University of Virginia Library

John Warburton's Lost Plays
by
John Freehafer

Folio 1 of British Museum MS Lansdowne 807 contains a two-part list of "Manuscripts" of about 56 old plays, which are said to have been owned by John Warburton (1682-1759), Somerset Herald. This list has appended to it a memorandum in which Warburton said, "After I had been many years Collecting these MSS Playes, through my own carelesness and the Ignora[n]ce of my Ser[vant] in whose hands I had lodgd them they was unluckely burnd or put under Pye bottoms, excepting ye three which followes."[1] This list and memorandum form part of a bound volume which contains three surviving plays and part of a fourth play. Warburton's collection has won fame because of its shameful fate and his claim to have owned many irreplaceable plays by eminent older English dramatists, including Shakespeare.[2] In a jocose letter prefixed to his Fortunes of Nigel (1822), Sir Walter Scott typically portrayed Warburton as "the painful collector, but ah! the too careless custodian, of the largest collection of ancient plays ever known."

Since Warburton ascribed a dozen of the lost plays to Massinger, it is not surprising that William Gifford, a notably quick-tempered editor, in his 1805 edition of Massinger denounced Warburton for having allowed "treasures which ages may not reproduce" to be "burnt from an economical wish to save him the charges of more valuable brown paper" (I, xviii-xix n). In his reissue of that edition in 1813, Gifford speculated that it would have taken at least ten years for Warburton's cook to use up fifty plays in covering "her pies" (I, vii). Gifford's story of Warburton, his cook, and her pies was not seriously challenged for a century. In 1911, after noting other possibilities, W. W. Greg suggested as his "own idea of what happened" that Warburton's list was a largely unfilled want list, "containing the titles of such pieces as he thought it might be possible to recover," compiled from "various sources," but chiefly from entries of unprinted plays in the Stationers' Register by the famous play publisher Humphrey Moseley, who died in 1660/61. Greg finally suggested that "we have undoubtedly to lament the loss of a few pieces, . . . but not by any means the dramatic

holocaust that has made famous the name of the 'pie-eating Somerset Herald.'" (II, 258-259). Since 1911 various commentators have agreed or disagreed with Greg's suggestion, but no one has examined it in detail, although it is important to know whether Warburton owned the numerous and significant plays that he listed because, as Greg said, "upon the answer depends most of the value of the evidence the list affords."[3]

Although Greg said, "There is no doubt that Warburton was acquainted with the bulk entrances made by Humphrey Moseley in the Register,"[4] it is not known that Warburton consulted the Stationers' Register. Bishop Tanner made extracts from the Register in 1699, but no other scholar is known to have consulted it until nineteen years after Warburton's death.[5] Furthermore, Warburton's list differs greatly from the Register entries of unprinted plays. Of about 145 unpublished plays of pre-Restoration date which were entered in the Register, only 36 appear in Warburton's list. If Warburton searched the Register to compile a want list, it is strange that he copied less than a quarter of the pertinent items, ignored entries by all stationers other than Moseley, and omitted more than half of the unpublished plays entered by Moseley. Furthermore, the order of the items in Warburton's list shows no correlation with the order of Moseley's Register entries; no two successive entries in Warburton's long list follow the same order as any two successive items in the Register. Whereas Warburton's list appears not to be systematically arranged, the plays in Moseley's long entries of September 9, 1653 and June 29, 1660 are massed by authors and, in the 1653 entry, the authors are arranged in recognizable, though not precise, alphabetical order. What is more, sixteen scattered items in Warburton's list can not have been copied from the Register, because they do not appear in it. In 1958 Greg acknowledged the difficulty he had created by asking us to "assume that the titles of the two classes got mixed in compiling the list," adding that "it is not altogether easy to see how this occurred."[6] Greg's suggestion that Warburton might have copied some of these sixteen listings from the "Registers from 1640 onwards" (II, 256) was proved valueless when publication of those Registers (in 1913-14) revealed no previously unknown entry that corresponded to any Warburton listing. Greg also saw the need to reject his original suggestion that Warburton might have derived some entries from the Master of the Revels' "Office Book."[7] Edmond Malone apparently was the first scholar to examine that

Office Book, thirty years after Warburton's death.[8] Warburton's list also lacks the deletions which he might have made if, as Greg suggested, it was originally a mere want list of items of which Warburton actually obtained some at later dates.

A comparison of details of items on Warburton's list with those of the corresponding Register entries further suggests that Warburton did not copy the Register. Although Greg cited Warburton's listing of The Second Maiden's Tragedy as "definite evidence" that Warburton copied the Register (II, 255), that listing can not be a mere copy of the Register, because Warburton specifies a supposed author of the play ("Geo. Chapman"), whereas the Register lists none. Chapman is the second of three authors named in the existing manuscript of this play, where his name is now deleted. Recognizing that Warburton could not have copied the attribution of the play to Chapman from a Register entry that does not mention Chapman, Greg said, "There surely can be little doubt that Warburton had this entry as well as the manuscript play before him when he entered the latter in his own list" (II, 256). It seems unlikely, however, that Warburton would have entered the play in a list of "such pieces as he thought it might be possible to recover" if he already had "the manuscript play before him." Warburton's listing of The Second Maiden's Tragedy is one of many listings that he can not have merely copied from the Register, because he adds significant information that is not to be found in the corresponding Register entries. In ten other such listings, Warburton designates as comedies, tragicomedies, or tragedies plays that are not so designated in the Register. In five of these ten cases, the type of play is indicated in no known original source other than Warburton's list.[9] Warburton's designations of The Bugbears as a comedy and The Queen of Corsica as a tragedy are corroborated by existing manuscripts. His designation of The Crafty Merchant as a comedy is confirmed by an entry in Herbert's Office Book, which Warburton did not see. Of the ten listings, one is apparently wrong, for The Governor is a tragicomedy, whereas Warburton called it a tragedy.[10] In a significant number of cases, therefore, Warburton supplies information about Moseley plays which appears to be correct and is not found in the Register. This suggests that Warburton had access to a source or sources of information other than the Register.

Warburton further reveals his independence of the Register by omitting a dozen spurious subtitles which appear in Register entries. In each of these

entries Moseley has listed two plays as if they were but one play with a title and subtitle. This practice, which apparently was intended to save on fees for entering plays, was made possible by a Star Chamber decree of 1637, which took the licensing of plays for the press out of the hands of the Masters of the Revels, who were unlikely to allow such frauds. John Okes made the first such spurious entry on November 28, 1637.[11] A typical example of this frugal but risky practice is provided by Moseley's entry of "Alexius the Chast Gallant or. The Bashfull Lover . . . . by Phill: Massinger," for this single entry covers two distinct plays, which had been separately licensed for the stage. The Bashful Lover is in print, and the alternate title Alexius the Chast Gallant does not suit it. Thus, the Register entry is spurious, whereas Warburton's separate listing of "Alexias or ye chast Glallant" (sic) substantially agrees with an official stage license. Moseley entered at least 26 plays in spurious entries. Of these Warburton lists twelve without reproducing a spurious subtitle.[12] Warburton could scarcely have omitted all of these twelve carefully concealed errors, if he copied his listings from the Register.

Warburton also has avoided another subtle error that appears in the Register. In 1660 Moseley entered "The Tale of Ioconda and Astolso. a Comedy . . . by Tho: Decker." This play presumably dealt with Jocundo and Astolfo in Ariosto's Orlando Furioso. Although the second name (at least) is wrongly given in the Register, Warburton correctly lists "Jocondo & Astolfo C. Tho. Decker." It is possible that Warburton knew the Ariostan source and silently corrected an error; but it is as likely that the Stationers' Clerk misread an "f" as a long "s" in an unfamiliar proper name, whereas Warburton copied from a different source in which "Astolfo" appeared correctly. It is unlikely that Warburton, in the course of copying 40 brief entries from the Register, could have made substantial changes in 28 cases, at least 16 of which involve additions or corrections to the information in the Register. Although Greg suggests that the Register "would account for perhaps three quarters" of Warburton's list (II, 256), there is not a single Warburton listing that agrees precisely with the Register, and only about a fifth of his listings are of a sort that even a careless copyist could have derived from the Register, because the remaining four fifths either do not appear in the Register, or are fuller than the Register entries, or omit errors in the Register. This suggests that Warburton did not derive his information about the Moseley plays (or any others) from the Register.

The presence of titles of printed works on Warburton's list of "Manuscripts" further suggests that it was not a want list of unprinted plays. "The

fair favourit" had been printed in the Davenant folio of 1673, and Warburton would have had no apparent occasion to include "Sr. Jon. Sucklings Workes" in a list of "such pieces as he thought it might be possible to recover." Warburton may also have had "Manuscripts" of two printed works by Elkanah Settle — Fatal Love (1680) and The Fairy Queen (1692). Since manuscripts of printed plays were regarded as virtually valueless in the seventeenth and eighteenth centuries,[13] it is unlikely that Warburton would have specifically sought them out. Presumably he obtained these "Manuscripts" of printed works as parts of manuscript collections that also included unpublished plays.

The testimony of Warburton's memorandum has been challenged on the ground that he was untruthful, but Warburton's contemporaries and immediate successors seem to have found him veracious; certainly they accepted his account of his play collection. Citing what he described as "general evidence of character," Greg deemed it "intrinsically probable" that Warburton was "a liar" (II, 252); but William Hutton, a younger contemporary of Warburton who had closely studied his work, pronounced him "judicious" and expressly praised him "for his veracity."[14] The suggestion that Warburton falsely claimed (in a private memorandum) to have owned old plays to gain fame as a collector takes no account of either the low valuation of play manuscripts in the eighteenth century or the fact that those plays could have formed only a small and seemingly unimportant part of his large collection. Warburton collected many valuable manuscripts relating to every part of England, which he classified and had bound in numerous volumes.[15] "He had an amazing collection of MSS. books, prints, &c. relating to the History and Antiquities of England, which were sold by auction, after his death," on six evenings in November 1759.[16] In 1720 Humphrey Wanley, the scholarly and tight-fisted librarian of the celebrated Harleian Collection, went "to Mr Warburton, & offer'd him 100 Guineas for his old MSS &c."[17] Warburton demanded three hundred guineas, but eventually accepted Wanley's offer,[18] at a time when a manuscript of an old play might sell for a shilling. Although Warburton thus had little reason to regard his play manuscripts as important, he reported the destruction of most of them with mortification, and bound together three of the surviving plays.

Greg and others have speculated that Warburton's servant and her destruction of his unbound play manuscripts might be creations of folklore.[19] About the time of Warburton's birth, however, Dryden, in "MacFlecknoe," chose Herringman — who came into possession of much of Moseley's stock — to be captain of the "Bilk'd stationers" whose unsold books and plays became "Martyrs of pies" (lines 101-105). This seemingly prescient reference to the subsequent fate of plays that may have passed through Herringman's hands on the way to Warburton's reveals not so much a gift of prophecy as accurate observation of an age in which grocers, tobacconists, and cooks sought scrap paper, which was neither cheap nor plentiful. In Act I of Peter Motteux' Farewel Folly (1707), Mimic says of a group of play manuscripts, "We have such heaps of Tragedies, Comedies, Farces, Masques, Opera's, and what not, in the House, that we had twenty pounds bidden for 'em by a Grocer and a Pastry Cook." Thus, the reduction of fifty unbound play manuscripts to scrap paper is horrendous, but not improbable. Indeed, a cheesemonger once offered £10 for fifty volumes of the papers of Sir Julius Caesar, Master of the Rolls to James I and Charles I, which now form part of the same collection as the volume of Warburton plays.[20] Warburton's memorandum thus offers a plausible explanation for the loss of fifty plays that once existed.

The fact that Warburton's younger contemporaries and immediate successors accepted his acount of his plays and their destruction has been obscured by Greg's statement that the Warburton list "was first printed, at a time when the manuscript was still in the possession of the first Marquis of Lansdowne, by Reed in his 'Variorum' Shakespeare of 1803" (II, 227). Actually, Reed had printed the same list ten years earlier and had mentioned a still earlier list in which Malone referred to some of the Warburton plays.[21] Malone's list, which appeared in January 1778 as part of An Attempt to Ascertain the Order in which the Plays Attributed to Shakespeare were Written,[22] included "the names of several ancient plays . . . which are not known to have been ever printed," but which might "be yet in being" in manuscript.[23] This list of 34 plays shows that Malone knew of the Warburton collection in 1778, because it includes Demetrius and Marina, a play which is not recorded prior to the sale of Warburton's library

in 1759,[24] and "The honoured Loves," which can scarcely be anything but a misreading of Warburton's "The Honr. Loves," in which "the superior letter resembles a 'd.'"[25] Malone's list of 1778 includes ten plays that have been associated with Warburton, but his listing of "The Nobleman, by Cyril Tourneur" must be disregarded, because Malone derived it from a schedule of plays "acted in the year 1613," which he found in a copy of Lord Stanhope's accounts in Bodleian MS Rawlinson A239.[26] Of the nine other Warburton plays listed by Malone in 1778 as perhaps "yet in being," five still exist in manuscript,[27] while another two survived Warburton's death, since they appear in the sale catalogue.[28] Since Malone owned the manuscript of The Parliament of Love,[29] which Warburton attributed to William Rowley, he may have assumed that the two other plays attributed by Warburton to the same writer — The Four Honourable Loves and The Nonesuch — were likely survivors from Warburton's collection. Although Malone may have erred in these two cases, it is clear that he had substantial information about Warburton's plays before 1778.

Since Malone did not include the remaining 46 Warburton items on a list that was intended to include all the old plays that he thought might be found in manuscript, it may be suspected that Malone had concluded, no later than 1778, that those 46 plays were no longer "in being" in manuscript form. In 1780 Malone published additions to his list of old plays that might have survived in manuscript, which he revised and expanded further for publication in 1785 and later;[30] but he never added to that list any of the Warburton plays that he had not included in 1778. The additions by Malone, Reed, and Steevens to the new edition of David E. Baker's Biographia Dramatica (1782)[31] include more than fifty unnoted references to Warburton's play collection, which constitute a substantial reprint of Warburton's memorandum and list. The first such separate entry of a Warburton play reckons Beauty in a Trance "among those destroyed by Mr. Warburton's servant" (II, 30), and 37 other entries of single plays include similar statements. "The general havock made by Mr. Warburton's

servant" is noted several times (see, for example, II, 148), and the destruction of Duke Humphrey, supposedly by Shakespeare, evokes the following comment: "Could we believe it to have been really written by him, what a subject of regret would its ill fate be to every admirer of our immortal poet!" (II, 95). Thus, in 1778, and more fully in 1782, Malone indicated acceptance of Warburton's statement that most of his play manuscripts had been destroyed, but correctly suggested that the surviving plays numbered more than the three specified in Warburton's memorandum.

The 1782 Biographia Dramatica lists the three plays that survive in Warburton's bound volume as "in MS. in the library of lord Shelburne" (II, 38, 296, 331); Shelburne did not become Marquis of Lansdowne until 1784. Greg noted that "Warburton and Shelburne book-plates" are mounted on a preliminary leaf in Lansdowne MS 807 (II, 230). Shelburne did not purchase this volume at the Warburton sale, however. In 1759 he was merely Lord Fitzmaurice, and three months prior to the sale he had fought in the battle of Minden. He returned to England in 1760, and became seriously interested in collecting manuscripts in 1765.[32] James West, Treasurer and later President of the Royal Society, probably was the unrecorded purchaser in 1759. As Malone noted, West was interested in literary works;[33] and he bought "part of Warburton's collection" of manuscripts.[34] After West's death, "his curious collection of MSS. were sold to William Earl of Shelburne (afterwards first Marquis of Lansdowne),"[35] who apparently made the Warburton volume available to Malone.

If Greg had been right in saying that information about Warburton's play collection was first made public 44 years after his death by Reed from the Warburton list and memorandum alone, that information should scarcely be regarded as an accurate reflection of the views of Warburton's contemporaries and immediate successors. In fact, however, the first printed account of Warburton's collection (after the sale catalogue) appeared less than nineteen years after his death, and it was first published not by Reed, but by Malone, a highly respected scholar. Furthermore, besides the list and memorandum, Malone had studied the whole Warburton volume, as he shows by indicating that The Bugbears is "a free translation from some Italian drama" and printing an extract from The Second Maiden's Tragedy, together with a discussion of its authorship and sources.[36] Malone also had studied the Warburton sale catalogue and had purchased one Warburton manuscript, The Parliament of Love. Furthermore, in his publication

of 1778 and its various reprints, Malone invited those who had information about manuscripts such as Warburton's to come forward, and it is likely that persons living in 1778 and 1782 could have disproved Warburton's claims if they had been basically false.

Since it is hard to see how Warburton's list could have been a fabrication based on the Stationers' Register, its sources should be sought elsewhere. Malone and other early commentators agreed that Warburton had owned the 56 play manuscripts he claimed, and the grounds hitherto used to disprove that Warburton had those manuscripts will not bear inspection. Therefore, if Warburton's list contained nothing that could not plausibly derive from such manuscripts, it would be convenient to suspect that they were the basis of the list. In the cases of the plays not associated with Moseley, manuscripts may indeed have been Warburton's only sources; but in the cases of certain Moseley plays, it seems likely that Warburton had an additional source from which he derived information which is not found in existing play manuscripts that were certainly, or probably, in Warburton's hands. Greg correctly pointed to "2d. pt. Maidens Trag." as a Warburton listing that can not derive from the manuscript, which refers three times to "The Second Maydens Tragedy," but never to a "second part." On the other hand, Warburton's listing of this play as by "Geo. Chapman" can not derive from Moseley's Register entry, which specifies no author. Since, however, the Register and Warburton both call the play a "second part" without known authority, their listings may have an unrecognized common source. Furthermore, since Moseley was the publisher of Beaumont and Fletcher, he would have had a commercial motive for representing his manuscript play as a sequel to their celebrated Maid's Tragedy.

Other evidence that the Register and Warburton copied from a common source other than the play manuscripts is provided by Believe as you List, in the Register entry of which "a Comedy" has been deleted and replaced by "a Tragedy."[37] Warburton later retains the original error, which can not derive from the manuscript, which mentions "A new playe" and "A Tragedy," but not "a Comedy."[38] Presumably the deleted Register entry and Warburton's listing derive from a common, uncorrected source. Further evidence of a common source is provided by two other manuscripts. Both the Register and Warburton attribute The Governor to a mysterious "Sir Cornelius Formido;" but neither of the fragmentary names in the manuscript can be read as that of this otherwise unknown knight. Both the Register and Warburton attribute The Parliament of Love to William Rowley, but the manuscript bears no author's name, and the Master of the Revels licensed the play as a work of Massinger. The Register and Warburton likewise agree in including the word "Gallant" in the title of Alexius,

where the Master of the Revels reads "Lover" instead; but Alexius is, on the other hand, one of the dozen significant cases in which Warburton omits spurious subtitles that Moseley inserted in the Register. It therefore may be that Warburton had a source for information known to Moseley but not included either in Moseley's play manuscripts or in his Register entries. This lost source may also have supplied Warburton other data about the Moseley plays that is not found in the Register, such as his designations of ten additional plays as comedies, tragicomedies, or tragedies; his attributions of The Second Maiden's Tragedy and The Noble Trial; and his titles for Jocondo & Astolfo and Love hath found out his Eyes.

The Warburton manuscripts that had once been in Moseley's stock may have been accompanied by a descriptive list, prepared by Moseley or one of his successors, and ultimately derived from Moseley's office records. That Moseley prepared lists of plays in the course of his business is shown by his bulk entries in the Register and an extant bill of sale, covering 76 books, mostly plays, that he sold in 1640.[39] Since Moseley presumably planned to publish his plays, he probably tried to secure, along with each manuscript he bought, the information he would later need for a printed title page, including title, author, and type of play. A conflation of the corresponding entries of the Register and Warburton's list will provide these three bits of information for 30 Moseley plays out of 40, whereas the Register alone supplies them in only 21 cases, Warburton alone in only 22. This suggests that these 40 entries may derive from a source that supplied this information as far as Moseley had obtained it, and that the Register entries and Warburton's list contain variant and independent transcripts from such a source — both incomplete, but each including considerable information not found in the other.

The seemingly haphazard order of the items in the two parts of Warburton's list may have followed the arrangement of his manuscripts on two shelves; a number of old lists of plays apparently made up in shelf order exist.[40] One item only on the first part of Warburton's list — "The Vestall A Tragedy by H. Glapthorn" — is clearly duplicated on the second part. Occasional duplications of this sort are characteristic of old play lists, both manuscript and printed, and it is, in any case, possible that Warburton had two manuscripts of The Vestal. Since the three plays referred to in Warburton's memorandum come from the second part of his list and are the only known survivors from that part of his list, Warburton, before writing his memorandum, may have checked only one of two parts of his play collection and found three survivors there. All the other separate plays that appear to have survived from Warburton's collection come from the first part of his list.

Warburton's list supplies three kinds of information, in addition to the facts of his collection, that are not available from the Stationers' Register. First, Warburton lists sixteen plays not entered by Moseley, about half of which are not known from other sources. Second, Warburton supplies classifications of plays, attributions, and other details that Moseley did not enter in the Register. Third, Warburton omits intentional errors that appear in corresponding Register entries. Thus, although other sources leave little doubt that Moseley entered plays with spurious subtitles, Warburton's list establishes Moseley's guilt and points specifically to most of the spurious entries. By eliminating spurious double entries, Warburton helps to reveal the actual number of manuscripts by such a major dramatist as Massinger that Moseley registered. Warburton also supplies, augments, or repeats statements, many of which may be accurate, about lost plays attributed to Ford, Middleton, Tourneur, Shakespeare, Greene, Rowley, Dekker, Chapman, Marlowe, and lesser writers. The Harleian and Rawlinson collections, which were assembled about the same time as Warburton's, survive, but unfortunately their plays are mostly academic or unacted pieces of limited interest or, with the conspicuous exception of Sir Thomas More, manuscripts of published works by professional dramatists. The loss of most of Warburton's collection is serious because he apparently owned many unpublished plays by dramatists of note. In the case of Massinger, and probably in that of Ford, the loss is severe.

It is unnecessary, however, to blame Warburton alone for the loss of these plays. If Moseley's successors had shown the same zeal for complete publication of the works of writers of note that he showed in his editions of Cartwright and Suckling, the loss might be less; but even their folios of Davenant (1673) and Beaumont and Fletcher (1679) were less complete than Moseley's holdings could have made them. Furthermore, very few plays have survived from the portion of Moseley's unpublished stock that Warburton did not claim to have obtained, or from the large collection of manuscript plays listed by Abraham Hill.[41] The unknown custodians of those manuscripts probably merit censure beyond any that Warburton deserves. In any case, initial publication of old plays was virtually unknown in the eighteenth century. The publication of Middleton's Witch in 1778 was justly regarded as an unparalleled event; and it was accomplished only at the private expense of Isaac Reed, and because of the Shakespearean interest of a piece that revealed a connection with Macbeth.

Thus, a new examination of Warburton's list, memorandum, sale catalogue and surviving manuscripts suggests that Warburton's list and memorandum record plays that he owned and information that he believed to be correct. Therefore, skepticism about Warburton's motives and veracity might well give way to a further study of his statements and collection that could add considerably to existing knowledge of the finest period of English drama.

Notes