The Painful Misadventures of Pericles Acts I-II
© Notes and Queries, ccxxxvi, 1991, 67-70
The Oxford Works  include the little-known five-act play called “Pericles, by William Shakespeare and George Wilkins”. The same page admits that this joint attribution, which credits Wilkins with the first two acts and leaves the rest to Shakespeare, is not a fact at all but merely “the hypothesis” on which “our edition is based”. But it is not even a hypothesis in any sense save the Oxford English Dictionary's definition 4: “something assumed to be true without proof… a mere assumption or guess”.
This usage is confirmed by Gary Taylor and MacD. Jackson in the Oxford Companion.  They define their assumption as follows: “We believe that no textual theory can make it credible that Shakespeare wrote the bulk of [Pericles I-11].” Thus they explicitly base their textual theory on their literary opinion, and treat both as infallible. The credo continues: “We believe… that a variety of evidence points to Wilkins as [Shakespeare's] probable collaborator.” Thus a mere belief provides the evidence, the advocacy, and the verdict. There is no defence, and no appeal.
It is high time for a retrial. Call George Wilkins, of the parish of St Sepulchre's without Newgate, who in 1612 was “of the age of thirty-six years or thereabouts”.  So he was born c. 1576, which makes him some twelve years younger than Shakespeare. Next, John Dryden deposes, in his Epilogue c. 1684 to a play  written by the son of William Davenant, who knew Shakespeare (and was prima facie the source of the information) that Pericles was Shakespeare's very first play, that is, it was written in the 1580s. In 1781, Thomas Warton gives the date of 1596 for a pamphlet  that speaks of Pericles as a play that had proved very popular. Already in 1629, Ben Jonson  had described Pericles as ‘mouldy’, that is, stale in style, as he also felt about the old-fashioned Titus Andronicus, written c. 1589.
Thus on the mutually corroborative affirmation of Jonson, Dryden (a close associate of Davenant), and Warton the infant Wilkins had no connection whatever with Pericles, which on this evidence was an early play later revised, in accordance with its famous author's well-documented practice. But modern text-editors have a short way with these four past poets laureate. All such positive support for an early de-Wilkinsed Pericles is simply ignored. So is the even stronger evidence for a wholly Shakespearian Pericles, whenever it was written or revised. Despite its failure to figure in the 1623 First Folio, its authenticity has been accepted for centuries, beginning with the Third and Fourth Folios (1664, 1685) and theComplete Works edited by a fifth poet laureate, Nicholas Rowe. 
The direct contemporary testimony includes explicit title-page attribution in seven successive quarto issues, incorporating textual corrections, from seven separate printers or booksellers. Three of them had close commercial dealings with Shakespeare himself, or his company, in the printing and publication of his own unaided works.  Not only Dryden (and by plain implication Jonson) but two other seventeenth-century poets ascribed Pericles to Shakespeare as sole author, whether in praise  or blame. 
The first three quarto editions  of Pericles had also named Shakespeare's company, the King's Men, and his theatre, the Globe on Bankside; the next four quarto issues  again ascribe the entire play to him alone. Such continued assurances of authorship and provenance spanned eight years of Shakespeare's own lifetime, without any recorded syllable of dissent from any quarter, not even from the alleged part-author George Wilkins himself, at any stage of his own lifetime.
Nor did Wilkins ever advance any claim to what is allegedly his own work, not even in his elaborate preface to his 1608 prose narrative version  of the play, prophetically called The Painful Adventures of Pericles. That phrase was filched from Lawrence Twine's translation (published c. 1594 or earlier) of a French version of the original source-story in the Gesta Romanorum. Indeed, Wilkins's whole text is freely stolen from Twine as well as Shakespeare. On copious and uncontested evidence, Wilkins's only known literary activities were confined to his early thirties, between 1606 and 1608; and they consisted largely of plagiarism.  Yet the very existence of his plagiarism from Pericles is adduced by the Oxford editors as ‘external evidence’ for their hypothesis about his part-authorship of Pericles, because his adaptation “associated Wilkins with the play a year before Shakespeare's name was linked with it”.
This blandly assumes, in silent defiance of Dryden and the rest, that Pericles was not an early play. Besides, Wilkins regularly set his own name on the work of others, including Shakespeare. His unacknowledged adaptation of Pericles is an undeniable and shameless example. Yet we are asked to accept that his prior publication of that plagiarism somehow implies his part-authorship of the original play, which he modestly managed to leave unmentioned, even in his own adaptation of what was supposedly his own work. What could conceivably have inhibited him from making such a claim, except its readily demonstrable falsity?
The Oxford editors point out that “Scholars have in the past been puzzled by Wilkins's apparent reliance on his memory, in compiling The Painful Adventures; if he were part-author [of Pericles] why did he not have a text of the play which he would copy as mechanically as he did Twine?” Why not, indeed? But the plain explanation of this non-problem, namely that Wilkins had no text of his share in the play because he had no share in the play, must somehow be rejected. So first of all he must be feigned to have forgotten his own recently written text. Then the theatre company “may have had good reason to ensure, in Wilkins's ease, that the playwright retained no copy”, because he was, as contemporary records show,  “an unscrupulous petty criminal”. This is no bar to his being a suitable collaborator for Shakespeare, who is serenely assumed to have received literary assistance from a petty criminal twelve years his junior; it is just a bar to the supposed collaborator's ownership of his own copy of his own supposed contribution, which the hypothesis forbids. In reality, however, how could Shakespeare's company possibly have enforced any such prohibition?
The internal evidence is just as self-refuting as the external. First, some unnamed scholars have “pointed to a surprising number of parallels between Pericles and Wilkins's (few) other works” - in other words, a known plagiarist's parallels with his known victim are evidence of the plagiarist's authorship, not the victim's. This is like honouring a forger's cheque because the signature looks so authentic. We are not told, conversely, that such parallels were dismissed three decades ago by J. C. Maxwell, in a source  actually cited by the Oxford editors, who however choose to ignore its refutation of the Wilkins hypothesis. Similarly they cite G. Bullough's magisterial work  on the sources of Pericles (1609) without mentioning his reasons for postulating an earlier version of the play and rejecting Wilkins's authorship of it at any stage.
The only other internal evidence offered by the Oxford editors is that “Lake (1969) identified  a series of peculiarities in rhyme technique in which [Acts I-II] of Pericles correspond more closely with the work of Wilkins than of any other candidate”. But what is ‘the work’ of the admitted plagiarist Wilkins? This is like cataloguing the property of a thief. Besides (pace Lake), these assonances are manifestly meant to convey the effect of archaism. Of course the jackdaw Wilkins would filch such finery to adorn his own nest. No wonder that, as Lake himself points out, another work to which Wilkins set his name  also contains “a Chorus or presenter whose rôle is very similar to that of Gower in Pericles” and who indulges in the very same assonances. The rational inference is that Wilkins has stolen these ideas, not invented them. And who was his victim, that is, the author of Pericles I-II? “He is not, of course, Shakespeare.” So Lake assures us. Why? Because “the first two acts are known [sic] to be non-Shakespearean”. Thus, by stern academic decree, the only playwright who cannot even be considered for the authorship of Pericles I-II is Shakespeare himself.
An analogous posture has been adopted for the past nine years by Dr M. W. A. Smith,  whose professional computer-aided methods of authorship investigation, like the Oxford editors' avowedly amateur methods, yield so-called evidence for Wilkins. But all these results rely on the same baseless literary assumption that the young Shakespeare  may safely be left out of all the calculations. Dr Smith explicitly restricts his many analyses to Shakespeare's mature usages, ignoring the far less familiar early style. This is like looking for a lost coin under a lit lamp-post regardless of where it was dropped.
Dr Smith has summed up his results in his latest article. When Wilkins is excluded, Acts III of Pericles prove “to be more akin to Shakespeare than to Chapman, Jonson, Middleton, Tourneur or Webster”. One clear cause of this kinship could be that Shakespeare, at any age, wrote his own works. His ghost writer Wilkins was arguably laid twelve years ago by the rival mathematical methods of Dr A. Q. Morton,  whose stylornetric tests showed “no statistically significant difference” between Pericles I–II and III-V. This result and its rationale have been repeatedly rejected by Dr Smith. Yet he has now cleaned his own conclusion as follows:
Pericles I-II was found as more likely to be Wilkins's work than [the mature] Shakespeare's but the closeness of the result encourages speculation that Shakespeare also had a hand in this part of the text, Indeed, the outcome for the same test of Acts III–V was even a little closer and could be interpreted as revealing the presence of Wilkins in this part. [Emphasis my own.]
Thus all the complex equations are now satisfied by Wilkins = 0. That solution was already required by the rational principle of economy, or Ockham's Razor, which has been left to rust unused in this case as in so very many others. In its absence, the doctors disagree just as sharply about mathematics as about literature. But nobody has yet published any tests of the garbled text of Pericles 1609, which is the actual question at issue. Morton used the anti-Wilkins Cambridge edition of J. C. Maxwell,  while Smith preferred the pro-Wilkins Arden edition of F. Hoeniger. 
Neither the recent Arden reissue nor the Oxford Companion finds it necessary to mention that Hoeniger himself has completely recanted, in a long and thoughtful article  attributing the entire play to Shakespeare. This was soon denounced by Sidney Thomas,  who somehow knew that Shakespeare could not conceivably have perpetrated the least syllable of Pericles I–II at any age or stage. This is where we came in, and where ‘George Wilkins’ came out. He is among the convenient scapegoats to be blamed for any style or text that the experts happen to hate. Yet Acts I–II have never given any such difficulty to ordinary readers, audiences, or viewers, over the centuries and by the million; and it is they who should decide, in the light of historical fact instead of academic prejudice.
Further, every rational Shakespeare-lover, lay or professional, will see that the documentary evidence permits only one hypothesis properly so-called, namely that Pericles was indeed an early Shakespeare play which he later revised. That harmonizes a thick dossier (much compressed here) of diverse data, and also permits objective tests.
 William Shakespeare: The Complete Works, ed. S. Wells, G. Taylor et al. (1986), 1167.
 William Shakespeare: A Textual Companion, ed. G. Taylor et al. (1988), 86-9, 130-1, 556-60.
 R. Prior, ‘The Life of George Wilkins’, Shakespeare Survey, xxv (1972), 137-52; see also his ‘George Wilkins and the Young Heir’, Shakespeare Survey, xxix (1976)
 C. Davenant, Circe, A Tragedy, written c. 1684, when as author was twenty-one.
 Anon., Pimlyco or Runne Red-Cap; see T. Warton, History of English Poetry (1774-81), ii.233, and T. Lowndes, Bibliographers' Manual (1861).
 B. Jonson, Ode to Myself (1629).
 N. Rowe, Works, 6 vols (1709).
 William White printed not only Pericles Q1 (1609) but also Love's Labour's Lost Q1 (1598) and Richard II Q4 16(18); Thomas Creede (not mentioned on the Pericles Q1 title-page, but identified by Peter Blayney, see n. 2 above, 556) also printed Richard III Q2 (1598), Q3 (1602), Q4 (1605), Q5 (1612) as well as Romeo and Juliet Q2 (1599), Henry V Q1 (1600), Q2 (1602) and Merry Wives of Windsor Q1 (1602); Simon Stafford printed not only Pericles Q2 (c. 1609) but also Henry IV Q2 (1599).
 S. Shephard, The Times Displayed, in Six Sestiads (1646).
 J. Tatham, in verses prefixed to Richard Brome's Jove's Crew (1652).
 ‘The Late / and much admired Play, / Called / Pericles, Prince / of Tyre etc / As it hath been divers and sundry times acted by / his Maiesties Seruants, at the Globe on / the Banck-side. / By William Shakespeare.’ So on Q1 (1609), Q2 (c. 1609), Q3 (1611).
 The same title and author appear on Q4 (1619), Q5 (1630, two issues), Q6 (1635).
 G. Wilkins, The Painful Adventures of Pericles, Prince of Tyre / Being / The true History of the play of Pericles, as it was / lately presented by the worthy and an- / cient Poet Iohn Gower (1608); its introduction confirms the King's Majesty's players, i.e. Shakespeare's company, as the actors concerned.
 Another example is The History of Justine (i.e. the Emperor Justinian) (1606), a translation freely plagiarized from an earlier version by Arthur Golding.
 Pericles, ed. J. Maxwell (Cambridge, 1956; rev. 1969).
 G. Bullough, Narrative and Dramatic Sources of Shakespeare (1966), vi.349-564.
 D. Lake, ‘Rhymes in Pericles’, N&Q , ccxiv (1969), 139-43.
 The Travailes of the Three English Brothers (1607), a play by John Day, William Rowley, and George Wilkins.
 M. Smith, ‘The Authorship of Pericles; an initial investigation’, The Bard, iii (1982), 143-76; ‘The Authorship of Pericles: collocations investigated again’, The Bard, iv (1983), 15-21; ‘An Initial Investigation of the Authorship ofPericles’, Shakespeare Newsletter xxiii, 3 (Fall, 1983), 32; ‘Critical Reflections on the Determination of Authorship by Statistics’, ibid., xxiv, I (Spring, 1984), 4-5, 28, 33, 47; ‘Stylometrics '84: A Workshop for Authorship Studies’, ibid., xxiv, 4 (Winter, 1984), 45; ‘An Investigation of Morton's Method to Distinguish Elizabethan Playwrights’, Computers and the Humanities, xix (1985), 3-21, 144; ‘An Investigation of the Basis of Morton's Method for the Determination of Authorship’, Style, xix (1985), 341-68; 'The Authorship of Pericles: New Evidence for Wilkins’, Literary and Linguistic Computing, ii, 4 (1987), 221-230; ‘The Authorship of Acts I-II of Pericles: A New Approach using First Words of Speeches’, Computers and the Humanities, xxii (1988), 23-41; ‘A Procedure to Determine Authorship using Pairs of Consecutive Words: More Evidence for Wilkins's Participation in Pericles’, ibid., xxiii (1989), 113-29; ‘Function Words and the Authorship of Pericles’, N&Q, ccxxxiv (1989), 333-6.
 i.e. in the 1580s, when (as his first biographers Aubrey and Rowe independently testify, after due enquiry in Stratford) he began writing popular plays, such as Titus Andronicus and the first version of 2-3 Henry VI. There is ample additional evidence of the earliest stylistic strata, e.g. in the other first versions currently misclassified as ‘Bad Quartos’ and ‘memorial reconstructions’. Their affinities withPericles I-II are manifest.
 A. Morton, Literary Detection (1978), 184-8.
 Pericles, ed. F. Hoeniger, Arden Shakespeare (1963; rev. 1966, 1984).
 F. Hoeniger, ‘Gower and Shakespeare in Pericles’, Shakespeare Quarterly, xxxiii (1982), 461-79.
 S. Thomas, ‘The Problem of Pericles’, Shakespeare Quarterly, xxxiv (1983), 448-50.