Is “Hand D” of Sir Thomas More Shakespeare’s? Thomas Bayes and the Elliott–Valenza Authorship Tests.
MacDonald P. Jackson
University of Auckland
Jackson, MacDonald P. "Is “Hand
D” of Sir Thomas More Shakespeare’s? Thomas Bayes and the Elliott–Valenza
Authorship Tests." Early Modern Literary Studies 12.3 (January,
2007) 1.1-36 <URL: http://purl.oclc.org/emls/12-3/jackbaye.htm>.
The chronicle-history play Sir Thomas More has come down to us in the form of a manuscript now in the British Library (Harleian MS 7368). The original script is agreed to be in the hand of Anthony Munday. Whether he was also at least part-author is in dispute. John Jowett offered evidence that he collaborated with Henry Chettle. The date of this first version is uncertain, but “between autumn 1592 and mid-1595” has been widely accepted as probable. But the play was heavily revised. Revisions are in several hands, including Chettle’s, Thomas Dekker’s, and probably Thomas Heywood’s. The three pages of additional material by “Hand D” are thought by most Shakespeareans to be Shakespeare’s autograph. If they are indeed his, their interest and value are enormous. Although a ramshackle structure, Sir Thomas More can still entertain an audience. But the possibility that Shakespeare contributed to it, and that his contribution survives in his own handwriting, is what makes the play special. There have, however, always been skeptics about this attribution.
In “Two Tough Nuts to Crack: Did Shakespeare Write the ‘Shakespeare’ Portions of Sir Thomas More and Edward III,” Ward E. Y. Elliott and Robert J. Valenza subject the works indicated in their title to their “silver bullet” methods of testing and calculate probabilities that Shakespeare was the author. For both “dubitanda” the odds are found to be heavily against Shakespeare’s authorship.
By now Elliott and Valenza’s impressive body of research is well known to students of attribution. They devised a series of mini-tests in which various stylistic features are counted. Elliot and Valenza determined, for each of these mini-test counts, ranges within which almost all plays undoubtedly and wholly Shakespeare’s fall but within which many non-Shakespeare plays do not. They have shown that all non-Shakespearean, collaborative, and apocryphal plays have far more “rejections” on these tests than any plays undoubtedly Shakespeare’s alone.
In order to assess the claims of Hand D-plus of Sir Thomas More to be Shakespeare’s, Elliott and Valenza derived from their core Shakespeare plays a sample of ninety verse blocks of about 750 words each, roughly the size of the verse portion of “Hand D-plus,” which includes “Addition III,” More’s soliloquy, which is actually in the scribal “Hand C.” They discovered that, on blocks of text of this small size, ten of their tests gave “good mass discrimination between Shakespeare and non-Shakespeare.” Only three of their ninety Shakespeare baseline verse blocks had two rejections, and so tested falsely as “non-Shakespearean,” whereas 75 per cent of non-Shakespeare blocks had two or more rejections, and so tested correctly as “non-Shakespearean.” Hand D (the verse without Addition III) and Hand D-plus both had two rejections when regarded as written post-1600, and three or four rejections (depending on whether automatic or manual counts of feminine endings were used) when regarded as written pre-1600.
In calculating probabilities Elliott and Valenza used “discrete composite probabilities” and “continuous composite probabilities,” and they applied them to results for all ten tests (or rather nine, since they rightly discarded one) and for a selection of the five that yield “strong disproof” rejections. We can restrict our attention to their data for post-1600 composition, since the evidence for a seventeenth-century date for Shakespeare’s contribution to Sir Thomas More is, in my view, overwhelming. Clearly, if Hand D-plus is Shakespeare’s, Elliott and Valenza’s findings support this belief, since they are more easily reconciled with a post-1600 than a pre-1600 date. For all ten (or nine) tests their discrete composite probability is less than one in twenty that Hand D-plus is Shakespeare’s post-1600 and their continuous composite probability is about one in forty. For the five selected tests it is less than one in twenty on the first way of reckoning (“discrete”) and one in 1,000 on the other (“continuous”). But editors wishing to know whether they should keep including Sir Thomas More in Shakespeare’s Collected Works should not, I think, be guided by any of these probabilities.
This is because in cases such as this it is appropriate to invoke the insights
of eighteenth-century cleric and mathematician Thomas Bayes. Bayesian analysis
of the Elliott-Valenza tests for Hand D and Hand D-plus of Sir Thomas
More produces results much more favorable to a Shakespeare ascription.
Piattelli-Palmarini illustrates how Bayes’s law applies to a case
of testing (or “screening”) for a particular illness, where the test is
less than a hundred per cent reliable.
He bases his discussion on a classic study of the effective diagnostic
capability of mammography in detecting a malignant tumour. A table shows
the percentages yielded by the test of (a) positive results where the
illness is in fact present, (b) negative results where the illness is
in fact present (or “false negatives”), (c) positive results where the
illness is in fact absent (or “false positives”), and (d) negative results
where the illness is in fact absent.
|Illness present||Illness absent|
WA = (0.79) x (0.01) = 0.0079
The next step is to calculate the product of (1) the probability,
in terms merely of the test, of the patient not having the illness,
though she tests positive, and (2) the probability that she does not have
the illness quite independent of the test. In the above table false positives
number 0.1 (10 per cent). The second figure to be multiplied is 0.99 (or
99 per cent), the result of subtracting 0.01 (the probability of having
the illness independent of the test) from 100. Piatelli-Palmarini calls
the product of these two numbers OE (for optimistic error). The result is:
OE = (0.1) x (0.99) = 0.099
Bayes’s formula combines WA and OE as follows, to give the real probability
of having the illness given that the test is positive: WA ÷ (WA + OE). Our
figures are 0.0079 ÷ (0.0079 + 0.099), or 0.0079 ÷ 0.1069, which works out
at 0.0739, or a little over 7 per cent. Despite the positive results on
a test described as “79 per cent reliable,” the chances are better than
thirteen to one that the patient does not in fact have the disease.
|Tests as non-Shakespearean||0.75||0.033|
|Tests as Shakespearean||0.25||0.967|
The importance of Dawson’s study has not been sufficiently appreciated.
His procedure is the logical one to adopt in order to assess whether Hand
D is Shakespeare’s. He began with scrutiny of Shakespeare’s signatures (and
the words “By me” which precede that on the last sheet of his will) in search
of peculiarities of letter formation, and then determined whether any of
them could be found in Hand D’s pages or the 250 early modern documents
used as a control. This is sensible, because the signatures provide such
a small amount of material with which to work and in several respects differ
among themselves. The significance of shared peculiarities can be evaluated,
whereas disparities between the signatures and Hand D cannot, since they
might be accounted for by the contrasting nature of the materials: Hand
D’s pages are the “foul papers” of a play-scene, whereas the signatures
are formal appendages to legal documents, and two of them are constrained
by the width of parchment rectangles, set each within a waxen seal, on which
they are penned. Signatures are, in any case, rather special pieces of writing.
WA = 0.75 x 0.001 = 0.00075
OE = 0.033 x 0.999 = 0.03297
WA ÷ (WA + OE) = 0.00075 ÷ (0.00075 + 0.03297) = 0.00075 ÷ 0.03372 = 0.022
This mode of analysis yields a probability that Hand D is not Shakespeare’s of 0.022 or 2.2 per cent. The probability that it is Shakespeare’s is 0.978 or 97.8 per cent. In other words, the probability that the scene is not Shakespeare’s, once Elliott and Valenza’s testing has been taken into account, is only one in forty-five. The odds remain heavily in favour of Shakespeare’s authorship.
1. Stanley Wells and Gary Taylor, with John Jowett and William Montgomery, William Shakespeare: A Textual Companion (Oxford: Clarendon Press, 1987), 124, where the main facts about Sir Thomas More are set forth (124–5). Vittorio Gabrieli and Giorgio Melchiori have edited the play for the Revels series (Manchester: Manchester University Press, 1990). For Chettle as Munday’s collaborator, see John Jowett, “Henry Chettle and the Original Text of Sir Thomas More,” in Shakespeare and “Sir Thomas More”: Essays on the Play and its Shakespearian Interest, ed. T. H. Howard-Hill (Cambridge: Cambridge University Press, 1989), 131–49. Thomas Merriam has repeatedly argued against Munday’s involvement as author and, sometimes by implication, in favour of Shakespeare’s; see, for example, “Correspondences in More and Hoffman,” Notes and Queries 248 (2003): 410–14, which challenges Jowett’s conclusions; “An Unwarranted Assumption,” Notes and Queries 245 (2000): 438–41; and “Feminine Endings and More,” Notes and Queries 246 (2001): 278–80; “The Misunderstanding of Munday as Author of Sir Thomas More,” Review of English Studies 51 (2000): 540–81.
2. The play was performed successfully by the Royal Shakespeare Company at the Swan Theatre in Stratford-upon-Avon in 2005.
3. Forthcoming in Shakespeare Yearbook and available on the following website: http://govt.claremontmckenna.edu/welliott/UTConference/2ToughNuts.pdf; my references are to this website version, as of September 2005. I am grateful to Ward Elliott for sending me a pre-publication text of the article and for answering a lot of questions about it; also for sending me an offprint of his and Valenza’s “Oxford by the Numbers: What are the Odds that the Earl of Oxford Could Have Written Shakespeare’s Poems and Plays?,” Tennessee Law Review 71 (2004): 323-454, to which they refer in “Two Tough Nuts.”
4. Key articles are “And Then There Were None: Winnowing the Shakespeare Claimants,” Computers and the Humanities 30 (1996): 191–245; “The Professor Doth Protest Too Much Methinks: Problems with the Foster ‘Response’,” Computers and the Humanities 32 (1999): 420–90; “Glass Slippers and Seven-League Boots: C-Prompted Doubts About Ascribing A Funeral Elegy and A Lover’s Complaint to Shakespeare,” Shakespeare Quarterly 48 (1997): 177–207.
5. Although in the hand of a scribe, designated “C,” More’s soliloquy has been judged (rightly in my view) to be in Shakespeare’s style, and is included, along with Hand D’s pages, in, for example, The Riverside Shakespeare: Second Edition, ed. G. Blakemore Evans (Boston and New York: Houghton Mifflin, 1997), and William Shakespeare: The Complete Works, general eds Stanley Wells and Gary Taylor (Oxford: Clarendon Press, 1988). Elliott and Valenza’s “750-word” blocks, restricted to verse, varied a little in actual length, and excluded verse lines that were not pentameters (with or without double endings). The restriction to verse meant that Hand D’s first page of dialogue was not tested.
6. Elliott and Valenza, “Two Tough Nuts,” 16
7. In setting the Shakespearean limits, Elliott and Valenza established the actual ranges for their blocks from core Shakespeare plays (essentially those Folio plays that are not collaborative) and trimmed “outliers.” Their rule of thumb was that the trimming should not place more than 5 per cent of genuinely Shakespearean material outside the Shakespearean limits. The effect was pretty much the same as trimming plays or blocks of material for which the score on a given test was over two standard deviations from the mean, but since scores did not always conform, even approximately, to a “normal” distribution producing a bell-shaped curve, a slightly more subjective procedure was used. The test for hyphenated compound words gave a score for Hand D of 30 and of Hand D-plus of 24, where the Shakespeare range for 750-word verse blocks was set at 26–236. But the scores of 24, 26, and 30 all represent a single instance, the score for HCW (hyphenated compound words) being calculated in terms of the raw figure, divided by the number of words in the block and multiplied by 20,000. Hand D had fewer than 750 words (namely) 660, and Hand D-plus more (832). So Elliott and Valenza were right to ignore Hand D-plus’s HCW score as “a technical rejection” only. Besides, the test strikes me as an unsatisfactory one, too dependent on editorial policy with respect to hyphenation.
8. The case is made by Gary Taylor, “The Date and Auspices of the Additions to Sir Thomas More,” in Shakespeare and “Sir Thomas More,” 101–29, and confirmed by MacDonald P. Jackson, “The Date and Authorship of Hand D’s Contribution to Sir Thomas More: Evidence from ‘Literature Online’,” Shakespeare Survey 59 (2006): 69–78.
9. I ignore Elliott and Valenza’s probability calculations derived from the five tests selected from the original ten, because, although this kind of post hoc analysis may be a graphic way of explaining what they call “the simple essence of our case” (16), it is, strictly speaking, illegitimate as a means of calculating probabilities.
10. Piattelli-Palmerini, 106.
11. Ibid., 204–7.
12. For the purpose of illustrating the application of Bayes’s theorem, it was assumed that “On average, this illness affects 1 percent of the [female] population in the same age group as the patient” and that nothing was known about the patient’s symptoms (81). Of course, were symptoms or knowledge about family history to be taken into account, the pre-test probabilities of the patient’s having the disease would change. Piattelli-Palmerini is simplifying for the sake of illustrating how Bayes’s formula works. Estimating pre-test probabilities for Hand D’s being or not being by Shakespeare more closely resembles the more complex estimating of pre-test probabilites of having an illness that do take account of relevant personal factors.
13. Piattelli-Palmerini, 206.
14. The information that 84 non-Shakespearean blocks were used, kindly provided me by Ward Elliott, is now available in “Oxford by the Numbers,” 441, 445–6.
15. Piattelli-Palmerini, 109.
16. A detailed case was first made in Shakespeare’s Hand in the Play of “Sir Thomas More”: Papers by Alfred W. Pollard, W. W. Greg, E. Maunde Thompson, J. Dover Wilson and R. W. Chambers (Cambridge: Cambridge University Press, 1923). A survey of attitudes to the attribution is given in G. Harold Metz’s “‘Voice and Credyt”: The Scholars and Sir Thomas More,” in Shakespeare and “Sir Thomas More,” 11–56. See also Metz’s Sources of Four Plays Ascribed to Shakespeare (Columbia: University of Missouri Press, 1989), 135-62, and Four Plays Ascribed to Shakespeare: An Annotated Bibliography (New York: Garland, 1982). A good recent summary, stressing the convergence of different kinds of evidence, is by Brian Vickers in Shakespeare Co-Author: A Historical Study of Five Collaborative Plays (Oxford: Oxford University Press, 2002), 35–43.
17. Thompson’s case was made most elaborately in “The Handwriting of the Three Pages Attributed to Shakespeare Compared with his Signatures,” in Shakespeare’s Hand, 57–112.
18. Textual Companion, 124.
19. Carol Chillington, “Playwrights at Work: Henslowe’s, not Shakespeare’s, Book of Sir Thomas More,” English Literary Renaissance 10 (1980): 439–79. This is rebutted by Charles R. Forker, “Webster or Shakespeare? Style, Idiom, Vocabulary, and Spelling in the Additions to Sir Thomas More,” in Shakespeare and “Sir Thomas More,” 151–70, and by Taylor, “Date and Auspices,” in the same volume.
20. Giles E. Dawson, “Shakespeare’s Handwriting,” Shakespeare Survey 42 (1990): 119-28.
21. The exact nature of the famous “spurred a” has often been misunderstood “by inexperienced commentators,” according to C. J. Sisson, who stated that “During some forty years of continuous reading of contemporary manuscripts, amounting perhaps to half a million sheets” he had found no instances of the deposition signature’s peculiar form of “ha” outside Hand D; “Postscript” to R. A. Hubler, “On Looking over Shakespeare’s ‘Secretarie’,” in B. A. W. Jackson (ed.), Stratford Papers on Shakespeare (Toronto: W. J. Gage, 1961), 70–77, at 74. Hubler’s article is on pages 52–70.
22. Dawson, 126.
23. J. Dover Wilson, “Bibliographical Links between the Three Pages and the Good Quartos,” in Shakespeare’s Hand, 113–41.
25. “Literature Online” reveals instances of “scilens” in a poem by John Lydgate (1370–1449 ), Say the Best and Never Repent, and in an anonymous fourteenth-century poem, The Reply of Friar Daw Topias, with Jack Upland’s Rejoinder. In Shakepeare Co-Author, Brian Vickers notes (42) that A. C. Partridge, Orthography in Shakespeare and Elizabethan Drama (London: Arnold, 1964), 62, pointed to instances of “scylens” in Rastell’s 1533 edition of John Heywood’s The Pardoner and the Frere and “scilence” in the manuscript play Thomas of Woodstock, but the latter lacks the crucial “-ns” ending and even the former is not identical to “scilens.” Similarly Metz (“Scholars,” in Shakespeare and Sir Thomas More,” 21, and Sources of Four Plays, 155) claims that “scilens” can be found in Thomas Middleton’s The Puritan (1607) and John Mason’s The Turk (1610), citing F. P. Wilson, Shakespeare and the New Bibliography, ed. Helen Gardner (Oxford: Oxford University Press, 1970), 111, n. 1. But the spelling does not appear in either of those works. Examples of “scilence” (without the “-ns” ending) are not uncommon, and John Heywood has “silens” in his The Four Ps (of about 1522), which is closer to Hand D’s “scilens” than the Heywood example noted by Partridge. But the exact spelling “scilens” has been discovered only in Hand D and 2 Henry IV among works later than the early fifteenth century.
26. Textual Companion, 351.
27. Three examples of the spelling in Thomas Deloney’s Thomas of Reading (1612) are of a proper name. Shakespeare’s Folio-only play Cymbeline, apparently set from a scribal transcript, has one instance of “Iarmen,” where the singular is intended: in fact “a Iarmen on” contains another Hand D spelling, “on” for “one,” which is not, however, uncommon among dramatists.
28. MacD. P. Jackson, “A Non-Shakespearian Parallel to the Comic Mispronunciation of ‘Ergo’ in Hand D of Sir Thomas More,” Notes and Queries 216 (1971): 139.
29. Paul Ramsey, “Shakespeare and Sir Thomas More Revisited: or, A Mounty on the Trail,” Papers of the Bibliographical Society of America 70 (1976): 333–46, at 339. It should be added that Dover Wilson’s evidence extends far beyond these pieces that can readily be assessed by appeal to “Literature Online,” and in “Shakespeare’s Handwriting” Dawson added to Wilson’s argument from misreadings and spellings.
30. Chambers contributed to Shakespeare’s Hand, but his essay achieved its most persuasive form as “Shakespeare and the Play of More” in his own book Man’s Unconquerable Mind (London: Cape, 1939), 204–49.
31. Ibid., 248.
32. Caroline F. E. Spurgeon, “The Imagery in the Sir Thomas More Fragment,” Review of English Studies 6 (1930): 257–70; Karl P. Wentersdorf, “Linkages of Thought and Imagery in Shakespeare and More,” Modern Language Quarterly 34 (1973): 384–305; William H. Matchett, “Shylock, Iago, and Sir Thomas More; with Some Further Discussion of Shakespeare’s Imagination,” PMLA 92 (1977): 217–30, Charles R. Forker, “Webster or Shakespeare?” (see n. 13 above); John W. Velz, “Sir Thomas More and the Shakespeare Canon: Two Approaches,” in Shakespeare and “Sir Thomas More,” 171–95; E. A. J. Honigmann, “Shakespeare, Sir Thomas More and Asylum Seekers,” Shakespeare Survey 57 (2004): 225–35; Honigmann also further supplements Dover Wilson’s case.
33. Thomas Merriam, “Some Further Evidence for Shakespeare’s Authorship of Hand D in Sir Thomas More,” Notes and Queries 250 (2006): 65–8. I am grateful to Dr Merriam for sending me a pre-publication copy of this. In “The Date and Authorship of Hand D’s Contribution to Sir Thomas More: Evidence from ‘Literature Online’”, Shakespeare Survey 59 (2006), 69–78, I offer evidence that, I believe, greatly increases the probability that Hand D is Shakespeare—at least a hundredfold, I should estimate. But I have not taken it into account in the Bayesian analysis undertaken here.
34. One skeptic to have written thoughtfully and at some length is Michael L. Hays, “Shakespeare’s Hand in Sir Thomas More: Some Aspects of the Paleographic Argument,” Shakespeare Studies 8 (1975): 241-53, but this was published long before Dawson’s expert analysis, as was Ramsey’s “Shakespeare and Sir Thomas More Revisited,” which pronounced against the attribution to Shakespeare. Both Hays and Ramsey mention police expert Hubler’s verdict in 1961 that the handwriting evidence “is not sufficiently strong to justify a positive identification of the poet” (“On Looking Over,” 70). Ramsey again reaches a negative conclusion in his more recent “The Literary Evidence for Shakespeare as Hand D in the Manuscript Play Sir Thomas More: A Re-re-consideration,” The Upstart Crow 11 (1991): 131–55, but, as the title indicates, this concentrates on only one aspect of the case, ignoring Dawson’s argument. The same objection applies to Arthur F. Kinney’s “Text, Context, and Authorship of The Booke of Sir Thomas Moore” in Pilgrimage for Love: Essays in Early Modern Literature in Honor of Josephine A. Roberts, ed. Sigrid King (Tempe, Arizona: Arizona Center for Medieval and Renaissance Studies, 1999; Medieval and Renaissance Texts and Studies, 213), 133–60. Kinney (135) quotes an early skeptic, L. L. Schücking, “Shakespeare and Sir Thomas More,” Review of English Studies 1 (1925): 40–59, as protesting that the capital “B” of “By me” in Shakespeare’s will “is so thoroughly unlike the capital B’s in the manuscript, that one is at a loss to understand how they should be written by the same hand.” In fact instances of capital “B” in Hand D’s pages are very like the “B” in the will, once allowance is made for a certain patchiness and lack of fluency in Shakespeare’s final signature. In seeking to counter Chambers’s arguments, Kinney greatly simplifies them. Paul Werstine ignores Dawson’s article in his deeply skeptical “Shakespeare, More or Less: A. W. Pollard and Twentieth-Century Shakespeare Editing,” Florilegium 16 (1999): 125–45.
35. I might add that in the Royal Shakespeare Company’s 2005 production of Sir Thomas More the sudden surge of poetic and dramatic power as the performance reached the “Shakespearean” scenes was obvious.
36. Elliott and Valenza, “Glass Slippers,” 202.
37. Elliott and Valenza, “Two Tough Nuts,” 20.
38. Again, the data, not in “Two Tough Nuts,” were supplied me by Ward Elliott.
39. Elliott and Valenza, “And Then There Were None,” Appendix One (A), 220; repeated in “The Professor Doth Protest,” 461.
40. Elliott and Valenza, “And Then There Were None,” 196; “The Professor Doth Protest,” 432; and especially “Two Tough Nuts,” 41, n. 49.
41. Elliott and Valenza, “Two Tough Nuts,” 20.
42. See “And Then There Were None,” Appendices One (S), and One (C), and One (A), 216–21.
43. “Oxford by the Numbers,” 443–4.
44. The 84 non-Shakespearean samples also draw on few plays: 56 are from Marlowe’s 2 Tamburlaine, Greene’s James IV, Daniel’s Cleopatra, Peele’s David and Bethsabe, Munday’s John a Kent and John a Cumber, Jonson’s Sejanus, and Fletcher’s The Woman’s Prize; and 28 are from non-dramatic verse blocks by fourteen different poets. None of the 90 Shakespearean blocks are from his poems.
45. Frederick Mosteller and David L. Wallace, Inference and Disputed Authorship: The Federalist (Reading, Mass.: Addison-Wesley, 1964); 2nd edn. as Applied Bayesian and Classical Inference: The Case of the “Federalist” Papers (New York: Springer-Verlag, 1984). There is a judicious assessment of the Elliott–Valenza tests by Gray Scott, “Signifying Nothing? A Secondary Analysis of the Claremont Authorship Debates,” Early Modern Literary Studies 12.2 (September, 2006) 6.1–50 <URL: http://purl.oclc.org/emls/12-2/scotsig2.htm>. But Scott does not consider the Clinic’s handling, in their most recent work, of 1500-word and 3000-word samples.
46. Elliott and Valenza, “Two Tough Nuts,” 34.
Responses to this piece intended for the Readers' Forum may be sent to the Editor at M.Steggle@shu.ac.uk.
© 2007-, Matthew Steggle (Editor, EMLS).