:: :: University of Virginia Library

5. MS. Folger 1.27, p. 43. Miss Latham (Ralegh, p. 116) says that this manuscript contains both parts of the poem, but it contains only the epigram.
6. MS. Huntington 198, p. 165.

The text of this poem and its ascription present the most difficult problem of the entire canon. Who wrote the poem and what text to print are questions which admit of no certain answer. At best one can advance solutions which appear to account for the facts, but such solutions must be taken as provisional.

The number of manuscript texts of the poem is very large, but must nevertheless represent a mere fragment of the number of copies which were made, for the texts that we have are contaminated, one with another, to such an extraordinary degree that the relationship of the texts does not allow of accurate definition. A collation of three miscellany, and six fragmentary, and seventeen complete manuscript versions (Nos. 16, 18, 19; 22-27; and 1-15) shows that, while the texts have a tendency to break down into groups, a large number of them shift from one group to another on different readings and display no consistent pattern in doing so. The texts which show a more or less constant affinity are Nos. 1-3; Nos. 4-6; Nos. 8-10; and Nos. 11-12. Nos. 1-3 are the only texts which show a constant uniformity in themselves (although No. 3 gives a number of unique readings). Naturally on some readings where the number of variants is small, the other groups or single manuscripts sometimes fall in with them. Nos. 4-6, for example, although they constitute a definite group of readings, are not consistent; No. 6 introduces, like No. 3, several readings peculiar to itself, and on occasion falls in with other groups. Nos. 7-9 are another fairly constant group with which No. 10 falls in as often as it diverges from it.

What I am arguing then is simply this: it is impossible to construct a definitive text for poems derived from such sources as we are considering by any method of collation and comparison of the texts themselves, for there is nothing to provide the principle that determines which reading is a variant and which the original. This is particularly true of MSS. that do not fall into definite groups, but it is also true of MSS. which form definite and apparently derivative groups.[2] For example, there are 16 MSS. and printed texts of another Ayton poem, "Thou sent to mee a heart was Crown'd," which fall into six groups, the "first and true" state represented by the Ayton MSS., and five states of progressive corruption. The corruptions of Group Two are not found in Group One. The corruptions of

195

Group Three include those of Group Two and new ones of its own. The corruptions of Group Four include those of Group Three and new ones of its own. And the Fifth Group derives from the Third without contamination from the Fourth. But it might be just as conveniently argued that Group Three was the "first and true" state and that three lines of corruption emerged from it. It is only by the assumption that the poem is by Ayton and that the version of the Ayton MSS. is the most authoritative text that the progressive line of corruption and the derivation of groups can be shown. There is no doubt about the authorship of this poem, so that the assumptions have a certain necessary force; without them, there would be no possibility of a definitive text.