Volume 1 — POEMS

ON SEVERAL

OCCASIONS.

Written by the Reverend

JOHN DONNE, D.D.

Late Dean of St. PAUL'S.

WITH

Elegies on the Author's Death.

To this Edition is added,

Some Account of the Life of the Author.

LONDON:

Printed for J. Tonson, and Sold by W. Taylor at the Ship in Pater-noster-Row. 1719.

Title Page

These, then, are the early editions of Donne's poems. But the printed editions are not the only form in which the poems, or the great majority of the poems, have come down to us. None of these editions, we have seen, was issued before the poet's death. None, so far as we can discover (I shall discuss this point more fully later), was printed from sources carefully prepared for the press by the author, as were for example the LXXX Sermons issued in 1640. But Donne's poems were well known to many readers before 1633. One of the earliest published references to them occurs in 1614, in a collection of Epigrams by Thomas Freeman, called Runne | And a great Cast | The | Second Book.

Epigram 84.

To Iohn Dunne.

This is a very interesting piece of evidence as to the manner in which Donne's poems had been preserved by his friends, and the form in which they were being distributed. There is no reference to publication. It is doubtless due to this activity in collecting and transcribing the poems of the now famous preacher that we owe the number of manuscript collections dating from the years before and immediately after 1630.

Had Donne undertaken the publication of his own poems, such of these manuscript collections as have been preserved—none of which are autograph, and few or none of which have a now traceable history—would have little importance for a modern editor. The most that they could do would be to show us occasionally what changes a poem had undergone between its earliest and its latest appearance. But Donne's poems were not published in this way, and the manuscripts cannot be ignored. They must have for his editor at least the same interest and importance as the Quartos have for the editor of Shakespeare. Whatever opinion he may hold, on a priori or a posteriori grounds, regarding the superior authority of the First Folio of Shakespeare's plays, no editor, not 'thirled to' a theory, will deny that a right reading has been preserved for us often by the Quartos and the Quartos only. No wise man will neglect the assistance even of the more imperfect of them. Before therefore discussing the relative value of the different editions, and the use that may be made of the manuscripts, it will be well to give a short description of the manuscripts which the present editor has consulted and used, of their relation to one another, their comparative value, and the relation of some of them to the editions. It is, of course, possible that there are manuscripts of Donne's poems which have not yet come to light; and among them may be some more correctly transcribed than any which has come into the present editor's hands. He has, however, examined between twenty and thirty, and with the feeling recently of moving in a circle—that new manuscripts were in part or whole duplicates of those which had been already examined, and confirmed readings already noted but did not suggest anything fresh.

I will divide the manuscripts into four classes, of which the first two, it will be seen at a glance, are likely to be the most important for the textual critic.

(1) Manuscript collections of portions of Donne's poems, e.g. the Satyres. The 'booke' to which Freeman refers in the epigram quoted above was probably a small collection of this kind, and we have seen that Jonson sent the Satyres to Lady Bedford, and Francis Davison lent them to his brother. Of such collections I have examined the following:

Q. This is a small quarto manuscript, bound up with a number of other manuscripts, in a volume (MS. 216) in the library of Queen's College, Oxford. It is headed Mr. John Dunnes Satires, and contains the five Satires (which alone I have accepted as Donne's own) followed by A Storme, A Calme, and one song, The Curse (see p. 41), here headed Dirae. As Mr. Chambers says (Poems of John Donne, i, p. xxxvi), this is probably just the kind of 'booke' which Freeman read. The poems it contains are probably those of Donne's poems which were first known outside the circle of his intimate friends.

What seems to be a duplicate of Q is preserved among the Dyce MSS. in the South Kensington Museum. This contains the five Satyres, and the Storme and Calme. The MSS. are evidently transcribed from the same source, but one is not a copy of the other. They agree in such exceptional readings as e.g. Satyres, I. 58 'Infanta of London'; 94 'goes in the way' &c.; II. 86 'In wringing each acre'; 88 'Assurances as bigge as glossie civill lawes'. The last suggests that the one is a copy of the other, but again they diverge in such cases as III. 49 'Crants' Dyce MS.; 'Crates' Q; and IV. 215-16 'a Topclief would have ravisht him quite away' Q, where the Dyce MS. preserves the normal 'a Pursevant would have ravisht him quite away'.

If manuscripts like Q and the Dyce MS. carry us back, as they seem to do, to the form in which the Satyres circulated before any of the later collections of Donne's poems were made (between 1620 and 1630), they are clearly of great importance for the editor. The text of the Satyres in 1633 and the later editions, which closely resembles that of one of the later MS. collections, presents many variants from the older tradition. It is a difficult matter to decide how far these may be the corrections of the author himself, or of the collector and editor.

W. This, the Westmoreland MS., belonging to Mr. Edmund Gosse, is one of the most interesting and valuable manuscripts of Donne's poems which have come down to us. It is bound in its original vellum, and was written, Mr. Warner, late Egerton Librarian, British Museum, conjectured from the handwriting, 'a little later than 1625'. This date agrees with what one would gather from the contents, for the manuscript contains sonnets which must have been written after 1617, but does not contain any of the hymns written just at the close of Donne's life.

It should be noted that though W as a whole may have been transcribed as late as 1625, it clearly goes back in portions to an earlier date. The letters are headed e.g. To Mr. H. W., To Mr. C. B., &c. Now the custom in manuscripts and editions is to bring these headings up to date, changing 'To Mr. H. W.' into 'To S^r Henry Wotton'. That they bear headings which were correct at the date when the poems were written points to their fairly direct descent from the original copies.

If Q probably represents the kind of manuscript which circulated pretty widely, W is a good representative of the kind which circulated only among Donne's friends. Some of the poems escaped being transcribed into larger collections and were not published till our own day. The value of W for the text of Donne's poems must stand high. For some of the letters and religious poems it is our sole authority. Though a unique manuscript now, it was probably not so always, for Addl. MS. 23229 in the British Museum contains a single folio which must have been torn from a manuscript identical with W. The handwriting is slightly different, but the order of the poems and their text prove the identity.

A23. This same manuscript (Addl. MS. 23229), which is a very miscellaneous collection of fragments, presented to the Museum by John Wilson Croker, contains two other portions of what seem to have been similar small 'books' of Donne's poems. The one is a fragment of what seems to have been a carefully written copy of the Epithalamion, with introductory Eclogue, written for the marriage of the Earl of Somerset. Probably it was one of those prepared and circulated at the time. The other consists of some leaves from a collection of the Satyres finely written on large quarto sheets.

G. This is a manuscript containing only the Metempsychosis, or Progresse of the Soule, now in the possession of Mr. Gosse, who (Life &c. of John Donne, i. 141) states that it 'belonged to a certain Bradon, and passed into the Phillipps Collection'. It is not without errors, but its text is, on the whole, more correct than that of the manuscript source from which the version of 1633 was set up in the first instance.

(2) In the second class I place manuscripts which are, or aim at being, complete collections of Donne's poems. Most of these belong to the years between 1620 and 1633. They vary considerably in accuracy of text, and in the care which has been taken to include only poems that are authentic. They were made probably by professional copyists, and some of those whose calligraphy is most attractive show that the scribe must have paid the smallest attention to the meaning of what he was writing.

Of those which I have examined, two groups of manuscripts seem to me especially noteworthy, because both show that their collectors had a clear idea of what were, and what were not, Donne's poems, and because of the general accuracy with which the poems in one of them are transcribed. Taken with the edition of 1633 they form an invaluable starting-point for the determination of the canon of Donne's poems.

The first of these is represented by three manuscripts which I have examined, D (Dowden), H49 (Harleian MS. 4955), and Lec (Leconfield).

D is a small quarto manuscript, neatly written in a thin, clear hand and in ordinary script. It was formerly in the Haslewood collection, and is now in the possession of Professor Edward Dowden, Trinity College, Dublin, by whose kindness I have had it by me almost all the time that I have been at work on my edition.

H49 is a collection of Donne's poems, in the British Museum, bound up with some by Ben Jonson and others. It is a large folio written throughout apparently in the same hand. It opens with some poems and masques by Jonson. A certain Doctor Andrewes' poems occupy folios 57-87. They are signed Franc: Andrilla. London August 14. 1629. Donne's poems follow, filling folios 88 to 144b. Thereafter follow more poems by Andrewes, Jonson, and others, with some prose letters by Jonson.

Lec. This is a large quarto manuscript, beautifully transcribed, belonging to Lord Leconfield and preserved at Petworth House. Many of the manuscripts in this collection were the property of Henry, ninth Earl of Northumberland (1564-1632), the friend who communicated the news of Donne's marriage to his father-in-law.

These three manuscripts are obviously derived from one common source. They contain the same poems, except that D has one more than H49, and both of these have some which are not in Lec. The order of the poems is the same, except that D and Lec show more signs of an attempt to group the poems than H49. The text, with some divergences, especially on the part of Lec, is identical. One instance seems to point to one of them being the source of the others. In the long Obsequies to the Ld. Harrington, Brother to the Countess of Bedford, the original copyist, after beginning l. 159 'Vertue whose flood', had inadvertently finished with the second half of l. 161, 'were [sic] blowne in, by thy first breath.' This error is found in all the three manuscripts. It may, however, have come from the common source of this poem, and there are divergences in order and text which make me think that they are thus derived from one common source.

A special interest attaches to this collection, apart from the relative excellence of its text and soundness of its canon, from the probability that a manuscript of this kind was used for a large, and that textually the best, part of the edition of 1633. This becomes manifest on a close examination of the order of the poems and of their text. Mr. Gosse has said, in speaking of the edition of 1633: 'The poems are thrown together without any attempt at intelligent order; neither date, nor subject, nor relation is in the least regarded.' This is not entirely the case. Satires, Elegies, Epigrams, Songs are grouped to some extent. The disorder which prevails is due to two causes: (1) to the fact that the printer set up from a variety of sources. There was no previous collected edition to guide him. Different friends supplied collections, and of a few poems there were earlier editions. He seems to have passed from one of these to another as was most convenient at the moment. Perhaps some were lent him only for a time. The differences between copies of 1633 show that it was prepared carefully, but emended from time to time while the printing was actually going on. (2) The second source of the order of the poems is their order in the manuscripts from which they were copied. Now a comparison of the order in 1633 with that in D, H49, Lec reveals a close connexion between them, and throws light on the composition of 1633.

It is necessary, before instituting this comparison with 1633, to say a word on the order of the poems in D, H49, Lec themselves, as it is not quite the same in all three. H49 is the most irregular, perhaps therefore the earliest, each of the others showing efforts to obtain a better grouping of the poems. All three begin with the Satyres, of which D and Lec have five, H49 only four; but the text of Lec differs from that of the other two, agreeing more closely with the version of 1633 and of another group of manuscripts. They have all, then, thirteen Elegies in the same order. After these H49 continues with a number of letters (The Storme, The Calme, To S^r Henry Wotton, To S^r Henry Goodyere, To the Countesse of Bedford, To S^r Edward Herbert, and others) intermingled with Funeral Elegies (Lady Markham, Mris Boulstred) and religious poems (The Crosse, The Annuntiation, Good Friday). Then follows a long series of lyrical pieces, broken after The Funerall by A Letter to the Lady Carey, and Mrs. Essex Rich, the Epithalamion on the Palatine marriage, and an Old Letter ('At once from hence', p. 206). The lyrical pieces are then resumed, and the collection ends with the Somerset Eclogue and Epithalamion, the Letanye, both sets of Holy Sonnets, a letter (To the Countesse of Salisbury), and the long Obsequies to the Ld. Harrington.

D makes an effort to arrange the poems following the Elegies in groups. The Funeral Elegies come first, and two blank pages are headed An Elegye on Prince Henry. The letters are then brought together, and are followed by the religious poems dispersed in H49. The lyrical poems follow piece by piece as in H49, and the whole closes with the two epithalamia and the Obsequies to the Ld. Harrington.

The order in Lec resembles that of H49 more closely than that of D. The mixed letters, funeral elegies, and religious poems follow the Elegies as in H49, but Lec adds to them the two letters (Lady Carey and The Countess of Salisbury) and the Letanie which in H49 are dispersed through the lyrical pieces. Lec does not contain any of the Holy Sonnets, but after The Letanie ten pages are left blank, evidently intended to receive them. Thereafter, the lyrical poems follow piece by piece as in D, H49, except that The Prohibition ('Take heed of loving mee') is omitted—a fact of some interest when we come to consider 1633. Lec closes, like D, with the epithalamia and the Obsequies to the Lo: Harrington.

Turning now to 1633, we shall see that, whatever other sources the editor of that edition used, one was a collection identical with, or closely resembling, D, H49, Lec, especially Lec. That edition begins with the Progresse of the Soule, which was not derived from this manuscript. Thereafter follow the two sets of Holy Sonnets, the second set containing exactly the same number of sonnets, and in the same order, as in D, H49, whereas other manuscripts, e.g. B, O'F, S, S96, which will be described later, have more sonnets and in a different order; and W, which agrees otherwise with B, O'F, S, S96, adds three that are found nowhere else. The sonnets are followed in 1633 by the Epigrams, which are not in D, H49, Lec, but after that the resemblance of 1633 to D, H49, Lec becomes quite striking. These manuscripts, we have seen, begin with the Satyres. The edition, however, passes on at once to the Elegies. Of the thirteen given in D, H49, Lec, 1633 prints eight, omitting the first (The Bracelet), the second (Going to Bed), the tenth (Loves Warr), the eleventh (On his Mistris), and the thirteenth (Loves Progresse). That the editor, however, had before him, and intended to print, the Satyres and the thirteen Elegies as he found them in his copy0 of D, H49, Lec, is proved by the following extract which Mr. Chambers quotes from the Stationers' Register:

13^o September, 1632

This note is intelligible only when compared with this particular group of manuscripts. In others the order is quite different.

This bar—which was probably dictated by reasons of propriety, though it is difficult to see why the first and the eleventh Elegies should have been singled out—was got over later as far as the Satyres were concerned. They are printed after all the other poems, just before the prose letters. But by this time the copy of D, H49, Lec had perhaps passed out of Marriot's hands, for the text of the Satyres seems to show that they were printed, not from this manuscript, but from one represented by another group, which I shall describe later. This is, however, not quite certain, for in Lec the version of the Satyres given is not the same as in D, H49, but is that of this second group of manuscripts. Several little details show that of the three manuscripts D, H49, and Lec the last most closely resembles 1633.

With The Dampe the manuscript which I am supposing the editor to have followed in the main probably came to an end. The poems which follow in 1633 are of a miscellaneous character and strangely conjoined. The Dissolution (p. 64), A Ieat Ring sent (p. 65), Negative Love (p. 66), The Prohibition (p. 67), The Expiration (p. 68), The Computation (p. 69), complete the tale of lyrics. A few odd elegies follow ('Language thou art,' 'You that are she,' 'To make the doubt clear') with The Paradox. A Hymne to Christ, at the Authors last going into Germany is given a page to itself, and is followed by The Lamentations of Jeremy, The Satyres, and A Hymne to God the Father. Thereafter come the prose letters and the Elegies upon the Author.

What this comparison of the order of the poems points to is borne out by an examination of the text. The critical notes afford the materials for a further verification, and I need not tabulate the resemblances at length. In Elegie IV, for example, ll. 7, 8, which occur in all the other manuscripts and editions, are omitted by 1633 and by D, H49, Lec. Again, when a song has no title in 1633 it has frequently none in the manuscript. When there are evidently two versions of a poem, as e.g. in The Good-morrow and The Flea, the version given in 1633 is generally that of D, H49, Lec. Later editions often contaminate this with another version of the poem. At the same time there are ever and again divergences between the edition and the manuscript which are not to be ignored, and cannot always be explained. Some are due to error in one or the other, but some point either to divergence between the text of the editor's manuscript and ours, or to the use by the editor of other sources as well as this. In the fifth elegy (The Picture), for example, 1633 twice seems to follow, not D, H49, Lec, but another source, another group of manuscripts which has been preserved; and in The Aniversarie ll. 23, 24, the version of 1633 is not that of D, H49, Lec but of the same second group, which will be described later. On the whole, however, it is clear that a manuscript closely resembling that now represented by these three manuscripts supplied the editor of 1633 with the bulk of the shorter poems, especially the older and more privately circulated poems, the Songs and Sonets and Elegies. When he is not following this manuscript he draws from miscellaneous and occasionally inferior sources.

After D, H49, Lec, the most carefully made collection of Donne's poems is one represented now by four distinct manuscripts:

A18. Additional MS. 18646, in the British Museum.

N. The Norton MS. in Harvard College Library, Boston, of which an account is given by Professor Norton in a note appended to the Grolier Club edition.

TCC. A manuscript in the Library of Trinity College, Cambridge.

TCD. A large manuscript in the Library of Trinity College, Dublin, containing two apparently quite independent collections of poems—the first a collection of Donne's poems with one or two additional poems by Sir John Roe, Francis Beaumont, Sir Thomas Overbury, and Corbet; the second a quite miscellaneous collection, put together some time in the thirties of the seventeenth century, and including some of Donne's poems. It is only the first of these which belongs to the group in question.

These four manuscripts are closely connected with one another, but a still more intimate relation exists between A18 and TCC on the one hand, N and TCD on the other. N and TCD are the larger collections; A18 and TCC contain each a smaller selection from the same body of poems. Indeed it would seem that N is a copy of TCD, A18 of TCC.

TCD, to start with it, is a beautifully written collection of Donne's poems beginning with the Satyres, passing on to an irregularly arranged series of elegies, letters, lyrics and epicedes, and closing with the Metempsychosis or Progresse of the Soule and the Divine Poems, which include the hymns written in the last years of the poet's life. N has the same poems, arranged in the same order, and its readings are nearly always identical with those of TCD, so far as I can judge from the collation made for me. The handwriting, unlike that of TCD, is in what is known as secretary hand and is somewhat difficult to read. What points to the one manuscript being a copy of the other is that in 'Sweetest Love, I do not go' the scribe has accidentally dropped stanza 4, by giving its last line to stanza 3, and passing at once to the fifth stanza. Both manuscripts make this mistake, whereas A18 and TCC contain the complete poem. In other places N and TCD agree in their readings where A18 and TCC diverge. If the one is a copy of the other, TCD is probably the more authoritative, as it contains some marginal indications of authorship which N omits.

Presumably TCC is the earlier collection, and when TCD was made, the copyist was able to add fresh poems. It is clear, however, that in the case of even those poems which the two have in common, the one manuscript is not simply a copy of the others. There are several divergences, and the mistake referred to above, in 'Sweetest Love, I do not go', is not made in TCC. Strangely enough, a similar mistake is made by TCC in transcribing Loves Deitie and is reproduced in A18.

Whoever was responsible for this collection of Donne's poems, it was evidently made with care, at least as regards the canon. Very few poems that are not certainly by Donne are included, and they are correctly initialled. The only uninitialled doubtful poems are A Paradox, 'Whoso terms Love a fire,' which in all the four manuscripts follows 'No Lover saith, I love', and Beaumont's letter to the Countess of Bedford, which begins, 'Soe may my verses pleasing be.' In N, TCD this follows Donne's letter to the same lady, 'You that are she and you.' It is regrettable that the text of the poems is not so good as the canon is pure. The punctuation is careless. There are numerous stupid blunders, and there are evidences of editing in the interest of more regular metre or a more obvious meaning. At times, however, it would seem that the copyist is following a different version of a poem or poems (e.g. the Satyres) from that given in D, H49, and other manuscripts, and is embodying corrections perhaps made by the author himself. It is quite credible that Donne, in sending copies of his poems at different times to different people, may have revised and amended them. It is quite clear, as my notes will show, that of certain poems more than one version (each correct in itself) was in circulation.

Was A18, N, TC, or a manuscript resembling it one of the sources of the edition of 1633? In part, I think, it was. The most probable case at first sight is that of the Satyres. These, we have seen, Marriot was at first prohibited from printing. Otherwise they would have followed the Epigrams, and immediately preceded the Elegies. As it is, they come after all the other poems; they are edited with some cautious dashes; and their text is almost identical with that of N, TCD. In the first satire the only difference between 1633 and N, TCD occurs in l. 70, where N, TCD, with all the other manuscripts read—

1633,

'high' is either a slip or an editorial emendation. There are other cases of similar editing, not all of which it is possible to correct with confidence; but a study of the textual notes will show that in general 1633 follows the version preserved in N, TCD, and also in L74 (of which later), when the rest of the manuscripts present an interestingly different text. But strangely enough this version of the Satyres is also in Lec. This is the feature in which that manuscript diverges most strikingly from D and H49. Moreover in some details in which 1633 differs from A18, N, TC it agrees with Lec. It is possible therefore that the Satyres were printed from the same manuscript as the majority of the poems.

Again in the Letters not found in D, H49, Lec there is a close but not invariable agreement between the text of 1633 and that of this group of manuscripts. Those letters, which follow that To Sir Edward Herbert, are printed in 1633 in the same order as in this edition (pp. 195-226), except that the group of short letters beginning at p. 203 ('All haile sweete Poet') is here amplified and rearranged from W. Now in A18, N, TC these letters are also brought together (N, TCD adding some which are not in A18, TCC), and the special group referred to, of letters to intimate friends, are arranged in exactly the same order as in 1633; have the same headings, the same omissions, and the same accidental linking of two poems. In the other letters, to the Countesses of Bedford, Huntingdon, Salisbury, &c., the textual notes will show some striking resemblances between the edition and the manuscripts. In the difficult letter, 'T'have written then' (p. 195), 1633 follows N, TCD where O'F gives a different and in some details more correct text. In 'This twilight of two yeares' (p. 198) the strange reading of l. 35, 'a prayer prayes,' is obviously due to N, TCD, where 'a praiser prayes' has accidentally but explicably been written 'a prayer praise'. In the letter To the Countesse of Huntingdon (p. 201) the 1633 version of ll. 25, 26 is a correct rendering of what N, TCD give wrongly:

On the other hand there are some differences, as e.g. in the placing of ll. 40-2 in 'Honour is so sublime' (p. 218), which make it impossible to affirm that these poems were taken direct from this group of manuscripts as we know them, without alteration or emendation. The Progresse of the Soule or Metempsychosis, as printed in 1633, must have been taken in the first instance from this manuscript. In both the manuscripts and the edition, at l. 83 of the poem a blank space is left after 'did'; in both, l. 137 reads, 'To see the Prince, and soe fill'd the waye'; in both, 'kinde' is substituted for 'kindle' at l. 150; in l. 180 the 'uncloth'd child' of 1633 is explicable as an emendation of the 'encloth'd' of A18, N, TC; and similarly the 'leagues o'rpast', l. 296 of 1633, is probably due to the omission of 'many' before 'leagues' in A18, N, TC—'o'rpast' supplies the lost foot. It is clear, however, from a comparison of different copies that as 1633 passed through the press this poem underwent considerable correction and alteration; and in its final printed form there are errors which I have been enabled to correct from G.

The paraphrase of Lamentations, and the Epithalamion made at Lincolns Inn (which is not in D, H49, Lec) are other poems which show, in passages where there are divergent readings, a tendency to follow the readings of A18, N, TC, though in neither of these poems is the identity complete. It is further noteworthy that to several poems unnamed in D, H49, Lec the editor of 1633 has given the title which these bear in A18, N, TCC, and TCD, as though he had access to both the collections at the same time.

These two groups of manuscripts, which have come down to us, thus seem to represent the two principal sources of the edition of 1633. What other poems that edition contains were derived either from previously printed editions (The Anniversaries and the Elegy on Prince Henry) or were got from more miscellaneous and less trustworthy sources.

A third manuscript collection of Donne's poems is of interest because it seems very probable that it or a similar collection came into the hands of the printer before the second edition of 1635 was issued. A considerable number of the errors, or inferior readings, of the later editions seem to be traceable to its influence. At least it is remarkable how often when 1635 and the subsequent editions depart from 1633 and the general tradition of the manuscripts they have the support of this manuscript and this manuscript alone. This is the manuscript which I have called

O'F, because it was at one time in the possession of the Rev. T. R. O'Flaherty, of Capel, near Dorking, a great student of Donne, and a collector. He contributed several notes on Donne to Notes and Queries. I do not know of any more extensive work by him on the subject.

This manuscript has been already described by Mr. R. Warwick Bond in the Catalogue of Ellis and Elvey, 1903. It is a large but somewhat indiscriminate collection, made apparently with a view to publication. The title-page states that it contains 'The Poems of D. J. Donne (not yet imprinted) consisting of

Divine Poems, beginning Pag. 1

Satyres 57

Elegies 113

Epicedes and Obsequies 161

Letters to severall personages 189

Songs and Sonnets 245

Epithalamions 317

Epigrams 337

With his paradoxes and problems 421

finished this 12 of October 1632.'

The reader will notice how far this arrangement agrees with, how far it differs from, that adopted in 1635.

Of the twenty-eight new poems, genuine, doubtful, and spurious, added in 1635, this manuscript contains twenty, a larger number than I have found in any other single manuscript. An examination of the text of these does not, however, make it certain that all of them were derived from this source or from this source only. The text, for example, of the Elegie XI. The Bracelet, in 1635, is evidently taken from a manuscript differing in important respects from O'F and resembling closely Cy and P. Elegie XII, also, His parting from her, can hardly have been derived from O'F, as 1635 gives an incomplete, O'F has an entire, version of the poem. In others, however, e.g. Elegie XIII. Julia; Elegie XVI. On his Mistris; Satyre, 'Men write that love and reason disagree,' it will be seen that the text of 1635 agrees more closely with O'F than with any of the other manuscripts cited. The second of these, On his Mistris, is a notable case, and so are the four Divine Sonnets added in 1635. Most striking of all is the case of the Song, probably not by Donne, 'Soules joy now I am gone,' where the absurd readings 'Words' for 'Wounds' and 'hopes joyning' for 'lipp-joyning' (or perhaps 'lipps-joyning') must have come from this source. One can hardly believe that two independent manuscripts would perpetrate two such blunders. Taken with the many changes from the text of 1633 in which 1635 has the support of O'F, one can hardly doubt that among the fresh manuscript collections which came into the hands of the printer of 1635 (often only to mislead him) O'F was one.

Still less valuable as an independent textual authority is

P. This manuscript is a striking example of the kind of collections of poems, circulating in manuscript, which gentlemen in the seventeenth century caused to be prepared, and one cannot help wondering how they managed to understand the poems, so full is the text of gross and palpable errors. P is a small octavo manuscript, once in the Phillipps collection, now in the possession of Captain C. Shirley Harris, Oxford. On the cover of brown leather is stamped the royal arms of James I. On p. 1 is written, '1623 me possidet Hen. Champernowne de Dartington in Devonia, generosus.' Two other members of this old, and still extant, Devonshire family have owned the volume, as also Sir Edward Seymour (Knight Baronett) and Bridgett Brookbrige. The poems are written in a small, clear hand, and in Elizabethan character. Captain Harris has had a careful transcript of the poems made, and he allowed me after collating the original with the transcript to keep the latter by me for a long time.

JC. The John Cave MS. is a small collection of Donne's poems now in the possession of Mr. Elkin Matthews, who has kindly allowed me to collate it. It was formerly in Mr. O'Flaherty's possession. The original possessor had been a certain John Cave, and the volume opens with the following poem, written, it will be seen, while Donne was still alive:

The manuscript is divided into three parts, the first containing the five Satyres, the Litany and the Storme and Calme. The second consists of Elegies and Epigrammes and the third of Miscellanea, Poems, Elegies, Sonnets by the same Author. The elegies in the second part are, as in D, H49, Lec, and W, thirteen in number. Their arrangement is that of W, and, like W, JC gives The Comparison, which, D, H49, Lec do not, but drops Loves Progress, which the latter group contains. The text of these poems is generally that of W, but here and throughout JC abounds in errors and emendations. It contains one or two poems which were published in the edition of 1650, and which I have found in no other manuscript except O'F. In these JC supplies some obvious emendations. The poems in the third part are very irregularly arranged. This is the only manuscript, professing to be of Donne's poems, which contains the elegy, 'The heavens rejoice in motion,' which the younger Donne added to the edition of 1650. It is not a very correct, but is an interesting manuscript, with very few spurious poems. At the other end of the manuscript from Donne's, are poems by Corbet.

What seems to be practically a duplicate of JC is preserved in the Dyce Collection at the South Kensington Museum. It belonged originally to a certain 'Johannes Nedlam e Collegio Lincolniense' and is dated 1625. Cave's poem 'Upon Doctor Donne's Satyres' is inscribed and the contents and arrangement of the volume are identical with those of JC except that one poem, The Dampe, is omitted, probably by an oversight, in the Dyce MS. After my experience of JC I did not think it necessary to collate this manuscript. It was from it that Waldron printed some of the unpublished poems of Donne and Corbet in A Collection of Miscellaneous Poetry (1802).

S. The Stephens MS., now in the Harvard College Library, Boston, is the manuscript on which Dr. Grosart based his edition (though he does not reproduce it either consistently or with invariable accuracy) in 1873—an unhappy choice even were it legitimate to adopt any single manuscript in preference to the edition of 1633. Of all the manuscripts I have examined (I know it only through the collation made for me and from Dr. Grosart's citations) it is, I think, without exception the worst, the fullest of obvious and absurd blunders. There are too in it more evidences of stupid editing than in P, whose blunders are due to careless copying by eye or to dictation, and therefore more easy to correct.

(3) In the third class I place manuscripts which are not primarily collections of Donne's poems but collections of seventeenth-century poems among which Donne's are included. It is not easy to draw a hard and fast line between this class and the last because, as has been seen, most of the manuscripts at the end of the last list contain poems which are not, or probably are not, by Donne. Still, in these collections Donne's work predominates, and the tendency of the collector is to bring the other poems under his aegis. Initials like J. R., F. B., J. H. disappear, or J. D. takes their place. In the case of these last collections this is not so. Poems by Donne are included with poems which the collector assigns to other wits. Obviously this class could be made to include many different kinds of collections, ranging from those in which Donne is a prominent figure to those which include only one or two of his poems. But such manuscripts have comparatively little value and no authority for the textual critic, though they are not without importance for the student of the canon of Donne's poetry. I shall mention only one or two, though I have examined a good many more.

C. A strange duplicate of certain parts of A25 is a small manuscript in the Cambridge University Library belonging to the Baumgartner collection. It is a thin folio, much damaged by damp, and scribbled over. A long poem, In cladem Rheensen ('Verses upon the slaughter at the Isle of Rhees'), has been used by the cataloguer to date the manuscript, but as this has evidently been inserted when the whole was bound, the rest of the contents may be older or younger. The collection opens with three of the Elegies contained in A25. It then omits eleven poems which are in A25, and continues with twenty Songs and Obsequies, following the order of A25 but omitting the intervening poems. Some nine more poems are given, following the order of A25, but many are omitted in C which are found in A25, and the poems in C are often only fragments of the whole poems in A25. Evidently C is a selection of poems either made directly from A25, or from the collection of Donne's poems (with one or two by Beaumont and others) which A25 itself drew from.

M. This is a manuscript bought by Lord Houghton and now in the library of the Marquis of Crewe. It is entitled

A Collection of

Original Poetry

written about the time of

Ben: Jonson

qui ob. 1637

A later hand, probably Sir John Simeon's, has added 'Chiefly in the Autograph of Dr. Donne Dean of St. Pauls', but this is quite erroneous. It is a miscellaneous collection of poems by Donne, Jonson, Pembroke, Shirley, and others, with short extracts from Fletcher and Shakespeare. Donne's are the most numerous, and their text generally good, but such a collection can have no authority. It is important only as supporting readings and ascriptions of other manuscripts. I cite it seldom.

Another collection frequently cited by Grosart, but of little value for the editor of Donne, is the Farmer-Chetham MS., a commonplace-book in the Chetham Library, Manchester, which has been published by Grosart. It contains one or two of Donne's poems, but its most interesting contents are the 'Gulling Sonnets' of Sir John Davies, and some poems by Raleigh, Hoskins, and others. Nothing could be more unsafe than to ascribe poems to Donne, as Grosart did, because they occur here in conjunction with some that are certainly his.

A similar collection, which I have not seen, is the Hazlewood-Kingsborough MS., as Dr. Grosart called it. To judge from the analysis in Thorpe's Catalogue, 1831, this too is a miscellaneous anthology of poems written by, or at any rate ascribed to, Shakespeare, Jonson, Bacon, Raleigh, Donne, and others. There is no end to the number of such collections, and it is absurd to base a text upon them.

Of the manuscripts of the first two classes, which alone could put forward any claim to be treated as independent sources of the text of an edition of Donne's poems, it would be impossible, I think, to construct a complete genealogy. Different poems, or different groups of poems in the same manuscript, come from different sources, and to trace each stream to its fountain-head would be a difficult task, perhaps impossible without further material, and would in the end hardly repay the trouble, for the difficulties in Donne's text are not of so insoluble a character as to demand such heroic methods. The interval between the composition of the poems and their first publication ranges from about forty years at the most to a year or two. There is no case here of groping one's way back through centuries of transmission. The surprising fact is rather that so many of the common errors of a text preserved and transmitted in manuscript should have appeared so soon, that the text and canon of Donne's poems should present an editor in one form or another with all the chief problems which confront the editor of a classical or a mediaeval author.

The manuscripts fall into three main groups (1) D, H49, Lec. These with a portion of 1633 come from a common source. (2) A18, N, TCC, TCD. These also come from a single stream and some parts of 1633 follow them. L74 is closely connected with them, at least in parts. (3) A25, B, Cy, JC, O'F, P, S, S96, W. These cannot be traced in their entirety to a single head, but in certain groups of poems they tend to follow a common tradition which may or may not be that of one or other of the first two groups. Of the Elegies, for example, A25, JC, O'F and W transcribe twelve in the same order and with much the same text. Again, B, O'F, S96, and W have taken the Holy Sonnets from a common source, but O'F has corrected or altered its readings by a reference to a manuscript resembling D, H49, Lec, while W has a more correct version than the others of the common tradition, and three sonnets which none of these include. Generally, whenever B, O'F, S96, and W derive from the same source, W is much the most reliable witness.

Indeed, our first two groups and W have the appearance of being derived from some authoritative source, from manuscripts in the possession of members of Donne's circle. All the others suggest, by the headings they give to occasional poems, their misunderstanding of the true character of some poems, their erroneous ascriptions of poems, that they are the work of amateurs to whom Donne was not known, or who belonged to a generation that knew Donne as a divine, only vaguely as a wit.

These being the materials at our command, the question is, how are we to use them to secure as accurate a text as possible of Donne's poems, to get back as close as may be to what the poet wrote himself. The answer is fairly obvious, though it could not be so until some effort had been made to survey the manuscript material as a whole.

Moreover, any careful examination of the later editions, of their variations from 1633, and of the text of the poems which they print for the first time, shows clearly that some method more trustworthy than individual preference must be found if we are to distinguish between those of their variations which have, and those which have not, some authority behind them; those which are derived from a fresh reference to manuscript sources, and those which are due to carelessness, to misunderstanding, or to unwarrantable emendation. Apart from some such sifting, an edition of Donne based, like Mr. Chambers', on an eclectic use of the editions is exactly in the same position as would be an edition of Shakespeare based on an eclectic use of the Folios, helped out by a quite occasional and quite eclectic reference to a quarto. A plain reprint of 1633 like Alford's (of such poems as he publishes) has fewer serious errors than an eclectic text.

The first result of a collation of the manuscripts is thus to vindicate 1633, and to provide us with a means of distinguishing among later variants those which have, from those which have not, authority. But in vindicating 1633 the agreement of the manuscripts vindicates itself. If B's evidence is found always or most often to support A, a good witness, on those points on which A's evidence is in itself most probably correct, not only is A's evidence strengthened but B's own character as a witness is established, and he may be called in when A, followed by C, an inferior witness, has gone astray. In some cases the manuscripts alone give us what is obviously the correct reading, e.g. p. 25, l. 22, 'But wee no more' for 'But now no more'; p. 72, l. 26, 'his first minute' for 'his short minute'. These are exceptionally clear cases. There are some where, I have no doubt, my preference of the reading of the manuscripts to that of the editions will not be approved by every reader. I have adopted no rigid rule, but considered each case on its merits. All the circumstances already referred to have to be weighed—which reading is most likely to have arisen from the other, what is Donne's usage elsewhere, what Scholastic or other 'metaphysical' dogma underlies the conceit, and what is the source of the text of a particular poem in 1633.

—which is quite certainly right. Only three manuscripts have the, to my mind, most probably correct reading in Satyre I, l. 58, p. 147:

and only two, Q and the Dyce MS. which is its duplicate, the tempting and, I think, correct reading in Satyre IV, l. 38, p. 160:

Lastly, there are poems for which 1633 is not available. The authenticity of these will be discussed later. Their text is generally very corrupt, especially of those added in 1650 and 1669. Here the manuscripts help us enormously. With their aid I have been able to give an infinitely more readable text of the fine Elegie XII, 'Since she must go'; the brilliant though not very edifying Elegies XVII, XVIII, and XIX; as well as of most of the poems in the Appendixes. The work of correcting some of these had been begun by Dr. Grosart and Mr. Chambers, but much was still left to do by a wider collation. Dr. Grosart was content with one or two generally inferior manuscripts, and Mr. Chambers mentions manuscripts which time or other reasons did not allow him to examine, or he could not have been content to leave the text of these poems as it stands in his edition.

One warning which must be borne in mind when making a comparison of alternative readings has been given by Mr. Chambers, and my examination of the manuscripts bears it out: 'In all probability most of Donne's poems existed in several more or less revised forms, and it was sometimes a matter of chance which form was used for printing a particular edition.' The examination of a large number of manuscripts has shown that it is not probable, but certain, that of some poems (e.g. The Flea, A Lecture upon the Shadow, The Good-Morrow, Elegie XI. The Bracelet) more than one distinct version was in circulation. Of the Satyres, too, many of the variants represent, I can well believe, different versions of the poems circulated by the poet among his friends. And the same may possibly be true of variants in other poems. Our analysis of 1633 has shown us what versions were followed by that edition. What happened in later editions was frequently that the readings of two different versions were combined eclectically. In the present edition, when it is clear that there were two versions, my effort has been to retain one tradition pure, recording the variants in the notes, even when in individual cases the reading of the text adopted seemed to me inferior to its rival, provided it was not demonstrably wrong.

In view of what has been said, the aim of the present edition may be thus briefly stated:

(1) To restore the text of 1633 in all cases where modern editors have abandoned or disguised it, if there is no evidence, internal or external, to prove its error or inferiority; and to show, in the textual notes, how far it has the general support of the manuscripts.

(2) To correct 1633 when the meaning and the evidence of the manuscripts point to its error and suggest an indubitable or highly probable emendation.

(3) To correct throughout, and more drastically, by help of the manuscripts when such exist, the often carelessly and erroneously printed text of those poems which were added in 1635, 1649, 1650, and 1669.

(4) By means of the commentary to vindicate or defend my choice of reading, and to elucidate Donne's thought by reference to his other works and (but this I have been able to do only very partially) to his scholastic and other sources.

As regards punctuation, it was my intention from the outset to preserve the original, altering it only (a) when, judged by its own standards, it was to my mind wrong—stops were displaced or dropped, or the editor had misunderstood the poet; (b) when even though defensible the punctuation was misleading, tested frequently by the fact that it had misled editors. In doing this I frequently made unnecessary changes because it was only by degrees that I came to understand all the subtleties of older punctuation and to appreciate some of its nuances. A good deal of my work in the final revision has consisted in restoring the original punctuation. In doing this I have been much assisted by the study of Mr. Percy Simpson's work on Shakespearian Punctuation. My punctuation will not probably in the end quite satisfy either the Elizabethan purist, or the critic who would have preferred a modernized text. I will state the principles which have guided me.

I do not agree with Mr. Chambers that the punctuation, at any rate of 1633, is 'exceptionally chaotic'. It is sometimes wrong, and in certain poems, as the Satyres, it is careless. But as a rule it is excellent on its own principles. Donne, indeed, was exceptionally fastidious about punctuation and such typographical details as capital letters, italics, brackets, &c. The LXXX Sermons of 1640 are a model of fine rhetorical and rhythmical pointing, pointing which inserted stops to show you where to stop. The sermons were not printed in his lifetime, but we know that he wrote them out for the press, hoping that they might be a source of income to his son.

But Donne did not prepare his poems for the press. Their punctuation is that of the manuscript from which they were taken, revised by the editor or printer. One can often recognize in D the source of a stop in 1633, or can see what the pointing and use of capitals would have been had Donne himself supervised the printing. The printer's man was sometimes careless; the printer or editor had prejudices of his own in certain things; and Donne is a difficult and subtle poet. All these circumstances led to occasional error.

With trifling differences that is how Chambers and the Grolier Club editor print them. But the lines might run, to my mind preferably—

'And the right' being taken as equivalent to 'And as to the right'. One might even print—

One of the semicolons is equivalent to a little more than a comma, the other to a little less than a full stop.

Another effect of this finely-shaded punctuation is that the question is constantly forced upon an editor, is it correct? Has the printer understood the subtler connexion of Donne's thought, or has he placed the semicolon where the full stop should be, the comma where the semicolon? My solution of these difficulties has been to face and try to overcome them. I have corrected the punctuation where it seemed to me, on its own principles, definitely wrong; and I have, but more sparingly, amended the pointing where it seemed to me to disguise the subtler connexions of Donne's thought or to disturb the rhetoric and rhythm of his verse paragraphs. In doing so I have occasionally taken a hint from the manuscripts, especially D and W, which, by the kindness of Mr. Gosse and Professor Dowden, I have had by me while revising the text. But if I occasionally quote these manuscripts in support of my punctuation, it is only with a view to showing that I have not departed from the principles of Elizabethan pointing. I do not quote them as authoritative. On questions of punctuation none of the extant manuscripts could be appealed to as authorities. Their punctuation is often erratic and chaotic, when it is not omitted altogether. Finally, I have recorded every change that I have made. A reader should be able to gather from the text and notes combined exactly what was the text of the first edition of each poem, whether it appeared in 1633 or a subsequent edition, in every particular, whether of word, spelling, or punctuation. My treatment of the last will not, as I have said, satisfy every reader. I can only say that I have given to the punctuation of each poem as much time and thought as to any part of the work. In the case of Donne this is justifiable. I am not sure that it would be in the case of a simpler, a less intellectual poet. It would be an easier task either to retain the old punctuation and leave a reader to correct for himself, or to modernize. With all its refinements, Elizabethan punctuation erred by excess. A reader who gives thought and sympathy to a poem does not need all these commands to pause, and they frequently irritate and mislead.

Loves Vsury, p. 13, l. 5:

let my body raigne 1633 let my body range 1635-69, Chambers

Aire and Angels, p. 22, l. 19:

Ev'ry thy hair 1633 Thy every hair 1650-69, Chambers

The Curse, p. 41, ll. 3, 10:

His only, and only his purse 1633-54 Him, only for his purse 1669, Chambers

who hath made him such 1633 who hath made them such 1669, Chambers

A Valediction, p. 50, l. 16:

Those things which elemented it 1633 The thing which elemented it 1669, Chambers

The Relique, p. 62, l. 13:

mis-devotion 1633-54 mass-devotion 1669, Chambers

Elegie II, p. 80, l. 6:

is rough 1633, 1669 is tough 1635-54, Chambers

Elegie VI, p. 88, ll. 24, 26:

and then chide 1633 and there chide 1635-69, Chambers

her upmost brow 1633 her utmost brow 1635-69, Chambers (an oversight).

Epithalamions, p. 129, l. 60:

store, 1633 starres, 1635-69, Chambers

Ibid., p. 133, l. 55:

I am not then from Court 1633 And am I then from Court? 1635-69, Chambers

Satyres, p. 169, ll. 37-41:

The Iron Age that was, when justice was sold, now Injustice is sold deerer farre; allow All demands, fees, and duties; gamsters, anon The mony which you sweat, and sweare for, is gon Into other hands: 1633 The iron Age that was, when justice was sold (now Injustice is sold dearer) did allow All claim'd fees and duties. Gamesters, anon The mony which you sweat and swear for is gon Into other hands. 1635-54, Chambers (no italics; 'that' a relative pronoun, I take it)

The Calme, p. 179, l. 30:

our brimstone Bath 1633 a brimstone bath 1635-69, Chambers

To Sr Henry Wotton, p. 180, l. 17:

dung, and garlike 1633 dung, or garlike 1635-69, Chambers

Ibid., p. 181, ll. 25, 26:

The Country is a desert, where no good, Gain'd, as habits, not borne, is understood. 1633 The Country is a desert, where the good, Gain'd inhabits not, borne, is not understood. 1635-54, Chambers.

CANON.

Bearing this in mind we find that in the edition of 1633 there are only two poems—Basse's Epitaph on Shakespeare and the Psalme 137, both already mentioned—for the genuineness of which there is not strong evidence, internal and external. But these two poems are the only ones not contained in D, H49, Lec or in A18, N, TC. In D, H49, Lec, on the other hand, there are no poems which are not, on the same evidence, genuine. There are, however, some which are not in 1633, seven in all. But of these, five are the Elegies which, we have seen above, the editor of 1633 was prohibited from printing. The others are the Lecture upon the Shadow (why omitted in 1633 I cannot say) and the lines 'My fortune and my choice'. There are poems in 1633 which are not in D, H49, Lec. These, with the exception of poems previously printed, as the Anniversaries and the Elegie on Prince Henry, are all in A18, N, TC. This last collection does contain some twelve poems not by Donne, but of these the majority are found only in N and TCD, and they make no pretence to be Donne's. Three are initialled 'J. R.' (in TCD), and two of these, with some poems by Overbury and Beaumont, are not part of the Donne collection but are added at the end. Another poem is initialled 'R. Cor.' The only poems which are included among Donne's poems as though by him are The Paradox ('Whoso terms Love a fire') and the Letter or Elegy, 'Madam soe may my verses pleasing be.' Of these, the first is in all four manuscripts, the second only in N and TCD. Neither is in D, H49, Lec, or 1633. The last is by Beaumont, and follows immediately a letter by Donne to the same lady, the Countess of Bedford. Doubtless the two poems have come from some collection in which they were transcribed together, ultimately from a commonplace-book of the Countess herself. The former may be by Donne, but has probably adhered for a like reason to his paradox, 'No lover saith' (p. 302), which immediately precedes it.

We have thus three collections, each of which has kept its canon pure or very nearly so, and in which any mistake by one is checked by the absence of the poem in the other two. It cannot be by accident that these collections are so free from the unauthentic poems which other manuscripts associate with Donne's. Those who prepared them must have known what they were about. Marriot must have had some help in securing a text on the whole so accurate as that of 1633, and in avoiding spurious poems on the whole so well. When that guidance was withdrawn he was only too willing to go a-gathering what would swell the compass of his volume. If then a poem does not occur in any of these collections it is not necessarily unauthentic, but as no such poem has anything like the wide support of the manuscripts that these have, it should present its credentials, and approve its authenticity on internal grounds if external are not available.

1635. A Lecture upon the Shadow.

1635. Elegie XI. The Bracelet.

1635. Elegie XVI. On his Mistris.

1669. Elegie XVIII. Love's Progresse.

1669. Elegie XIX. Going to Bed.

(These are the five Elegies suppressed in 1633—at such long intervals did they find their way into print.)

1635. On himselfe.

We may add to these, without lengthy investigation, the four Holy Sonnets added in 1635:—

I. 'Thou hast made me.'

III. 'O might those sighs and tears.'

V. 'I am a little world.'

VIII. 'If faithfull soules.'

For these (though in none of the three collections) we have, besides internal probability, the evidence of W, clearly an unexceptionable manuscript witness. Walton, too, vouches for the authenticity of the Hymne to God my God, in my sicknesse, which indeed no one but Donne could have written.

This leaves for investigation, of poems inserted in 1635, 1649, 1650, or 1669, the following:—

1. Song. 'Soules joy, now I am gone.'

2. Farewell to love.

3. Song. 'Deare Love, continue nice and chaste.'

4. Sonnet. The Token.

5. 'He that cannot chuse but love.'

6. Elegie (XIII in 1635). 'Come, Fates; I feare you not.'

7. Elegie XII (XIIII in 1635). His parting from her. 'Since she must goe, and I must mourne.'

8. Elegie XIII (XV in 1635). Julia. 'Harke newes, ô envy.'

9. Elegie XIV (XVI in 1635). A Tale of a Citizen and his Wife. 'I sing no harme.'

10. Elegie XVII. Variety. 'The heavens rejoice.'

11. Satyre (VI in 1635, VII in 1669). 'Men write that love and reason disagree.'

12. Satyre (VI in 1669). 'Sleep, next society and true friendship.'

13. To the Countesse of Huntington. 'That unripe side of earth, that heavy clime.'

14. A Dialogue between Sr Henry Wotton and Mr. Donne. 'If her disdayne least change in you can move.'

15. To Ben Iohnson, 6. Jan. 1603. 'The state and mens affaires.'

16. To Ben Iohnson, 9. Novembris, 1603. 'If great men wrong me.'

17. To Sir Tho. Roe. 1603. Deare Thom: 'Tell her, if she to hired servants shew.'

18. Elegie on Mistresse Boulstred. 'Death be not proud.'

19. On the blessed Virgin Mary. 'In that, ô Queene of Queenes.'

20. Upon the translation of the Psalmes by Sir Philip Sydney and the Countesse of Pembroke his Sister. 'Eternall God, (for whom who ever dare).'

21. Ode. 'Vengeance will sit.'

22. To Mr. Tilman after he had taken Orders. 'Thou, whose diviner soule hath caus'd thee now.'

23. On the Sacrament. 'He was the Word that spake it.'

Of these twenty-three poems there is none which does not seem to me fairly open to question, though of some I think Donne is certainly the author.

Now this statement of Jonson's is confirmed by some at any rate of the manuscripts which contain the poem (see textual notes) since these append the initials 'J. R.' But all the manuscripts which contain the one poem contain also the next, 'If great men wrong me,' and though none have added the initials 'J. R.', B, in which it has been separated from 'The state and mens affairs' by two other poems, appends 'doubtfull author' (the whole collection being professedly one of Donne's poems). The third poem, To Sr Tho. Roe, 1603 (p. 410), is in the same way found in all the manuscripts (except two, which are one, H40 and RP31) which contain the epistles to Jonson, generally in their immediate proximity, and in B initialled 'J. R.' In the others the poem is unsigned, and in L74 a much later hand has added 'J. D.'

Of the other poems, the first—the poem which was in 1669 printed as Donne's seventh Satyre, was dropped in 1719 but restored by Chalmers, Grosart, and Chambers—is said in B to be 'By Sir John Roe', and it is initialled 'J. R.' in TCD. Even an undiscriminating manuscript like O'F adds the note 'Quere, if Donnes or Sr Th: Rowes', the more famous Sir Thomas Roe being substituted for his (in 1632) forgotten relative. Of the remaining five poems only two, 'Dear Love, continue nice and chaste' (p. 412) and 'Shall I goe force an Elegie?' (p. 410) are actually initialled in any of the manuscripts in which I have found them.

But the presence or absence of a name or initials is not a conclusive argument. It depends on the character of the manuscript. That 'Sleep next Society' is initialled 'J. R.' in so carefully prepared a collection of Donne's poems as TCD is valuable evidence, and the initials in a collection so well vouched for as HN, Drummond's copy of a collection of poems in the possession of Donne, can only be set aside by a scepticism which makes all historical questions insoluble. But no reliance can be placed upon the unsupported statement of any other of the manuscripts in which some or all of these poems occur, any more than on that of the 1635 and later editions. The best of them (H40, RP31) are often silent, and the others are too often mistaken to be implicitly trusted. If we are to get the truth from them it must be by cross-examination.

For the second proof on which my ascription of the poems to Roe is based is the singular regularity with which they adhere to one another. If a manuscript has one it generally has the rest in close proximity. Thus B, after giving thirty-six poems by Donne, of which only one is wrongly ascribed, continues with a number that are clearly by other authors as well as Donne, and of ten sequent poems five are 'Sleep next Society,' 'The State and mens affairs,' 'True love finds witt,' 'If great men wrong mee,' 'Dear Thom: Tell her if she.' A fragment of 'Men say that love and reason disagree' comes rather later. H40 and RP31 give in immediate sequence 'The State and mens affairs,' 'If great men wrong me,' 'True Love finds witt,' 'Shall I goe force an elegie,' 'Come Fates; I fear you not.' L74, a collection not only of poems by Donne but of the work of other wits of the day, transcribes in immediate sequence 'Deare Love continue,' 'The State and mens affairs,' 'If great men wrong mee,' 'Shall I goe force an elegie,' 'Tell her if shee,' 'True love finds witt,' 'Come Fates, I fear you not.' Lastly A10, a quite miscellaneous collection, gives in immediate or very close sequence '[Dear Thom:] Tell her if she,' 'True love finds witt,' 'Dear Love continue nice and chaste,' 'Shall I goe force an elegie,' 'Men write that love and reason disagree.' 'Come Fates; I fear you not' follows after a considerable interval.

To my mind they have a strong family resemblance, and very little resemblance to Donne's work. They are witty, but not with the subtle, brilliant, metaphysical wit of Donne; they are obscure at times, but not as Donne's poetry is, by too swift and subtle transitions, and ingeniously applied erudition; there are in them none of Donne's peculiar scholastic doctrines of angelic knowledge, of the microcosm, of soul and body, or of his chemical and medical allusions; they are coarse and licentious, but not as Donne's poems are, with a kind of witty depravity, Italian in origin, and reminding one of Ovid and Aretino, but like Jonson's poetry with the coarseness of the tavern and the camp. On both Jonson's and Roe's work rests the trail of what was probably the most licentious and depraving school in Europe, the professional armies serving in the Low Countries.

For a brief account of Roe's life will explain some features of his poetry, especially the vivid picture of life in London in the Satire, 'Sleep next Society,' which is strikingly different in tone, and in the aspects of that life which are presented, from anything in Donne's Satyres. Roe has been hitherto a mere name appearing in the notes to Jonson's and to Donne's poems. No critic has taken the trouble to identify him. Gifford suggested or stated that he was the son of Sir Thomas Roe, who as Mayor of London was knighted in 1569. Mr. Chambers accepts this and when referring to Jonson, Epigram 98, on Roe the ambassador, he adds, 'there are others in the same collection to his uncles Sir John Roe and William Roe.' Who this uncle was they do not tell us, but Hunter in the Chorus Vatum notes that, if Gifford's conjecture be sound, then he must be John Roe of Clapham in Bedfordshire, the eldest son of the Lord Mayor.

Such in brief outline is the life of the man who in 1603, possibly between his Irish and Low Country campaigns, appears in London as one, with his more famous cousin Thomas, of the band of wits and poets whose leader was Jonson, whose most brilliant star was Donne. Jonson's epigrams and conversations enable us to fill in some of the colours wanting in the above outline. The most interesting of these shows Roe to have been in Russia as well as Ireland and the Low Countries, and tells us that he was, like 'Natta the new knight' in his Satyre, a duellist:

XXXII.

ON SIR IOHN ROE.

In his conversations with Drummond Jonson as usual gave more intimate and less complimentary details: 'Sir John Roe was an infinite spender, and used to say, when he had no more to spend he could die. He died in his (i.e. Jonson's) arms of the pest, and he furnished his charges 20lb., which was given him back,' doubtless by his brother William. Morant states that 'Sir John the eldest son, having no issue, sold this Manor (i.e. Higham-hill) to his father-in-law Sir Reginald Argall, of whom it was purchased by the second son—Sir William Rowe'.

Such a career is much more likely than Donne's to have produced the satire 'Sleep, next Society', with its lurid picture of cashiered captains, taverns, stews, duellists, hard drinkers, and parasites. It is much more like a scene out of Bartholomew Fair than any of Donne's five Satyres. Nor was Donne likely at any time to have written of James I as Roe does. He moved in higher circles, and was more politic. But Roe had ability. 'Deare Love, continue nice and chaste' is not quite in the taste of to-day, but it is a good example of the paradoxical, metaphysical lyric; and there are both feeling and wit in 'Come, Fates; I feare you not', unlike as it is to Donne's subtle, erudite, intenser strain.

Returning to the list of poems open to question on pp. cxxviii-ix we have sixteen left to consider. Of some of these there is very little to say.

Nos. 1 and 14 are most probably by the Earl of Pembroke, and the Earl of Pembroke collaborating with Sir Benjamin Rudyard. Both were wits and poets of Donne's circle. The first song,

But both the contemptuous tone and the Platonic thought were growing common. We get it again in Lovelace's

The thought is Donne's, but not the airy note, the easy style, or the tripping prosody. Donne never writes of absence in this cheerful, confident strain. He consoles himself at times with the doctrine of inseparable souls, but the note of pain is never absent. He cannot cheat his passionate heart and senses with metaphysical subtleties.

The song Farewell to love, the second in the list of poems added in 1635, is found only in O'F and S96. There is therefore no weighty external evidence for assigning it to Donne, but no one can read it without feeling that it is his. The cynical yet passionate strain of wit, the condensed style, and the metaphysical turn given to the argument, are all in his manner. As printed in 1635 the point of the third stanza is obscured. As I have ventured to amend it, an Aristotelian doctrine is referred to in a way that only Donne would have done in quite such a setting.

and that rather enigmatical poem The Expostulation, which found its way into Jonson's Underwoods:

All of these poems bear the imprint of some actual experience, and to this cause we may perhaps trace the comparative rareness with which His parting from her is found in manuscripts, and that it finally appeared in a mutilated form. The poet may have given copies only to a few friends and desired that it should not be circulated. In the Second Collection of poems in TCD it is signed at the close, 'Sir Franc: Wryothlesse.' Who is intended by this I do not know. The ascriptions in this collection are many of them purely fanciful. Still, that the poem is Donne's rests on internal evidence alone.

Of the other two elegies, Julia, which is found in only two manuscripts, B and O'F, is quite the kind of thing Donne might have amused himself by writing in the scurrilous style of Horace's invectives against Canidia, frequently imitated by Mantuan and other Humanists. The chief difficulty with regard to the second, A Tale of a Citizen and his Wife, is to find Donne writing in this vein at so late a period as 1609 or 1610, the date implied in several of the allusions. He was already the author of religious poems, including probably La Corona. In 1610 he wrote his Litanie, and, as Professor Norton points out, in the same letter in which he tells of the writing of the latter he refers to some poem of a lighter nature, the name of which is lost through a mutilation of the letter, and says, 'Even at this time when (I humbly thank God) I ask and have his comfort of sadder meditations I do not condemn in myself that I have given my wit such evaporations as those, if they be free from profaneness, or obscene provocations.' Whether this would cover the elegy in question is a point on which perhaps our age and Donne's would not decide alike. Donne's nature was a complex one. Jack Donne and the grave and reverend divine existed side by side for not a little time, and even in the sermons Donne's wit is once or twice rather coarser than our generation would relish in the pulpit. But once more we must add that it is possible Donne has in this case been made responsible for what is another's. Every one wrote this occasional poetry, and sometimes wrote it well.

(1) The poem was not included in the 1633 edition, nor is it found in either of the groups D, H49, Lec and A18, N, TCC, TCD. It was added in 1635 with four other spurious poems, the dialogue ascribed to Donne and Wotton but assigned by the great majority of manuscripts to the Earl of Pembroke and Sir Benjamin Rudyard, the two epistles to Ben Jonson, and the Elegy addressed to Sir Thomas Roe, which we have assigned, for reasons given above, to Sir John Roe. The poem is found in only two manuscript collections, viz. P and the second, miscellaneous collection of seventeenth-century poems in TCD. In both of these it is headed Sr Walter Ashton (or Aston) to the Countesse of Huntingtone, and no reference whatsoever is made to Donne. I do not attach much importance to this title. Imaginary headings were quite common in the case of poems circulating in manuscript. Poems are inscribed as having been written by the Earl of Essex or Sir Walter Raleigh the night before he died, or as found in the pocket of Chidiock Tichbourne. Editors have occasionally taken these too seriously. Drayton's Heroicall Epistles made it a fashion to write such letters in the case of any notorious love affair or intrigue. The manuscript P contains a long imaginary letter from Sir Philip Sidney to Lady Mary Rich and a fragment of her reply. In the same manuscript the poem, probably by the Earl of Pembroke, 'Victorious beauty though your eyes,' is headed The Mar: B to the Lady Fe: Her., i.e. the Marquis of Buckingham to—I am not sure what lady is intended. The only thing which the title given to the letter in question suggests is that it was not an actual letter to the Countess but an imaginary one.

(2) Of Donne's relations with Elizabeth Stanley, who in 1603 became the Countess of Huntingdon, his biographers have not been able to tell us very much. He must have met her at the house of Sir Thomas Egerton when her mother, the dowager Countess of Derby, married that statesman in 1600. Donne says:

Donne's friend, Sir Henry Goodyere, seems to have had relations with her either directly or through her first cousin, the Countess of Bedford, for Donne writes to him from Mitcham, 'I remember that about this time you purpose a journey to fetch, or meet the Lady Huntington.' This fact lends support to the view of Mr. Chambers and Mr. Gosse that she is 'the Countesse' referred to in the following extract from a letter to Goodyere, which has an important bearing on the poem under consideration. Very unfortunately it is not dated, and Mr. Chambers and Mr. Gosse differ widely as to the year in which it may have been written. The latter places it in April, 1615, when Donne was on the eve of taking Orders, and was approaching his noble patronesses for help in clearing himself of debt. But Mr. Chambers points to the closing reference to 'a Christning at Peckam', and dates the letter 1605-6, when Donne was at Peckham after leaving Pyrford and before settling at Mitcham. I am not sure that this is conclusive, for in Donne's unsettled life before 1615 Mrs. Donne might at any time have gone for her lying-in or for a christening festival to the house of her sister Jane, Lady Grimes, at Peckham. But the tone of the letter, melancholy and reflective, is that of the letters to Goodyere written at Mitcham, and the general theme of the letter, a comparison of the different Churches, is that of other letters of the same period. The one in question (Letters 1651, p. 100; Gosse, Life, ii. 77) seems to be almost a continuation of another (Letters, 1651, p. 26; Gosse, Life, i. 225). Whatever be its date, this is what Donne says: 'For the other part of your Letter, spent in the praise of the Countesse, I am always very apt to beleeve it of her, and can never beleeve it so well, and so reasonably, as now, when it is averred by you; but for the expressing it to her, in that sort as you seeme to counsaile, I have these two reasons to decline it. That that knowledge which she hath of me, was in the beginning of a graver course then of a Poet, into which (that I may also keep my dignity) I would not seeme to relapse. The Spanish proverb informes me, that he is a fool which cannot make one Sonnet, and he is mad which makes two. The other strong reason is my integrity to the other Countesse' (i.e. probably the Countess of Bedford. The words which follow seem to imply a more recent acquaintance than is compatible with so late a date as 1615), 'of whose worthinesse though I swallowed your words, yet I have had since an explicit faith, and now a knowledge: and for her delight (since she descends to them) I had reserved not only all the verses which I should make, but all the thoughts of womens worthinesse. But because I hope she will not disdain, that I should write well of her Picture, I have obeyed you thus far as to write; but intreat you by your friendship, that by this occasion of versifying, I be not traduced, nor esteemed light in that Tribe, and that house where I have lived. If those reasons which moved you to bid me write be not constant in you still, or if you meant not that I should write verses; or if these verses be too bad, or too good, over or under her understanding, and not fit; I pray receive them as a companion and supplement of this Letter to you,' &c. If this was written in 1615 it is incompatible with the fact (supposing the poem under consideration to be by Donne) that he had already written to the Countess of Huntingdon a letter in a very thinly disguised tone of amatory compliment. If, however, it was written, as is probable, earlier, the reference may be to this very poem. Perhaps Goodyere thought it 'over or under' the Countess's understanding and did not present it.

(3) Certainly, looking at the poem itself, one has difficulty in declaring it to be, or not to be, Donne's work. Its metaphysical wit and strain of high-flown, rarefied compliment suggest that only he could have written it; in parts, on the other hand, the tone does not seem to me to be his. It is certainly very different from that of the other letters to noble ladies. It carries one back to the date of the Elegies. If Donne's, it is a further striking proof how much of the tone of a lover even a married poet could assume in addressing a noble patroness. Would Donne at any time of his life write to the Countess of Huntingdon in the vein of p. 418, ll. 21-36, or the next paragraph, ll. 37-76? One could imagine the Earl of Pembroke, or some one on a level of equality socially with the Countess, writing so; not a dependent addressing a patroness. The only points of style and verse which might serve as clues are (1) the peculiar use of 'young', e.g. l. 84 'youngest flatteries', l. 13 'younger formes'. With which compare in the Letter to Wotton, here added, at p. 188:

(2) A recurring pattern of line to which Sir Walter Raleigh drew my attention:

35. Who first looked sad, griev'd, pin'd, and shew'd his pain.

61. Love is wise here, keeps home, gives reason sway.

88. You are the straight line, thing prais'd, attribute.

113. Such may have eye and hand, may sigh, may speak.

I have not found this pattern elsewhere, and indeed the versification throughout seems to me unlike that of Donne. Donne's decasyllabic couplets have two quite distinctive patterns. The one is that of the Satyres. In these the logical or rhetorical scheme runs right across the metrical scheme—that is, the sense overflows from line to line, and the pauses come regularly inside the line. A good example is the paragraph beginning at p. 156, l. 65.

In the Elegies and in the Letters the structure is not so irregular and unmusical, but is periodic or paragraphic, i.e. the lines do not fall into couplets but into larger groups knit together by a single sentence or some closely connected sentences, the full meaning or emphasis being well sustained to the close. Good examples are Elegie I. ll. 1 to 16, Elegie IV. ll. 13 to 26, Elegie V. l. 5 to the end, Elegie VIII. ll. 1 to 34. Excellent examples are also the letter To the Countesse of Salisbury and the Hymn to the Saints and the Marquesse Hamylton. Each of these is composed of three or four paragraphs at the most. Now in the poem under consideration there are two, or three at the most, paragraphs which suggest Donne's manner, viz. ll. 1 to 10, ll. 11 to 16, and ll. 37 to 46. But the rest of the poem is almost monotonously regular in its couplet structure. To my mind the poem is not unlike what Rudyard might have written. Indeed a fine piece of verse by Rudyard, belonging to the dialogue between him and the Earl of Pembroke on Love and Reason, is attributed to Donne in several manuscripts. The question is an open one, but had I realized in time the weakness of the positive external evidence I should not have moved the poem. I have been able to improve the text materially.

With regard to the Elegie on Mistris Boulstred (18 on the list) I cannot expect readers to accept at once the conjecture I have ventured to put forward regarding the authorship, for I have changed my own mind regarding it. Two Elegies, both perhaps on Mris. Boulstred, Donne certainly did write, viz.

and another, entitled Death, beginning

Both of these are attributed to Donne by quite a number of manuscripts and are very characteristic of his poetry in this kind, highly charged with ingenious wit and extravagant eulogy. It is worth noting that in the Hawthornden MS. the second bears no title (it is signed 'J. D.'), and that it is not included in D, H49, Lec. It is certainly Donne's; it is not quite certain that it was written on Mris. Boulstred. Indeed, as I have pointed out elsewhere, the reference to Judith in a verse letter which seems to have been sent to Lady Bedford with the poem, and the tenor of the poem, suggest that Lady Markham is the subject of the elegy. Jonson, in speaking of Mris. Boulstred, says, 'whose Epitaph Done made,' which points to a single poem; but he may have been speaking loosely, or be loosely reported.

In contrast to these two elegies that beginning 'Death be not proud' is found in only five manuscripts, B, H40, O'F, P, RP31. Of these H40 and RP31 are really one, and in them the poem is not ascribed to Donne. In two others, O'F and P, the poem is given in a very interesting and suggestive manner, viz. as a continuation of 'Death I recant'. What this suggests is the fairly obvious fact that the second poem is to some extent a reply to the first. 'Death I recant' is answered by 'Death be not proud'. If O'F and P are right in their arrangement, then Donne answers himself. Beginning in one mood, he closes in another; from a mood which is almost rebellious he passes to one of Christian resignation. This was the view I put forward in a note to the Cambridge History of Literature (iv. 216). I had hardly, however, sent off my proofs before I felt that there was more than one objection to this view. There is in the first place nothing to show that 'Death I recant' is not a poem complete in itself; there is no preparation for the recantation. In the second place, 'Death be not proud' is as a poem slighter in texture, vaguer in thought, in feeling more sentimental and pious, than Donne's own Epicedes. Whoever wrote it had a warmer feeling for Mris. Boulstred than underlies Donne's rather frigid hyperboles. This suggested to me that the poem was indeed an answer to 'Death I recant', but by another person, another member of Lady Bedford's entourage. In this mood I came on the ascription in H40, viz. 'By C. L. of B.' This indicated no one whom I knew; but in RP31 it appeared as 'By L. C. of B.,' i.e. Lucy, Countess of Bedford. We know that the Countess did write verses, for Donne refers to them. In a letter which Mr. Gosse dates 1609 (Gosse's Life, &c., i. 217; Letters, 1651, p. 67) he speaks of some verses written to himself: 'They must needs be an excellent exercise of your wit, which speak so well of so ill.' That the Countess of Bedford could have written 'Death be not proud', we cannot prove in the absence of other examples of her work; that if she could she did, is very likely. She had probably asked Donne for some verses on the death of her friend. He replied with 'Death I recant'. The tone, which if not pagan is certainly not Christian, while it is untouched by any real feeling for the subject of the elegy, displeased her, and she replied in lines at once more ardent and more resigned. At any rate, whether by Lady Bedford or not, the poem is not like Donne's work, and the external evidence is against its being his. B attributes it to 'F. B.', i.e. Francis Beaumont. It is right, on the other hand, to point out that Donne opens one of the Holy Sonnets with the exclamation used here:

I have left the question of authorship an open one. Personally I cannot bring myself to think that it is Donne's.

The sonnet On the Blessed Virgin Mary (19 on the list), 'In that O Queene of Queenes, thy birth was free,' is included among Donne's poems in 1635 and in B, O'F, S, S96. There is little doubt that it is not Donne's but Henry Constable's. It is found in a series of Spiritual Sonnets by H. C., in Harl. MS. 7553, f. 41, which were first published by T. Park in Heliconia, ii. 1815, and unless all of these are to be given to Donne this cannot. It is not in his style, and Donne more than once denies the Immaculate Conception in the full Catholic sense of the doctrine. Nothing could more expressly contradict this sonnet than the lines in the Second Anniversarie:

and the passage quoted in the note to this poem.

The poem Upon the translation of the Psalmes by Sir Philip Sydney, and the Countesse of Pembroke his Sister, if by Donne, was probably written late in his life and never widely circulated. It occurred to me that the author might be John Davies of Hereford, who was a dependent of the Countess and her two sons, and who made a calligraphic copy of the Psalms of Sidney and his sister, from which they were printed by Singer in 1823. But Professor Saintsbury considers, I think justly, that the 'wit' of the opening lines,