As it turns out, though, all of the above claims are false. Specifically:
There are two separate lists, to make it easier to test the claim that the names of "the Stratford man" and the playwright were distinct and spelled differently. The first list consists of non-literary references to William Shakespeare of Stratford; these include records from Stratford and its environs, as well as various London records, including those relating to his career as an actor and shareholder in the Globe and Blackfriars theatres. The second list consists of references to Shakespeare as a poet and/or playwright up to the publication of the First Folio in 1623; this includes both direct references to Shakespeare by other writers as well as mentions of his name on title pages of quartos, with each edition of a work counting as a separate reference. Some references could be taken as belonging on either list; I have generally put in the second ("literary") list any references which allude to Shakespeare as a playwright/poet, regardless of other content it may have. For instance, the references in "The Return from Parnassus, Part II" and in John Davies's epigram of 1611 are in the "literary" list because they praise Shakespeare as a playwright/poet, even though they also allude to him as an actor and member of the Chamberlain's/King's Men. The following tables give a breakdown of how many times each of the various spellings of Shakespeare's name occurs in the documents that have survived, from 1564 through 1616, the year of his death. [note1] Abbreviated versions are listed only if they include more than half of the name; thus "Shakespe" is included, but "Shak" is not. I should emphasize that the figures in Table 1 refer only to non-literary references which can be reasonably taken to refer to William Shakespeare of Stratford, and that literary references to William Shakespeare as a poet/playwright are summarized in Table 2. Each of the tables is divided into three columns. In Table 1, the first column lists the total occurrences of each spelling, and the other two columns break these down into non-London (mostly Stratford) and London occurrences. In Table 2, the first column also lists total occurrences, while the other two columns break these down into printed vs. handwritten occurrences. At the bottom of each table I have tabulated the total occurrences of spellings with the first 'e' (e.g. "Shakespeare") vs. spellings without the first 'e' (e.g. "Shakspere").
Outside In Total London London ----- ------- ------ Shakespeare 71 8 63 Shakespere 27 25 2 Shakespear 16 16 0 Shakspeare 13 9 4 Shackspeare 12 11 1 Shakspere 8 7 1 Shackespeare 7 7 0 Shackspere 6 5 1 Shackespere 5 5 0 Shaxspere 3 3 0 Shexpere 2 2 0 Shakspe~ 2 0 2 Shaxpere 1 1 0 Shagspere 1 1 0 Shaksper 1 1 0 Shaxpeare 1 1 0 Shaxper 1 1 0 Shake-speare 1 0 1 Shakespe 1 0 1 Shakp 1 0 1 with first 'e' 128 (71%) 61 (59%) 67 (87%) w/o first 'e' 52 (29%) 42 (41%) 10 (13%)
Hand- Total Printed written ----- ------- ------- Shakespeare 119 107 12 Shake-speare 21 21 0 Shakspeare 10 5 5 Shaxberd 4 0 4 Shakespere 4 1 3 Shakespear 3 1 2 Shak-speare 2 2 0 Shakspear 2 0 2 Shakspere 1 0 1 Shaksper 1 0 1 Schaksp. 1 0 1 Shakespheare 1 1 0 Shakespe 1 0 1 Shakspe 1 0 1 with first 'e' 149 (87%) 131 (95%) 18 (55%) w/o first 'e' 22 (13%) 7 (5%) 15 (45%)
It is clear from Tables 1 and 2 that, at least by modern standards, there was considerable variation in the spelling of Shakespeare's name in both literary and non-literary contexts. In both contexts, though, "Shakespeare" was the most common spelling by a wide margin, and in both contexts, variants with the first 'e' ("Shakespeare," etc.) greatly outnumber variants without the first 'e' ("Shakspere," etc.). The latter type are somewhat more prevalent in non-literary contexts (29% vs. 13%), a fact which Oxfordians might seize upon as some small support for their theories. However, there are a variety of factors which need to be taken into account, factors which upon examination make it clear that the evidence gives no support to the claim that "Shakespeare" and "Shakspere" were distinct names, referring to playwright and Stratford citizen respectively.
First, spelling in London tended to be more uniform and modern than in the rest of the country, where regional dialects and lower literacy rates made it more idiosyncratic. Note that among the non-literary references from outside London, "Shakspere" spellings are fairly prominent (41%), and this group includes the more idiosyncratic spellings often ridiculed by Oxfordians, such as "Shagspere" and "Shaxper." Among non-literary references from London, though, "Shakespeare" is the overwhelmingly preferred spelling (63 out of 77). The percentage of "Shakespeare"-spellings among non-literary references in London (87%) is comparable to the percentage among literary references (95%), nearly all of which are also from London.
Another very significant factor is handwritten vs. printed spellings. Spelling in Elizabethan printed texts was much more uniform and closer to modern practice than in handwritten ones, because compositors tended to normalize idiosyncratic features of the manuscripts they worked from. This factor can clearly be seen in the breakdown of literary references in Table 2. The spelling "Shakespeare" and its hyphenated variant "Shake-speare" were overwhelmingly preferred in printed texts, but among handwritten references they were much less prevalent, with "Shakspere"-spellings (without the first 'e') almost as common as "Shakespeare"-spellings (15 versus 18). Given this pattern, it is actually somewhat surprising that there are not more "Shakspere"-type spellings among the non-literary references, all but two of which are handwritten.
A specific example illustrates more forcefully the difference between printed and handwritten spellings. We have four surviving contemporary records where someone recorded his purchase of one of Shakespeare's printed works while noting the author's name; in each case the writer spelled the name without the first 'e', even though in three of the four cases the corresponding printed work spells the name "Shakespeare":
Surely these entries indicate that "Shakspere," "Shaksper," "Shakspear," and "Schakspe(a)re," when they happened to appear, were just seen as variants of "Shakespeare," and that nobody gave them a second thought.
Still another factor to be considered is the nature of the documents themselves. Legal documents such as contracts, which might have to be referred to decades later, tended to be more carefully written than informal jottings such as those of Stonley, Alleyn, and Drummond. It is thus worth noting that the documents relating to Shakespeare's property purchases, both in Stratford and in London, invariably spell the name "Shakespeare" or its close variants. For example, both of the documents relating to Shakespeare's purchase of New Place in 1597 consistently have "Shakespeare"; the two documents concerning his 1602 purchase of the Old Stratford freehold consistently have "Shakespere"; the indenture for his 1605 purchase of tithes has "Shakespear" throughout; both the conveyance and the mortgage from his 1613 purchase of the Blackfriars Gatehouse consistently have "Shakespeare." It is clear that from a legal standpoint, the man's name was considered to be "William Shakespeare," not "Shaksper." [note 2]
I have tried to be as complete as possible in compiling the accompanying lists; others have sometimes tried to make arguments based on selective or incomplete lists of references, which can lead to a distorted picture. I have also tried to be fair and reasonable, but inevitably some people may object to one aspect or another of the procedure; I will not attempt here to rebut every possible objection. [note3]Arguments about the spelling of the name are ultimately less important than it might seem, though, because as I will argue in the next section, there is no evidence that "Shake-" and "Shak-"-type spellings represented different pronunciations, while there considerable evidence that they were seen as interchangeable.
Similarly, all indications are that the spellings "Shakespeare" and "Shakspere" (and their variants) did not represent a consistent pronunciation difference, despite our intuition based on modern spelling rules. The foremost authority on the subject, Fausto Cercignani, says in Shakespeare's Works and Elizabethan Pronunciation (p. 1) that "we do not even know how Shakespeare pronounced his own surname, since the first part of his signature (Shakspere and Shakspeare) may imply either the antecedent of present shake or a variant pronounced like shack, while both -spere and -speare may conceal either the antecedent of present spear or a form rhyming with pear." It is entirely possible that the name was pronounced differently in Stratford and London, and this may be one more factor among several contributing to the greater occurrence of "Shak-"-type spellings in Stratford. (Though recall that in London, "Shake-"-type spellings were used more often in non-literary referernces to "the Stratford man" than in literary references to Shakespeare.) If we look at Stratford and London references separately, though, the distribution of "Shake-" and "Shak-" spellings gives no indication that they represented any consistent difference in pronunciation.
Furthermore, there is independent evidence that "Shake-" and "Shak-" in this context were seen in Shakespeare's day as interchangeable spelling variants. The anonymous play Arden of Feversham was first printed in 1592. One of the villians of the play is a ruffian and hired murderer whose name is given in modern editions as "Shakebag," based on an actual historical figure named George Shakebag; the similarity to "Shakespeare" makes comparison relevant. In the 1592 Quarto, this character's name appears 43 times (not counting abbreviations), and it is spelled six different ways, broken down as follows:
Shakbag (25 times) Shakebag (10 times) Shakebagge (5 times) Shakbagge (1 time) Shackbag (1 time) Shakabage (1 time)
No rhyme or reason is apparent in the use of these various spellings; their distribution is essentially random. The name is spelled "Shakbag" on the title page, but in the character's first appearance in the text, in a stage direction in line 665, he is "Shakebagge"; in line 829 the name is "Shakebagge," and in the very next line it is "Shakbag"; in line 1234 it is "Shakebag," but two lines later we get "Shakbag." (Line numbers are taken from the 1947 Malone Society facsimile of the 1592 Quarto.) There is no indication at all that any of these spellings represent different pronunciations; it is apparent that the first syllable of the name was pronounced identically (presumably like the word "shake," however that may have been pronounced in Elizabethan times) whether it was spelled "Shake," "Shak," or "Shack."
Matus (p. 26) provides some similar evidence, involving the playwright Shakerley Marmion. On the title page and in the author's dedication of Marmion's Holland's Leaguer, his first name is spelled "Shackerley." The next year, on the title page of his A Fine Companion, the name is spelled "Shakerley," but the author's dedication is signed "Shack: Marmyon." On the engraved title page of Marmion's poem Cupid and Psyche, his first name is spelled "Shakerley," but on the typeset title page it is "Shackerley." We cannot be certain whether Marmion pronounced the first syllable of his own name as "shake" or "shack," but whatever the pronunciation, it is evident that "Shack-" and "Shak-" were seen as equally valid spellings for it.
There is little or no evidence to support the common Oxfordian assertion that "Shakspere" always required a "short" 'a' pronunciation while "Shakespeare" always required a "long" 'a.' Rather, the evidence indicates (as near as we can tell 400 years later) that these were alternate spellings for a single pronunciation, though that pronunciation may have varied regionally. The variant spellings may have sometimes indicated different pronunciation, but we have no way of reconstructing with any confidence when this was.
Despite Oxfordian claims to the contrary, it was not at all unusual for proper names of real people to be hyphenated in print in Elizabethan times. The following is a very partial list of names of real people which were hyphenated on title-pages of printed works between 1570 and 1640; some of these are taken from Irvin Matus's Shakespeare, In Fact(pp. 28-30), while others are culled from my own research with the Short-Title Catalogue, the National Union Catalogue, and the OCLC database.
In fact, the pattern seems to be that a name could be hyphenated (according to the whim of the printer) if it could be seen as divided into two parts; most often one or both of these parts were English words, though in "Fitz-Geffrey" the "Fitz-" is a Norman French prefix meaning "son of." As the above examples show, whether a name was real or pseudonymous had nothing to do with the matter; real names were hyphenated just as easily as pseudonyms (perhaps even more easily). "Shakespeare" obviously fits this pattern; the name was taken to consist of "shake" + "spear," even though E. K. Chambers (EKC II: 374-5) doubts that this is its actual historical origin. Contemporary evidence that this was how people thought of the name can be found in William Camden's Remaines, first published in 1605. Camden had a long section on the origins of English names, and at one point he says that some men derived their names "from that which they commonly carried, as Palmer, that is, Pilgrime, for that they carried Palme when they returned from Hierusalem, Long-sword, Broad-speare, Fortescu, that is, Strong-shield, and in some such respect, Breake-speare, Shake-Speare, Shotbolt, Wagstaffe." Camden here not only confirms that the name was thought of as "shake" + "spear," but he hyphenates it along with several of the other names. At least one more tidbit of evidence that the family name had the "spear-shaking" interpretation long before William was born can be found in the records of Richard Shakespeare, William's paternal grandfather. Richard was called "Richard Shakstaff" in a 1533 record; some scribe apparently wrote "staff" instead of the semantically similar "spear" as part of the name.
In sum, there is no evidence to support the Oxfordian assertion that the occasional hyphenation of the name Shakespeare means that people thought of it as a pseudonym. Real names were occasionally hyphenated when they could be divided into two parts; the same is true of fictitious names. The best-known pseudonym of the time, Martin Marprelate, was only occasionally hyphenated, while names of several real people (such as Charles Fitz-geffrey and Robert Walde-grave) were hyphenated with great regularity.