XV COMPARATIVE PHILOLOGY THE field which is covered by Comparative Philology is so wide, that it may be well to begin by breaking it up into smaller portions, which can be dealt with in detail. These sections are: (1) the characteristics of the individual IndoGermanic, Indo-European or Aryan languages (for as yet there is no agreement on the terminology of this study in this matter), in respect of sounds, formation, and syntax; (2) the development of the meaning of words and the principles on which change of meaning depends; (3) the number and relationship of these languages and the original home and civilisation of the people who spoke the earlier language from which the individual languages are descended. In the century which has elapsed since Comparative Philology came into existence, these different sections have not been all tilled with equal industry at all times. The earlier investigators were naturally most occupied with determining what languages belonged to the family, and with ascertaining the principles of their formation. It was only when this work had been to some extent completed, that the more minute enquiries which are necessary for a thorough thorough investigation of the system of sounds were possible. These enquiries were prosecuted with the greatest ardour in the seventies and eighties of last century, and the results were codified in Brugmann's Grundriss der vergleichenden Grammatik, the first volume of which, dealing with the sounds, appeared in 1886, and in a second and much enlarged edition in 1897. An abridged form, dealing only with the more important languages, was published in 1902-1904, and has been translated recently into French, under the the title of Abrégé de Grammaire comparée (Klincksieck; 20fr.). A slighter, but more readable, book was published some years ago by A. Meillet, who has now succeeded M. Bréal (Introduction à l'étude comparative des langues indo-européennes, Hachette, 1903; 10fr.). A Dutch work, by Dr. Jos. Schrijnen, was published at Leiden in 1905 (Sijthoff) on Greek, Latin, and the Teutonic languages (see C. R., May, 1906, p. 229 ff., where some account is given also of an important book on the Greek gutturals by Dr. J. Mansion). The main facts and theories regarding Greek and Latin are given in P. Giles's Short Manual of Comparative Philology for Classical Students (Macmillan, 2nd ed., 1901; 14s.). Most of these books postulate a considerable amount of knowledge in their readers. But for Latin, an admirable little compendium for beginners has been published by Max Niedermann, entitled Précis de phonétique historique du Latin (Paris, Klincksieck, 1906; 2fr. 50). The first sentences of M. Meillet's introduction to this little work deserve to be quoted entire. "La grammaire comparée," says Prof. Meillet, "qui a renouvelé toutes les conceptions linguistiques n'a pour ainsi dire pas pénétré dans l'enseignement scolaire des langues anciennes, et n'a presque pas modifié les doctrines des grammaires qui sont entre les mains des élèves; nulle part, certes, le divorce n'est plus absolu entre des doctrines scientifiques désormais incontestées et l'enseignement usuel. Et c'est, sans doute, ce qui contribue le plus à rendre stérile et ennuyeux l'enseignement grammatical qu'on donne d'ordinaire; séparé de toutes les idées modernes, resté scolastique dans son fond et dans sa forme, il est devenu un élément mort; il se borne à des faits qu'aucune idée ne relie, ou que relient seulement des idées aujourd'hui périmées et contraires à tout ce qu'apprennent par ailleurs les élèves. Aussi l'opinion devient-elle chaque jour plus hostile à la grammaire, et les heures sont-elles chaque année plus étroitement mesurées à un enseignement, dont chacun sent obscurément la stérilité. Il semble pourtant qu'une science dont l'objet essentiel est d'étudier le principal moyen d'expression de la pensée humaine mérite l'attention des élèves et soit capable, si elle est convenablement enseignée, de provoquer un vif intérêt." The bearing of these remarks seems not to be confined to education south of the English Channel. During the last fifteen years investigations into the history of sounds have been pushed back into the pro-ethnic language mainly by Streitberg and Hirt, who have constructed complex and far-extending theories as to the forms of the words in this language before the peoples separated. These theories can be tested only by the agreement with them of characteristics in the separate languages, for investigations into the pro-ethnic tongue transcend Comparative Philology. The most remarkable treatise on this subject is Hirt's Ablaut (Strassburg, Trübner, 1900; 6m.). Stated briefly, the theory postulates for the explanation of many words disyllabic roots. As a root is only that part of a word which defies analysis, there is no reason why roots should not be disyllabic; it is hardly possible to explain the relationship between forms like Oávaτos and Ovnτós without postulating some such form, varying according to the position of the accent. The tacit assumption of the older philologists, that roots were monosyllabic, was one of the theories taken over from Sanskrit grammar, when Sanskrit was supposed to show in all respects an earlier condition of things than the sister tongues. It is, however, now clear that in its vowel system Greek represents more faithfully an early state of the language than do any of the other languages. It seems unlikely, however, that much further advance is possible in this department till the nature and action of accent in early times have been thoroughly investigated. All scholars admit that there are two kinds of accent, pitch and stress, the former being a rise or fall of the musical tone in the voice, the latter appearing in the greater or less degree of emphasis with which a particular syllable is pronounced. Hirt, who has taken the whole of Comparative Philology for his province, produced a work on this subject also in 1895 (Der indo-germanische Accent; Trübner; 9m.). The book came too early; general conclusions were not really possible while so much had still to be done in clearing up the accentual conditions of individual languages. Dr. J. Vendryes, following up his elaborate treatise of 1902 (Recherches sur l'histoire et les effets de l'intensité initiale en Latin; Paris, Klincksieck), published in 1904 a brief account of Greek accentuation, Traité d'accentuation grecque (Paris, Klincksieck; 3fr.). A short paper on the origin of Greek verb-accent by Professor Bezzenberger appears in vol. xxx. of his Beiträge (see also GRAMMAR). Professor Skeat has treated, along with other subjects, the problems of gradation for the classical languages and English in a very clear and concise manner in his little book entitled A Primer of Classical and English Philology (Clarendon Press, 1905; 28.). A brightly written Introduction to Comparative Philology for Classical Students, intended for scholarship candidates at public schools, has been prepared by Mr. J. M. Edmonds (Cambridge University Press, 1906; 48.). For the formation of the classical languages by far the most important work that has appeared is the second and much enlarged edition of the first part of vol. ii. of Brugmann's Grundriss (Strassburg, Trübner, 1906; 17m. 50), which deals with the formation of the noun. A contribution to verb-formation is E. Fraenkel's Griechische Denominativa in ihrer geschichtlichen Entwicklung (Teubner; 8m.). Meantime, more attention has been devoted to Syntax, though for the last forty years Professor Delbrück has almost singlehanded carried on the investigation. In this section investigation is much harder than in either sounds or forms. In these, research is possible by means of dictionary and grammar only; in Syntax, nothing can be done till the investigator has got a thorough grip of the idiom of the language, And Comparative Syntax is not possible till the investigator has got a thorough grip of the idiom of several languages. No wonder therefore that Delbrück, in the three large volumes (1893-1900) with which he completed Brugmann's Grundriss, limited himself in the main to Sanskrit, Greek, Latin, Teutonic, and Slavonic. Hermann and Lobeck were no doubt thinking mainly of syntax when they poured contempt upon Comparative Philology in its early days as "a refuge for sciolists." Yet when the syntax of a single language is studied by itself, it cannot rise beyond rule of thumb, for it has no historical basis, and, as an intellectual exercise, if learnt from the grammar alone, is on the same level as the learning of the multiplication table by rote. Hermann's own treatment of languageforcing it into the Procrustes bed of Kant's categories—came to an end when, in 1847, Madvig, as Professor Goodwin unkindly says, restored "Syntax to the realms of commonsense." A very interesting Syntax of Greek, from the comparative point of view, has been given by Brugmann in the third edition of his Griechische Grammatik (Munich, 1899; 11m.). The syntax of Latin is much more difficult to treat comparatively, and from this point of view Schmalz's treatment in the Lateinische Grammatik, parallel to Brugmann's Greek Grammar already mentioned, is a failure. In this respect Blase's treatment in the Historische Grammatik der lateinische Sprache is much superior, but as yet he has not proceeded beyond the simple sentence. Lindsay and Sommer (see GRAMMAR) eschew Syntax altogether. In Delbrück's volumes also Greek and Sanskrit have the lion's share, and Latin comes off with somewhat scurvy treatment. The fact is, we know still too little of the early history of Latin. When we come to know Latin, it has already suffered many changes which have made it markedly different from the other languages; and till the various schools established in Rome dig up for us more early remains, much in Latin must remain uncertain. kindred dialects, Oscan and Umbrian, are in like case; but we may hope, for the former at any rate, that if excavation Its |