The fifteenth century, which is so dismal in the works of the more ambitious poets (Lydgate, and Occleve, e.g.), is rich in popular carols which by this time have drawn close to the modern meaning of the name. They are Christmas carols, and the name loses its old general application to any song that went with dancing in a round. In the carols, the art is generally much more simple than in the lyrics which have just been quoted; they belong more truly to the common people, and their authors are less careful. Yet the difference is one of degree. The only difference which is really certain is between one poem and another.

Speaking generally about the carols one may say truly they are unlike the work of the Chaucerian school; the lyrics of the Harleian book in the reign of Edward I are nearer the Chaucerian manner. It is hardly worth while to say more, for the present.

And it is not easy to choose among the carols. Some of them are well known to-day—

When Christ was born of Mary free

In Bethlehem that fair city

Angels sang loud with mirth and glee

In excelsis gloria.

Ballads in the ordinary sense of the term—ballads with a story in them, like Sir Patrick Spens or The Milldams of Binnorie—are not found in any quantity till late in the Middle Ages, and hardly at all before the fifteenth century. But there are some early things of the kind. A rhyme of Judas (thirteenth century) is reckoned among the ballads by the scholar (the late Professor Child) who gave most time to the subject, and whose great collection of the English and Scottish Popular Ballads has brought together everything ascertainable about them.

By some the ballads are held to be degenerate romances; and they appear at a time when the best of romance was over, and when even the worst was dying out. Also, it is quite certain that some ballads are derived from romances. There is a ballad of the young Hynd Horn which comes from the old narrative poem of King Horn or of Horn Childe. There is a ballad version of Sir Orfeo, the ‘Breton lay’ which has been described in another chapter. But there are great difficulties in the way of this theory. In the first place, there are many ballads which have no romance extant to correspond to them. That may not prove much, for many old romances have been lost. But if one is to make allowance for chances of this sort, then many old ballads may have been lost also, and many extant ballads may go back to the thirteenth century or even earlier for their original forms. Again, there are ballads which it is scarcely possible to think of as existing in the shape of a narrative romance. The form of the ballad is lyrical; all ballads are lyrical ballads, and some of them at any rate would lose their meaning utterly if they were paraphrased into a story. What would the story of Sir Patrick Spens be worth if it were told in any other way—with a description of the scenery about Dunfermline, the domestic establishment of the King of Norway, and the manners at his Court? Further, the theory that the ballads are degenerate romances is unfair to those ballads which are known to be descended from romances. The ballad of Hynd Horn may be derived from an older narrative poem, but it is not a corruption of any old narrative; it is a different thing, in a lyrical form which has a value of its own. ‘Corruption’, ‘degeneracy’, does not explain the form of the ballads, any more than the Miracle Plays are explained by calling them corruptions of the Gospel.

The proper form of the ballads is the same as the carole, with narrative substance added. Anything will do for a ring dance, either at a wake in a churchyard, or in a garden like that of the Roman de la Rose, or at Christmas games like those described in Sir Gawayne and the Green Knight. At first, a love-song was the favourite sort, with a refrain of douce amie, and so on. But the method was always the same; there was a leader who sang the successive verses, the fresh lines of the song, while the other dancers came in with the refrain, most often in two parts, one after the first verse, the second after the second—