308 PROBLEMS OF THE ENGLISH I think, be safely dismissed, and for the moment we may treat C and x as a single witness. Now since C (or x) sometimes follows Y and sometimes W, it is clear that it cannot be direftly derived from either. Two possibilities present themselves : either C is derived from an earlier form of W, or else C and W are independently derived from an earlier form of Y. In the latter case C will point to the original reading where Y and W differ, in the former the agreement of C and W is only of equal authority with Y. If, therefore, it can be shown that C and W agree in unoriginal readings, it will follow that C does not go back to an earlier Y, but only to an earlier W. It is difficult if not impossible to demonstrate "this with certainty, but I think the following instances of agreement between W and C (and x) will be found significant : (i) Y : Nowe herken yone barne with his brandyng, He wenes he kens more than we knawes . . . 89-90 W : Hark, yonder barn with his bowrdyng, He wenys he kens more then he knawys . . . 65-6 C : This besse bweye of his tong All secrettis surely he thynkith he knois. 902-3 X : Harkes this childe in his bourding, He wenes he kennes more then he knowes ..241-2 In the first line c brandyng ' is an error of Y. In beginning of the line. In C, however, the line is practically re- written, so that its evidential value is seriously impaired, and the fact that it begins with the same word as Y may very well be an accident due to the general construction of the sentence, which in a manner suggests it. This case is typical of the group. For the other instances see Y 134, 147, 197, 198, W no, 123, 189, 190, C 954, 9 6 7> I0 5> 1006, X 268 > 2 79> 261, 262.
