be judged by the context, as, standing alone, it is meaningless.” This is a good instance of the way in which a savage dialect may grow up.

[121] Waitz, iii. 477.

[122] Mi took the place of o in old Japanese, hence the title of the Mi-kado, or “high Gate” (Grande Porte).

[123] See Hoffmann: “Japanese Grammar,” pp. 72 sq. and § 111-120.

[124] Endlicher: “Chinesische Grammatik,” pp. 258 sq.

[125] Pott: Humboldt’s “Verschiedenheit des menschlichen Sprachbaues,” i. p. 395.

[126] Max Müller: “Lectures,” i. p. 56.

[127] “On the best Method of Studying the American Languages,” p. 11.

[128] P. 152 (4th edition).

[129] Andree: “Wendische Wanderstudien” (1874).

[130] Appendix to Mielcke’s edition of Ruhig’s “Wörterbuch.”

[131] “Pasigraphical Dictionary and Grammar” (1871). Galliani’s “Dictionnaire télégraphique, économique, et secret” contains 15,576 groups of only three letters, each of which expresses a word or a whole sentence.

[132] “Lectures on the Science of Religion,” pp. 161, 154.

[133] Bréal: “La Langue Indo-Européenne,” in his “Mélanges de Philologie Comparée” (1878).

[134] “History of the European Languages,” pp. 31, 32. The following are the significations assigned to these nine rudiments of speech:—

“I. To strike or move with swift, equable, penetrating or sharp effect was Ag! Ag!

If the motion was less sudden, but of the same species, WAG.

If made with force and a great effort, HWAG.

These are varieties of one word, originally used to mark the motion of fire, water, wind, darts.

“II. To strike with a quick, vigorous, impelling force, BAG or BWAG, of which FAG and PAG are softer varieties.

“III. To strike with a harsh, violent, strong blow, DWAG, of which THWAG and TWAG are varieties.

“IV. To move or strike with a quick, tottering, unequal impulse, GWAG or CWAG.

“V. To strike with a pliant slap, LAG and HLAG.

“VI. To press by strong force or impulse so as to condense, bruise, or compel, MAG.

“VII. To strike with a crushing, destroying power, NAG and HNAG.

“VIII. To strike with a strong, rude, sharp, penetrating power, RAG or HRAG.

“IX. To move with a weighty strong impulse, SWAG.”

[135] Cfer. Wullschlaegel: “Kurzgefasste Neger-Englische Grammatik” (1854), and “Deutsch-Neger-Englisches Wörterbuch, nebst einem Anhang Neger-Englische Sprüchwörter enthaltend” (1856).

[136] Helmholtz: “Die Lehre von den Tonempfindungen,” 3rd edition (1870).

[137] Speaking accurately a musical note is a tone only in so far as its quality or timbre is taken account of.

[138] The fact can easily be tested by Captain Galton’s whistle. According to some authorities, however, it requires less than eight and more than 24,000 vibrations per second, to produce no effect upon the auditory nerve.

[139] “Lectures,” ii. 8th edition (1875), p. 111.

[140] Helmholtz: “Die Lehre,” &c., p. 143.

[141] Serpents have no voice in the proper sense of the word, as they have no vocal chords; the hissing sounds they produce being caused by a mere forcible breathing through a soft glottis.

[142] That is, “U-shaped bone.”

[143] Of course, if an opening is made in the trachea, voice is impossible unless it is closed, and division or injury of the laryngeal nerves will equally destroy voice by paralyzing the muscles of the vocal chords.

[144] In men the average length is about eleven lines.

[145] Faidherbe: “Essai sur la Langue Poul,” in the “Revue de Linguistique,” January and April, 1875.

[146] Bleek: “A brief Account of Bushman Folklore, and other Texts,” p. 6. “A most curious feature in Bushman folklore is formed by the speeches of various animals, recited in modes of pronouncing Bushman, said to be peculiar to the animals in whose mouths they are placed. It is a remarkable attempt to imitate the shape or position of the mouth of the kind of animal to be represented. Among the Bushman sounds which are hereby affected, and often entirely commuted, are principally the clicks. These are either converted into other consonants, as into labials (in the language of the Tortoise), or into palatals and compound dentals and sibilants (as in the language of the Ichneumon), or into clicks otherwise unknown in Bushman (as far as our present experience goes), as in the language of the Jackal, who is introduced as making use of a strange labial click, which bears to the ordinary labial click a relation in sound similar to that which the palatal click bears to the cerebral click. Again, the Moon—and it seems also the Hare and the Anteater—substitute a most unpronounceable click in place of all others, excepting the lip click. Another animal, the Blue Crane, differs in its speech from the ordinary Bushman, mainly by the insertion of a tt at the end of the first syllable of almost every word.”

[147] “Lectures,” ii. p. 141 (8th edition).

[148] Max Müller: “A Sanskrit Grammar for Beginners,” 2nd edition, p. 23, note. See Mr. Ellis’s examination of the “Rules of the Indian Phonologists,” as given by Whitney (“Atharva-Vêda Prâtiçâkhya”), in “Early English Pronunciation,” pt. 4, pp. 1336-1338.

[149] See Sievers: “Grundzüge d. Lautphysiologie” (1876), p. 19. This explanation of the causes of this difference between the two kinds of voice (true and falsetto) is due to the observations of Garcia. Various theories had previously been put forward to account for it. J. Müller thought that in producing chest notes, the whole breadth of the vocal chords vibrated, only their thin inner margins in producing falsetto notes. Mayo and Magendie held that the falsetto notes are produced by the vibrations of only one-half the length of the vocal chords, when the glottis is partially closed; G. Weber that they are due to the vibration of the chords in segments, separated by nodal points, so that harmonics of the fundamental note are formed. Pétrequin and Diday maintained that they are produced by the vibration of the air itself in the glottis, without any movement on the part of the vocal chords, while Wheatstone thought that they are formed by the division of the air in the trachea into harmonic lengths, the tone produced by the vocal chords being thus reciprocated, since, besides vibrating by reciprocation with a sonorous body, the vibrations of which are isochronous with its own, a column of air may also vibrate by reciprocation in its several lengths, the number of its vibrations being in this case a multiple of those of the sonorous body.

[150] “Lectures,” ii. p. 128 (8th edition).

[151] Bindseil: “Abhandlungen zur allgemeinen vergleichenden Sprachlehre,” p. 212 (1838), quoted by Max Müller: “Lectures,” ii. p. 136.

[152] For the difference between these two sounds, see Sweet: “Handbook of Phonetics,” § 14, p. 9.

[153] “Grundzüge d. Lautphysiologie,” p. 46 (1876).

[154] “Lectures,” ii. p. 133.

[155] “Grundzüge d. Physiologie u. Systematik d. Sprachlaute,” p. 16 (1856).

[156] “Wiener Sitzungsberichte,” lii. 2, pp. 623 sq.

[157] “Grundzüge der Lautphysiologie,” p. 88 (1876).

[158] Sievers, loc. cit. p. 50.

[159] Haldeman in the “Proceedings of the American Oriental Society,” 1874, p. xlv. According to Professor Rhys the Welsh ll has resulted from the meeting of two l’s, each sounded independently up to the ninth century. Like the Welsh pronunciation of dd, the pronunciation of ll may have been originally borrowed from English.

[160] Mr. Sweet has proved that the pronunciation of these two Anglo-Saxon letters was originally the same, but it would be convenient to use them to distinguish the different sounds of the modern English th.

[161] Hübschmann: “Z. d. D. M. G.” xxx. pp. 53, 57.

[162] That is, the guttural sonant in question.

[163] As in nach.

[164] This is the guttural sonant in question.

[165] “Die Kerenzer Mundart,” p. 5.

[166] Bleek: “Comparative Grammar of South African Languages,” pp. 12-15. The lateral click is sounded by the Kafirs, as by Europeans, by placing the tongue against the side-teeth, and then withdrawing it, whereas the Nama Hottentots produce it as far back as possible, covering the whole of the palate with the tongue. The palatal click of the Hottentots, which is very difficult to imitate, seems to be found in one or two Kafir words. The clicks, it must be noted, only occur in the Kafir dialects adjoining the Hottentots, and the Kafir clicks “are only found in the place of other consonants, and are used like consonants at the beginning of syllables, whilst in Hottentot a guttural explosive consonant (k, kh, or g), the faucal spirant h and the nasal n, can be immediately preceded by a click, and form together with it the initial element of the syllables.”

[167] See above, [p. 243]. The compound dental click is produced, according to Wuras, by pressing the air through the upper and lower teeth, which stand slightly apart. Dr. Bleek says that “the Bushman word for ‘to sleep’ seems to be ǀphkoĩnyé, beginning with a combination of dental click, aspirated labial and guttural tenuis in which three letters are sounded together.”

[168] Ellis: “Early Engl. Pron.” p. 1349.

[169] Professor Mahaffy notices that “old women among us express pity by a regular palatal click.”

[170] Sievers holds that our th (as in the) is sometimes “reduced” to a glide (“Grundzüge,” p. 91).

[171] “The Japanese ṛ,” says Mr. Sweet, “seems to be formed by first bringing the tip of the tongue against the gums without any emission of breath, and then passing on to an untrilled r, allowing voiced breath to pass at the moment of removing the tongue.” The sound has been mistaken for r, l, or even d, and as it is substituted for all foreign l’s and r’s, the Japanese tendency to change l into r has been contrasted with the Chinese tendency to change r into l. It is possible that the Old Egyptian possessed the same curious sound.

[172] The compound tone in Swedish, according to Mr. Sweet, “only occurs in words of more than one syllable,” and “consists of a falling tone on the first (the accented syllable), followed by a high tone on the next. The high tone seems to be reached by a jump rather than by a glide. The compound is, therefore, a compound rise distributed over two syllables.” The other Swedish accent, the simple tone, is the negative of the compound one, and answers to the “glottal catch” or stöd tone of Danish.

[173] See Wackernagel in Kuhn’s “Zeitschrift,” 23 (1877).

[174] See Sayce in “Journal of Royal Asiatic Society,” x. 2 (1878), pp. 251, 252.

[175] Sweet: “Handbook,” p. 35.

[176] “Kerenzer Mundart,” pp. 142 sq.

[177] Sweet: “Handbook,” pp. 59, 60.

[178] In many instances, however, guṇa and vṛiddhi seem to be due to the presence of the vowel a in the following syllable, which has been anticipated, as in the case of the German umlaut or the Greek epenthesis (as in λόγοις for the locative λόγο-σι, and then, by false analogy, λόγοισι).

[179] The existence of these velar gutturals was first pointed out by Ascoli, and since by Fick and Havet.

[180] Diez: “Grammatik d. romanisch. Spr.” (2nd edition), i. 248, 254.

[181] “Comparative Grammar of the Anglo-Saxon Language” (1870), p. 28.

[182] See Robin and Verdeil: “Chimie anatomique,” ii. p. 44.

[183] Appendix to Greenwell’s “British Barrows” (1877), p. 645.

[184] Daa: “On the Languages of the Northern Tribes of the Old and New Continents,” in the “Transactions of the Philological Society” (1856), p. 256.

[185] “Sir George Grey’s Library,” i. p. 167; and A. Kaufmann: “Das Gebiet des weissen Flusses und dessen Bewohner” (1861), quoted by Max Müller: “Lectures,” 8th edition, ii. p. 178.

[186] See, however, Ascoli’s ingenious attempt to remove the phonological difficulties in his “Studj Critici,” ii. (1877), pp. 386-396.

[187] For a recent English examination of the subject see Douse: “Grimm’s Law: a Study” (1876), and Rhŷs’s review in the “Academy,” Jan. 12, 1878; also Murray and Nicol in the “Academy,” Feb. 23, March 2 and 16, 1878.

[188] The table of consonants is taken from Rhŷs: “Lectures on Welsh Philology,” p. 17.

[189] Before υ.

[190] In the middle of a word, e.g. ruber (ἐρυθρός).

[191] P did not exist in the early Keltic languages; hence proper names like Menapia must be treated as non-Aryan, or at all events as non-Keltic.

[192] Pratt: “Grammar and Dictionary of the Samoan Language” (edited by Whitmee), p. 1, 1878.

[193] To Brugman belongs the credit of first demonstrating the existence of these three distinct vowel-sounds in the Parent-Aryan (Kuhn’s “Zeitschrift,” 1877). Brugman has been criticized by Collitz in Bezzenberger’s “Beiträge zur Kunde der indogermanischen Sprachen,” ii. 4 (1878), who maintains that the three primitive sounds were really ĕ, ŏ, ă, and not the indeterminate a¹, a², a³. On the other hand, an able article by De Saussure in the “Mémoires de la Société de Linguistique de Paris,” iii. 5 (1878), accepts Brugman’s nomenclature, while criticizing and modifying some of his conclusions. His a, Brugman’s a¹, became e in West Aryan, and is never weakened into i or u in Sanskrit. His a² (also Brugman’s) is the West Aryan o², and in Sanskrit is lengthened in an open syllable (e.g. jajâna = γέγονε). In Latin ŏ often became ĕ, as in genu (Greek γόνυ, Sansk. jânu). Different from this o² is another o¹, standing in the same relation to a that o² does in Latin to e, and answering to a Sanskrit i or î. Besides this short o¹ is also a long ō, which appears also as ā, and corresponds with Sansk. â. De Saussure further points out that velar k in Sanskrit is palatalized (becomes ch) when followed by a (= ĕ) and a² (= ŏ).

[194] “Lectures,” ii. pp. 184 sq. (8th edition).

[195] “The Polynesian,” October, 1862.

[196] Agnel: “Observations sur la Prononciation et le Langage rustique des environs de Paris,” pp. 11, 28; Terrien Poncel: “Du Langage,” p. 49.

[197] “Sir George Grey’s Library,” i. p. 135. Professor Mahaffy informs us of a child of three years of age who invariably substitutes n for l, and cannot be made to feel the difference between them.

[198] Sievers: “Lautphysiologie,” p. 127.

[199] Kuhn’s “Zeitschrift,” xxiii. pp. 97-130 (1877).

[200] The termination of the participles of German weak verbs, such as the Goth. tami-da (“domitus”), answers to the Vedic dami-tás (like the Greek κλυτός) where the accent is oxytone. Verner sums up his conclusions as follows: (1) The original accentuation was preserved in Teutonic even after the introduction of those changes of sound characteristic of the Teutonic branch of the Aryan family; (2) the accent, however, was no longer purely tonic, it had become also an accent of stress; (3) the exceptional representation of an Aryan k, t, and p at the beginning of a syllable by a Teutonic g, d, b is due to the original accentuation of the words in which it occurs; and (4) this is also the case with z or r in the place of s.

[201] Sanskrit jîv, Latin vivo (vixi), English quick, presupposing an original reduplicated qwi-gwi.

[202] “Lautphysiologie,” p. 137.

[203] “Jenaer Literaturzeitung” (1874), p. 767; quoted by Sievers.

[204] “Zur Geschichte des indogermanischen Vocalismus,” ii.

[205] According to the current theory the sonant or vocalic n, l, and r develop out of a consonantal n, l, r. Fick (Bezzenberger’s “Beiträge,” iv., 1878) has shown that Greek aorists like ἔδρακε or παθών owe their α-vowel to this cause. The accentuation of the last syllable occasioned the loss of the vowel of the present-stem (which Fick proves to represent the oldest form of the verb), and out of the resulting ἔδρκε or πθών grew ἔδρακε and παθών. The corresponding Swarabhakti vowel in Teutonic is u (cfer. ἑκατόν, i.e. ἑκαντόν for ἑκντόν, and Gothic hund (our hundred), ἄρκτος, and Gothic vulfs, -ματός and Gothic participial -munds). According to the Indian grammarians the Sanskrit ṛi = ¼ a + ½ r + ¼ a (Greek ᾰρᾰ).

[206] “Beiträge zur Kenntniss der Sicilianischen Mundart,” p. 154.

[207] See Joh. Schmidt: “Ueber Metathesis von Nasalen,” in Kuhn’s “Zeitschrift,” xxiii. pp. 266-302 (1877).

[208] “Grundzüge” (2nd edition), p. 650.

[209] Brugman in Kuhn’s “Zeitschrift,” xxiv. p. 81 (1878).

[210] Recent researches seem to have shown that the parent-Semitic possessed two dentals, which may be written ṭ and ḍ, and are represented in Arabic by th and dh (ذ), and in Assyrian, Hebrew, and Ethiopic by s (sh) and z. Consequently the table of sound-shiftings will be—

Parent-Semitic.	Arabic.	Assyrian.	Hebrew.	Ethiopic.	Aramaic.
ṭ	th	s	sh	s	th
ḍ	dh	z	z	z	d
t	t	t	t	t	t (th)
d	d	d	d	d	d
sh	s	s	sh	s	s
s	sh	s	s	sh	sh
´s	s	´s	´s	s	sh

[211] Logan: “Indian Archipelago,” iii. p. 665.

[212] “Magyarische Grammatik” (1858).

[213] De Ujfálvy, in the “Revue de Philologie et d’Ethnographie,” i. 1, pp. 20-50.

[214] Reprinted by I. Pitman in 1850.

[215] The “Grammatica Linguæ Anglicanæ,” to which is prefixed a treatise “De Sonorum omnium loquelarium formatione: Tractatus Grammatico-physicus.”

[216] Swift’s “Works,” ix. pp. 137-139 (edit. Sir W. Scott).

[217] The monthly journal of these reformers, published at Bremen, is entitled: “Reform. Zeitšrift des algemeinen fereins zur einfürung einer fereinfahten deutšen rehtšreibung.”

[218] “Etymologische Forschungen,” v. p. xxii.

[219] “Language and the Study of Language,” p. 106.

[220] See his “Etymologische Forschungen,” v. i., Introd. (2nd edition).

[221] Donner: “Z. D. M. G,” xxvii. 4 (1873).

[222] “Lectures,” ii. pp. 268 sq. (8th edition).

[223] That is, hard-explosive.

[224] These characters represent the palæotype symbols employed by Mr. Ellis.

[225] Soft-explosive.

[226] Nasal-explosive.

[227] Hard-continuous.

[228] Soft-continuous.

[229] Hard-trill.

[230] Soft-trill.

[231] Nasal-continuous.

[232] Soft-liquid.

[233] Hard-liquid.

[234] For Mr. Ellis’s own Palæotype Alphabet, see “Early English Pronunciation,” part i. pp. 3-12, where also a list of signs denoting clicks, pitch, whisper, glide, &c., is given.

[235] In “A Handbook of Phonetics,” pp. xv-xvii.

[236]

Old Chinese.	Mandarin.	Old Middle Dialect.	Hakka Dialect.	South Fukien.	Canton.
g	c’h, k’ (h)	g (dj)	k’	k’, k	k’
d	t’(l)	d	t’	t’, t	t’
b	p’ (f)	b (v)	p’	p’, p (h)	p’

This table applies only to words which have the fifth tone (Edkins: “Introduction to the Study of the Chinese Characters,” p. 185).

[237] Sayce: “Principles of Comparative Philology,” Preface to 2nd edition, p. ix.

[238] See “Contemporary Review,” April, 1876.

[239] Fr. Müller: “Grundriss der Sprachwissenschaft,” i. 2, p. 2.

[240] Sayce: “Principles of Comparative Philology” (2nd edition), p. 229.

[241] “Transactions of the Philological Society” (1856), pp. 40, 41.

[242] “La Philosophie de la Science du Langage étudiée dans la formation des mots” (1875), p. 83.

[243] Tylor: “Primitive Culture,” i. pp. 199-201.

[244] See Sayce: “Principles of Comparative Philology,” 2nd edition, p. 253.

[245] Wilson: “Grammar,” p. 32.

[246] Schiefner: “Tibetische Studien” (1851), p. 30.

[247] Platzmann: “Brasilianische Grammatik,” chap. xv.

[248] In Curtius’ “Studien,” vii. pp. 185 sq.

[249] Buschmann: “Abhandlungen d. Berlin. Akad.,” 1869, i. p. 122.

[250] The whole subject of reduplication has been exhaustively treated by Professor Pott, to whose work reference should be made: “Doppelung als eines der wichtigsten Bildungsmittel der Sprache.”

[251] “On the Origin of Civilization,” pp. 403-405.

[252] “Ethnography and Philology of the Hidatsa Indians,” p. 77.

[253] See Pott: “Humboldt’s Verschiedenheit des menschlichen Sprachbaues” (1876), i. pp. 305, 306.

[254] Aristoph. “Eccl.” 1169.

[255] See Sayce: “The Tenses of the Assyrian Verb,” in “J. R. A. S.,” Jan. 1877.

[256] Not dhâ, “to place,” like the perfect in the Teutonic languages, since Old Irish has b (e.g. caru-b = “amabo”), and in Keltic b cannot come from dh.

[257] Earle: “Philology of the English Tongue” (2nd edition), p. 305.

[258] The only difficulty here is that the base of the feminine in Sanskrit is tiśar.

[259] Brachet, however, holds that justice and justesse are collateral forms, both from the Latin -itia.

[260] “Archæologia Americana,” ii. pp. 25, 166, 169.

[261] Bleek: “Comparative Grammar of the South African Languages,” pp. 97, 98.

[262] Matthews: “Ethnography and Philology of the Hidatsa Indians” (1877), p. 95.

[263] Buschmann: “Abhand. d. Berliner Akademie” (1869), i. p. 103.

[264] The Sanskrit equivalent of humus, however, has had to submit to the prevailing analogy, and in the form of bhûmi assume what has become the feminine suffix.

[265] “Ethnography and Philology of the Hidatsa Indians,” p. 96.

[266] “Abhandlungen d. Berlin. Akad.” (1869), i. p. 122.

[267] “Eléments de la Grammaire Othomi,” in the “Revue Orientale et Américaine,” p. 21.

[268] Friedrich Müller: “Grundriss der Sprachwissenschaft,” i. 2, p. 27.

[269] “Trans. of the Ethnological Society,” i. p. 304.

[270] Latham in the “Proceedings of the Philological Society” (1852), p. 59.

[271] “Du Rôle de la Dérivation dans la Déclinaison indo-européenne,” in the “Mémoires de la Société de Linguistique de Paris,” ii. 5.

[272] “Grundriss der Sprachwissenschaft,” i. 2, p. 2.

[273] See Sayce: “The Tenses of the Assyrian Verb,” in the “J. R. A. S.,” Jan. 1877.

[274] Max Müller: “Lectures,” i. p. 106.

[275] See, for instance, Wiedemann: “Grammatik der Wotjakischen Sprache” (1851), pp. 268-271.

[276] Böhtlingk: “Jakutische Grammatik,” p. 160.

[277] In Bunsen’s “Philosophy of Universal History,” i. p. 265.

[278] “La Grèce avant les Grecs,” p. 45. According, however, to M. Dozon (“Grammaire albanaise,” 1878), the postfixed Albanian article is really a termination like that of the German adjective, and not a relic of the demonstrative pronoun.

[279] “De la Construction Grammaticale,” in the “Mémoires de la Société de Linguistique de Paris,” iii. 1, 2, 3.

[280] It is possible that the position of the article in the greater number of the Romanic languages may have been influenced by Teutonic usage.

[281] Exod. xxxii. 1, Josh. ix. 12 sq., Is. xxiii. 13.

[282] Jer. xvi. 16, Ps. xxxii. 10, lxxxix. 51.

[283] “Chips,” iv. p. 39.

[284] “Primer of Philology” (1877), p. 112.

[285] “De la Construction Grammaticale,” in the “Mémoires de la Société de Linguistique,” iii. 1, 2, 3.

[286] Prologue, 396.

[287] Orosius, i. 1, 23.

[288] Gaussin: “Du dialecte de Tahiti” (1853).

[289] See Jolly: “Ueber die einfachste Form der Hypotaxis im Indo-germanischen,” and Windisch: “Untersuchungen über den Ursprung des Relativpronomens,” in Curtius’s “Studien,” vi. 1 and ii. 2.

END OF VOL. I.

CHISWICK PRESS:—CHARLES WHITTINGHAM, TOOKS COURT,
CHANCERY LANE.