Richard III 1597, by William Shakespeare (1564-1616)

A quatercentenary contribution

1997, previously unpublished; © the estate of eric sams

Richard III was first published anonymously in a 1597 Quarto edition [1] (Q1). It proved popular, and was duly reprinted seven times, between 1598 and 1634, as the work of William Shakespeare. But nowadays it is universally condemned [2] as a bogus corruption compiled by Shakespeare's own company from their collective memory of the very different version unknown until the 1623 First Folio. For this purpose, editors have always felt free to antedate that latter text by some twenty years, and stigmatise hundreds of Tudor worthies (including Shakespeare himself, though this is never mentioned) as conniving knaves or duped fools. All this happens at the behest of a notion that no one so much as mentioned for three centuries, and which derives solely from literary theory, without reference to any known fact or actual person at any time.    

     So far, however, all refutations have proved futile. Such theories are deemed too technical for the laity to consider and too settled for the profession to reconsider. But it is time to try again, starting from the documents. These record [3] that Shakespeare's own accredited signatures use three different symbols for each one required. The same Protean profusion also occurs in the manuscript insurrection scene he contributed to the play Sir Thomas More, [4] where that surname appears as moor/more/moore and the word shreeve (=sheriff) as shreef/Shreiff/shreeve/Shreiue/ Shreue, all within five lines. About one word in every ten is thus variably spelt. But its pronunciation surely stayed the same. If so, the writer regarded o/oo, e/ee/ei and f/ff/ue/ve as phonetic equivalents. The same applies to a/au in Comand/Comaund or in seriant/Seriant/seriaunt (=sergeant). So Shakespeare predictably spoke broad vowels and soft consonants, like a Warwickshire countryman. As those examples show, he also felt free to double or halve his medial consonants, or capitalise initial letters. The More manuscript also abounds in old-fashioned forms like y for i, and in abbreviations found in legal documents, such as a single symbol for per or par, eccentrically inserted as the first syllable of 'parsnips'. Shakespeare also used that same symbol, just as quirkily, to signify the second syllable of his own name in one of his signatures.

     Old-fashioned eccentric phonetic provincial spellers born in the 1560s were likely to have been educationally disadvantaged, just as Shakespeare's contemporaries and first biographers said he was [5] Conformably with the same clear consensus from the same sources, reinforced by modern palaeographic expertise, the author of the More scene had been a legal penman. [6] Such spellings as 'aucthoryty', 'maiestrate' or 'scilens' seem to confirm his lack of Latin; a practised Latinist might well have recalled auctoritas, magistratus, silens.

     Of course many of Shakespeare's idiosyncrasies were normalised, as the early printings prove. Such changes would be made by actual compositors, rather than hypothetical 'copyists'. But both categories would be paid to reproduce the manuscript before them, not to introduce their own personal preferences. So the many variants found throughout the early editions are prima facie authorial; they had either remained unnoticed or (at a time of widely diverse orthography) been adjudged acceptable.

     Those that occur in the 1597 text of Richard III exemplify the same broad vowels (a/au, o/ou) and other equivalents (ee/ei, f/ff/ue, i/y, arbitrary capitalisation) as the More manuscript, sometimes in the same words (Comaund, lyue, wyth). But the crucial evidence for a Shakespeare holograph source behind Richard III 1597 consists in that edition's remarkably close kinship with the first edition of Richard II, [7] also published in 1597, after licensing by the same senior stationer (Warden Mann) and issue by the same superior Shakespeare printer (Valentine Sims) and the same irreproachable publisher (Andrew Wise). Their hundreds of identical spellings have escaped editorial attention, because Richard III1597 has been rejected as a mere reported piracy; [8] but they can hardly be mere coincidence. The true coincidences occur when both plays happen to use the same word; then the shared variant spellings become evident, and evidential.

     They include a dozen triple variants, such as Coosen/coosin/Cosen, daie/day/daye, leige/liege/Liege, pitie/pittie/pitty, stir/stirre/stur, told/tolde/tould. Both plays thus contain the same unpredictable admixtures as the More scene: capital letters, broadened vowels, doubled consonants, interchange of ie with ei or y, and so forth. Further, both plays sharenearly two hundred other alternate spellings, from aire/ayre to yong/young; and many of those alternatives would be rare singly, let alone in pairs, e.g. adiew/adue, cease/seaze [=seize], inforce/inforst. These examples illustrate other Shakespearean equivalents, also found in the More scene, such as c/s/z, ea/ie, and en/in, as well as his formation of past tenses in -t. All these usages are further exemplified many times in both the Richard plays. And what of all their unusual yet shared spellings of single words, such as allegeance, atturnies, caitiue (=caitiff), fact (=faced), maner (=manner), modle (=model), Norffolke, of (=off), sculs (=skulls), and as many as one hundred and seventy others, including the broad vowels in affoord, foorth, auncestors, daungerous, demaund, Fraunce, launce, slaunder, bould, could (=cold), souldier? Further, such shared spellings as annointed, forreine, mettall, mittigate, prophane may again confirm a certain unfamiliarity with the corresponding Latin roots (inunctus, foras, metallum, mitigare, profanus).

     All such occurrences in Richard II 1597 are readily explicable as typical authorial eccentricities. But they are equally copious in Richard III 1597, which also includes such Shakespearean spellings as a farre (afar), alablaster, Albones (Albans), a sleepe (asleep), ceaze/seaze (=seize), centinell, Dutchesse, ghesse (guess), hower (hour), Phisitions (physicians), president (precedent), sabaoth (=sabbath), scrowle (scroll), tayler (tailor), waight (wait), wrinckle.

     Yet more massive reinforcements have now arrived from an unexpected quarter. Emeritus Professor Ernst Honigmann has recently published a seminal account [9] of Shakespearean spellings in Othello 1622, on the strength of which that edition is inferred to have been printed from copy close to the poet's own foul papers or indeed from his own manuscript. In an appendix, Honigmann lists eighty-seven specific spellings and five typical general categories. He concedes that not all the spellings are uncommon, but claims that their large overlap with Hamlet Q2, agreed to have been printed from authorial copy, cannot be merely coincidental. 'To put it another way', he explains, 'a decided preference for shew (instead of show), or for vertue (instead of virtue), or for sence (instead of sense) was not unusual, taking each word individually. But how many other writers shared Shakespeare's preference for shew and vertue and sence and all the other strong or occasional preferences listed above?'.

     One immediate and manifest answer is the writer of the copy used for the first edition of Richard III 1597. Shakespeare was an extremely variable speller; but of the forty words common to Othello 1622 and Richard III 1597, no fewer than twenty-six correspond letter for letter with Honigmann's Shakespearisms, including such oddities as 'ghesse'. In particular, the Richard III writer exhibits exactly the preference described, for shew (and shewes), vertue (and vertuous) as well as sence. As to Honigmann's five general categories of characteristic spelling, the correspondence could hardly be closer. The first is the 'very frequent substitution of y for modern i', as exemplified by ayme and ayre in Othello 1622; Richard III has twenty-five such words, including ayming and ayre. Next comes -oo- for -o-, as in approoue; Richard III has eleven examples, including prooue. The third item is 'the doubling of consonants'; only medial rr is cited, of which there are a dozen examples. Fourthly, Shakespeare is said to put in- for en-, as in indure or inforce; Richard III contains thirteen such examples, including those two, and if im- for em- were added the tally would rise to sixteen. The fifth and last category is -full for -ful, as in faithfull, fearefull, lawfull, lustfull, pittifull, powerfull, etc.; Richard III has all six of those examples and twenty others. Honigmann also points out that the 1623 Folio text of Julius Caesar, like Othello 1622, contains Battaile and Battell, both Shakespearean spellings of battle; but so does Richard III1597. Hongimann further draws attention to 'Shakespeare's nd- ending' in past tenses; Richard III 1597 has dand (=damned).

     Now, how are all these precise and profuse parallels to be explained, save by the simple self-evident solution that Richard III 1597 represents an authentic Shakespeare play, derived from his own manuscript? The latter cannot, therefore, be a 'collective report made from memory by the entire company', unless of course Shakespeare had consented not only to participate in this venal conspiracy to sell a bogus imitation of his own play but also to copy out the resulting corruption in his own hand. But not even Academia will venture to advance any such absurdity. So modern editors should now abandon their futile fantasy that Richard III 1597 is a 'memorial reconstruction'. All it proves is their own collective inability to tell the difference between a Shakespeare masterpiece and a botched corruption. And indeed this defect is openly conceded by the profession itself, which has confessed [10] to making the same mistake, for most of this century, about the 1608 text of King Lear.

     In that case, a majority verdict has been returned; the first edition was in fact an early version. But what of all the many other so-called memorial reconstructions and apocrypha? Their spellings and variants can be objectively analysed in accordance with this proposed new approach, which confirms for example the integral authenticity of Edward III 1596 (now included in the Harvard Riverside Shakespeare). All such texts deserve a new look in a new light, after so many years laid in darkness.


[1] often reprinted in facsimile, beginning with Richard the Third, by William Shakespeare, ed. P. A. Daniel, London, c.1885.

[2] e.g. by Taylor and Wells, in A Textual Companion to the Oxford Shakespeare, Oxford, 1988, 228-232, and by H. Baker, in The Riverside Shakespeare 2/1997, 748.

[3] see e.g. E.B. Everitt, The Young ShakespeareCopenhagen 1954, 82-96.

[4] often reprinted, notably 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 & R. W. ChambersCambridge, 1923. 

[5] in testimony recorded from 1592, as summarised in E. Sams, The Real Shakespeare, New Haven and London 1995, 2/1997, 201-2.

[6] e.g. by Sir E. Maunde Thompson, as cited at note 3 above, 70 etc. 

[7] first reproduced as King Richard the Second by William Shakespeare, ed. P.A. Daniel, London, 1890, and concorded in The Oxford Shakespeare Concordances, ed. T. Howard-Hill, Oxford, 1971.

[8] and hence never publicly concorded; the present essay relies on private endeavour.

[9] E. A. J. Honigmann, The Texts of Othello and Shakespearean Revision, London, 1996.

[10] in A Textual Companion, cited at note 2 above, 509.