Ernest De Selincourt (ed.), The Poetical Works of Edmund Spenser: In Three Volumes, Vol. 1: Spenser's Minor Poems
pg 506APPENDIX II. ON THE TEXT OF SPENSER FROM 1617–1884.
I have not collated throughout editions of Spenser later than F (1611), but I have compared with earlier copies long passages chosen from them at haphazard and have consulted them on all doubtful and interesting lines. I am thus able to form some opinion as to their respective characteristics and their relation to one another.
F 2 appeared in 1617. In pagination it is identical with its predecessor, and is in the main a reprint of it, using for the last time the original woodcut illustrations. It is noticeable that the Bodleian and British Museum copies contain copies of Mother Hubberds Tale with the date 1613. (Bodl. K. 4. 23 Art. has two, one before the Shep. Cal. and one after. B.M. 1346. l. 1, on the other hand, has no copy of Mother Hubberds Tale). Possibly no special reprint of the poem was made for F 2. In very few instances does F 2 correct F's deviations from the earlier Qq, and in each case that I have noticed it was probably an independent correction (e.g. Shep. Cal. Ep. 214 rare Q 1, F 2: rath F; Feb. 121 girlonds Q 1, F 2: garlands F). Conversely, its reversions to the mistakes of later Qq, corrected by F, are probably accidental (e.g. Jan. 51 'sighes' for 'sithes'; April 36 'turned' for 'tuned'). Its best emendations are Colin Clout 861 'life giving' for 'like giving' and Elegie p. 357, l. 3 'glasse' for 'grasse';1 but against these we must set many bad errors due to careless printing, e.g. Jan. 77 'sunny' for 'sonned'; April 63 'white' for 'sweet'; July 46 'faine' for 'saine'; Aug. 79 'royde' for 'rovde'; Oct. 51 'lustlihead' for 'lustihead'; July 12 'cumbers' for 'climbers'; Feb. 159 'plant' for 'plainte'. It carries still further the process of modernization.
In 1653 appeared a Latin translation of the Shepheardes Calender by Theodore Bathurst, Fellow of Pembroke Coll., Camb., with the English on the opposite page. The number of instances in which the English text reproduces the errors of F 2 (though it occasionally corrects them) shows it to be in the main a reprint of F 2; but it is obvious that the translator had by him also an early Q copy. Thus the last line of the Envoy is printed with the error of Q 5, Ff 1, 2, The better please, the worse displease, etc. but is translated Sperne malos, placeasque bonis, etc. So also the 'sunny' sheep of F 2 (Jan. 77) he translates 'apricatus', the 'fain' of F 2 (July 46) he translates 'dicunt', and the 'talke' of Ff 1, 2 (Sept. 158) he translates 'oberrat', in each case showing his use of a Q copy.
Moreover the stanza omitted at June 89 by Q 5, Ff I, 2 is not found in its place, but is inserted at the end of the volume preceded by this note:
'Reader, Be pleased to take notice that in the later Editions of Spencer's Poems in Folio (which should have been the best) there is wanting one whole pg 507Stanza in the Month of June, which out of the first Edition of the Shepherds Calendar in Quarto may be thus supplyed, and is to come in after page 70. l. 10. of the Book.' (The stanza follows)
Todd is my authority for the statement that this volume was twice reprinted, once with a Latin dissertation De Vita Spenseri, et Scriptis, prefixed, and a glossary subjoined (not dated), and again with another title-page, dated 1732. I have not examined these copies.
F 3 followed in 1679. In type and general appearance it is far superior to the earlier Ff, but its text presents a still further degeneration. Indeed, it makes no pretence of going behind the authority of the preceding Ff, but is content in the main to follow F 2, though it sometimes prefers F. There is no sign that the Qq were ever referred to; for the corrections of 'glitter and' to 'glitterand' (July 177), 'beene' to 'beene not' (May 35), and 'Miracles' to 'Oracles' (ib. Gloss 340), the only ones I have noted, are probably accidental conjectures. At Dec. 76 it gives the new reading 'season' for 'reason', which many editors have accepted; other attempts to emend the text, 'her' for 'him' (Jan. 12), 'thinking' for 'thinken' (Feb. 24), 'his' for 'her' (April 14), are due to misunderstanding, and there are many meaningless misprints such as 'wine' for 'wind' (Feb. 7), 'underfond' for 'underfong' (Nov. 22), 'cough' for 'couth' (June 41), 'longring' for 'long lingring' (Oct. 3), 'arfen' for 'a frenne' (April 28). Like Q 5, Ff 1, 2, it omits the June stanza and only discovers the error when it comes to print, at the close of the volume, Bathurst's Latin translation. Then the missing lines are added with the note already quoted, the words 'which should have been the best' significantly altered to 'which we now followed'.
In 1715 Hughes brought out his edition of the Works of Spenser, with a Glossary, Life of the author, and an Essay on Allegorical Poetry (6 vols. 8vo).
'It is hoped,' he states in his introduction, 'that the reader will find it much more correct than some former Editions. The Shepherd's Calender had been so extremely corrupted that it is now in a manner wholly restor'd. Care has been taken not only to collect everything of this Author … and to preserve the Text entire, but to follow likewise, for the most part, the old spelling. This may be thought by some too strict and precise; yet there was a necessity for it, not only to show the true state of our language, as Spenser wrote it, but to keep the exact sense, which would sometimes be chang'd by the Variation of a syllable or letter.'
But, as was pointed out by Church and Warton, there are in Hughes's edition far too many of these changes to allow us to regard his text as anything but a compromise, and Todd is almost justified in speaking of it as 'reduced to modern orthography'. It is true that often Hughes goes behind the Folios to restore a Quarto reading, e.g. Feb. 280, Sept. 18 and 40, Oct. 39 and 65; but he took F 3 as his basis, accepting many of its errors and modernizations, and only fitfully and capriciously carried out his original intention. Many of his emendations were due to an inadequate knowledge of the earlier stages of the language; and by such changes as 'wonned' to 'wonted' (Feb. 119), 'assott' to 'a sot' (March 25), 'cryen' to 'crying' (April 95), 'sits' to 'fits' (May 77), 'more' to 'mere' (June 29), he falls an easy prey to Todd. But he was a capable editor up to the standard required in the early pg 508eighteenth century. Thus he notes the metrical deficiency of June 98, due to an error in Q 2 and ignored by all subsequent editions; he detects the omission of the Embleme to December, and supplies one himself; he sees that at Nov. 85 Spenser intended to write 'hath displayde' though all previous copies had printed 'doth displaye'. In his text too I first find the smoothing of metre in Dec. 95, and the alteration of 'tell' to 'till' (Daphnaida, 391), both of which have been attributed to earlier copies. There can be no doubt that Hughes's edition did much for the reputation of Spenser in the eighteenth century; it was reprinted in 1750, and all later editions before Todd were content with reprinting their text from Hughes.
Todd's valuable and fully annotated edition appealed in 1805 (8 vols. 8vo). 'The Text of this edition,' he says, 'is given from a careful collation of the various poems, which were published while the author lived; and from an attention to the mutual help in regard to correction, as well as to the choice of phraseology and orthography, which the several editions of those poems afford. … Of the Miscellaneous Poems every edition, subsequent to the original publications, has been more or less distinguished by innovation and errour; by innovation, which often perplexes what is originally perspicuous; by errour, which has sometimes converted what is serious into ridicule.'
But his text, though it is a great improvement upon that of the Folios and Hughes, hardly reaches the standard that he claims for it. A careful examination of those lines in his Shepheardes Calender where the different Quartos vary, proves, I think conclusively, that he printed from Q 3, collating with Qq 2 and 5, but with no reference to Qq 1 or 4. Possibly the absence of date from the title-page of Q 2 led him into mistaking it for Q 1: anyhow, wherever Q 1 and 2 differ Todd shows no acquaintance with the reading of the earlier Q.1 Similarly, though he often quotes F 1611, there is no evidence that he ever consulted it. For wherever F and F 2 vary, it will be found that it is F 2 and not F that he had before him.1 Here again he may have been misled by the binding up of a 1611 Faerie Queene with 1617 copies of the Miscellaneous Poems. But he laid himself open to attacks from Collier, who abused him for inaccuracy without realizing the source of his readings. And if he examined F 3 also it was only fitfully, for it is noticeable that wherever he mentions its errors or corrections he attributes them to Hughes, who had copied them, and not to their ultimate source.
The edition published in America in 1855, edited by F. J. Child, contains some excellent emendations, several of which have been adopted by later editors, and gives some evidence of acquaintance with the early editions. But like Todd, Child seems to have based his text of the Shepheardes Calender on Q 3, collating it with Q 2, but showing no knowledge of Q 1 except where it coincides with Q 2. Indeed, it is highly probable that he printed from his own corrected copy of Todd. The variae lectiones at the end of his volumes are very few in number, and they are to a large extent inaccurate; his citations in particular of Q 1 are quite untrustworthy. And the fact that three times he attributes to Q 1 readings which are mere misprints of Todd's inclines one to believe that his independent researches were not extensive. It pg 509might, of course, be urged that both Todd and Child had access to a copy of Q 1 which contained these readings; but one of them, 'gutted' for 'glutted' (Sept. 185), makes nonsense of the passage, and would certainly have been noted and emended by Todd, if he had found it in Q; and the accumulative evidence of all his citations from Q 1 makes it impossible to believe that a copy exists varying to such an extent from the Bodleian and British Museum copies.
J. P. Collier's edition, in five volumes, appeared in 1862. 'In these volumes,' writes Collier, 'attention has been most of all, and very anxiously directed to the purity and accuracy of Spenser's text.' He asserts that he has consulted and collated every edition from 1579–1679, and he makes some capital out of his preference for the original reading to emendations both of himself and others, even when they are obviously correct. His text as a whole is certainly the most accurate that had as yet appeared, and he is the first editor to show a careful study of Q 1 of the Shepheardes Calender. But his editorial arrogance has little justification in fact. Not only is his critical apparatus scanty, but in direct opposition to his avowed principles he often follows a late Quarto or Folio, without noting his departure from Q 1 (e. g. April 135, May 186, June 23, July 99, Nov. 33, Amoretti xv. 3). He is particularly bitter in his attacks upon Todd's accuracy, but he often charges him with the invention of readings to be found in F 2, which he himself professes to have collated, though his notes show no acquaintance either with the readings of F 2 or F 3 (cf. Colin Clout, 861). His desire to expose Todd has in one place led him into an error in his own text. In April, Gloss 328, he reads 'blended' and adds '"Blinded", for "blended" was a misprint for which Todd alone was responsible'. Yet 'blinded' is not only the Qq reading, but is also obviously right.
Collier's reputation for honesty in dealing with early texts is now a little tarnished, but most Spenser editors appear, up to the present, to have accepted his statements without reservation.1 This is no longer possible; but it can freely be admitted that under his hands the text of Spenser was much improved, and that some of his emendations, both those which he accepted himself and those merely offered as suggestions, were brilliant and scholarly.
In 1869 appeared the first issue of the Globe Spenser, edited by Richard Morris, by far the most accurate text that had appeared. Morris exercised a wise conservatism in his treatment of the text, and though he introduced rather more changes than a modern scholar would be inclined to admit, his book may be regarded as a careful reprint of the first editions of the poems, which he follows as a rule in all details of spelling, only allowing himself some latitude in matters of punctuation. But any one who has attempted a collation of the text of Spenser will be inclined to re-echo Grosart's despairing criticism, that Morris's 'abounding errors of omission and commission in his Various Readings have perplexed and worried me'. Some quite trivial changes he records; far more, of real importance, he omits. In dealing with the Shepheardes Calender he rarely refers to more than Q 1 and Q 5 or F, and his references are quite often inaccurate. His treatment of the pg 510other poems is little better. He is often content with a vague allusion to 'some old copies' or 'some modern editions', not specifying the books to which he refers; and he attributes to the wrong source, or gives as emendations of his own, suggestions which had been made by previous editors. Morris, indeed, states in his Preface that in this part of his work he has received some assistance from the labours of his predecessors; and many of his errors are certainly due to too implicit a trust in their authority. As his variae lectiones stand, however, they are both incomplete and untrustworthy, and are thus on quite a different level of scholarship from the admirable text which they accompany.
In 1882–4 Dr. Grosart published his Library Edition of Spenser in eight volumes, privately printed. The Shepheardes Calender he elected to print from Q 5, 'convinced that Spenser himself over-saw the successive editions.' That this view is quite untenable I have already shown in my introduction. Similarly Grosart preferred Q 2 of Daphnaida to Q 1. He is often guilty of a similar lack of critical judgement in his treatment of the text that he has elected to follow, both in unnecessary departures from it and in preserving it where it is obviously corrupt. In his apparatus criticus he set out to record the minutest changes of text in the editions published in the poet's lifetime; and his record, though not absolutely complete, is very full, and, considering its bulk, remarkably accurate. But he underrated the value and the interest of the Folio, of which he made no systematic collation, and he paid little attention to the alterations of the text in later copies.
It should be frankly acknowledged that the errors of all editions, from the Folio to Grosart, no less than their corrections, have been of the greatest help to me in checking the accuracy of my own text, and have called my attention to many points of interest which would otherwise have escaped me. My object in stating their several characteristics is not to expose their faults, but rather to show clearly that my divergences from their readings are not accidental. Where I differ from them I have referred once more to the original copies, and where I have accepted an emendation I have been able to credit it to its true originator.
Since this volume has been in the press, my attention has been called to Spenser's Poetical Works, ed. by R. E. Neil Dodge, Cambridge ed. (U.S.A.), one vol. 1908. The editor does not profess to give a complete collation of the early copies. His aim is rather to reproduce the first editions of the poems, except in punctuation and 'evident misprints', and to record every other departure from them. The task is carried out with great care and with a scholarship that inclines to conservatism (as e. g. in the retention of Q 1's reading in Shep. Cal., Dec. 75 and Q's of Teares of Muses 232, Colin Clout 168); and his departures from the original texts seem to me in nearly every case judicious. But the number of those departures made without comment shows his interpretation of the phrase 'evident misprints' to be somewhat too liberal. It is not always easy to determine when a word is misprinted, especially when the writer affects an archaic and dialectal style; and even if it appear 'evident' to the modern editor, it was not necessarily so to the Elizabethan; for it is in a measure by reason of the miscorrection of earlier errors, or what were taken for errors, that the text degenerated.