Page:John Dowland - First Book of Airs.djvu/10

 remembered that time-signatures at the beginning of a piece of Elizabethan music never indicated the number of beats in a bar but merely the relative value of the longer notes to the shorter. Rhythms of $4⁄4$, $3⁄2$, $3⁄4$, and $6⁄8$, not to mention more complex rhythms, were introduced promiscuously by these composers, and they were indicated by the true accentuation of the words rather than by any system of time-signatures or of barring. It is, in consequence, necessary that a singer should acquire perfect facility in dealing with the sudden and frequent changes of rhythm in order to interpret Elizabethan music satisfactorily.

In a few instances the songs have been transposed with the object of bringing them all into line for voices of medium compass. Those who desire to do so can of course transpose them to any key that suits their own individual requirements.

The accompaniment in this alternative version has been arranged by the Editor exclusively from the composer's own material, but the comparatively narrow compass.of the lute has been suitably extended, especially in its upper ranges, and the idiom of the lute has been adapted to that of the Pianoforte as far as possible without sacrificing the character of the music. A short introductory phrase to each song has also been added by the Editor where none existed in the original Edition, and these phrases are designed upon material to be found in the songs themselves. Modern key-signatures have been added, and also expression-marks, none whatever being found in the original Editions.

In this last particular the Editor has no desire to hamper the individual discretion of those who may sing these songs; and, indeed, his suggestions mainly apply only to the first-verse of the songs, for it would give a confused appearance to print different expressions-marks for each verse. It should be remembered that the first verse alone was, as a general rule, printed with the music in the original editions, the remaining verses appearing in metrical form at the foot of the page. Although, of course, the metre of the subsequent verses is identical with that of the first, it is not infrequently necessary to re-arrange the position of the words of the later verse in order to fit them to some of the smaller phrases of the music. A good example of this will be found in No. 20 of Dowland's first book.

The authorship of the words was seldom recorded in the original Editions. Whenever it has been possible to identify the poet, his name has been given in the present Edition.

No complete Edition of the Lutenist songs has ever been issued since the original editions appeared, now more than three hundred years ago. It is hoped that the present Edition may be instrumental in making these superb English songs known, as they deserve to be, throughout the world. No library, either public or private, which devotes any space to the English poets, should lack a copy of these song-books; the songs themselves should find a place on a large proportion of Concert-programmes.

I desire to express my cordial thanks to my friend the Rev. A. Ramsbotham for valuable advice in connection with the Pianoforte accompaniments, and for help in reading the proofs. EDMUND H. FELLOWES

2em