take the liberty of recommending to him to be less pastoral in his vignettes and preface; and rather to address the public, than the shepherds of the Roman Arcadia. We always thought this fiction puerile and undignified, even for Italians. His exertions in this branch of literature, which has been too much neglected in Great Britain, may become very valuable to the public. A perfect selection would be extremely desireable; but one incompletely made is hurtful. The uninformed reader relies on the selector, and, thinking that he sees the flower of Italian poetry, is contented to look no further. We are far from intending to discourage the editor, whose undertaking we think very meritorious; but wish to stimulate him to render his work less imperfect. We hope it may reach a second edition, and then perhaps our observations (however slight or superficial) may be of some assistance to him: and if we shall have pointed out any poetry, concerning which his judgment may coincide with ours, we trust he will insert it. We should prefer seeing the lighter lyric poetry published as an additional volume, and not a separate work; for no distinctive line can be accurately drawn. The canzone, the ode Toscana, and the canzonetta, have been blended together by infinite variations: the metre of the latter has been adapted to the highest subjects; and the rersi sdruccioli, which more properly belong to it, occur in the canzoni of those who followed the Greek school.

An Inquiry into the Principles of Harmony in Language, and of the Mechanism of Verse, Modern and Ancient. By WILLIAM MITFORD, Esquire. 8vo. 1804.

1805.

THIS interesting treatise, which was first published anonymously, has been greatly enlarged, and reprinted with the name of its learned author. Mr. Mitford has stated in his dedication, perhaps somewhat too boastfully, that more book-learning, acquaintance with living tongues, leisure, and industry, may probably not soon meet together in another person. We give him full

credit for his learning; but the subject, as he has handled it, did not necessarily demand any extraordinary extent of reading. It did, however, require nice powers of discrimination, and those he appears to possess ; it also required a very wide acquaintance with living languages, in which, we suspect him to be somewhat deficient; for it does not appear that he is acquainted with any but French, Italian, and the sister tongue of Spain. The work, however, is unquestionably valuable; though there are many points upon which we are compelled to differ from the author; and we are not without hopes, that upon reconsideration, he may assent to some of our statements; and that we may suggest to him some new and curious subjects of investigation, which

we have not sufficient learning, leisure, or industry, to explore thoroughly. The most important points of difference between us, relate to the number and sounds of vowels, and the mechanism of Latin verse, which, according to Mr. Mitford, is regulated solely by certain dispositions of quantities; whereas, in opposition both to that statement and to the established creed of modern writers, we are humbly of opinion that, in various instances, it was regulated by accent also. This point is of some importance; because, if our opinion be not ill founded, it will appear, that many modern verses, which have been esteemed good and classical metre, are fundamentally faulty. We have not leisure to pursue these observations to their utmost extent: we shall deem it sufficient, if we can establish, that some accentual rules were observed; and we shall leave for the future investigation of those who have more industry and learning, to inquire into the truth of other rules, which we suspect to have existed; and to trace that which we suspect also, but of which the proofs are probably lost with the vulgar poetry of the dark ages, namely, the gradual progress by which accent superseded quantity in European metre.

The introductory chapter of Mr. Mitford's book, contains several definitions, which will be useful and even necessary to those who are inclined to consider the principles of harmony in language, and the mechanism of verse. We shall quote the most important, without commenting on them.

Of the articulated sounds of which language is composed, the smallest integral or portion which can exist as whole by itself, is what we call a syllable. Of the

unarticulated sounds, which constitute music, the smallest integral is what we call a note. Among the properties of syllables, then, we must seek the principles of the harmony of language, as those of the harmony of music among its notes. A musical note exists only in its specific tone. By tone, we mean sound as it is characterized, not by any varieties of articulation, but by those indicated by the terms loud, soft, high, low, sharp, flat, shrill, deep. Among our grammarians, instead of the word tone, derived from the Greek, the word accent, derived from the Latin, is in common use. Syllable exists by articulation; with which tone or accent, is necessarily co-existent, but not a specific tone. Any among the numerous varieties of tone, may co-exist with any among the innumerable varieties of syllable. Varying tones or accents, in pleasing succession, equally in music and in language, constitute melody. The time employed in pronouncing a syllable, compared with the time employed in pronouncing other syllables, at the same rate of delivery, is called in grammatical phrase, the quantity of the syllable. The same thing, in speaking of a musical note, is called simply the time. The term measure, derived from the Latin, or metre, as we often call it, from the Greek, means measure of time, or quantity, as it is indicated by one note, or combination of notes in music, compared with other notes or combination of notes, and by one syllable or by a combination of syllables in speech, compared with other syllables or combinations of syllables,' p. 4.

• Harmony in language, is the result of a happy combination of measure and melody.' p. 9.

Having explained himself thus far with precision, Mr. Mitford proceeds, in the second section, to survey the sounds of the English language, and in a later part of the work, those of the Greek and Latin. In this, we find material cause to differ from him: we think he has confounded some sounds, and overlooked others; and are of opinion that he has hazarded some assertions, which are quite unfounded. For instance, he has contended, that the Greeks could not have used the sound of the French u, because it is to be found only in the Parisian dialect; and in three different passages he asserts that no other people use this sound, or have any character to express it (see p. 366). The fact however is, that it is very frequent in the Danish and the Icelandic; in both which, it is expressed by the letter y, which is not applied to any other purpose by the Danes. The same sound is always given to oo, in part of Devon and Somersetshire, and in Norfolk, as well as to o in to and do, which is spoken shorter: with us in Scotland, it is still more familiar. It is likewise given in Gaelic to ao; and we understand that it occurs also in Welsh. Mr. Mitford is equally mistaken in saying, that the nasal n forms one of the characteristical features of the French language, which distinguish it at a wide distance from all others.' This nasal sound, which we express by ng, is perhaps much more frequent amongst the Portuguese; and from their manner of expressing it by a mark over the vowel, without writing the n, the French, perhaps, adopted the absurd notion, that the nasal n is a vowel. Le son nasal étant une voyelle.' Levisac. It is possible, however, that this sound, which is not used in Spanish, was introduced into Portugal by