# Double Falsehood / Cardenio — annotated claims register of the authorship scholarship

Companion index of everything fetched: `../sources/scholarship/SOURCES.md`.
Scope: every study I could identify on (a) who wrote the text printed by Lewis Theobald as *Double Falshood; or, The Distrest Lovers* (Drury Lane 13 Dec 1727; 8vo 1728, two issues), and (b) whether that text descends from the King's Men's *Cardenno/Cardenna* (court performances paid 20 May 1613 and 9 July 1613 for 8 June 1613) entered by Humphrey Moseley in the Stationers' Register on 9 Sept 1653 as "The History of Cardenio, by Mr Fletcher. & Shakespeare."

## 0. Conventions used throughout

* **Access tags** (stated for every study):
  * `[DIRECT]` — full text read in this project; file saved under `sources/scholarship/`.
  * `[ABSTRACT]` — only the publisher's abstract / a preview page read verbatim.
  * `[SECONDARY]` — known only from quotation or summary in another study that I did read directly (mostly Alotaibi's 2016 Liverpool thesis, Taylor 2016, the NOS *Authorship Companion* 2017, Graham 1920, Bradford 1910, Oliphant 1927). Page numbers given for the *secondary* source are marked "via".
  * `[WEB-SUMMARY]` — known only from a web-fetch or web-search auto-summary (paraphrase, not verbatim; lowest reliability).
* **Scene numbering** = Arden Third Series (Hammond 2010): 1.1, 1.2, 1.3, 2.1, 2.2, 2.3, 2.4, 3.1, 3.2, 3.3, 4.1, 4.2, 5.1, 5.2 (fourteen scenes; the 1728 quarto's scene divisions, which Bradford/Graham/Oliphant cite as "III, 3", "IV, 1" etc., map one-to-one onto these). Line numbers are Arden unless a page of the 1728 quarto is given ("DF, p. 48").
* **Quotations**: verbatim, from OCR'd copies where marked [DIRECT]; eighteenth-century long-s normalised; OCR slips silently corrected only where the reading is certain.
* **Abbreviations**: Sh = Shakespeare; Fl = Fletcher; Th = Theobald; Dav = Davenant; Mass = Massinger; *Quest* = Carnegie & Taylor 2012; *Creation* = Bourus & Taylor 2013; NOS = New Oxford Shakespeare; LION = Literature Online; SR = Stationers' Register.

## 1. Documentary anchors that the scholarship argues about (summary only — the documentary record is in `documents.md`)

1. Treasurer of the Chamber accounts (Bodleian MS Rawlinson A.239, f. 47): payment 20 May 1613 to John Heminges for six plays performed at court in the 1612–13 season, one called "Cardenno"; payment 9 July 1613 for "Cardenna" performed 8 June 1613 before the Duke of Savoy's ambassador. No author named. Stern 2011 (555–57) notes the spellings could in principle denote the Spanish town Cardeña, not Cervantes's Cardenio [SECONDARY via Folger LPD and Alotaibi]; Taylor 2012 (*Quest* 11–61 and 283–99) argues the court performance most likely fell in February 1613 and the June performance at the Lord Mayor's [WEB-SUMMARY + NOS Companion 583–84, DIRECT].
2. SR 9 Sept 1653, Moseley: "The History of Cardenio, by Mr Fletcher. & Shakespeare." (with a full stop after Fletcher). Bradford 1910: "Moseley was far from reliable" (p. 55) [DIRECT]. Freehafer 1969 (p. 508, via Alotaibi): the stop suggests Shakespeare's name "was added as an afterthought" and Moseley entered "in alphabetical order of authorship", implying he "might have preferred to have a play by Fletcher alone" [SECONDARY]. Stern 2011 (557–58): Moseley "seems to take liberties with some names, particularly those of authors who could be published lucratively" and misascribed other plays to Shakespeare [SECONDARY via Alotaibi, thesis p. 14].
3. Assignment of Moseley's copyrights to Jacob Tonson the younger, 5 April 1718 (Hammond 2010, p. 81) [SECONDARY].
4. Charles Gildon, letter written c.1710, printed 1719 (pp. 267–8): a manuscript play by "the immortal Shakespear; in the Maturity of his Judgment, a few Years before he dy'd", with "infallible Proof that the Copy is genuine" [SECONDARY via NOS Companion p. 585, DIRECT].
5. Theobald's Preface (1728): three manuscripts; one "of above sixty years standing, in the handwriting of Mr Downes, the famous old prompter", earlier owned by Betterton, who "designed to have ushered it into the world"; a "Noble Person" who supplied a copy; the tradition that Shakespeare gave it "as a Present of Value, to a Natural Daughter of his, for whose Sake he wrote it"; a third copy "much more perfect". Second issue of 1728 changes "in our Tongue" to "in that other Great Poet" (Graham 1920, p. 17 n., citing R. F. Jones) and adds that Theobald "had once design'd a Dissertation" on the play's authenticity but reserved it "to my Defence" [DIRECT via Graham; Alotaibi pp. 10–11].
6. Royal licence (5 Dec 1727) calling it "an original play of William Shakespeare".
7. *Gazetteer* 31 March 1770: "the original Manuscript of this play is now treasured up in the Museum of Covent Garden Playhouse"; Covent Garden burned 19 Sept 1808 [WEB-SUMMARY; Hammond 2010, 48–58 cited by LPD].

## 2. Register of studies (chronological)

### 2.1 Contemporary reactions, 1727–1733 `[DIRECT via Theobald's preface as quoted by Bradford 1910 and Graham 1920; Pope via Nicholl 2010]`
* Critics at once urged Fletcher. Theobald's Preface: "Others again, to depreciate the Affair, as they thought, have been pleased to urge, that tho' the Play may have some Resemblance of Shakespeare, yet the Colouring, Diction, and Characters, come nearer to the Style and Manner of Fletcher. This, I think, is far from deserving any Answer; I submit it to the Determination of better Judgments; tho' my Partiality for Shakespeare makes me wish, that Everything which is good, or pleasing, in our Tongue, had been owing to his Pen." (quoted Graham 1920, pp. 16–17) [DIRECT].
* Pope and the Scriblerians: forgery (*Dunciad*; Mallet's *Epistle to Mr Pope* 1733: "See him on Shakespeare pore, intent to steal / Poor farce, by fragments, for a third-day meal" — quoted Nicholl 2010) [DIRECT web page].
* Theobald's own claim to have written 1.3.10–14 ("Strike up, my masters; / But touch the strings with a religious softness…") and his insistence that 3.1.17 ("None but itself can be its parallel") was Shakespeare's (Stern 2011, 588–89, via Alotaibi p. 201) — the pair of claims that Farmer, Malone and Stern all exploit.
* Why it matters later: Bradford (1910, p. 55) treats the early "detection of Fletcher in the play by persons who had no reason whatever to look for him there" as evidence of a genuine old manuscript; Oliphant (1927, p. 283) treats Theobald's denial of Fletcher as "the strongest proof of his bona fides".

### 2.2 Richard Farmer, *An Essay on the Learning of Shakespeare* (Cambridge, 1767; 2nd ed. 1767, pp. 26–29) `[DIRECT — four archive.org copies + ECCO-TCP div; file farmer-copy-onlearningofshak00farm.txt]`
* **Method**: a single linguistic "minutia" (accentuation) plus impressionistic "Style and Manner"; also cites Langbaine on Shirley's unpublished manuscripts.
* **Argument** (2nd ed. pp. 26–29): "Sometimes a very little matter detects a forgery. You may remember a Play called the Double Falshood, which Mr. Theobald was desirous of palming upon the world for a posthumous one of Shakespeare … Mr. Pope himself, after all the strictures of Scriblerus, in a Letter to Aaron Hill, supposes it of that age; but a mistaken accent determines it to have been written since the middle of the last century." He quotes "This late example / Of base Henriquez, bleeding in me now, / From each good Aspect takes away my trust" and "You have an Aspect, Sir, of wondrous wisdom": "The word Aspect, you perceive, is here accented on the first Syllable, which, I am confident, in any sense of it, was never the case in the time of Shakespeare".
* "After all, the Double Falshood is superior to Theobald. One passage, and one only in the whole Play, he pretended to have written: … 'Strike up, my Masters; / But touch the Strings with a religious softness: / Teach sound to languish thro' the Night's dull Ear, / 'Till Melancholy start from her lazy Couch, / And Carelessness grow Convert to Attention.' These lines were particularly admired; and his vanity could not resist the opportunity of claiming them: but his claim had been more easily allowed to any other part of the performance."
* Answers Benjamin Victor's objection that the Cervantes source was "not published 'till the year after Shakespeare's death": "unluckily the same Novel appears in a part of Don Quixote, which was printed in Spanish, 1605, and in English by Shelton, 1612."
* **Conclusion**: "But you want my opinion:—and from every mark of Style and Manner, I make no doubt of ascribing it to Shirley. Mr. Langbaine informs us, that he left some Plays in MS.—These were written about the time of the Restoration, when the Accent in question was more generally altered. Perhaps the mistake arose from an abbreviation of the name." (Stern 2011, 590, reads Farmer's abbreviation as "W.Sh." = William Shirley [SECONDARY via Alotaibi p. 199].)
* **Critiques**: Bradford 1910, p. 52: Farmer forgot "that Theobald confessed to having revised the original with 'great labour and pains'"; "I have myself studied Shirley's plays carefully and I detect none of these 'marks of style and manner' in Double Falsehood" [DIRECT]. Graham 1920, p. 14: D. Nichol Smith repeated the "aspect" argument "as if Theobald, among other emendations, would not certainly have changed any words having an accent not current in 1727" [DIRECT]. Followers of Farmer: Dyce (1833, would have printed it in Gifford–Dyce *Shirley* "had he possessed one of the manuscripts"), O. G. Gilchrist (*Quarterly Review* 7: "the internal evidence of that play strongly confirms his decision"), A. W. Ward (Graham 1920, pp. 13–14; Bradford 1910, p. 52) [DIRECT]. No modern study has tested Shirley (Stern 2012 and the *Renaissance and Reformation* reviewer both note Massinger and Shirley "have not been addressed adequately").

### 2.3 George Steevens and Edmond Malone, *Supplement to the Edition of Shakspeare's Plays* (1780); Malone's annotated copy of the 1728 second issue `[SECONDARY — via Alotaibi 2016, pp. 66–67 and 14–15, and Ivan Lupić, 'Malone's Double Falsehood', Quest 95–114, as cited there; Lupić's chapter itself inaccessible (OUP Cloudflare)]`
* Steevens (1780) first connected the 1613 "Cardenno/Cardenna" payments (extracts he obtained from the Treasurer of the Chamber accounts and copied into his Langbaine) with Moseley's 1653 entry; Reed (1782) then suggested the identity with *Double Falsehood* (Alotaibi p. 15; Lupić) [SECONDARY]. Bradford 1910, p. 55, believed Malone "does not show the most distant sign of associating that play with either Shakespeare or Fletcher"; Lupić shows the Steevens/Reed/Malone sequence in detail (pp. 98–99, per my earlier notes).
* Steevens rejected Shakespeare: "had it … been written by our author, it would surely have been published in the folio of 1623, or at least would have been ascribed to him in some ancient catalogue" (via Alotaibi p. 66).
* Malone → Massinger: listed from the Stationers' Books "eleven plays that were written by Massinger but never published"; cited the parallel between the line Pope ridiculed ("None but Itself can be its Parallel", 3.1.17) and Massinger's *Duke of Milan* ("Her goodness does disdain comparison, / And but itself admits no parallel"); and *Duke of Milan* "yet be wise; / Soar not too high to fall, but stoop to rise" ~ DF 1.3.72–73 "I must stoop to gain her; / Throw all my gay Comparisons aside"; but noted the stronger parallel "lay his gay comparisons apart" (*Antony and Cleopatra* 3.13.26), which, if not interpolated by Theobald, "would serve to confirm Massinger's title to this play, he having very frequently imitated Shakespeare" (Alotaibi pp. 66–67) [SECONDARY].
* Malone's marginalia in his 1728 copy (Lupić, p. 107, via Alotaibi p. 66): "Theobald asserted that he was the author of these five lines & they were the only lines in the play that he wrote. I believe both assertions to have been false."
* **Status**: the only serious historical case for H3 (another Jacobean/Caroline dramatist). Nobody since has run Massinger tests except Proudfoot 2012 (early Massinger plays give "over 20 per cent" line-end-polysyllable matching, which he discounts) and Pascucci 2017 (detects Massinger in 2.2, 2.3, 2.4 — see §2.34(b)).

### 2.4 Isaac Reed (ed.), Baker's *Biographia Dramatica* (1782), vol. II `[DIRECT — reed-1782-biographia-dramatica-vol2.txt]`
* D-section no. 104: "THE DOUBLE FALSHOOD, or, The Distrest Lovers. Trag. by L. Theobald, Acted at Drury-Lane, 8vo. 1727. This piece Theobald endeavoured to persuade the world was written by Shakspeare. How true his assertion might be, I cannot pretend to determine, but very few I believe gave any credit to it. The play, however, was acted with considerable success, and was the last piece in which Mr. Booth appeared. Dr. Farmer is of opinion, that it is a production of Shirley's, or at least not earlier than his time. Mr. Malone inclines to believe it written by Massinger."
* H-section no. 103: "THE HISTORY OF CARDENIO. A Play, by Mr. Fletcher and Shakspeare; entered on the books of the Stationers' Company Sept. 9, 1653; but I believe never printed. It has been suggested, that this play may possibly be the same as The Double Falshood; afterwards brought to light by Mr. Theobald."
* Graham 1920 (p. 14): Jaggard "considers Reed's conjecture the 'shrewdest' — that Theobald was the author" — i.e. Reed's private view was forgery, while his entry is the first printed identification of DF with *Cardenio*.

### 2.5 Nineteenth-century opinion (via Graham 1920, pp. 13–15, and Bradford 1910, pp. 52–53) `[DIRECT for the summaries; the originals not consulted]`
Gifford (1813, *Massinger* ed.) denied Theobald's authorship on the strength of the verse and connected the play with Elizabethan drama "because of the use of comparison for caparison"; Dyce, Gilchrist, Ward → Shirley; Halliwell-Phillipps discredited Theobald "merely because the history of the manuscript … was not satisfactory"; Churton Collins: "might be founded on some old play, but is for the most part from Theobald's own pen"; Sidney Lee: "there is nothing in the play as published by Theobald to suggest Shakespeare's hand, but Theobald doubtless took advantage of a tradition that Shakespeare and Fletcher had combined…"; Jaggard: forgery; D. Nichol Smith (1903): Shirley. Bradford's verdict on the lot (p. 52–53) and Graham's (p. 15): "With two or three exceptions, the scholars cited made no careful study of Double Falsehood before passing judgment on it."

### 2.6 Gamaliel Bradford Jr., "The History of Cardenio by Mr. Fletcher and Shakespeare", *Modern Language Notes* 25.2 (Feb. 1910), 51–56 (first stated in *The Nation* 88 (1909), 328) `[DIRECT — bradford-1910-mln-history-of-cardenio.txt]`
* **Method**: connoisseurial verse-and-diction tests: double (feminine) endings, especially two-word feminine endings and endings in "now/too/yet"; favourite words ("extremely", "now"); parallels of imagery (cloud figures); dramatic structure (Fletcherian climaxes, garrulous old men, Julio's madness ~ Jailer's Daughter). Corpus: Fletcher canon read as a whole; Shakespeare by ear.
* **Results**: "an examination of the different scenes which I had attributed, on independent grounds, to Fletcher and to his fellow author, shows 47% of double endings in Fletcher's portion and only 32% in the remainder. Of course, owing to Theobald's revision, this is much less than Fletcher's usual percentage. Nevertheless the difference is significant." Examples: "Yes, I am that Lord Roderick, and I lie not" (3.3); "She's stol'n away; and whither gone, I know not" (3.3); "Make up your Malice, and dispatch his Life too". "The adverb 'extremely,' occurring twice in this play, is a favorite with Fletcher. It occurs only five times in all Shakespeare, two of those being in the Fletcherian Henry VIII. It is found at least twenty-eight times in Fletcher". (p. 54)
* **Scene assignment** (p. 55): "These marked Fletcher peculiarities, then, do not appear at all before III, 3. In that scene they are abundant. Fletcher's also is the speech of the shepherds at the beginning of IV, but the business of Julio's madness takes us back to the firmer, stronger hand: — 'It puzzles my Philosophy, to think / That the rude Blast, hot Sun, and dashing Rains / Have made no fiercer War upon thy Youth.'" Graham 1920 (p. 17) paraphrases Bradford's Fletcher share as "II, iii [sic — Bradford's own text says III, 3]; IV (except about fifty lines near the beginning); and V (except a small portion which he does not define)".
* **Conclusion** (p. 55): forgery excluded by Theobald's denial of Fletcher — "This paragraph seems to me absolutely incompatible with any possibility of Theobald's having set himself deliberately to imitate Fletcher. I do not see, then, how we can avoid the conclusion that Theobald was really in possession of an old play, and that, in view of its subject and of Fletcher's part authorship, that play was 'The History of Cardenio.'" On Shakespeare: "For external evidence we have Moseley's attribution … but Moseley was far from reliable. We have Theobald's tradition, which is worth little, but something." "The internal evidence is certainly not of great importance. Numerous passages resembling Shakespeare's acknowledged work … can be produced; but none is close enough to have much weight." "Many lines and phrases in the non-Fletcherian portion of the play have, to my ear something of a Shakespearean turn or resonance, as distinguished from either Beaumont or Massinger". Final: "All these arguments are certainly far from sufficient to associate Shakespeare's name in any positive fashion with The History of Cardenio, not to speak of Double Falsehood as it stands … a play which was almost indisputably Fletcher's, which had at least some claim to be classed among Fletcher's collaborations with his greatest predecessor" (p. 56).
* **Critiques**: Schevill 1911 (below) attacked the Elizabethan-source premise; Graham 1916 confirmed Bradford's division statistically; Oliphant 1927 "almost complete agreement"; Jackson (NOS Companion ch. 8, p. 123) and Taylor 2016 treat Bradford/Oliphant as the start of the modern consensus; Stern 2012 objects that none of this checked Theobald's own habits.

### 2.7 Rudolph Schevill, "Theobald's Double Falsehood?", *Modern Philology* 9.2 (Oct. 1911), 269–285 `[DIRECT — schevill-1911-theobalds-double-falsehood.txt]`
* **Claim**: the play's immediate source is not Shelton (1612) but the prose "novel" *The Adventures on the Black Mountains* in Samuel Croxall's *Select Collection of Novels* (1720–22; 2nd ed. 1729), a version "new translated" by an "eminent hand"; hence the play is Theobald's own work of the 1720s ("Double Falsehood was taken from The Adventures, etc., as printed by Croxall", p. 285).
* **Method**: three-column parallel of Cervantes/Shelton, Croxall's novel and DF (pp. 271–284).
* **Critiques**: Graham 1916 and 1920 (pp. 8–11) show seven verbal parallels to Shelton absent from Croxall (Shelton IV.i.2, IV.i.3, IV.i.5 and 11, III.xi.267, III.xi.269, Cardenio's song III.250 ~ DF 4.1; IV.i.1 ~ DF 3.3) and that Croxall's own version "furbished up Shelton"; Graham's first established "fact": "The immediate source of Double Falsehood was Shelton's Don Quixote" (1916, p. 280). Schevill's thesis has had no defenders since, though Pujante 1998 and Stern 2011 revive the different point that Shelton could have been used by Theobald directly (a 1725 reprint of Shelton existed: Stern 582, via Alotaibi p. 26).

### 2.8 Walter Graham, "The Cardenio–Double Falsehood Problem", *Modern Philology* 14.5 (Sept. 1916), 269–280; "Note on the Cardenio–Double Falsehood Problem", *MP* 14 (Jan. 1917), 184; *Double Falsehood by Lewis Theobald* (Western Reserve Studies I.6, Cleveland, May 1920: introduction pp. 1–24 + reprint of the 1728 text) `[DIRECT — all three files]`
* **Method (1916)**: verse tests on Bradford's division and on Theobald's own tragedies: feminine endings, run-on lines, weak and light endings, feminine caesural pauses, mid-stopped speeches, two-monosyllable feminine endings; definitions from Neilson/Thorndike. Corpus: DF (1447 verse lines) v. Theobald's *Persian Princess* (1278 lines), *Electra* (1376), *Orestes* (1606); TNK and H8 as the Sh/Fl comparators.
* **Results (1916, tables pp. 275–276, 278)**: feminine endings — *Persian Princess* 20.5%, *Electra* 16%, DF 38.1%, *Orestes* 22.6%; run-on lines 15 / 9 / 18.3 / 10; weak-and-light endings .08 (10) / .06 (8) / .15 (22) / .025 (4). "There are in all 137 such [two-monosyllable] endings in Double Falsehood, whereas Electra and Orestes have 20 and 27 respectively … 'Sir' and 'not' are found 11 and 10 times used in this way … [in *The Chances*] 'Sir' is found 22 times and 'not' 11 times." Bradford's division: "Parts possibly by Fletcher" 754 lines — feminine endings 44.2%, run-on 15.2%, weak/light 3; "By Another" 693 lines — feminine 32%, run-on 21.5%, weak/light 19 (feminine caesural pauses 23.3% after the 1917 correction). "Nineteen to three is a very significant difference." "It seems hardly possible that the portion of the latter which contains 44.2 per cent of feminine endings was composed in 1727 by the same hand that wrote Orestes three years later with only 22.6 per cent. Nor does it appear any more likely that the first part of Double Falsehood which contains 19 weak and light endings was written by the same author as the play of 1730 in which there are only 4 altogether." (pp. 276–277)
* **1917 Note** corrects two table errors (feminine caesural pauses "by another" 23.3 not 13.3; TNK weak/light 0.23 not 23).
* **Conclusion (1916, p. 280)**: "(1) The immediate source of Double Falsehood was Shelton's Don Quixote. (2) The style of the play as a whole, and the second part in particular, differs appreciably from Theobald's acknowledged work. (3) There are unmistakable evidences of two styles. (4) These distinct styles show a general similarity to those in Two Noble Kinsmen and Henry VIII, which are now recognized as belonging to Shakspere and Fletcher." But also (p. 279): "the shortness of it, the lifeless verse, some of the language, and the distinctly inferior dramatic quality are strong arguments against its being the work of Shakspere or Fletcher either … It is not impossible … that the manuscripts were revised several times before they reached him."
* **1920 introduction** (pp. 23–24): "The presence of Fletcher's hand in the play seems clearly evident … Since this external evidence makes Shakespeare Fletcher's collaborator, it renders probable the supposition that the other original writer … is our great master-dramatist. Yet … no scholar is yet ready to declare unreservedly his belief in Shakespeare's participation." Notes that "One scholar has suggested that II, ii, III, iii, and V, ii, have something reminiscent of Beaumont" and that Forsythe wanted Daborne, Davenport and William Rowley tried; Jones wanted DF read against Theobald's *Richard II* and *Duchess of Malfi* adaptations (both programmes later executed by Jackson 2012 and Carnegie 2012).
* **Critiques**: none specific; Tarlinskaja 2017 and Jackson 2012 effectively re-run and confirm the two-part contrast; Stern 2012 (*Quest* 129) complains that no one before Hope checked Theobald's own plays — untrue of Graham, who used three.

### 2.9 E. H. C. Oliphant, "'Double Falsehood': Shakespeare, Fletcher and Theobald", *Notes and Queries* 12th ser. 5 (1919), 30–32, 60–62, 86–88 `[not accessible]`; *The Plays of Beaumont and Fletcher* (Yale UP, 1927), §39 pp. 282–302 `[DIRECT — oliphant-1927-bwb.txt lines 12542–13130]`
* **Method**: ear-based scene-by-scene assignment on "the imagery, the habit of thought, the diction, and the sentence-building"; acknowledges that "the verse's mechanism is more or less ruined … definite results are not to be looked for" (p. 291). Retracts his earlier verdict that DF contained "nothing that could have been written by Fletcher or Shakspere" ("I confess my fault, and retract", p. 295).
* **Assignment** (pp. 291–92, 295): "the hand of the eighteenth-century reviser is to be found in every scene, and … two scenes (II. 3 and 4) are wholly his." "Mr. Bradford was the first critic to note … that a new voice becomes audible in III. 3. It is the voice of Fletcher, who is thenceforward dominant. In this first scene we see his work overwritten by Theobald in the first nine speeches, though the editor's other work is limited to the provision of a closing couplet. The other Fletcherian portions of the play are that part of IV.1 preceding Julio's entry (the text having perhaps undergone some revision by Theobald); that part of the same scene between Violante's reëntry and the entry of Roderick; IV. 2, as far as 'And those to come shall sweetly sleep together' (with the exception of the song), the remainder of the scene showing Fletcher overwritten by Theobald; V. 1 (as revised by Theobald); and, also revised, V. 2, from 'Thou art a right one,' though as far as 'Duke. Weep not, child' has been left as originally written, the persistent Theobald, to atone for this, having left nothing whatever of Fletcher from 'Leon. The righteous powers at length have crown'd our loves' onward." The "firmer, stronger hand … Overlaid by Theobald's, this hand is to be seen in the first two speeches of the opening scene, in I. 2 (with the exception of ten speeches beginning 'Leon. What do you mean?'), in I. 3, in II. 1, in II. 2, in III. 1, in the first four speeches of III. 2 and in that portion of the scene between 'Scene opens to a large hall' and 'Most perjur'd if I do,' and in five speeches of IV. 1 immediately succeeding Julio's entry … As for the rest of I. 1 and 2, the balance of III. 2, and the remainder of IV. 1 to the reëntry of Violante, as well as the conclusion of that scene, they seem to be entirely the work of the reviser, though his portions of III. 2 and IV. 1 may possibly hold some fragments of the work of the older writer, from whom may perhaps also come the opening portion of the final scene. This writer I believe to have been Shakespeare; but in coming to a determination on the point one has to be guided largely by the external evidence and the probabilities, for the original work has been too heavily overlaid for the writer's touch to be clear and unmistakable."
* **Forgery**: "This suggestion [Fletcher] was indignantly denied by Theobald—a fact which … affords the strongest proof of his bona fides, as does also his obviously genuine belief that the work had never been staged." (p. 283)
* **Later use**: Jackson 2012 and Pascucci 2017 use Oliphant's "Shakespeare revised by Theobald" segments as their sampling frame; Taylor 2016 (n. 18): "since Oliphant (1927), attribution scholars have found the most concentrated evidence of Theobald's hand in Act Two."

### 2.10 Kenneth Muir, *Shakespeare as Collaborator* (Methuen, 1960), ch. "Cardenio", pp. 148–160; "Cervantes, Cardenio and Theobald", in *Spain and its Literature: Essays in Memory of E. Allison Peers* (Liverpool UP, 1997), 141–150 `[SECONDARY — archive.org copy lending-only (401); via Alotaibi pp. 29–30 and web summaries]`
* 1960: accepts the Bradford/Oliphant two-hands division and Shelton as source; argues Theobald's revisions fell hardest on Shakespeare's share. 1997 (via Alotaibi p. 29, quoting Muir): "Shakespeare's scenes were vandalized" while Fletcher's "were left relatively undamaged"; "the division of authorship Muir presents is in agreement with previous scholarship" (Alotaibi). No numbers recorded. Verbal-parallel lists in Muir 1960 are the basis for Hammond's commentary (Hammond 2010, p. 148 refers to the "range and scale of Shakespearean and Fletcherian allusion").

### 2.11 John Freehafer, "Cardenio, by Shakespeare and Fletcher", *PMLA* 84.3 (May 1969), 501–513, DOI 10.2307/1261138 `[ABSTRACT — verbatim from Cambridge Core; body via Alotaibi and NOS Companion]`
* **Abstract (verbatim)**: "Cardenio, a 'lost' play of 1612-13 based on Shelton's translation of Don Quixote (1612), is partly preserved, in altered form, in Lewis Theobald's Double Falshood of 1727. Theobald probably altered Cardenio drastically, as he did Richard II. To internal evidence that Cardenio was written by Shakespeare and Fletcher can be added chronological evidence and reluctant testimony to the joint authorship by the publisher Moseley, Charles Gildon, and Theobald. One of Theobald's three manuscripts can be traced in the hands of Moseley, the Davenants, Betterton, Gildon, Theobald's patron (Charles Boyle), and Theobald. A Restoration performance probably was intended. Theobald persisted in claiming that the original play was by Shakespeare alone after he learned otherwise, from Gildon and the Stationers' Register, to save his reputation as a Shakespeare scholar and please his patron. To protect his 'Shakespeare' play, Theobald obtained a unique royal license, which Pope ridiculed in the Dunciad. Theobald probably sold his play and Shakespeare manuscripts outright to his publisher. He was prevented from publishing Cardenio by the copyright Act of 1710, the practice of regarding copyright as perpetual, and his quest of money and patronage."
* **Key claims**: first proposal that the Downes manuscript was Davenant's Restoration adaptation (NOS Companion p. 585: "John Freehafer (1969) suggested that Theobald possessed, in the handwriting of Downes, William Davenant's adaptation of the play, and Hume (2016) agrees"). On the SR stop: p. 508 (see §1.2).
* **Critiques**: Fleissner (*Neuphilologische Mitteilungen* 1984?) doubts the "afterthought" reading (via Alotaibi p. 14); Stern 2011 attacks the whole manuscript-provenance chain as unevidenced (563–66); Hammond 2010 "pursued enthusiastically" the Davenant hypothesis (48–58) [WEB-SUMMARY].

### 2.12 Harriet C. Frazier, "The Rifling of Beauty's Stores: Theobald and Shakespeare", *Neuphilologische Mitteilungen* 69 (1968), 232–256; *A Babble of Ancestral Voices: Shakespeare, Cervantes and Theobald* (The Hague/Paris: Mouton, 1974), 161 pp. `[SECONDARY — archive.org copy lending-only; via Alotaibi pp. 3, 13 n. 13, and LPD]`
* Position: forgery — Theobald composed the play himself, borrowing Shakespearean phrasing ("rifling"). LPD: "The case for forgery … is upheld by Harriet C. Frazier, Jeffrey Kahan and, most fully, by Tiffany Stern." Alotaibi (p. 3) counts her, with Kahan, among those who "not only question Shakespeare's involvement in the play, but also suggest the possibility that Theobald actually forged it." Method: parallel-hunting between DF and Shakespeare's canon read as evidence of imitation rather than authorship. No scene assignments. Standard rebuttal (Hammond 2010; Taylor 2013; Jackson 2012): parallels to Fletcher and Shelton, which a Shakespeare-imitator had no reason to plant.

### 2.13 G. Harold Metz, *Four Plays Ascribed to Shakespeare: An Annotated Bibliography* (Garland, 1982; Routledge reissue 2019; Cardenio section pp. 119–131) and *Sources of Four Plays Ascribed to Shakespeare* (Univ. of Missouri Press, 1989; Cardenio chapter with reprint of Shelton) `[SECONDARY — not accessible; via Alotaibi p. 14 n. 45 (Metz 1989, p. 227) and web summaries]`
* Metz accepts DF as a genuine adaptation of the Shakespeare–Fletcher *Cardenio* with Shelton 1612 as source; treats the anomalies of Moseley's entry as products of its "overall hurried effect" (Alotaibi p. 14). Bibliographical rather than analytical; no new attribution data.

### 2.14 Stephan Kukowski, "The Hand of John Fletcher in Double Falsehood", *Shakespeare Survey* 43 (1991 [vol. dated 1990]), 81–89 (Cambridge Core gives 81–90), DOI 10.1017/CCOL0521395291.007 `[SECONDARY — Cambridge Core landing page only; via Alotaibi pp. 29–30 and Pascucci 2017 (NOS Companion p. 416)]`
* **Method**: metre, collocations and stylistic mannerisms compared against the Fletcher canon; explicitly a Fletcher study.
* **Findings** (as quoted by Alotaibi, p. 30, with Kukowski's page cited as 89): "the metre, the collocation of certain words, and the stylistic mannerisms of large parts of the play" are "distinctively Fletcherian". Pascucci (2017, p. 416) records that Kukowski (1991, p. 88) suggested that the Shakespearean residue in 3.2 "could well correspond to lines 39-43".
* **Position**: strengthens H2 (Fletcher's presence certain); agnostic-to-sceptical on Shakespeare (Alotaibi uses him for "Fletcher as the (sole?) author"). Hammond 2010 cites him as the modern basis for the Fletcher half.
* **Critiques**: none published; Stern 2012's "negative check" objection applies (no Theobald control).

### 2.15 Jonathan Hope, *The Authorship of Shakespeare's Plays: A Socio-linguistic Study* (Cambridge UP, 1994), pp. 89–100 `[NOT ACCESSED — archive.org item `authorshipofshak00hope` is lending-only (HTTP 401 on the full text); Google Books API returned 429 throughout; known only from search snippets and citations]`
* **Method** (from the book's design, chapters "The auxiliary 'do'" and "Relative markers"): rate of "regulated" v. "unregulated" auxiliary *do* (Fletcher's regulation rate is much higher than Shakespeare's), and relative-marker choice (*who/which/that*), on a two-section division of DF (section A = early part; section B = Fletcherian part). The comparison samples: Shakespeare, Fletcher, and — uniquely before 2012 — one Theobald play, *The Persian Princess* (Stern, *Quest* p. 129, via Alotaibi p. 92: "only Jonathan Hope had checked to see whether any of the habits being identified as Shakespearean or Fletcherian were also Theobaldian—and he limited his comparison to one 'pure' Theobald play, The Persian Princess").
* **Results actually recoverable**: (i) "section B shows a strikingly good fit to the Fletcher comparison sample" [search snippet quoting Hope; WEB-SUMMARY]; (ii) 'that'-relativisation in DF at a "low rate of 39 per cent", with Hope's remark that by Theobald's day *who* had become standardised and that Theobald when editing Shakespeare replaced *which* with *who*, so an increase in *who* is what adaptation would predict [search snippet, pp. 89–100; WEB-SUMMARY]; (iii) overall: Hope finds "strong evidence that Theobald's play is based on Elizabethan manuscripts and also finds evidence for a co-authored play", Fletcher's involvement "quite strong", Shakespeare's debatable (Wikipedia/poemshape paraphrase) [WEB-SUMMARY]. Pascucci 2017 (p. 407, my earlier reading) lists Hope 1994, 91–100 among the studies that retain the Shakespeare–Fletcher collaboration. **The actual auxiliary-*do* percentages for DF sections A/B were not obtained** — the book must be consulted physically or via a library e-copy.
* **Critiques**: Jackson 2012 (133–34, via Alotaibi p. 76) re-runs auxiliary *do*, double endings and *hath/doth* v. *has/does*; Alotaibi objects that all such counts are made on a text "subjected to more than one stage of adaptation" (p. 76). Poemshape (2010): Hope's study "cannot be used to favor Fletcher over Middleton" [WEB-SUMMARY].

### 2.16 A. Luis Pujante, "Double Falsehood and the Verbal Parallels with Shelton's Don Quixote", *Shakespeare Survey* 51 (1998), 95–105/106, DOI 10.1017/CCOL0521632250.008 `[SECONDARY — landing page; via Alotaibi p. 26 nn. 72–73]`
* **Method**: systematic verbal collation of DF with Shelton 1612 (extending Graham 1916/1920).
* **Finding**: many parallels are verbal, not merely narrative, and — the point Stern later exploits — "Shelton's translation could have been the immediate source for Double Falsehood (rather than for Cardenio) is a possibility we must not dismiss" (Pujante pp. 104, 102, as quoted by Alotaibi p. 26). Taylor 2012 (*Quest* 14–15) answers with a Shelton-only phrase in the play's *title* tradition ("The History of Cardenio" echoes Shelton's rendering of "el cuento de Cardenio" "and nowhere else", NOS Companion p. 583) and the NOS canon entry with the claim that Shelton was "a text not in Theobald's library catalogue, and which he never used elsewhere, or showed any awareness of" (p. 585) [DIRECT].

### 2.17 Ward E. Y. Elliott & Robert J. Valenza (Claremont Shakespeare Clinic), 1996–2010 `[DIRECT for the 2010 LLC preprint and Gray Scott's 2006 EMLS secondary analysis; 1996 and 2004 papers via summaries]`
* **Corpus/method**: 48 (earlier 51) "new-optics" tests on 3,000-word or whole-play blocks — function-word ratios, "BoB" (bundles of badges/flukes), hyphenated compounds, feminine endings, open/closed lines, grade-level, relative clauses, contractions, etc.; 29 core Shakespeare plays as baseline; a play "rejected" on a test if outside the Shakespeare profile; discrete rejections counted; composite probabilities computed.
* **1996** "And Then There Were None: Winnowing the Shakespeare Claimants" (*Computers and the Humanities* 30): 100% rejection of all claimant and apocryphal plays in combined tests [WEB-SUMMARY].
* **2004** "Oxford by the Numbers" (*Tennessee Law Review* 72, 323–453; DF discussed c. pp. 348–58) [not accessed].
* **2006** Gray Scott, "Signifying Nothing? A Secondary Analysis of the Claremont Authorship Debates", *EMLS* 12.2, Table 4 "Dubitanda and apocrypha" [DIRECT — web/emls-scotsig.txt]: *Double Falsehood* — Round 1 rejections 3, Round 2 9, Round 3 2, Total 14; "Non-Shakespeare Probability" 0.40725; "Shakespeare Probability" 0 (for comparison: *Sir Thomas More* 7 rejections, Sh-prob 0.000022; *Second Maiden's Tragedy* 25; *Mucedorus* 8).
* **2010** "Two Tough Nuts to Crack: Did Shakespeare Write the 'Shakespeare' Portions of Sir Thomas More and Edward III?", *Literary and Linguistic Computing* 25.1 (2010), 67–83 (preprint dated Aug. 2009) [DIRECT — elliott-valenza-2010-two-tough-nuts.txt], Table 2 "25 Shakespeare Apocrypha Plays Ranked by Shakespeare Rejections": **Double Falsehood (FALS): 11 discrete rejections; discrete composite probability 1.376E-09; continuous composite probability <1.0000E-15** (Shakespeare thresholds: ≤2 rejections; 2.316E-01; 3.6895E-03). Neighbours: *Sir Thomas More* 7, *Leir* 8, *Arden* 10, *Mucedorus/Oldcastle/Birth of Merlin/Merry Devil* 11, *Ironside* 12. Text (p. 6): "all of the 25 Apocrypha Plays we tested were on different statistical planets or galaxies (and so were all 51 plays by Shakespeare 'claimants' like Marlowe, Jonson, Middleton, and Fletcher). None had fewer than seven rejections." Which tests fail is not itemised for DF in the 2010 paper (it lists only counts).
* **Caveat by the authors themselves**: whole-play testing cannot detect a Shakespeare share in a collaborative or adapted text; CMC news release 22 April 2010 (Elliott): "Shakespeare's hand in the first half of Double Falsehood and four scenes from Arden of Faversham has not been ruled out" and the 2010 Clinic was to test "the old suspicion, argued in Brean Hammond's forthcoming Double Falsehood, that Double Falsehood is a reworked version of Shakespeare's [and John Fletcher's] long-lost Cardenio" [DIRECT — web/cmc-news…]. No published result of that half-play test was found.
* **Critiques**: Scott 2006 (EMLS) on Poisson modelling and false negatives; the anti-Stratfordian *Oxfordian* 12 (2010) "Auditing the Claremont Shakespeare Clinic" (not DF-specific) [DIRECT]; Stern 2012 (*Quest* 115–30) treats the 11–14 rejections as a mark against Shakespeare; Taylor and Jackson answer that the tests presuppose an unadapted whole play. Poemshape blog (2010) uses the "11 Discrete Rejections" as placing DF "far outside the realm of Shakespearean authorship" while noting E&V "did not separately analyze putative Shakespeare vs. Fletcher portions" [WEB-SUMMARY].

### 2.18 Jeffrey Kahan (ed.), *Shakespeare Imitations, Parodies and Forgeries, 1710–1820*, 3 vols (Routledge, 2004) `[SECONDARY — via Alotaibi p. 3 n. 13; LPD]`
* Prints DF among forgeries/imitations; Kahan's introduction treats it as Theobald's imitation of Shakespeare. No stylometric data. Counted by LPD and Alotaibi among the forgery school (with Frazier and Stern).

### 2.19 Brean Hammond (ed.), *Double Falsehood or The Distressed Lovers*, Arden Shakespeare Third Series (Methuen Drama, 2010), 443 pp. `[SECONDARY — archive.org copy lending-only; via Alotaibi (extensive quotation), Nicholl LRB 2010, Hamlet Weblog 2010, London Magazine, Wikipedia, Taylor 2016 and NOS Companion]`
* **Position** (general editors' preface, p. xvi): "this edition makes its own cautious case for Shakespeare's participation in the genesis of the play". Introduction p. 6: linguistic/stylistic analysis "lends support" to the detection of Shakespeare and Fletcher; p. 8: the edition "reinforces the accumulating consensus that the lost play has a continuing presence in its eighteenth-century great-grandchild" (Wikipedia citing pp. 6, 8) [SECONDARY].
* **Transmission model** (pp. 48–58, 95–96; Alotaibi p. 30 quoting Hammond pp. 53–4): between the original and Theobald "there was probably 'an intermediate version prepared in the Restoration'"; the text is a "palimpsest" or "pentimento" "that contains elements dating back to c.1611-12, elements dating to the mid-1660s, and elements first introduced in the mid-to-late 1720s" with "nothing that is straightforwardly Shakespeare-Fletcher". Davenant/Betterton/Downes chain (Freehafer's hypothesis "pursued enthusiastically", per search summary): Downes was prompter of Davenant's Duke's Company; Betterton and Davenant staged H8 (1663) and TNK as *The Rivals* (1664), so a *Cardenio* revival c.1665–67 is plausible (Nicholl) [DIRECT web page].
* **Scene-level view**: not a formal table. Nicholl (LRB 2 Dec 2010, reviewing the edition): "It is broadly agreed that Shakespeare's hand is more discernible in the first two acts of Double Falsehood, though we have no idea how the original collaboration with Fletcher might have worked, and this is not at all a watertight division, just a general consensus between old-style connoisseurs … and new-fangled stylometrists" [DIRECT]. Hamlet Weblog (June 2010), summarising Hammond's Nottingham vodcast: "Shakespeare has to have written a large proportion of the first half of the play" [DIRECT]. Search summary of the introduction: Shakespeare's voice "particularly in Act 1 but also in Act 2 and in the first couple of scenes of Act 3, while Fletcher appears to take over the writing in the second half" [WEB-SUMMARY].
* **Shelton parallels**: commentary claims "a full attempt to assess the range and scale of Shakespearean and Fletcherian allusion" (p. 148, via Alotaibi p. 29). Alotaibi (ch. 4) faults it for preferring Shakespearean to Theobaldian or Shelton parallels, e.g. at 4.2.38–39 Hammond cites *AYL* 2.1.3–4 and *Cymbeline* rather than the Shelton passage Graham had identified (p. 273) [SECONDARY]. Taylor 2016 (n. 13) disputes Hammond's claim (pp. 179, 232, 274) that "Leonora" can be trisyllabic.
* **On Stern**: Hammond 2010 wrote that Stern "built up a case convincing enough to render any editor of the play cautious"; in 'After Arden' (*Quest* 62–78, at 71–72) he re-rated her case as "the most sceptical response" but one "holding less weight than the comments in [his] introduction might suggest" (Alotaibi pp. 242–43) [SECONDARY].
* **Reviews/critiques**: Brian Vickers (2010/11): "the arguments claiming that Theobald's text preserves something of Shakespeare and Fletcher's original seem to me unconvincing" [WEB-SUMMARY, two independent hits]; Robert Folkenflik, "'Shakespearesque': The Arden Double Falsehood", *Huntington Library Quarterly* 75.1 (2012), 131–143 (review essay; not accessible — JSTOR); Charles Nicholl, "Cardenio's Ghost", *LRB* 32.23 (2 Dec 2010) [DIRECT — balanced; accepts the Restoration-intermediary model]; Kate Maltby (via Wikipedia) on over-selling; Stern 2011 (below) is the principal rebuttal; Proudfoot (Arden general editor, quoted in reviews): "the balance of doubt lies in favour of its claim being authentic rather than a total fabrication" [from my earlier notes; source page not re-verified].

### 2.20 Tiffany Stern, "'The Forgery of some modern Author'?: Theobald's Shakespeare and Cardenio's Double Falsehood", *Shakespeare Quarterly* 62.4 (Winter 2011), 555–593, DOI 10.1353/shq.2011.0074 `[SECONDARY — OUP/MUSE/academia.edu all blocked; known from Alotaibi 2016 (who adopts her position and cites pages), Taylor 2016, Hammond 2014 (abstract only), LPD, NOS Companion]`
* **Thesis**: DF is, or may well be, a forgery by Theobald — "an eighteenth-century forgery" in Tarlinskaja's words (NOS Companion 405) — and even if not, the documentary chain linking it to a Shakespeare–Fletcher *Cardenio* is unsound. Component claims, with her page numbers as cited by Alotaibi and LPD:
  1. 555–57: the 1613 payments are the sole hard evidence; the spellings "Cardenno/Cardenna" could theoretically denote the Spanish town Cardeña, so need not be Cervantes's story at all (LPD, citing 556–57).
  2. 557–58: Moseley's 1653 ascription is unreliable ("Moseley seems to take liberties with some names, particularly those of authors who could be published lucratively in quartos or en masse in large folios"); she suggests reading the entry's "F" as standing for Fletcher alone in a hurried list (my earlier notes: "'F for Fletcher', 557").
  3. 558: Fletcher's known Cervantes-based plotting habits (Love's Pilgrimage, The Chances, The Coxcomb, Knight of the Burning Pestle by Beaumont): *Cardenio*, "if it descends from Cervantes's Don Quixote (as its title suggests), mimics the particular plotting habits [Fletcher] displays elsewhere"; Fletcher "the more obviously plot-reliant of the two" and likely supplied the plot (Alotaibi p. 21).
  4. 563–66: "the most comprehensive reading of Theobald's manuscripts' account" (Alotaibi p. 9): Theobald claimed three, "possibly four", manuscripts; no one else ever saw them; the story of the natural daughter; the inconsistency between the Downes copy "of above sixty years standing" and a Betterton design to stage it; Theobald's failure to print the play in his 1733 *Works* or to produce the promised "Dissertation"; the 1770 *Gazetteer* notice and the 1808 fire as an unfalsifiable provenance. Alotaibi (p. 12) paraphrases her: had a Shakespeare manuscript sat in Covent Garden from before the 1770s to 1808 "it is barely conceivable that the great editors … Malone, Steevens, Johnson, and Capell, would not have gone to see it, or that … David Garrick would not have devoted his life to acquiring it."
  5. 576–79: Theobald as a practised imitator of Shakespeare (*The Cave of Poverty*, 1715, "an imitation of Shakespeare's early narrative style"; *Orestes* admitted in letters to Warburton to be Shakespeare-plagiary; *The Perfidious Brother* plagiarism dispute with Mestayer); hence the Shakespearean "feel" of DF proves nothing.
  6. 582: Shelton's 1612 translation was reprinted (1725) and available to Theobald.
  7. 588–89: Theobald's contradictory claims — that he wrote 1.3.10–14 himself, but that the derided 3.1.17 "was definitely Shakespeare's".
  8. 590: Farmer's "W.Sh." = William Shirley reading.
* **No scene assignments** (the article denies that authorial layers can be recovered).
* **Rebuttals** (the NOS Companion, p. 585, lists "more than a dozen scholars"): Taylor 2012a/b, Jackson 2012, Proudfoot 2012, King 2012, Carnegie 2012, Taylor & Nance 2012, Pascucci 2012, Taylor & Wagschal 2013, Baker 2013, Nance 2013, Spiller 2013, Taylor 2013, Tarlinskaja 2014 (203–11), Hammond 2014, Boyd & Pennebaker 2015, Hume 2016, Marsden 2016, Taylor 2015/2016, Solomon 2016; Kirwan 2015 (177) "also rejects Stern's case for forgery" and explains the 1733 omission as "a defensively 'safe' move designed to preserve his editorial 'reputation' against further attacks by Pope" [DIRECT]. Taylor 2016 (353–54) on the UCLA symposium of 31 Jan–1 Feb 2014 (Hume, Folkenflik, Marsden, Payne, Solomon, Pennebaker, Hammond): "none of the speakers endorsed Stern's argument" and "the accusation of forgery no longer needs to be addressed" [DIRECT]. **Supporters**: Alotaibi 2016/2021 (Fletcher-only source rather than pure forgery); Vickers (unconvinced by Shakespeare); Kahan; Frazier; poemshape.

### 2.21 Tiffany Stern, "'Whether one did Contrive, the Other Write, / Or one Fram'd the Plot, the Other did Indite': Fletcher and Theobald as Collaborative Writers", *Quest* (2012), 115–130 `[SECONDARY — via Alotaibi pp. 92, 129, 245; RenRef review]`
* Argues the stylometric case is circular: "until [The Quest for Cardenio], only Jonathan Hope had checked to see whether any of the habits being identified as Shakespearean or Fletcherian were also Theobaldian—and he limited his comparison to one 'pure' Theobald play, The Persian Princess … Yet most of the 'unique' hallmarks of Shakespeare/Fletcher identified by other stylometricians can be found throughout Theobald's oeuvre and thus negate themselves as 'proof' of anything" (p. 129, via Alotaibi p. 92). Insists on M. St Clare Byrne's (1932) "negative check". The RenRef reviewer: Stern "offers a compelling critique of stylometric analyses … noting that the styles of other authorship candidates—Massinger and Shirley, for example—have not been addressed adequately" [DIRECT]. Alternative model: Theobald "collaborating" with a Fletcher (or Fletcher-and-other) manuscript.

### 2.22 *Shakespeare* 7.3 (2011), Cardenio cluster `[ABSTRACT/citation level only]`
Richard Meek & Jane Rickard, "'This orphan play': Cardenio and the construction of the author", 269–283; Lori Leigh, "'Tis no such killing matter': Rape in Fletcher and Shakespeare's Cardenio and in Lewis Theobald's Double Falsehood", 284–296; Peter Kirwan, "Chasing Windmills: An Identity Crisis in Double Falsehood at the Union Theatre, Southwark", 329–334, DOI 10.1080/17450918.2011.589063 (WRAP abstract [DIRECT]: "considers the relationship between the lost play Cardenio … and the surviving eighteenth-century play Double Falsehood supposedly derived from it … At stake are the ways in which we apportion creative agency"); David Carnegie 344–351; Ivan Lupić 369–376. Also Bernard Richards, "Cardenio", *Essays in Criticism* 61.1 (2011), 79–88 (on his own completion). None adds attribution data; Meek & Rickard (p. 270) are cited by Alotaibi as treating "authorship" as a construct.

### 2.23 David Carnegie & Gary Taylor (eds), *The Quest for Cardenio: Shakespeare, Fletcher, Cervantes, and the Lost Play* (OUP, 2012), xiv + 420 pp. `[SECONDARY — every chapter page blocked by OUP Cloudflare; chapter pagination from CrossRef; contents via Alotaibi, RenRef 36.3 review (DIRECT), Taylor 2016, NOS Companion]`
* **Gary Taylor, "A History of The History of Cardenio"** (11–61) and "The Embassy, the City, the Court, the Text: Cardenio Performed in 1613" (283–299 with Carnegie's chronology 283–85): dating of the court performance to February 1613 (NOS 584: "it is unlikely that Cardenio could have been performed for the grieving King … [before] February 1613 (Taylor 2012b, 293-9)"); the Shelton-only title phrase (14–15); Edmund Gayton's *Pleasant Notes upon Don Quixot* (1654) as possible witness (33–36); "Lute sounds within" stage direction shared with Fletcher's *Chances* (45); Restoration adaptation (40–44 on 2.1 as Theobald/Davenant material, per Taylor 2016 n. 18) [SECONDARY].
* **Edmund G. C. King, "Cardenio and the Eighteenth-Century Shakespeare Canon"** (81–94); **Ivan Lupić, "Malone's Double Falsehood"** (95–114) — see §2.3.
* **Stern** (115–130) — §2.21.
* **MacDonald P. Jackson, "Looking for Shakespeare in Double Falsehood: Stylistic Evidence"** (131–161). OUP abstract [WEB-SUMMARY]: "assesses stylistic evidence for supposing that Lewis Theobald's Double Falsehood had its basis in at least one manuscript descended from the Cardenio known to have been performed by the King's Men in 1613, and for crediting Humphrey Moseley's 1653 attribution of Cardenio to Fletcher and Shakespeare." **Method**: LION phrase-and-collocation searches over 1576–1642 drama plus Theobald's works; auxiliary *do*, double endings, *hath/doth* v. *has/does* (133–34, 144); three-word-sequence tests. **Corpus** (Alotaibi p. 77 quoting Jackson): "fifty plays by Fletcher (of which fifteen were solely his), thirty-nine by Shakespeare, yet only fifteen works by Theobald, 'most of [which] are very short entertainments'" — "Fletcher's output, including collaborations, is more than seven times the size of Theobald's". **Results** (via Alotaibi pp. 30–31, 79–80): 'o' my conscience' (3.3) — never Shakespeare or Theobald, twice in Fletcher's H8 (2.1.50, 3.1.30) and TNK 2.4.12; "'Stand off' (DF 4.2, p. 48) 'appears thirty times in plays of the extended Fletcher canon (including once in a Fletcher scene of Two Noble Kinsmen) but never in a Shakespeare play'"; 'all the miseries' (4.2) 7 Fletcher : 1 Shakespeare; 'as I have a soul' (5.1) 10× in Fletcher's unaided plays and non-Shakespearean collaborations, once in H8 4.1.44; Theobald-only parallels unmatched in LION: 'fair appeal' (DF p. 54; *Persian Princess* 5.137–9; *Richard II* preface) and one other; Beaumont check — "19 links to Beaumont and 328 to Shakespeare" in Oliphant's "Shakespeare revised by Theobald" segments, and, single-play v. single-play, "thirty-four verbal links to Beaumont [Knight of the Burning Pestle] and forty-eight to Shakespeare [Coriolanus]" ("the sequences are trivial in the extreme … at least they were objectively identified"), so Beaumont is excluded (159–61). **Conclusion** (161, quoted by RenRef review and Alotaibi p. 79): Theobald "probably" worked from a Shakespeare–Fletcher *Cardenio* "manuscript (or manuscripts)", but "scarcely a line of Shakespeare's verse survives intact into Double Falsehood"; the prose is more intact than the verse (Taylor 2016, 358). Jackson also endorses Muir's point that Shakespeare's "complex, and packed dramatic verse was more apt to be discarded or recast than Fletcher's" and that a Downes/Betterton/Davenant script "might already in the 1660s have been cut and altered" (Alotaibi p. 77). **Critiques**: Alotaibi (pp. 76–80): counts made on an adapted text; canon-size imbalance; parallels sought only inside Oliphant's segments. Jackson himself (NOS Companion ch. 8, 'A Review of Reviews', p. 123): the Cardenio case "has been hugely strengthened (Taylor 2013b) in response to elaborate expressions of scepticism (Stern 2011)".
* **Richard Proudfoot, "Can Double Falsehood Be Merely a Forgery by Lewis Theobald?"** (162–179). **Method** (Alotaibi pp. 72–74 quoting Proudfoot): line-end polysyllables (words of three or more syllables ending a pentameter — rare enough to yield "manageably small quantities of material"); the 100-odd polysyllabic line-endings of DF matched against line-end polysyllables in samples of Fletcher (15 plays), Shakespeare (13) and Theobald (9 works). **Results**: matching rates — Theobald 32%, Fletcher 65%, Shakespeare 63%, both Jacobeans combined 79%; Theobald's "was clearly the minor role"; Proudfoot concedes the Sh/Fl rates "may well be somewhat inflated in comparison to Theobald['s], as it covers a range of thirty plays against [Theobald's nine]"; Massinger's early plays show "high levels (over 20 per cent) of matching" but "perhaps [Massinger's] close connections and indebtedness to Fletcher have something to do with this"; other contemporaries (Beaumont, Chapman, Day, Dekker, Field, Middleton) below 20% ("unfortunately he does not provide the data behind these conclusions" — Alotaibi p. 78). Stated aim: "to dispel the notion that DF in its entirety could possibly be a forgery by Theobald (or any other writer of the early eighteenth century)"; the feature is one "most unlikely to occupy the attention of a forger". Theobald's line-endings concentrated in 2.3 (Proudfoot 173–74, per Taylor 2016 n. 18). RenRef review: "the play's style bears traces of three authors: Shakespeare, Fletcher, and Theobald." Search summary of his rhetorical question: if Theobald forged it, "why on earth did he leave in the forged play significant linguistic evidence for scholars … that there's Fletcher's writing in the play and Shakespeare?" [WEB-SUMMARY].
* **David Carnegie, "Theobald's Pattern of Adaptation: The Duchess of Malfi and Richard II"** (180–191): Theobald's *Fatal Secret* (1733) and *Richard II* (1720) as models of how he cut, regularised and moralised — used by everyone since as the control for what adaptation (not forgery) looks like; also "Chronology" (283–85).
* **Taylor & John V. Nance, "Four Characters in Search of a Subplot: Quixote, Sancho, and Cardenio"** (192–213; Taylor 2016 n. 18 cites 198–212 on 2.1): identifies Theobaldisms, notably 'at present' — "does not occur in the works of Shakespeare, Fletcher, Beaumont or Massinger, nor 'in any extant English play before 1638'" but "thirty-six times" in Theobald (Alotaibi p. 84, who adds further Theobald instances and accepts the attribution of DF pp. 13 and 23 to Theobald); also 'brutal' violence collocations (my earlier notes).
* **Brean Hammond, "After Arden"** (62–78) — see §2.19.
* **Huw Griffiths** (239–255), **Lori Leigh** (256–266), **Roger Chartier** (309–317; on Guillén de Castro's *Don Quixote de la Mancha* and Pichou's *Folies de Cardenio* as continental Cardenios), **Bernard Richards** (344–351 on his Cambridge completion), **Peter Kirwan** (356–359, reviews of two productions), **Gregory Doran** (on preparing the RSC *Cardenio*); Terri Bourus on staging Taylor's unadaptation. RenRef reviewer (36.3, 2013, 174–76) [DIRECT]: the volume's three of five sections "focus on the authorship question" and "more or less rescue Theobald from the charge of forgery levelled at him by Alexander Pope".

### 2.24 Giuliano Pascucci, "Double Falsehood/Cardenio: A Case of Authorship Attribution with Computer-Based Tools", *Memoria di Shakespeare* 8 (2012) `[SECONDARY — no copy located; described in Taylor 2016 (353) as "independent analysis of data-compression (Pascucci 2012)" that "disproved" some of Stern's claims; superseded by his 2017 chapter, §2.34(b)]`

### 2.25 Terri Bourus & Gary Taylor (eds), *The Creation and Re-Creation of Cardenio: Performing Shakespeare, Transforming Cervantes* (Palgrave Macmillan, 2013) `[SECONDARY — Springer chapter pages loop to an IdP cookie page; contents from CrossRef, Taylor 2016 and NOS Companion]`
* Taylor & Steven Wagschal, "Reading Cervantes, or Shelton, or Phillips? The Source(s) of Cardenio and Double Falsehood" (15–29): the play depends on Shelton specifically (not Phillips 1687 or Motteux) — a source Theobald did not own or otherwise use.
* David L. Gants, "The Printing, Proofing and Press-Correction of Shelton's 1612 Don Quixote" (31–46): Shelton was on sale by spring 1612 (p. 43, my earlier notes), fixing the earliest date for the play; NOS canon entry "Original Date Range: June 1612–January 1613; Best Guess: second half of 1612".
* John V. Nance, "Shakespeare, Theobald, and the Prose Problem in Double Falsehood" (109–123) [ABSTRACT via search summary]: examines the prose of 1.2.179–224 — "just 36 lines and 334 words" — and finds that "an analysis of the distinctive dramatic function and vocabulary of these prose lines strongly suggests the presence of Shakespeare more than any other author"; presented as evidence against Stern's forgery claim; (my earlier notes: attribution to Shakespeare stated at p. 117).
* Gary Taylor, "Sleight of Mind: Cognitive Illusions and Shakespearian Desire" (125–169): tests Theobald's *Cave of Poverty* (his one avowed Shakespeare imitation) with the same LION method — the poem "reeks of Theobald" (2 Shakespeare v. 8 Theobald unique parallels in its opening; table 10.1, pp. 134–36 — my earlier notes); DF 1.1's first eight lines assigned to Shakespeare (p. 137); at 157–61 shows Theobald "utterly incapable of imitating Shakespeare with anything remotely resembling this level of concentrated linguistic similarity" (Taylor 2016, 357, citing Taylor 2013, 157–161) [DIRECT for the 2016 citation].
* Lori Leigh (171–184); Gary Taylor, *The History of Cardenio, 1612–2012* (text, 241–316): the "unadaptation" — Theobald's text with Theobaldian passages removed and gaps supplied in imitation of Shakespeare/Fletcher from Shelton; the NOS canon entry cross-refers to this as Taylor 2013b.
* Gerald Baker 2013 and Elizabeth Spiller 2013 (in the same volume) are listed by NOS among the refutations of Stern.

### 2.26 Will Sharpe, "Authorship and Attribution", in Jonathan Bate & Eric Rasmussen (eds), *William Shakespeare and Others: Collaborative Plays* (RSC Shakespeare/Palgrave, 2013) `[WEB-SUMMARY only]`
* Ninety-page essay giving "a clear and measured judgement about the claims of each play"; the volume prints DF among ten plays. My earlier search notes record Sharpe's verdict on DF as ranging from "almost certainly" (a Fletcher–Shakespeare *Cardenio* underlies it) to "very likely" (that some Shakespeare survives) — the exact phrasing was not re-verified in this pass. Michael Billington on the RSC production: "more suggestive of Fletcher than Shakespeare" (Wikipedia, History of Cardenio) [DIRECT web page].

### 2.27 Roger Chartier, *Cardenio between Cervantes and Shakespeare: The Story of a Lost Play* (Polity, 2013; French 2011); "Mythical ghost…", *Cahiers Élisabéthains* 90 (2016), 141–150 `[WEB-SUMMARY; uchicago MP review blocked]`
* Chartier "does not try to answer the authorship questions definitively, although he thinks it unlikely that the play was an out-and-out fraud"; interested in "a paradoxical early modern relationship between collaboration, adaptation, authorship, and reading". No attribution data. Contributes the continental analogues (Guillén de Castro; Pichou).

### 2.28 Brean Hammond, "Double Falsehood: The Forgery Hypothesis, the 'Charles Dickson' Enigma and a 'Stern' Rejoinder", *Shakespeare Survey* 67 (2014), 165–179, DOI 10.1017/SSO9781107775572.012 `[SECONDARY — Cambridge Core "A summary is not available"; via Alotaibi p. 243 and Taylor 2016]`
* A paper from the UCLA symposium (Taylor 2016, 354). Refutes "individual strands of arguments pursued in [Stern's] articles" (Alotaibi, who complains he does not address "her most convincing observations"). The "Charles Dickson" enigma concerns a manuscript annotation/ownership puzzle raised by Stern (content not recoverable from what I could access — flag).

### 2.29 Robert Folkenflik, "'Shakespearesque': The Arden Double Falsehood", *HLQ* 75.1 (2012), 131–143 `[not accessible — JSTOR; ResearchGate 403]`. UCLA symposium co-organiser; Taylor 2016 counts his paper among the "devastating rebuttals" of Stern.

### 2.30 Ryan L. Boyd & James W. Pennebaker, "Did Shakespeare Write Double Falsehood? Identifying Individuals by Creating Psychological Signatures With Text Analysis", *Psychological Science* 26.5 (May 2015), 570–582, DOI 10.1177/0956797614566658, PMID 25854277 `[ABSTRACT verbatim via NCBI efetch (boyd-pennebaker-2015-pubmed-abstract.txt); body via UT Austin press release (DIRECT), Notes in the Margin (DIRECT), Laurie Johnson in *The Conversation* 14 May 2015 (DIRECT); SAGE 403]`
* **Abstract (verbatim)**: "More than 100 years after Shakespeare's death, Lewis Theobald published Double Falsehood, a play supposedly sourced from a lost play by Shakespeare and John Fletcher. Since its release, scholars have attempted to determine its true authorship. Using new approaches to language and psychological analysis, we examined Double Falsehood and the works of Theobald, Shakespeare, and Fletcher. Specifically, we created a psychological signature from each author's language and statistically compared the features of each signature with those of Double Falsehood's signature. Multiple analytic approaches converged in suggesting that Double Falsehood's psychological style and content architecture predominantly resemble those of Shakespeare, showing some similarity with Fletcher's signature and only traces of Theobald's. Closer inspection revealed that Shakespeare's influence is most apparent early in the play, whereas Fletcher's is most apparent in later acts. Double Falsehood has a psychological signature consistent with that expected to be present in the long-lost play The History of Cardenio, cowritten by Shakespeare and Fletcher."
* **Corpus**: 33 Shakespeare plays, 12 Theobald works, 9 Fletcher plays (UT press release). **Method**: LIWC (Linguistic Inquiry and Word Count) function-word categories (pronouns, articles, prepositions…) and psychological content categories (emotion, family, sensory, religion); the Meaning Extraction Method (factor analysis of content words) for "content architecture"; three classification approaches; "all three models designating Shakespeare as the best authorial candidate … when considering the eight general function-word classes" (search summary quoting the paper) [WEB-SUMMARY]. Personality inference: the author was "probably sociable and well educated — findings that don't jibe with accounts of Theobald as well educated but also rigid and abrasive" (press release).
* **Act-level result**: "Boyd and Pennebaker identified Shakespeare as the likely author for the first three acts; the measures varied between Shakespeare and Fletcher for the fourth and fifth acts" (press release) [DIRECT]. Johnson's tally (*Conversation*): "Five types of word classes or categories were tested, each using three different computational methods (so there were 15 different tests for each play) … In 39 out of 45 tests on the first three Acts, Shakespeare comes out on top as the closest match. In the last two Acts, though, Shakespeare is closest on only 11 out of the 30 tests." Theobald's contribution "very minor" (Wikipedia quoting the paper).
* **Critiques**: Laurie Johnson (Univ. of Southern Queensland), *The Conversation*, 14 May 2015: "we cannot independently verify that the psychological signature they attribute to Shakespeare is in fact his — we simply don't know enough about him outside of his writing"; two unknowns each used to prove the other; the results "varied" across acts [DIRECT]. General stylometric objections (not DF-specific, but applicable): act-level rather than scene-level units mix Fletcher's 3.3 with Shakespeare's 3.1–3.2; no Davenant or Massinger candidate; no control for adaptation; unequal corpora. **Replications**: none found. Boyd's later Behn paper (*PLOS ONE* 2018, 10.1371/journal.pone.0200588) cites the DF study for "near-100% accuracy" author discrimination but does not re-analyse DF [DIRECT]; the Frontiers 2018 RPAS paper (10.3389/fpsyg.2018.00289) does not analyse DF [DIRECT]. Taylor 2016 and NOS Companion cite it as one more refutation of forgery; Tarlinskaja/Pascucci implicitly contradict its "Shakespeare throughout Acts 1–3" by finding Davenant/Theobald layers there.

### 2.31 Gary Taylor, "Fake Shakespeare", *Journal of Early Modern Studies* 5 (2016), 353–379, DOI 10.13128/JEMS-2279-7149-18095 `[DIRECT — taylor-2016-fake-shakespeare-jems5.pdf/.txt]`
* **Position** (p. 353): "Lewis Theobald's Double Falsehood is a Georgian adaptation of a Jacobean play by Fletcher and Shakespeare. It is not a forgery, as Tiffany Stern (2011) contends." (354): "I will therefore assume, in this essay, that the accusation of forgery no longer needs to be addressed."
* **Method**: "comprehensive search of digital databases" (LION drama 1576–1642 plus Theobald's works; appendix pp. 371–79 gives the language data; asterisked items unparalleled in LION) applied to three Leonora speeches, plus theatrical/dramaturgical argument (Wilks's and Mrs Porter's parts).
* **Passage-level attributions**: (a) 5.2.251–257 ("The righteous Pow'rs at length have crown'd our Loves…") → Theobald: Fletcher never collocates 'righteous' and 'power(s)'; the adjective appears sixteen times in Theobald's smaller canon, ten times of deities (Appendix A); 'Julio' in the added line — "The name 'Julio' occurs another twenty times in Leonora's speeches, and all those twenty lines must be either (a) written in their entirety, like this one, by Theobald, or (b) rewritten by Theobald to accommodate the changed name" (p. 362); "With 'Cardenio' we can be certain, and with 'Fernando' reasonably confident, of the original names." (b) 5.2.94–101 ("For such sad Rites must be perform'd…") → Fletcher: "Anyone who has read and studied all of John Fletcher's work will recognize this as Fletcherian (and scholars have done so for a century) … Fletcher, not Theobald or Shakespeare, overwhelmingly dominates the language of this speech (Appendix B)." (p. 356) (c) the Julio–Leonora exchange "No Impediment / Shall bar my Wishes … My Flames are in the Flint … Let Freedom quit me" (cited "5.2.109-116" but, as the text makes clear, from "the encounter of Julio … and Leonora … in the middle of 1.2") → Shakespeare: "Shakespeare, not Theobald or Fletcher, overwhelmingly dominates the language of these lines (Appendix C) … no other early modern playwright comes anywhere near the number of unique links between this passage and Shakespeare." (p. 357). (d) 5.1.25–37 (Henriquez–Leonora): "Leonora's entire encounter with Henriquez in 5.1 is Theobald's melodramatic addition to the scene" (p. 365) — 'I throw me at your feet' (5.1.26) "appears nowhere in English [drama before 1642]", 'Rise, rise' (5.1.36) Theobaldian; (e) 4.1.212–257: Henriquez's presence "theatrically awkward and his speeches reek of Theobald"; (f) "Much of his soliloquy in 2.1, and of his speeches in 2.3, is also Theobald's" (n. 18: "since Oliphant (1927), attribution scholars have found the most concentrated evidence of Theobald's hand in Act Two") (p. 365).
* **On reconstructions** (pp. 357–58): Greenblatt & Mee's *Cardenio* (2008) preserves "only about sixty-three of the lines published by Theobald … The wording has been changed in eleven of those lines (21%) … ruins the meter of seventeen lines (27%)"; "less than four percent of Double Falsehood"; includes Theobaldisms ('let the gay scene', 'by proxy', 'her charms', 'love is contagious') and skips the Shakespearean 1.2 exchange; Doran "cannot distinguish Theobald from Shakespeare" (having praised the Theobald speech as El Greco-like); "Jackson (2012) and Nance (2013) have demonstrated that the prose is much more authentically Jacobean and Shakespearian than the verse."
* **Critiques**: Alotaibi 2016 (ch. 3–4) re-searches Taylor's parallels in ECCO/LION and finds many "Shakespearean" phrases also in Theobald; argues Taylor cites only Theobald's plays on LION, ignoring his prose (*Censor*, letters), and that the *Cave of Poverty* test is biased. Vickers (unconvinced, general). No published stylometric replication of the appendix data.

### 2.32 Naseem A. Alotaibi, *Lewis Theobald's Double Falsehood: The Authorship Question Reconsidered* (PhD, Liverpool, July 2016), 300+ pp. `[DIRECT — alotaibi-2016-liverpool-phd-double-falsehood.pdf/.txt]`; "'[T]hat Other Great Poet': Double Falsehood, the Arden Shakespeare, and Further Evidence for Fletcher", *Notes and Queries* 68.2 (2021), 184–189 `[ABSTRACT via search]`
* **Position**: the Stern line, modified: "there is little reason to believe that Shakespeare was ever involved in the lost Cardenio", while Fletcher is "possibly its prime (if not its sole) author" (N&Q abstract); Theobald's own contribution is much larger than the Arden edition allows; the Arden "has essentially taken Theobald's word for it".
* **Method**: negative-check parallel hunting (Byrne 1932) in LION + ECCO over the *whole* Theobald corpus (plays, *Censor*, translations, letters — abbreviations listed pp. viii–ix); comparison of DF with Theobald's admitted adaptations of *Richard II* (1720) and *The Duchess of Malfi* (*The Fatal Secret*); reassessment of every parallel in Hammond's commentary.
* **Results**: e.g. 'at present' occurrences in Theobald exceed Taylor's 36; 'fair appeal' and other Theobald-only phrases; Hammond's Shakespearean glosses often have closer Theobald or Shelton parallels (4.2.38–39); Proudfoot's corpus imbalance (15 Fl / 13 Sh / 9 Th) and Jackson's (50/39/15) flagged; "the study considers…" Stern's provenance objections restated (pp. 9–26); Malone/Steevens history (pp. 66–67).
* **Value**: the most thorough hostile audit of the pro-Shakespeare parallels, and the only modern full-length case for H2; also the most complete bibliography (Oliphant's 1919 N&Q articles, Fleissner 1984, Muir 1997, Kukowski, Pujante, Lupić, Stern 2012).

### 2.33 Deborah C. Payne (ed.), *Revisiting Shakespeare's Lost Play: Cardenio/Double Falsehood in the Eighteenth Century* (Palgrave Macmillan, Dec. 2016; © 2017), 138 pp. `[ABSTRACT — Hume's chapter abstract verbatim from Springer; TOC from Google Books]`
* Contents: Fiona Ritchie, Introduction (1–6); **Robert D. Hume, "Believers versus Skeptics: An Assessment of the Cardenio/Double Falsehood Problem"** (7–56), DOI 10.1007/978-3-319-46514-2_2; Diana Solomon (57–73, eighteenth-century rape culture); Jean I. Marsden (75–92, adaptation practice); Deborah C. Payne (93–124, DF versus the Scriblerians).
* **Hume abstract (verbatim)**: "This chapter maintains that 'the Cardenio problem' raises interesting questions about the handling of scanty, contradictory, and often inadequate evidence. Shakespeare apparently collaborated with Fletcher on the original 1613 play (lost). An adaptation (by Davenant?) from circa 1667 is also lost. Theobald's later adaptation, entitled Double Falsehood (1728), apparently combines generic features and language from both the Jacobean period and the 1720s. Quantitative stylistics suggest the presence of three authorial hands. While I conclude that Double Falsehood is an adaptation—longstanding charges of forgery ignore facts and historical circumstances—the play contains little or no unadulterated Shakespeare." NOS Companion (585) quotes Hume: "we have no grounds for asserting that it was not performed circa 1665 or 1667. All we can say with assurance is that no evidence of performance is known to survive" and notes Hume 1976 (240–2) on the 1663–68 vogue for Spanish romance after Tuke's *Adventures of Five Hours*.
* **Significance**: the leading theatre historian's verdict — anti-forgery but minimal-Shakespeare — is the best short statement of the current "sober" consensus.

### 2.34 *The New Oxford Shakespeare: Authorship Companion*, ed. Gary Taylor & Gabriel Egan (OUP, 2017) `[DIRECT — nos-authorship-companion-2017-archiveorg.txt (OCR)]`
**(a) Marina Tarlinskaja, ch. 23, "The Versification of Double Falsehood Compared to Restoration and Early Classical Adaptations", pp. 385–406.**
* **Method**: stress-profile analysis (which syllable positions lose stress: Elizabethan/Restoration/18th-c. dip on 6 v. Jacobean–Caroline dip on 8); location of strong syntactic breaks (after 4 v. 6/7); enclitic phrases and heavy feminine endings per 1,000 lines; unstressed grammatical monosyllables in position 10; "rhythmical italics"; pleonastic *do* and grammatical inversions; disyllabic *-ion*. Controls: Shakespeare (MM, WT, Mac, R2, H8/TNK shares), Fletcher (Valentinian, TNK/H8 shares), Davenant (*Unfortunate Lovers*, *Law Against Lovers*, *Macbeth*, *The Rivals*), Tate (*Sicilian Usurper*, *Injured Love*, *Brutus of Alba*), Theobald (*Persian Princess*, *Orestes*, *Richard II*, *Fatal Secret*), Webster (*Malfi*).
* **Figures** (pp. 402–405): pleonastic *do* "4.3 and 4.7 per 1,000 lines in Theobald's own compositions The Persian Princess … and Orestes … but at the rate of 16.2 per 1,000 lines in the first act of Double Falsehood" (Tate's and Theobald's *Richard II* adaptations 25 and 27.7; Tate's own *Brutus of Alba* 1.9): "On this feature, then, the first act of Double Falsehood looks like adaptation of Shakespeare." Disyllabic *-ion* at 1.3.7, 3.3.4, 3.3.29 ("Mended with strong imagination"; "Have trusted me with strong Suspicions"; "Bought my poor Boy out of Possession"). Rhythmical italics in Act 1: 133.6 per 1,000 lines — "much higher than found in Davenant's, Tate's, or Theobald's own compositions or adaptations, but typical of Shakespeare's writing" (MM 126.2, WT 120.4, H8-Sh 168.8, TNK-Sh 106.3; Davenant's *Macbeth* 65.7 v. Shakespeare's 194.7; *Law Against Lovers* 46.1). Stressing: "In Act 2 and Act 4 the dip in stressing on positions 6 and 8 is almost equal, as in Davenant's adaptation of Shakespeare's Macbeth and the non-Fletcher-derived parts of … The Rivals … In Act 3 and Act 5 … the strongest dip in stressing is on syllable 8, which is what we find in Jacobean drama generally." Enclitics: "In Double Falsehood Acts 1 and 2 and in scenes 3.1 and 3.2 there are 36.4, 33.2, and 52.9 enclitics per 1,000 lines, respectively. In Double Falsehood 3.3, the rate jumps to 200 (close to Fletcher's rate in his play Valentinian, 247) and it stays high to the end of the play." Comparators: Webster *Malfi* 127.6 → Theobald's *Fatal Secret* 41.7; TNK Fletcher 290.1 / Shakespeare 89.2 → *Rivals* Fletcher-derived 179.3 / other 58.3; R2 52.5 → Tate 41.4 / Theobald 32.2. "The low ratio of enclitic phrases in Act 1 and Act 2 and scenes 3.1 and 3.2 … (36.4 to 52.9) might, then, be a sign of Davenant adapting Shakespeare". Unstressed monosyllables at 10: 3.1–3.2 2.4% ("They might be a trace of late Shakespeare, or perhaps a sign of Davenant's hand"; Davenant's *Macbeth* 3.0%, *Rivals* Fletcher-derived 0.5% v. other 3.7%; Theobald "virtually none"). Syntactic breaks: positions 4 and 6 "about equal … much like Davenant's habit when adapting Fletcherian rhythms". Symmetrical-structure lines with dip on 6 (18th-c. type): numerous in Acts 1, 2, 3.1–2 (e.g. 2.3.94) "but just one case in the remainder of the play". Disyllabic *-ion* "13.3 such disyllabic suffixes per 1,000 lines in Double Falsehood 3.3 is typical of Fletcher, whose share of All Is True/Henry VIII has 13.9 per 1,000 lines. This is a strong indication of a Jacobean original".
* **Conclusion (405–406, verbatim)**: "Theobald's claim has been in large part justified: he had indeed worked from an existing older script." "A counterfeit Shakespeare would not contain signs of an eighteenth-century rewrite in the first half of the play and signs of Fletcher in the second half. If it were an eighteenth-century forgery, as Stern thinks, its creator was remarkably skilful in thinking to use enclitic phrases and heavy feminine endings at all, and impossibly subtle to think to put them in just half of the play, the Fletcher scenes." "Thus, a seventeenth-century manuscript was indeed in Theobald's possession. It was probably based on the lost Cardenio, a collaboration. From what we know of Shakespeare–Fletcher collaborations, one of the co-authors of the lost play was clearly Fletcher. The other is harder to identify solely on internal evidence. There are few signs of Shakespearean verse style in Acts 1 and 2, but his was in any case a style of versification—with its run-on lines and light feminine endings—that was easy for either Davenant or Theobald to rewrite. There are numerous signs of Fletcher's idiosyncratic style. There are signs of an eighteenth-century author in Acts 1 and 2 and scenes 3.1 and 3.2. The versification suggests that what Theobald had in his possession was not the original Jacobean manuscript but a later, post-Restoration rewrite, possibly by Davenant. Thus Double Falsehood seems to contain three layers from different times: Shakespeare–Fletcher, Davenant, and Theobald. The original text was probably the lost Cardenio."
* Earlier version: Tarlinskaja, *Shakespeare and the Versification of English Drama, 1561–1642* (Ashgate, 2014), 203–11 (cited by NOS as a refutation of Stern); later: "Post-Restoration Adaptations…" in *Medieval and Renaissance Drama in England* 36 (2023), 285–312 [not accessible; search summary attributes to her the dating of the Davenant layer to "1670–1680" and the finding of "few traces of Shakespeare's versification but definite signs of Fletcher's" — WEB-SUMMARY].

**(b) Giuliano Pascucci, ch. 24, "Using Compressibility as a Proxy for Shannon Entropy in the Analysis of Double Falsehood", pp. 407–416.**
* **Method**: relative entropy via a modified LZ77 compressor (Benedetto–Caglioti–Loreto): each DF scene appended to control sections of 32/16/8/4 KB from candidate authors; lower added compressed length = more similar. Candidates: Shakespeare, Fletcher, Massinger, Davenant, Theobald (p. 410: "William Shakespeare, John Fletcher, Philip Massinger, William Davenant, and Lewis Theobald"); controls include Massinger's *New Way*, *Renegado*, *Bondman*, *Bashful Lover*, Davenant's *Fair Favourite*, Theobald's *Fatal Secret* etc. "over half a million" string comparisons. Notes the method counts repetitions but "their distributions across the texts are not" and cannot distinguish conscious phrase-repetition from sub-lexical repetition.
* **Davenant-v-Theobald-only test** (p. 411): Davenant wins 2.1, 2.4, 3.1 — "Davenant left his fingerprint in 8 per cent of the whole play (144 lines out of 1815), or 21.5 per cent if we judge by the number of scenes he worked on (3 out of 14 in the whole play)"; the ratio "increases significantly once all the plausible candidates are included".
* **Scene-by-scene result (pp. 414–415, verbatim)**: "1.1 Theobald heavily revised Shakespeare / 1.2 Shakespeare / 1.3 Theobald revised Davenant who revised Shakespeare / 2.1 Davenant revised Shakespeare (Possibly slight revision by Theobald too) / 2.2 Shakespeare and Massinger collaborated / 2.3 Theobald revised Massinger / 2.4 Shakespeare and Fletcher and Massinger collaborated (Unreliable results) / 3.1 Davenant revised Shakespeare / 3.2 Davenant revised Shakespeare / 3.3 Davenant revised Fletcher / 4.1 Shakespeare and Fletcher collaborated / 4.2 Theobald revised Shakespeare / 5.1 Theobald revised Davenant / 5.2 Shakespeare and Fletcher collaborated".
* **Interpretation (415–416)**: "Double Falsehood scenes 2.2, 2.3, and 2.4 are the ones that most divide attribution scholars … we are confident of Massinger's contribution here. For scene 2.4 … our results are equivocal, although notably there are no signs of post-Jacobean writing." 3.1 and 3.2 "were so heavily revised by Davenant that they hardly retain their original Shakespearean elements, which in 3.2 could well correspond to lines 39-43, as Stephen Kukowski (1991, 88) suggested." "Although revised by Davenant, Double Falsehood scene 3.3 is the only one in which Fletcher's hand is strikingly apparent, whereas in scenes 4.1 and 5.2 it emerges only in the 4KB-section tests." "Scene 5.2 … seems to be mostly Shakespeare's and to a lesser extent Fletcher's. Our results sketch a slightly different scenario from the nowadays widely accepted belief that Shakespeare's presence is limited to the first half of the play." **Conclusion**: "On the evidence presented here, the possibility that Theobald forged Double Falsehood is eliminated. Theobald had a manuscript of a play containing contributions by Shakespeare and Fletcher, as many studies have shown, and, we believe, contributions by Massinger too. The likeliest explanation, then, is that Theobald had a manuscript of the lost play Cardenio." He notes (p. 407, my earlier reading) that the result contradicts Matthews & Merriam's 1993 neural-network finding that DF is "strongly Fletcherian" throughout, and that his method also detects Massinger in *All Is True/Henry VIII*, "that most other investigators" do not.
* **Critiques**: none published that I found; internal caveats as above. Massinger in 2.2–2.4 is unreplicated and sits oddly with every other study (Tarlinskaja finds an 18th-c./Davenant profile in Act 2; Oliphant/Taylor find Theobald).

**(c) Gary Taylor & Rory Loughnane, "The Canon and Chronology of Shakespeare's Works", entry "The History of Cardenio 1612", pp. 583–586.**
* Heading (verbatim): "John Fletcher and William Shakespeare, adapted by William Davenant, Lewis Theobald, and Colley Cibber / Original Date Range: June 1612–January 1613 / Best Guess: second half of 1612 / First Adaptation Date Range: 1663–7 / Second Adaptation Date Range: 1727 / Text: Reference pp. 3681–769" — i.e. the play is printed (as Theobald's text) in the old-spelling *Critical Reference Edition*, not in the *Modern Critical Edition* ("Complete Works", 2016). Loughnane (ch. 19, p. 308) states the operative reason: "Cardenio, which cannot, in its adapted state of Double Falsehood, be included in our search" of Shakespeare's post-1607 lexicon; Craig's canon word-count for Shakespeare (740,207 words) likewise "excluded Arden of Faversham and Double Falsehood (and the Additions to Spanish Tragedy)" (Taylor's Spanish Tragedy chapter, p. 247). Editorial rationale (586): "What editors of Shakespeare should do, in this situation, remains a matter of debate … The inclusion of the play in the New Oxford Shakespeare is designed to serve 'as a productive stimulus for debate, rather than as a final statement' (Kirwan 2015b, 178)."
* Text of the entry (583–86): the 1613 payments; Stern's denial "that these three documents refer to the same play" rejected; the Shelton-only title; Gants on the 1612 printing; the King's Men's Cervantes vogue; "Thirdly, Cardenio was the only play that was performed twice at royal expense in the 1612-13 fiscal year; if it was the first Fletcher-Shakespeare collaboration, that initial success would explain why they continued to collaborate" (584); Gildon 1719; "Stern's claims have since been refuted by more than a dozen scholars" (list as in §2.20); "Collectively, these studies demonstrate that the text of Double Falsehood is indeed what Theobald claimed: a Jacobean play adapted for an early eighteenth-century theatre. Although Theobald … was undoubtedly responsible for some passages of independent writing (and for structural and verbal tampering throughout), the text preserves writing by both Shakespeare and Fletcher, and its primary source was clearly Shelton's translation of Don Quixote (a text not in Theobald's library catalogue, and which he never used elsewhere, or showed any awareness of). Consequently, the Jacobean play that Theobald adapted can be confidently identified as The History of Cardenio, by Fletcher and Shakespeare." Downes manuscript "no later than 1667"; Freehafer's Davenant hypothesis endorsed with Hume 2016; "Tarlinskaja (Chapter 23) and Pascucci (Chapter 24) independently supply new stylometric evidence of Davenant's hand in Double Falsehood (especially in Act 2); Taylor's commentary in Reference provides additional Davenant parallels (especially in Act 2). Taylor also, for the first time, identifies Colley Cibber (one of the managers of Drury Lane in 1727) as the author of some of the comic prose in 2.3." Final: "It now seems clear that Double Falsehood is an adaptation, primarily by Theobald with some help from Cibber, of Davenant's adaptation of a Jacobean play by Fletcher and Shakespeare."
* **(d) MacDonald P. Jackson, ch. 8, "Shakespeare, Arden of Faversham, and A Lover's Complaint: A Review of Reviews"** (p. 123): "The case for believing that the King's Men's lost play Cardenio was a collaboration by Shakespeare and John Fletcher, adapted by Lewis Theobald as Double Falsehood, has been hugely strengthened (Taylor 2013b) in response to elaborate expressions of scepticism (Stern 2011)" — offered as an example of how "Through argument and counter-argument we gradually progress towards a new orthodoxy", alongside the eighty-year Middleton/*Revenger's Tragedy* case that began with Oliphant.
* **Reviews of the Companion**: Lukas Erne (Geneva) review [DIRECT — unige-review…]: quotes Tarlinskaja's and Pascucci's conclusions without endorsement; the MP review (uchicago) and MUSE review blocked.

### 2.35 2017–2026: what exists and what does not
* **New stylometric work on DF**: only Tarlinskaja 2023 (MRDE 36, 285–312; not accessible) and Alotaibi 2021 (N&Q). Hugh Craig & Brett Greatley-Hirsch, *Style, Computers, and Early Modern Drama* (CUP 2017) — no DF/Cardenio chapter located (TOC checked). Swisher & Shamir, "A data science and machine learning approach to continuous analysis of Shakespeare's plays", *JDMDH* 2023 (arXiv 2301.06024) — Shakespeare only, no DF. Boyd 2018 (PLOS) and the Frontiers 2018 RPAS paper — no DF. Repeated searches for 2019–2026 machine-learning or n-gram studies naming DF returned only re-descriptions of Boyd & Pennebaker 2015. **Conclusion: no post-2017 quantitative attribution of DF has been published that I could find** (Rizvi's 2018 methodological pieces and Vickers's collocation work do not treat DF).
* **New documentary finds**: none located (searches for Downes/Betterton/Theobald manuscript discoveries 2018–2026 negative). The Stern–Hammond exchange ended with Hammond 2014; Stern has not, to my knowledge, published a reply.
* **Editions**: Arden (2010) remains the only critical edition; NOS Reference Edition prints the text; RSC *Collaborative Plays* (2013) prints it; Taylor's unadaptation (2013) and Doran's version (2011, Nick Hern) are the performance texts.

### 2.36 Reference-work consensus (as of Sept 2026)
* **Folger Lost Plays Database** [DIRECT]: "Comparison of the collaborative The Two Noble Kinsmen with William Davenant's 1668 adaptation … indicates the scale of revision deemed necessary for a Restoration audience, excising Shakespeare almost entirely. Few critics claim that substantial chunks of Shakespeare or Fletcher's poetry remain in Double Falsehood." "The case for forgery … is upheld by Harriet C. Frazier, Jeffrey Kahan and, most fully, by Tiffany Stern." Folger's *Shakespeare Documented* page on collaborations (403) not read.
* **Arden** (Hammond 2010): cautious pro-Shakespeare, Restoration intermediary (§2.19).
* **RSC** (Bate & Rasmussen 2013, Sharpe): prints DF; graded verdict (§2.26).
* **Britannica** (David Bevington) [WEB-SUMMARY]: Hammond "argues Shakespeare was co-dramatist with Fletcher"; the controversy "remained in contention for nearly three centuries. It remains contentious still"; the hoax hypothesis "also plausible" (my earlier notes).
* **Wikipedia** ('Double Falsehood', saved 2026-09-14) [DIRECT]: "There appears to be agreement among scholars that the 18th century Double Falsehood is not a forgery, but is based on the lost Cardenio of 1612–13, and that the original authors of Cardenio were John Fletcher and possibly William Shakespeare"; the article also quotes Hammond pp. xvi, 6, 8 and Boyd & Pennebaker (Theobald's contribution "very minor").
* **ODNB** ("Theobald, Lewis", Peter Seary): 403 — not read.
* **NOS**: Fletcher & Shakespeare, adapted by Davenant, Theobald and Cibber (§2.34c).

### 2.37 Reconstructions — only their claims about what is original
* **Charles Hamilton, *Cardenio, or The Second Maiden's Tragedy* (1994)**: claims the 1611 manuscript play (BL Lansdowne 807) is *Cardenio* in Shakespeare's hand; universally rejected (Middleton's authorship established; "argued, with passion but little rigour" — search summary). Irrelevant to DF except as noise.
* **Stephen Greenblatt & Charles Mee, *Cardenio* (2008)**: a Curious-Impertinent play with a DF play-within; keeps c. 63 Theobald lines, mostly from the Julio–Leonora exchange "in the middle of 1.2", plus 4.1 lines (my earlier notes: 4.1.49–61), reworded 21% (Taylor 2016, 358) — no authorship claims beyond the "lost play by Shakespeare" billing.
* **Bernard Richards, *Cardenio* (Queens' College, Cambridge, March 2009; *Essays in Criticism* 2011; *Quest* 344–51)**: keeps Theobald's text, adds scenes "to explicate the early relations of Julio and Henriquez and to bring into close focus the rape of Violante and the abduction of Leonora from the nunnery in a coffin" [WEB-SUMMARY]; no line-level attribution.
* **Gregory Doran, RSC *Cardenio* (Swan, 2011; Nick Hern text; *Shakespeare's Lost Play: In Search of Cardenio*, 2012)**: "basically Theobald's text with additions" (Shakespeare Blog) [DIRECT]; missing scenes rebuilt from Shelton; "where extra lines were needed, I tried to limit myself to plundering only those Jacobean plays in which John Fletcher had drawn upon Cervantes" (Nick Hern blog) [DIRECT]; Doran's own attribution reasoning (Plays International, Feb 2011) [DIRECT]: "The more I read [Shelton], the more I realised that Double Falsehood did indeed contain Shakespeare and Fletcher because phrases from the Shelton just sprang about … 'This is clearly not just the source but Theobald has got the Shelton open while he's writing.'" — which is, of course, equally an argument that Theobald used Shelton directly. Taylor 2016 (356): Doran singled out the Theobaldian 5.2.251–57 as Shakespearean.
* **Gary Taylor, *The History of Cardenio, 1612–2012* (in *Creation*, 241–316; staged Indianapolis 2012; revised for NOS)**: the only reconstruction with explicit line-level claims — removes the passages listed in §2.31 (5.2.251–57; 5.1.25–37 Henriquez–Leonora; 4.1.212–57; much of 2.1 soliloquy and 2.3; Cibber prose in 2.3), restores 'Cardenio/Fernando/Luscinda', keeps the 1.2 prose and Julio–Leonora verse, 1.1's opening, 5.2.94–101 (Fletcher) etc.; supplies new lines "in the style of the original authors" [WEB-SUMMARY + Taylor 2016 DIRECT].

## 3. Scene/act-level attribution table (Arden numbering)

Cell codes: **Sh** Shakespeare; **Fl** Fletcher; **Th** Theobald; **Dav** Davenant; **Mass** Massinger; **Cib** Cibber; "Sh/Th" = Shakespeare overlaid by Theobald; "mixed" = more than one original hand within the scene; "?" = uncertain / not assigned; "—" = study makes no scene-level claim. Column sources: Th28 = Theobald's own claims (1728); Br10 = Bradford 1910 [DIRECT]; Gr16 = Graham 1916 (adopts Bradford's division for his counts) [DIRECT]; Ol27 = Oliphant 1927 [DIRECT]; Ku91 = Kukowski 1991 [SECONDARY]; Ho94 = Hope 1994 sections A/B [WEB-SUMMARY]; Ha10 = Hammond 2010 as reported by Nicholl/Hamlet Weblog/search summaries [SECONDARY]; Ja12 = Jackson 2012 [SECONDARY]; Pr12 = Proudfoot 2012 [SECONDARY]; Ta12–16 = Taylor 2012/2013/2016 + NOS canon entry [DIRECT for 2016 and NOS]; Na13 = Nance 2013 [ABSTRACT]; BP15 = Boyd & Pennebaker 2015 (act level) [ABSTRACT + press]; Tr17 = Tarlinskaja 2017 [DIRECT]; Pa17 = Pascucci 2017 [DIRECT].

| Scene | Th28 | Br10 | Gr16 | Ol27 | Ku91 | Ho94 | Ha10 | Ja12 | Pr12 | Ta12–16 | Na13 | BP15 | Tr17 | Pa17 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1.1 | Sh | ? (non-Fl, "Shakespearean turn") | "another" | mixed: first two speeches Sh/Th; rest Th | ? | A: ? (co-authored) | Sh (via Dav/Th) | Sh/Th (traces) | — | Sh opening 8 lines; rest Th-heavy | — | Sh (Act 1) | Sh/Th: Sh-type pleonastic do, rhythmical italics; 18th-c. symmetry lines | Th heavily revised Sh |
| 1.2 | Sh | ? (non-Fl) | "another" | Sh/Th except ten speeches from "Leon. What do you mean?" = Th | ? | A | Sh (via Dav/Th) | Sh (prose intact; verse links) | — | Sh (Julio–Leonora verse; prose) | Sh (1.2.179–224 prose) | Sh | Sh/Th | Sh |
| 1.3 | Th claims ll. 10–14; rest Sh | ? (non-Fl) | "another" | Sh/Th | ? | A | Sh (via Dav/Th) | Sh/Th (1.3.53–6 Sh per my notes) | — | Sh/Th (Farmer/Malone doubted Th's claim to 10–14) | — | Sh | Sh/Th (disyllabic -ion 1.3.7) | Th revised Dav who revised Sh |
| 2.1 | Sh | ? (non-Fl) | "another" | Sh/Th | ? | A | Sh (via Dav/Th) | Sh/Th | — | much Th (Henriquez soliloquy); Dav parallels | — | Sh (Act 2) | Dav/Th profile (dip 6≈8); low enclitics | Dav revised Sh (slight Th) |
| 2.2 | Sh | ? (non-Fl) | "another" | Sh/Th | ? | A | Sh (via Dav/Th) | Sh/Th | — | Sh/Th (Act 2 = most Th) | — | Sh | Dav/Th profile | Sh + Mass |
| 2.3 | Sh | ? (non-Fl) | "another" | **Th wholly** | ? | A | Sh?/Th | Th (Theobald parallels) | Th line-endings concentrated (173–74) | Th (+ Cib in comic prose; Mrs Porter part) | — | Sh | 18th-c. symmetry lines (e.g. 2.3.94) | Th revised Mass |
| 2.4 | Sh | ? (non-Fl) | "another" | **Th wholly** | ? | A | Sh?/Th | Th | — | Th-heavy | — | Sh | Dav/Th profile | Sh + Fl + Mass (unreliable) |
| 3.1 | Sh (3.1.17 "definitely Sh") | ? (non-Fl) | "another" | Sh/Th | ? | A | Sh (via Dav/Th) | Sh/Th | — | Sh/Th | — | Sh (Act 3) | 18th-c. signs; 2.4% unstressed monosyllables at 10 (Sh or Dav trace); Jacobean dip on 8 | Dav revised Sh |
| 3.2 | Sh | ? (non-Fl) | "another" | mixed: first four speeches + "large hall"…"Most perjur'd if I do" Sh/Th; rest Th (fragments?) | Sh residue at 3.2.39–43 | A | Sh (via Dav/Th) | Sh/Th | — | Sh/Th | — | Sh | as 3.1 | Dav revised Sh |
| 3.3 | Sh | **Fl** (first Fl scene) | Fl | Fl (first nine speeches overwritten Th; closing couplet Th) | Fl | B: Fl fit | Fl | Fl ('o' my conscience') | Fl/Sh | Fl | — | Sh | **Fl** (enclitics 200/1000; disyllabic -ion 13.3/1000) via Dav | Dav revised Fl |
| 4.1 | Sh | mixed: shepherds Fl; Julio's madness "firmer, stronger hand" | Fl (approx.) | mixed: Fl before Julio's entry; five speeches after Julio's entry Sh/Th; Th to Violante's re-entry; Fl to Roderick's entry; conclusion Th | Fl | B | Fl | Fl/Th | — | Th at 4.1.212–57; Fl elsewhere | — | Sh/Fl (Act 4) | Fl (Act 4 dip Davenant-like; enclitics high) | Sh + Fl |
| 4.2 | Sh | Fl | Fl | Fl to "sweetly sleep together" (song excepted); rest Fl/Th | Fl | B | Fl | Fl ('Stand off'; 'all the miseries') | — | Fl/Th | — | Sh/Fl | Fl | Th revised Sh |
| 5.1 | Sh | Fl | Fl | Fl/Th (revised) | Fl | B | Fl | Fl ('as I have a soul')/Th | — | Th at 5.1.25–37 (Henriquez–Leonora); Fl elsewhere | — | Sh/Fl (Act 5) | Fl (Jacobean dip on 8) | Th revised Dav |
| 5.2 | Sh | Fl (except undefined small portion) | Fl | mixed: opening perhaps Sh; Fl from "Thou art a right one" (unrevised to "Weep not, child"); Th from "The righteous powers" onward | Fl | B | Fl | Fl/Th | — | Fl at 94–101; Sh at 109–16 [sic; 1.2]; Th at 251–57 | — | Sh/Fl | Fl | Sh + Fl ("mostly Shakespeare's") |

Whole-play verdicts not reducible to scenes: Farmer 1767 = Shirley (whole); Malone 1780 = Massinger (whole); Steevens 1780, Frazier 1974, Kahan 2004, Stern 2011 = Theobald (forgery/imitation); Elliott & Valenza = not Shakespeare as a whole (11–14 rejections), Sh share in first half "not ruled out"; Matthews & Merriam 1993 (neural net) = "strongly Fletcherian" (via Pascucci); Alotaibi 2016/2021 = Fletcher prime/sole author of the source + large Theobald share; Hume 2016 = adaptation with "little or no unadulterated Shakespeare"; Proudfoot 2012 = Sh 63% / Fl 65% / Th 32% polysyllable matching, three hands.

**Points of near-universal agreement**: (1) a stylistic break at 3.3 (Bradford, Graham, Oliphant, Kukowski, Hope, Hammond, Jackson, Tarlinskaja, Pascucci — and, at act resolution, Boyd & Pennebaker); (2) Theobald's hand densest in Act 2 (Oliphant, Proudfoot, Taylor, Tarlinskaja; Pascucci differs: Massinger); (3) if any Shakespeare survives verbatim it is in 1.2 (Oliphant, Jackson, Nance, Taylor, Greenblatt & Mee's instinct, Pascucci "1.2 Shakespeare"). **Points of disagreement**: whether Acts 1–2 show *Shakespeare* (Hammond, Boyd & Pennebaker, Pascucci) or an *eighteenth-century/Restoration* profile with only traces (Tarlinskaja: "few signs of Shakespearean verse style in Acts 1 and 2"); whether 5.2 is mostly Fletcher (everyone before 2017) or "mostly Shakespeare's" (Pascucci); Massinger anywhere (Pascucci only).

## 4. Strongest points for each hypothesis

### H1 — Fletcher–Shakespeare *Cardenio* underlies DF
1. External chain: 1613 court payments for a *Cardenno/Cardenna* by the King's Men (Shakespeare's and Fletcher's company) in the very season of their other collaborations; Moseley's 1653 "by Mr Fletcher. & Shakespeare"; the 1718 Tonson assignment; Gildon c.1710/1719 on a manuscript play of Shakespeare's maturity; Theobald's Downes/Betterton manuscript "of above sixty years standing" (Freehafer 1969; Hammond 2010; NOS 583–85).
2. The source is Shelton 1612 specifically — a text "not in Theobald's library catalogue, and which he never used elsewhere, or showed any awareness of" (NOS 585; Taylor & Wagschal 2013; Graham 1916 against Schevill); the title formula "History of Cardenio" is Shelton's phrase (Taylor 2012, 14–15).
3. Two distinct Jacobean styles in the text, with the break at 3.3, measured repeatedly: Bradford 47%/32% double endings; Graham 44.2%/32% feminine endings, weak-and-light endings 3/19; Tarlinskaja enclitics 36.4–52.9 v. 200 per 1,000 lines and disyllabic -ion 13.3/1,000 in 3.3 ≈ Fletcher's H8 13.9; "impossibly subtle" for a forger to plant Fletcher features only in half a play (Tarlinskaja 405–406).
4. Theobald denied Fletcher's presence (Preface) and never mentioned Moseley's entry — "absolutely incompatible with any possibility of Theobald's having set himself deliberately to imitate Fletcher" (Bradford 55; Oliphant 283).
5. Fletcher-specific lexis unknown to Theobald: 'o' my conscience', 'Stand off' (30× extended Fletcher canon, 0× Shakespeare), 'as I have a soul', 'all the miseries' (Jackson 2012); Fletcherian 5.2.94–101 (Taylor 2016 App. B).
6. Positive Shakespeare signals in the first half: 1.2 prose (Nance 2013: 334 words "strongly suggest the presence of Shakespeare more than any other author"); Julio–Leonora verse in 1.2 (Taylor 2016 App. C: unique links to Shakespeare unmatched by any other dramatist); Act 1 pleonastic *do* 16.2/1,000 (Theobald 4.3–4.7) and rhythmical italics 133.6/1,000 "typical of Shakespeare" (Tarlinskaja 402); Boyd & Pennebaker: Shakespeare closest in 39/45 tests on Acts 1–3; Pascucci: 1.2 "Shakespeare", Sh in 4.1, 5.2.
7. Line-end polysyllables — a feature "most unlikely to occupy the attention of a forger" — match Sh 63% / Fl 65% / both 79% v. Theobald 32% (Proudfoot 2012).
8. Theobald's known adaptations (*Richard II*, *Fatal Secret*) show exactly the kind of cutting, regularising and moralising visible in DF (Carnegie 2012; Tarlinskaja's *Fatal Secret* enclitic reduction 127.6 → 41.7); his one avowed imitation, *Cave of Poverty*, "reeks of Theobald" and fails the same LION test that DF's Shakespearean passages pass (Taylor 2013, 2016).
9. Consensus of specialists after 2012: "refuted by more than a dozen scholars" (NOS 585); "none of the speakers endorsed Stern's argument" at UCLA 2014 (Taylor 2016, 354); Hume 2016 — the leading sceptic-minded theatre historian — rejects forgery.

### H2 — Fletcher only, or Fletcher plus someone other than Shakespeare
1. Fletcher's presence is the one thing every stylistic study confirms (Bradford, Graham, Oliphant, Kukowski, Hope B, Jackson, Tarlinskaja, Pascucci 3.3, Matthews & Merriam); Shakespeare's is inferred largely from the external record — Bradford: "The internal evidence is certainly not of great importance … none is close enough to have much weight" (55); Oliphant: "one has to be guided largely by the external evidence and the probabilities" (296); Tarlinskaja: "There are few signs of Shakespearean verse style in Acts 1 and 2" (405–406).
2. Moseley's entry is unreliable and Fletcher-first: the full stop after "Fletcher", "alphabetical order of authorship", Moseley's other false Shakespeare ascriptions (Freehafer 508; Stern 557–58; Bradford 55 "far from reliable"); Moseley's own TNK notice called it Beaumont and Fletcher's.
3. Cervantes/Shelton material is Fletcher's known hunting-ground (Love's Pilgrimage, The Chances, The Coxcomb, Rule a Wife…; Hatcher 1905 via Graham; Stern 558; Alotaibi); Fletcher "the more obviously plot-reliant of the two" (Stern).
4. Whole-play stylometrics reject Shakespeare: E&V 11 discrete rejections (composite probability 1.376E-09), 14 in Scott's tabulation; Matthews & Merriam 1993 "strongly Fletcherian".
5. Every "Shakespearean" verbal parallel has been challenged by a fuller Theobald negative-check (Alotaibi 2016; Stern 2012, 129); the Beaumont check (Jackson) turned up 34 v. 48 three-word links, "trivial in the extreme".
6. Hume's verdict — "little or no unadulterated Shakespeare" — and Jackson's — "scarcely a line of Shakespeare's verse survives intact" — mean H1 and H2 differ only about the lost play, not about DF; the observable text supports "Fletcher + heavy later rewriting" at least as well as "Fletcher + Shakespeare + heavy later rewriting".
7. Pascucci's Massinger detections in 2.2–2.4 (and in H8) — if replicated — would revive Malone's candidate for the non-Fletcher hand.

### H3 — another Jacobean/Caroline play (Shirley; Massinger)
1. Farmer's accentual argument: "aspect" stressed on the first syllable twice, "never the case in the time of Shakespeare" (2nd ed. 26–27) — and a Shirley whose manuscripts Langbaine records as unpublished.
2. Malone's Massinger parallels (*Duke of Milan* "but itself admits no parallel"; "stoop to rise") and eleven lost Massinger plays in the SR; Gifford's independent sense of an Elizabethan/Jacobean hand superior to Theobald.
3. Proudfoot's own data: Massinger's early plays reach "over 20 per cent" line-end-polysyllable matching (higher than every other contemporary tested), explained away as Fletcherian indebtedness; Pascucci 2017 finds Massinger in 2.2, 2.3, 2.4.
4. Nobody has ever run a Shirley test; Stern 2012 and the RenRef reviewer note that Massinger and Shirley "have not been addressed adequately"; Forsythe (via Graham 1920) wanted Daborne, Davenport and William Rowley tried and nobody has.
5. If Moseley's ascription is discounted (H2.2), nothing external ties DF's source to 1613 rather than to a later Fletcher-circle play; the "sixty years" manuscript only reaches back to c.1667.
   (Counter-weight: Bradford found "none of these 'marks of style and manner'" of Shirley; the Shelton-1612 dependence and King's Men 1613 *Cardenno* make a Fletcher–Massinger or Shirley *Cardenio* an unnecessary multiplication of plays; Farmer's accent test ignores Theobald's admitted revision.)

### H4 — Theobald forged it (with or without a Restoration source)
1. No one ever saw the manuscripts; Theobald's account of them is internally inconsistent (Downes copy c.1667 yet "designed" by Betterton for the stage; natural-daughter legend; three-then-four copies), he never printed the play in his 1733 *Works*, never delivered the promised "Dissertation", and the 1770 "Museum of Covent Garden" notice is unverifiable and conveniently ends in the 1808 fire (Stern 563–66; Alotaibi 9–26; Halliwell-Phillipps).
2. The 1653 entry was public (Steevens found the 1613 records only in 1780, but the SR was searchable); Theobald could have learned the title from Gildon/Moseley/Tonson (Freehafer himself says Theobald "learned otherwise, from Gildon and the Stationers' Register") — so "he never claimed Fletcher" cuts both ways.
3. Theobald was a practised imitator and adapter (*Cave of Poverty*; *Orestes* "plagiary" letters to Warburton; *Perfidious Brother* dispute; his *Richard II*), with Shelton available in the 1725 reprint (Stern 577–82; Frazier 1974; Kahan 2004; Alotaibi ch. 3).
4. Theobald's contradictory ownership claims — 1.3.10–14 his own, 3.1.17 "definitely Shakespeare's" — and Malone's "I believe both assertions to have been false" (Lupić 107).
5. Whole-play stylometrics reject Shakespeare (E&V), and the "Shakespearean" parallels dissolve under a full Theobald negative check (Stern 2012; Alotaibi).
6. Every "authenticating" feature (Fletcherian enclitics, Shelton phrasing) is consistent with Theobald "collaborating" with any old Fletcher-circle manuscript, or with a Restoration adaptation he possessed — the weak form of H4 that Hammond's own "palimpsest" model concedes.
   (Counter-weight: the strong form — no old manuscript at all — is now rejected by every quantitative study since 2012, by Hume, and by Stern's own critics as requiring a forger "impossibly subtle" in Jacobean prosody, Shelton phrasing and Fletcher lexis he had no motive to imitate.)

## 5. What each side says would overturn its position

* **H1 proponents** (Taylor, Jackson, Proudfoot, Tarlinskaja, Pascucci, Hammond, NOS): would be overturned by (a) a demonstration that Theobald's own oeuvre contains the Fletcher-specific and Shakespeare-specific features at comparable rates (the Byrne/Stern "negative check" — Taylor's *Cave of Poverty* test is their pre-emptive answer); (b) documentary proof that the 1613 *Cardenno* was not a Cervantes play or not the play Moseley entered (Stern's Cardeña reading); (c) a Massinger/Shirley/other test outperforming Shakespeare on the non-Fletcher scenes with equal corpora (Proudfoot and Jackson explicitly invite it; Pascucci's Massinger result is the nearest thing); (d) discovery of the Restoration text showing a different source play. Taylor 2016 (354) states that refutation of Stern is now so complete that forgery "no longer needs to be addressed" — i.e. only new documents or new negative-check failures would move him.
* **H2 proponents** (Alotaibi; Stern 2012 in part; Hume implicitly): would be overturned by (a) scene-level, equal-corpus tests in which Shakespeare beats Fletcher, Massinger, Beaumont, Davenant *and* Theobald on the first-half scenes (Nance's prose test and Taylor's App. C are the existing attempts; Alotaibi rejects them as corpus-biased); (b) an independent replication of Pascucci's 1.2/4.1/5.2 Shakespeare signals with a different method; (c) any document confirming Moseley's double attribution (e.g. the Tonson 1718 assignment's wording, or a Betterton/Davenant record naming Shakespeare).
* **H3 proponents** (Farmer/Malone tradition; nobody living): would be overturned by direct Shirley/Massinger tests failing (Proudfoot's ">20%" for Massinger is the only data point, and Bradford's negative reading of Shirley the only Shirley test); would be *revived* by replication of Pascucci's Massinger detections.
* **H4 proponents** (Stern 2011; Frazier; Kahan; Alotaibi in weak form): Stern's article frames the question as one of evidence, so a verified pre-1727 witness to the manuscripts (the "Noble Person", Betterton's or Downes's papers, the Covent Garden "Museum" inventory), or a demonstration that Theobald could not have known Shelton 1612 (as opposed to 1725) would end the strong form; Alotaibi concedes that "there is some external evidence, which although it does not confirm Theobald's claims, nevertheless suggests that there might be some truth behind them" (thesis p. 26). The weak form (Restoration-adaptation source) is not falsifiable by stylometry alone — it needs the 1660s text.

## 6. What an investigator should know

1. Every quantitative study since 1910 finds two Jacobean styles with a break at 3.3; Fletcher after that point is the single most secure finding in the whole literature.
2. Shakespeare's presence rests on the external chain (1613 payments + Moseley 1653 + Shelton-1612 source) plus a handful of internal signals (1.2 prose and verse; Act 1 versification; 1.1 opening; Pascucci's 1.2/5.2) — not on any large body of intact Shakespearean verse. Jackson: "scarcely a line … survives intact"; Hume: "little or no unadulterated Shakespeare"; LPD: "Few critics claim that substantial chunks … remain".
3. The strong forgery thesis (Stern 2011) has no current defenders among stylometricians or theatre historians; the NOS lists "more than a dozen" refutations. Treat it as answered, but note that most refutations were written by the *Quest*/NOS group (Taylor, Jackson, Proudfoot, Nance, Pascucci, Tarlinskaja); the independent voices are Hume, Boyd & Pennebaker, Hammond.
4. The live alternative is not forgery but H2: a Fletcher (± Massinger/Beaumont) play, or a Fletcher–Shakespeare play in which Shakespeare's share is unrecoverable. Alotaibi 2016/2021 is the only sustained modern case; nobody has run Shirley or a proper equal-corpus Massinger test.
5. The consensus transmission model is three layers — Jacobean original → Davenant-era adaptation (c.1663–67; Downes/Betterton) → Theobald 1727 (+ Cibber in 2.3 per NOS). Tarlinskaja and Pascucci independently locate the Davenant layer in Acts 2 and 3.1–3.2; this is why "Shakespeare's" first half looks eighteenth-century in verse profile.
6. Theobald's hand is densest in Act 2 (2.3 and 2.4 "wholly his" — Oliphant; Proudfoot; Taylor), in 4.1.212–57, 5.1.25–37 and 5.2.251–57 (Taylor 2016) — the very passages Doran and Greenblatt–Mee admired as "Shakespearean".
7. Boyd & Pennebaker 2015 is the most-cited but methodologically weakest pro-Shakespeare study (act-level units, no Davenant/Massinger candidates, unverifiable "psychological signature"); no replication exists.
8. The only truly hostile modern audit of the pro-Shakespeare parallels (Alotaibi) shows that Theobald's *full* corpus (prose, letters, translations) must be the negative check; Jackson, Proudfoot and Taylor used mainly his plays.
9. Corpus imbalance is endemic: Jackson 50 Fl / 39 Sh / 15 Th; Proudfoot 15 / 13 / 9; Boyd 33 Sh / 9 Fl / 12 Th.
10. Pascucci 2017 is the outlier on both ends: it eliminates forgery but finds Massinger (2.2–2.4) and "mostly Shakespeare" in 5.2; unreplicated.
11. Elliott & Valenza's 11 rejections say only that DF *as a whole* is not solo Shakespeare — which nobody claims; their half-play test was announced (2010) but never published.
12. Hope 1994's actual auxiliary-*do* figures were not obtainable here; the book must be consulted (pp. 89–100).
13. Not accessible in full: Stern 2011, Hammond 2010/2012/2014, all *Quest* and *Creation* chapters, Kukowski, Pujante, Muir, Frazier, Metz, Kahan, Lupić, Folkenflik, Tarlinskaja 2014/2023, ODNB; their content above is second-hand (chiefly via Alotaibi's thesis, Taylor 2016 and the NOS Companion, all read directly).
14. No new documentary evidence and no new quantitative study of DF has appeared 2018–2026 that I could find; the field is static since NOS 2017.
15. Reconstructions (Doran, Greenblatt–Mee, Richards) make no defensible line-level authorship claims; Taylor's unadaptation is the only one built on explicit attribution arguments and is itself contested (Alotaibi).
16. Reference works now converge on "adaptation of a Fletcher–Shakespeare *Cardenio* via a Restoration version" (Arden, NOS, RSC, Wikipedia, LPD), with Britannica still calling the hoax "plausible".
17. If the project wants a decisive new test, the gap is clear: equal-sized, scene-level comparison of DF 1.1–3.2 against Shakespeare, Fletcher, Massinger, Beaumont, Davenant *and* the complete Theobald corpus, with the 1660s adaptations of TNK/H8 (*The Rivals*) as calibration — no such study exists.

## 7. Access summary (details in ../sources/scholarship/SOURCES.md)

Read in full [DIRECT]: Farmer 1767; Reed 1782; Bradford 1910; Schevill 1911; Graham 1916/1917/1920; Oliphant 1927; Elliott & Valenza 2010 (LLC preprint); Scott 2006 (EMLS); Taylor 2016 (JEMS); Alotaibi 2016 thesis; NOS Authorship Companion 2017 (Tarlinskaja, Pascucci, canon entry, Jackson ch. 8, Loughnane, Taylor on the Spanish Tragedy canon count); Boyd & Pennebaker abstract (efetch); Hume 2016 abstract; Freehafer 1969 abstract; RenRef review of *Quest*; Erne review of NOS Companion; Nicholl LRB 2010; Johnson *Conversation* 2015; UT press release; CMC 2010 release; LPD; Wikipedia (both articles); Doran interviews (Plays International; Nick Hern; Shakespeare Blog).
Abstract/landing page only: Stern 2011 (OUP page), Kukowski, Pujante, Hammond 2014, Kirwan 2011 (WRAP), Nance 2013 (search summary), Jackson 2012 (OUP abstract via search).
Not accessible (recorded as such): Hope 1994; Muir 1960/1997; Frazier 1974; Metz 1982/1989; Kahan 2004; Hammond 2010 (Arden) and 2012; all OUP *Quest* chapters; all Palgrave *Creation* chapters except abstracts; Pascucci 2012; Tarlinskaja 2014/2023; Lupić 2012; Folkenflik 2012; Sharpe 2013; Chartier 2013/2016; ODNB; Vickers's review; Matthews & Merriam 1993; Oliphant 1919 N&Q; Fleissner 1984; Alotaibi 2021 N&Q (OUP 403).