minded mere words, mere assertions; still less, mere abuse. If I may call attention to a few of the more important passages where the reader of this new edition will find new information, I should point out the following. In the first volume, p. 242 seq., the statements on the relation of Pehlevi to Zend have been re-written in accordance with the new results that have been obtained by a more careful study of Pehlevi texts and inscriptions. In the second volume, pp. 15-23, the question of the origin of the participle in -ing has been more fully treated. On p. 33 will be found an interesting letter on ceremonial pronouns in Chinese, by M. Stanislas Julien. The analysis and classification of vowels and consonants, on pp. 123-168, has been carefully revised in accordance with the latest researches on this interesting subject. On pp. 139141 will be found my reply to Professor Czermak's important essay, Über den Spiritus asper und lenis. His independent testimony (p. 143, note 79), that the emissions of breath (the sibilants, etc.) are to be subdivided, exactly like the checks of breath (the muta), into soft and hard, will show that my own division of these sounds was not unfounded, while his experiment, described on pp. 159 and 160, explains, and to a certain extent justifies, the names of hard and soft by the side of surd and sonant.1 In the Fifth Lecture, 1 As a specimen of the over-confident and unsuspecting criticism described above, I quote some extracts from the North American, in many respects, I believe, one of the best American reviews: 'But specially Professor Max Müller's account of the spiritus asper and the spiritus lenis, and his explanation of the difference between such sounds as z, v, b, on the one hand, aud s, ƒ, p, on the other, is to be rejected. We have On Grimm's Law, I have endeavoured to place my explanation of the causes which underlie that law in a clearer light, and I have answered some important arguments that had been advanced against my theory, particularly that founded on the historical changes in the names of places, such as Strataburgum and Strazpuruc. My derivation of Earl, Graf, and King, which had been challenged, have been defended on pp. 280, 281, and 284, and the question whether a right to be astonished that he revives for these two classes of letters the old names "soft" and "hard," which have happily for some time been going out of use, and fully adopts the distinction which they imply, although this distinction has been so many times exploded, and the difference of the two classes shown to consist in the intonation or nonintonation of the breath during their utterance. It is in vain that he appeals to the Hindu grammarians in his support: they are unanimous against him—not one of them fails to see and define correctly the difference between "sonant" and "surd" letters.' I do not blame a writer in the North American Review for not knowing that I myself have run full tilt against the terminology of 'hard' and 'soft' consonants as unscientific (unwissenschaftlich), and that I was one of the first to publish and translate in 1856 the more scientific classification into 'surd' and 'sonant,' consonants as contained in the Rigveda-prâtisâkhya. But the Reviewer might surely have read the Lecture which he reviewed, where on page 130 (now page 144), I said: The distinction which, with regard to the first breathing or spiritus, is commonly called asper and lenis, is the same which, in other letters, is known by the names of hard and soft, surd and sonant, tenuis and media.' The same Review says: "The definition of the wh in when, as a simple whispered counterpart of w in wen instead of a w with a prefixed aspiration, is, we think, clearly false.' Now on a question concerning the correct pronunciation of English, it might seem impertinence in me were I not at once to bow to the authority of the North American Review. Still the writer might have suspected that on such a point a foreigner would not write at random, and if he had consulted the highest authorities on phonetics in England, and, I believe, in America too, he would have found that they agree with my own description of the two sounds of w and wh. See Lectures, vol. ii. p. 148, note 55. the reported initial digamma in the name of Helena renders a comparison between Helena and Saramâ impossible has been fully discussed on pp. 516 seq. Lastly, I wish to call attention to a letter with which I have been honoured by Mr. Gladstone (vol. ii. pp. 440-444), and in which his opinions on the component elements of Greek Mythology, which I had somewhat misapprehended, will be found stated with great precision. M. M. OXFORD: April 1871. CONTENTS. PAGE Name of the science of language The physical sciences three stages, empirical, classificatory, theoretical - The empirical stage Practical character of the science of language - Language the barrier between man and beast - The classificatory stage- Objections — Language the work of man — Has language a 28-87 Language studied in India and Greece — Empirical stage Grammar Study of foreign languages - Interpreters - Travels of Greek philosophers — Barbarians learning Greek - Berosus, Menander, Manetho Scholars at Alexandria -The article in Greek-Cases-Dionysius Thrax-Teachers of Greek at Rome— Greek influences at Rome - Crates of Pergamus - Carneades Grammatical study of Sanskrit The facts of grammar Grammar in Chinese- Grammar in Finnish-The origin of grammatical forms - Historical study of languages - Lineal rela- tionship Collateral relationship — Classification of languages Barbarians - Influence of Christianity-Work done by mis- sionaries-Semitic languages - Hebrew, the primitive language— Imperfect classification The language of India - Vedic Sanskrit Asoka's inscriptions-Grammatical and ungrammatical Prâkrits Grammatical Prâkrits - Ungrammatical Prâkrits, Asoka's inscriptions The Gâthâ dialect - Ancient Apabhram- sas - Two classes of Asoka's inscriptions - Introduction of writing - Difficulty of writing a spoken language - Renaissance of Sanskrit |