Nations,” 1710, Leibniz says:—“The study of languages must not be
conducted according to any other principles but those of the exact
sciences. Why begin with the unknown instead of the known? It stands
to reason that we ought to begin with studying the modern languages
which are within our reach, in order to compare them with one
another, to discover their differences and affinities, and then to
proceed to those which have preceded them in former ages, in order
to show their filiation and their origin, and then to ascend step by
step to the most ancient tongues, the analysis of which must lead us
to the only trustworthy conclusions.”
119 Nicolaes Witsen, Burgomaster of Amsterdam, travelled in Russia,
1666-1677; published his travels in 1672, dedicated to Peter the
Great. Second edition, 1705. It contains many collections of words.
120 Catherinens der Grossen Verdienste um die Vergleichende Sprachkunde,
von F. Adelung. Petersburg, 1815. Another letter of his to the
Vice-Chancellor, Baron Schaffiroff, is dated Pirmont, June 22, 1716.
121 Collectanea Etymologica, ii. 255. “Malim sine discrimine Dialectorum
corrogari Germanicas voces. Puto quasdam origines ex superioribus
Dialectis melius apparituras; ut ex Ulfilæ Pontogothicis, Otfridi
Franciscis.”
122 Monde primitif analysé et comparé avec le monde moderne: Paris,
1773.
123 Catalogo, i. 63.
124 “Mas se deben consultar gramaticas para conocer su caracter proprio
por medio de su artificio gramatical.”—_Catalogo_, i. 65. The same
principle was expressed by Lord Monboddo, about 1795, in his Ancient
Metaphysics, vol. iv. p. 326. “My last observation is, that, as the
art of a language is less arbitrary and more determined by rule than
either the sound or sense of words, it is one of the principal
things by which the connection of languages with one another is to
be discovered. And, therefore, when we find that two languages
practise these great arts of language,—derivation, composition, and
flexion,—in the same way, we may conclude, I think, with great
certainty, that the one language is the original of the other, or
that they are both dialects of the same language.”
125 Catalogo, ii. 468.
126 Ibid. i. 49. Witsen, too, in a letter to Leibniz, dated Mai 22,
1698, alludes to the affinity between the Tataric and Mongolic
languages. “On m’a dit que ces deux langues (la langue Moegale et
Tartare) sont différentes à peu près comme l’Allemand l’est du
Flamand, et qu’il est de même des Kalmucs et Moegals.”—_Collectanea
Etymologica_, ii. p. 363.
127 Leibniz held the same opinion (see Hervas, Catalogo, i. 50), though
he considered the Celts in Spain as descendants of the Iberians.
128 Catalogo, i. 30. “Verá que la lengua llamada _malaya_, la qual se
habla en la península de Malaca, es matriz de inumerables dialectos
de naciones isleñas, que desde dicha península se extienden por mas
de doscientos grados de longitud en los mares oriental y pacífico.”
Ibid. ii. 10. “De esta península de Malaca han salido enjambres de
pobladores de las islas del mar Indiano y Pacífico, en las que,
aunque parece haber otra nacion, que es de negros, la _malaya_ es
generalmente la mas dominante y extendida. La lengua malaya se habla
en dicha península, continente del Asia, en las islas Maldivas, en
la de Madagascar (perteneciente al Africa), en las de Sonda, en las
Molucas, en las Filipinas, en las del archipiélago de San Lázaro, y
en muchísimas del mar del Sur desde dicho archipiélago hasta islas,
que por su poca distancia de América se creian pobladas por
americanos. La isla de Madagascar se pone á 60 grados de longitud, y
á los 268 se pone la isla de Pasqua ó de Davis, en la que se habla
otro dialecto malayo; por lo que la extension de los dialectos
malayos es de 208 grados de longitud.”
129 Catalogo, ii. 134.
130 Ibid. ii. 135.
131 The first volume appeared in 1806. He died before the second volume
was published, which was brought out by Vater in 1809. The third and
fourth volumes followed in 1816 and 1817, edited by Vater and the
younger Adelung.
132 Evidence of this is to be found in Strahlenberg’s work on the “North
and East of Europe and Asia,” 1730; with tabula polyglotta, &c.; in
Messerschmidt’s “Travels in Siberia,” from 1729-1739; in
Bachmeister, “Idea et desideria de colligendis linguarum
speciminibus:” Petropoli, 1773; in Güldenstädt’s “Travels in the
Caucasus,” &c.
133 The empress wrote to Nicolai at Berlin to ask him to draw up a
catalogue of grammars and dictionaries. The work was sent to her in
manuscript from Berlin, in 1785.
134 “Glossarium comparativum Linguarum totius Orbis:” Petersburg, 1787.
A second edition, in which the words are arranged alphabetically,
appeared in 1790-91, in 4 vols., edited by Jankiewitsch de Miriewo.
It contains 279 (272) languages, _i.e._ 171 for Asia, 55 for Europe,
30 for Africa, and 23 for America. According to Pott,
“Ungleichheit,” p. 230, it contains 277 languages, 185 for Asia, 22
for Europe, 28 for Africa, 15 for America. This would make 280. It
is a very scarce book.
135 The Singhalese call Pali, Mungata; the Burmese, Magadabâsâ.
136 Works, vol. iii. p. 72.
137 M. M.’s Buddhism and Buddhist Pilgrims, p. 23.
138 Méthode pour déchiffrer et transcrire les noms Sanscrits qui se
rencontrent dans les livres chinois, inventée et démontrée par M.
Stanislas Julien: Paris, 1861, p. 103.
139 “Fan-chou (brahmâkshara), les caractères de l’écriture indienne,
inventée par Fan, c’est-à-dire Fan-lan-mo (brahmâ).”—_Stanislas
Julien, Voyages des Pèlerins Bouddhistes_, vol. ii. p. 505.
140 Sir Henry Elliot’s Historians of India, p. 259.
141 See Professor Flügel, in Zeitschrift der D. M. G., xi., s. 148 and
325.
142 Elliot’s Historians of India, p. 96. Al Birúni knew the Harivanśa,
and fixes the date of the five Siddhântas. The great value of Al
Birúni’s work was first pointed out by M. Reinaud, in his excellent
“Mémoire sur l’Inde,” Paris, 1849.
143 In the Persian work Mujmalu-t-Tawárikh, there are chapters
translated from the Arabic of Abu Saleh ben Shib ben Jawa, who had
himself abridged them, a hundred years before, from a Sanskrit work,
called “Instruction of Kings” (Râjanîti?). The Persian translator
lived about 1150. See Elliot, l. c.
144 Sâlotar is not known as the author of such a work. Śâlotarîya occurs
instead of Śâlâturîya, in Rája Rádhakant; but Śâlâturîya is a name
of Pâņini, and the teacher of Suśruta is said to have been Divodâsa.
An Arabic translation of a Sanskrit work on veterinary medicine by
Châņakya is mentioned by Háji Chalfa, v. p. 59. A translation of the
Charaka from Sanskrit into Persian, and from Persian into Arabic, is
mentioned in the Fihrist, finished 987 A. D.
145 See Vans Kennedy, “Notice respecting the Religion introduced by
Akbar:” Transactions of the Literary Society of Bombay: London,
1820, vol. ii. pp. 242-270.
146 Elliot, Historians of India, p. 249.
147 Müllbauer, Geschichte der Katholischen Missionen Ostindiens, p. 134.
148 Elliot, Historians of India, p. 248.
149 Ibid. pp. 259, 260. The Tarikh-i-Badauni, or Muntakhabu-t-Tawárikh,
written by Mulla Abdu-l-Kádir Maluk, Shah of Badáún, and finished in
1595, is a general history of India from the time of the Ghaznevides
to the 40th year of Akbar. The author is a bigoted Muhammedan and
judges Akbar severely, though he was himself under great obligations
to him. He was employed by Akbar to translate from Arabic and
Sanskrit into Persian: he translated the Râmâyaņa, two out of the
eighteen sections of the Mahâbhârata, and abridged a history of
Cashmir. These translations were made under the superintendence of
Faizi, the brother of the minister Abu-l-Fazl. “Abulfacel, ministro
de Akbar, sevalió del Amarasinha y del Mahabhárata, que traduxo en
persiano el año de 1586.”—_Hervas_, ii. 136.
150 See M. M.’s History of Ancient Sanskrit Literature, p. 327.
151 History of the Settlements of the Europeans in the East and West
Indies, translated from the French of the Abbé Bernal by J.
Justamond: Dublin, 1776, vol. i. p. 34.
152 Müllbauer, p. 67.
153 Ibid. p. 80. These Brahmans, according to Robert de Nobili, were of
a lower class, not initiated in the sacred literature. They were
ignorant, he says, “of the books Smarta, Apostamba, and
Sutra.”—_Müllbauer_, p. 188. Robert himself quotes from the
Âpastamba-Sûtra, in his defence, ibid. p. 192. He also quotes Scanda
Purâna, p. 193; Kadambari, p. 193.
_ 154 The Ezour-Veda_ is not the work of Robert de Nobili. It was
probably written by one of his converts. It is in Sanskrit verse, in
the style of the Pûraņas, and contains a wild mixture of Hindu and
Christian doctrine. The French translation was sent to Voltaire and
printed by him in 1778, “L’Ezour Vedam traduit du Sanscritam par un
Brame.” Voltaire expressed his belief that the original was four
centuries older than Alexander, and that it was the most precious
gift for which the West had been ever indebted to the East. Mr.
Ellis discovered the Sanskrit original at Pondichery. (Asiatic
Researches, vol. xiv.) There is no evidence for ascribing the work
to Robert, and it is not mentioned in the list of his works.
(Bertrand, la Mission du Maduré, Paris, 1847-50, t. iii. p. 116;
Müllbauer, p. 205, _note_.)
155 In 1677 a Mr. Marshall is said to have been a proficient in
Sanskrit. Elliot’s Historians of India, p. 265.
156 See an excellent account of this letter in an article of M. Biot in
the “Journal des Savants,” 1861.
157 Sidharubam seu Grammatica Samscrdamica, cui accedit dissertatio
historico-critica in linguam Samscrdamicam, vulgo Samscret dictam,
in qua hujus linguæ existentia, origo, præstantia, antiquitas,
extensio, maternitas ostenditur, libri aliqui in ea exarati critice
recensentur, et simul aliquæ antiquissimæ gentilium orationes
liturgicæ paucis attinguntur et explicantur autore Paulino a S.
Bartholomæo. Romæ, 1790.
158 The earliest publications were the “Bhagavadgîta,” translated by
Wilkins, 1785; the “Hitopadeśa,” translated by Wilkins, 1787; and
the “Sakuntalâ,” translated by W. Jones, 1789. Original grammars,
without mentioning mere compilations, were published by Colebrooke,
1805; by Carey, 1806; by Wilkins, 1808; by Forster, 1810; by Yates,
1820; by Wilson, 1841. In Germany, Bopp published his grammars in
1827, 1832, 1834; Benfey, in 1852 and 1855.
159 Halhed had published in 1776 the “Code of Gentoo Laws,” a digest of
the most important Sanskrit law-books made by eleven Brahmans, by
the order of Warren Hastings.
160 “On the Origin and Progress of Language,” second edition, Edinburgh,
1774. 6 vols.
161 “I have supposed that language could not be invented without
supernatural assistance, and, accordingly, I have maintained that it
was the invention of the Dæmon kings of Egypt, who, being more than
men, first taught themselves to articulate, and then taught others.
But, even among them, I am persuaded there was a progress in the
art, and that such a language as the Shanskrit was not at once
invented.”—_Monboddo, Antient Metaphysics_, vol. iv. p. 357.
162 Origin and Progress of Language, vol. vi. p. 97.
163 Antient Metaphysics, vol. iv. p. 322.
164 Conjugationssystem: Frankfurt, 1816.
165 New edition in 1856, much improved.
166 Second edition, 1859 and 1861. Pott’s work on the Language of the
Gipsies, 1846; his work on Proper Names, 1856.
167 “Although the Old Friesian documents rank, according to their dates,
with Middle rather than with Old German, the Friesian language
appears there in a much more ancient stage, which very nearly
approaches the Old High-German. The political isolation of the
Friesians, and their noble attachment to their traditional manners
and rights, have imparted to their language also a more conservative
spirit. After the fourteenth century the old inflections of the
Friesian decay most rapidly, whereas in the twelfth and thirteenth
centuries they rival the Anglo-Saxon of the ninth and tenth
centuries.”—_Grimm_, _German Grammar_ (1st ed.), vol. i p. lxviii.
168 The dialects of Swabia (the Allemannish), of Bavaria and Austria, of
Franconia along the Main, and of Saxony, &c.
169 Über das Leben und die Lehre des Ulfila, Hannover, 1840. Über das
Leben des Ulfila von Dr. Bessell, Göttingen, 1860.
170 Bessell, l. c. p. 38.
171 Sozomenus, H. E. vii. 6.
172 Auxentius thus speaks of Ulfilas, (Waitz, p. 19:) “Et [ita
prædic]-ante et per Cristum cum dilectione Deo Patri gratias agente,
hæc et his similia exsequente, quadraginta annis in episcopatu
gloriose florens, apostolica gratia Græcam et Latinam et Goticam
linguam sine intermissione in una et sola eclesia Cristi
predicavit.... Qui et ipsis tribus linguis plures tractatus et
multas interpretationes volentibus ad utilitatem et ad
ædificationem, sibi ad æternam memoriam et mercedem post se
dereliquid. Quem condigne laudare non sufficio et penitus tacere non
audeo; cui plus omnium ego sum debitor, quantum et amplius in me
laboravit, qui me a prima etate mea a parentibus meis discipulum
suscepit et sacras litteras docuit et veritatem manifestavit et per
misericordiam Dei et gratiam Cristi et carnaliter et spiritaliter ut
filium suum in fide educavit.
“Hic Dei providentia et Cristi misericordia propter multorum salutem
in gente Gothorum de lectore triginta annorum episkopus est
ordinatus, ut non solum esset heres Dei et coheres Cristi, sed et in
hoc per gratiam Cristi imitator Cristi et sanctorum ejus, ut
quemadmodum sanctus David triginta annorum rex et profeta est
constitutus, ut regeret et doceret populum Dei et filios Hisdrael,
ita et iste beatus tamquam profeta est manifestatus et sacerdos
Cristi ordinatus, ut regeret et corrigeret et doceret et ædificaret
gentem Gothorum; quod et Deo volente et Cristo aucsiliante per
ministerium ipsius admirabiliter est adinpletum, et sicuti Josef in
Ægypto triginta annorum est manifes[tatus et] quemadmodum Dominus et
Deus noster Jhesus Cristus Filius Dei triginta annorum secundum
carnem constitutus et baptizatus, cœpit evangelium predicare et
animas hominum pascere: ita et iste sanctus, ipsius Cristi
dispositione et ordinatione, et in fame et penuria predicationis
indifferenter agentem ipsam gentem Gothorum secundum evangelicam et
apostolicam et profeticam regulam emendavit et vibere [Deo] docuit,
et Cristianos, vere Cristianos esse, manifestavit et multiplicavit.
“Ubi et ex invidia et operatione inimici thunc ab inreligioso et
sacrilego indice Gothorum tyrannico terrore in varbarico
Cristianorum persecutio est excitata, ut Satanas, qui male facere
cupiebat, nolens faceret bene, ut quos desiderabat prevaricatores
facere et desertores, Cristo opitulante et propugnante, fierent
martyres et confessores, ut persecutor confunderetur, et qui
persecutionem patiebantur, coronarentur, ut hic, qui temtabat
vincere, victus erubesceret, et qui temtabantur, victores gauderent.
Ubi et post multorum servorum et ancillarum Cristi gloriosum
martyrium, imminente vehementer ipsa persecutione, conpletis septem
annis tantummodo in episkopatum, supradictus sanctissimus vir beatus
Ulfila cum grandi populo confessorum de varbarico pulsus, in solo
Romanie a thu[n]c beate memorie Constantio principe honorifice est
susceptus, ut sicuti Deus per Moysem de potentia et violentia
Faraonis et Egyptorum po[pulum s]uum l[iberav]it [et Rubrum] Mare
transire fecit et sibi servire providit, ita et per sepe dictum Deus
confessores sancti Filii sui unigeniti de varbarico liberavit et per
Danubium transire fecit, et in montibus secundum sanctorum
imitationem sibi servire de[crevit] ..... eo populo in solo Romaniæ,
ubi sine illis septem annis, triginta et tribus annis veritatem
predicavit, ut et in hoc quorum sanctorum imitator erat [similis
esset], quod quadraginta annorum spatium et tempus ut multos .....
re et .... a[nn]orum ..... e vita.” .. “Qu[i] c[um] precepto
imperiali, conpletis quadraginta annis, ad Constantinopolitanam
urbem ad disputationem ..... contra p ... ie ... p. t. stas
perrexit, et eundo in .... nn .. ne. p ... ecias sibi ax ..... to
docerent et contestarent[ur] .... abat, et inge . e .... supradictam
[ci]vitatem, recogitato ei im .... de statu concilii, ne arguerentur
miseris miserabiliores, proprio judicio damnati et perpetuo
supplicio plectendi, statim cœpit infirmari; qua in infirmitate
susceptus est ad similitudine Elisei prophete. Considerare modo
oportet meritum viri, qui ad hoc duce Domino obit Constantinopolim,
immo vero Cristianopolim, ut sanctus et immaculatus sacerdos Cristi
a sanctis et consacerdotibus, a dignis dignus digne [per] tantum
multitudinem Cristianorum pro meritis [suis] mire et gloriose
honoraretur.”
“Unde et cum sancto Hulfila ceterisque consortibus ad alium
comitatum Constantinopolim venissent, ibique etiam et imperatores
adissent, adque eis promissum fuisset conci[li]um, ut sanctus
Aux[en]tius exposuit, [a]gnita promiss[io]ne prefati pr[e]positi
heretic[i] omnibus viribu[s] institerunt u[t] lex daretur, qu[æ]
concilium pro[hi]beret, sed nec p[ri]vatim in domo [nec] in publico,
vel i[n] quolibet loco di[s]putatio de fide haberetur, sic[ut]
textus indicat [le]gis, etc.”
173 Theodoret. H. E. V., 30.
174 For instances where Old High-German is more primitive than Gothic,
see Schleicher, Zeitschrift für V. S., b. iv. s. 266. Bugge, ibid.,
b. v. s. 59.
175 See Schleicher, Deutsche Sprache, p. 94.
176 Ibid. s. 60.
177 Weinhold, Altnordisches Leben, p. 27; Gunnlaugssaga, c. 7.
178 See Dasent’s Burnt Njal, Introduction.
179 The name Edda is not found before the fourteenth century. Snorri
Sturluson does not know the word Edda, nor any collection of ancient
poems attributed to Saemund; and though Saemund may have made the
first collection of national poetry, it is doubtful whether the work
which we possess under his name is his.
180 The people whom we call Wallachians, call themselves Romàni, and
their language Romània.
This Romance language is spoken in Wallachia and Moldavia, and in
parts of Hungary, Transylvania, and Bessarabia. On the right bank of
the Danube it occupies some parts of the old Thracia, Macedonia, and
even Thessaly.
It is divided by the Danube into two branches: the Northern or
Daco-romanic, and the Southern or Macedo-romanic. The former is less
mixed, and has received a certain literary culture; the latter has
borrowed a larger number of Albanian and Greek words, and has never
been fixed grammatically.
The modern Wallachian is the daughter of the language spoken in the
Roman province of Dacia.
The original inhabitants of Dacia were called Thracians, and their
language Illyrian. We have hardly any remains of the ancient
Illyrian language to enable us to form an opinion as to its
relationship with Greek or any other family of speech.
219 B. C., the Romans conquered Illyria; 30 B. C., they took Moesia;
and 107 A. D., the Emperor Trajan made Dacia a Roman province. At
that time the Thracian population had been displaced by the advance
of Sarmatian tribes, particularly the Yazyges. Roman colonists
introduced the Latin language; and Dacia was maintained as a colony
up to 272, when the Emperor Aurelian had to cede it to the Goths.
Part of the Roman inhabitants then emigrated and settled south of
the Danube.
In 489 the Slavonic tribes began their advance into Mœsia and
Thracia. They were settled in Mœsia by 678, and eighty years later a
province was founded in Macedonia, under the name of Slavinia.
181 The entire Bible has been published by the Bible Society in
Romanese, for the Grisons in Switzerland; and in Lower Romanese, or
Enghadine, as spoken on the borders of the Tyrol.
182 “Ed il primo, così Dante, che cominciò a dire come poeta volgare, si
mosse, perocchè volle far intendere le sue parole a donna alla quale
era malagevole ad intendere versi Latini.”—_Vita Nuova_.
183 Schleicher, Beiträge, i. 19.
184 Oldest dated MS. of 1056, written for Prince Ostromir. Some older
written with Glagolitic letters. Schleicher, Beiträge, b. i. s. 20.
185 Schleicher, s. 22.
186 Schleicher, Deutsche Sprache, s. 77.
187 1 Kings viii. 21.
188 1 Kings ix. 26.
189 1 Kings x. 11.
_ 190 Gutta_ in Malay means _gum_, _percha_ is the name of the tree
(Isonandra gutta), or of an island from which the tree was first
imported (Pulo-percha).
191 See Lassen, Indische Alterthumskunde, b. i. s. 537.
192 See also Sir Henry Elliot’s Supplementary Glossary, s. v. Aheer.
193 The arguments brought forward by Quatremère in his “Mémoire sur le
Pays d’Ophir” against fixing Ophir on the Indian coast are not
conclusive. The arguments derived from the names of the articles
exported from Ophir were unknown to him. It is necessary to mention
this, because Quatremère’s name carries great weight, and his essay
on Ophir has lately been republished in the Bibliothèque Classique
des Célébrités Contemporaines. 1861.
194 Job xxii. 24.
_ 195 Zend-avesta_ is the name used by Chaqâni and other Muhammedan
writers. The Parsis use the name “_Avesta_ and _Zend_,” taking
_Avesta_ in the sense of text, and _Zend_ as the title of the
Pehlevi commentary. I doubt, however, whether this was the original
meaning of the word _Zend_. _Zend_ was more likely the same word as
the Sanskrit _chhandas_ (scandere) a name given to the Vedic hymns,
and _avesta_, the Sanskrit _avasthâna_, a word which, though it does
not occur in Sanskrit, would mean settled text. _Avasthita_, in
Sanskrit, means laid down, settled. The Zend-avesta now consists of
four books, Yasna, Vispered, Yashts, and Vendidad (Vendidad =
vidaeva dâta; in Pehlevi, Juddivdad). Dr. Haug, in his interesting
lecture on the “Origin of the Parsee Religion,” Bombay, 1861, takes
_Avesta_ in the sense of the most ancient texts, _Zend_ as
commentary, and _Pazend_ as explanatory notes, all equally written
in what we shall continue to call the Zend language.
196 “According to the Kissah-i-Sanján, a tract almost worthless as a
record of the early history of the Parsis, the fire-worshippers took
refuge in Khorassan forty-nine years before the era of Yezdegerd
(632 A. D.), or about 583. Here they stayed 100 years, to 683, then
departed to the city of Hormaz (Ormus, in the Persian Gulf), and
after staying fifteen years, proceeded in 698 to Diu, an island on
the south-west coast of Katiawar. Here they remained nineteen years,
to 717, and then proceeded to Sanján, a town about twenty-four miles
south of Damaun. After 300 years they spread to the neighboring
towns of Guzerat, and established the sacred fire successively at
Barsadah, Nauśari, near Surat, and Bombay.”—_Bombay Quarterly
Review_, 1856, No. viii. p. 67.
197 Alc. i. p. 122, _a_. Ὁ μὲν μαγείαν διδάσκει τὴν Ζωροάστρου τοῦ
Ὠρομάζον; ἔστι δὲ τοῦτο θεῶν θεραπεία.
198 In the inscriptions we find, nom. _Auramazdâ_, gen. _Auramazdâha_,
acc. _Auramazdam_.
199 Gen. _Ahurahe mazdâo_, dat. _mazdâi_, acc. _mazdam_.
200 Haug, Lecture, p. 11; and in Bunsen’s Egypt.
201 Berosus, as preserved in the Armenian translation of Eusebius,
mentions a Median dynasty of Babylon, beginning with a king
Zoroaster, long before Ninus; his date would be 2234 B. C.
Xanthus, the Lydian (470 B. C.), as quoted by Diogenes Laertius,
places Zoroaster, the prophet, 600 before the Trojan war (1800 B.
C.).
Aristotle and Eudoxus, according to Pliny (Hist. Nat. xxx. 1),
placed Zoroaster 6000 before Plato; Hermippus 5000 before the Trojan
war (Diog. Laert. proœm.).
Pliny (Hist. Nat. xxx. 2) places Zoroaster several thousand years
before Moses the Judæan, who founded another kind of Mageia.
202 Printed at the end of these Lectures.
203 See Schleicher, Deutsche Sprache, s. 81.
204 Farrar, Origin of Languages, p. 35.