Shakespeare and the Early Modern History Play

This post is a long excerpt from an essay of mine published a few months ago, “‘But, what euer you do, Buy’: Richard II as Popular Commodity” (in Jeremy Lopez, ed., Richard II: New Critical Essays, London and New York: Routledge, 2012, 223-44). You’ll find the concluding sections and full citations in the book — a volume anyone interested in Richard II or the history play in general should have a look at in any case, for Jeremy’s incredibly thorough introduction if nothing else. I thought it would make sense to post my opening salvoes here, though, as I hope they’re revisionist enough to stir up a bit of debate. Comments would be emphatically welcome!

Here goes.

On 6 February 1601, the eve of the Earl of Essex’s rebellion, some members of his circle asked the Lord Chamberlain’s Men to stage a special performance of a play about “the deposyng and kyllyng of Kyng Rychard the second.” When told that the play was “so old & so long out of vse as that they shold have small or no Company at yt,” they offered a top-up payment of 40 shillings, finally persuading the players to revive the dusty and unpopular show the next day (Chambers 1930: 2:325). Most scholars assume that Essex’s friends were asking for Shakespeare’s Richard II. The anecdote is well known, likely the most famous historical reference to a play about Richard II. It has occasioned much debate on the relationship between early modern theatre and politics, the relative subversiveness or conservatism of the Chamberlain’s Men (and Shakespeare as their leading playwright), and the power or failure of censorship. But to my mind, what is most telling and remarkable about the episode is not what it says about the political impact or powerlessness of the theatre, but what it can tell us about early modern notions of popularity.

It is by no means clear that the 1601 play was Shakespeare’s. We have only a very limited and fragmentary picture of the Chamberlain’s Men’s repertory, after all, and it is not inconceivable that they owned other works dramatizing the king’s deposition and death.[1] But as E. K. Chambers argued (1930: 1:354), Shakespeare’s play fits the description “from the theatrical point of view,” in which a five or six year old text would have to be considered “stale” and likely in need of revision. Then again, if Richard II was “long out of vse” by February 1601, why would Essex’s fellows request it? If their objective was to stir up revolutionary sentiments among Londoners, why choose a long-forgotten play? The answer may be that from perspectives other than that of the theatre, Richard II was anything but a distant memory. As a printed playbook, it was outselling almost all of Shakespeare’s other works by 1601, having reached its third edition a few years earlier in 1598. The only play of his that could keep pace was 1 Henry IV, whose third quarto had appeared in 1599. In fact, most booksellers in February 1601 would likely have identified Shakespeare’s two histories as near-unrivalled bestsellers among plays, matched only by Marlowe’s Tamburlaine and Kyd’s Spanish Tragedy, and most recently by Ben Jonson’s unlikely blockbuster Every Man Out of His Humour, reprinted three times in 1600.

Plays from the professional theatres were no longer the unusual commodity in stationers’ stalls and shops in 1601 that they would have been ten years earlier. 96 editions of 67 individual texts had appeared over the previous decade, and a market for printed drama had come into existence (see Lesser, forthcoming). But this market was young enough that few plays had established themselves as enduringly attractive to readers. By 1601, 22 editions had sold out and been reprinted, but only the five above (and Lyly’s Campaspe) were popular enough to have been issued a third time. Remarkably, of those 22 reprints, nine—over 40 per cent—were history plays.[2] Not only did such plays make up a larger proportion of the printed drama than in later decades (38 per cent, or 22 out of 67 published plays), they also sold better than other genres (40 per cent of those first printed in the 1590s had been reissued by 1601, whereas only 27 per cent of other plays were reprinted in the same time frame).

The Essex anecdote thus highlights a crucial aspect of the reception history of early modern drama in general, and of the Elizabethan history play in particular: success on stage and success in print did not necessarily develop along parallel lines. What was old and “long out of vse” in one format or venue might be newly or enduringly popular in another. The Spanish Tragedy is another case in point: Kyd’s play also seems to have fared rather less well in the theatre than in Paul’s Churchyard. By the late 1590s, the former box office champion had dwindled into near-insignificance, generating below average revenues for Philip Henslowe, the owner of the Rose Theatre, where the Admiral’s Men had revived the play in 1597 (see Syme 2010: 510). In printed form, however, The Spanish Tragedy sold well enough to merit a first reissue in 1594 and a third edition, two years after its theatrical demise, in 1599.[3]

In what follows, I will consider the popularity of Richard II from two separate but related angles, first by broadening my vista to discuss in more detail the enduring success of history plays in general, and of Shakespeare’s history plays in particular; and secondly by asking what specific features of Richard II might have predestined it for success as a book.

II

The dominant narrative of the history play’s development as a genre has largely remained unchanged and unchallenged in the twentieth century: a peculiarly English kind of drama, it was invented by playwrights in the 1580s, initially associated primarily with the Queen’s Men; it came into its own when Shakespeare became the most prolific author of histories; and it faded out of fashion and into irrelevance soon after James I came to the throne. Theatre historians and literary scholars agree on this account to an unusual extent. To cite just a few representative examples: Leonard Tennenhouse (1994: 110) argues that

a whole set of literary genres fell out of favour with the accession of James I…. Along with such forms as romantic comedy, Petrarchan poetry and prose romance, the chronicle history play enjoyed a period of unprecedented popularity during the 1590s. And, like so many other literary forms that had been popular in the last two decades of the sixteenth century, chronicle history plays with few exceptions simply ceased to be written after 1599, the year Henry V was produced.

Phyllis Rackin (1990: 31) maintains that

the English history play ceased to be a popular genre soon after Shakespeare abandoned it, and there is good evidence that the genre itself is largely Shakespeare’s creation…. [T]he taste for historical drama reached its peak around the end of the sixteenth century and declined soon after the accession of James I.

And most recently, Benjamin Griffin (2001: 145) has affirmed the “genre’s decline in terms of new compositions,” arguing that it “was no longer popular because its sense of dramatic shaping leaves certain desires unsatisfied.” For him, the question is not why (or if) “the histories died out in the 1600s, but why they flourished in the 1580s and 1590s.”[4]

At first blush, publication patterns support this account. In the 1590s, 19 history plays were printed for the first time; that number fell to 14 in the first decade of the seventeenth century; and by the 1610s, only a single new history play seemed worth publishing. At the same time, other plays became more and more dominant, rising from 34 in the 1590s to 81 in the 1600s, and then dropping sharply, but less sharply than chronicle drama, to 31 playtexts in the 1610s. In relative terms, history plays made up 36 per cent of new plays first published in the 1590s, 15 per cent in the 1600s, and a mere three per cent in the 1610s (fig. 1).

Figure 1

This count, however, does not tell the whole story. Although the number of first editions of new chronicle plays plummeted after 1605, histories retained a healthy share of the overall market in playbooks, because, as I already noted above, they were reprinted at higher rates than others—and their popularity as reprinted texts did not dwindle drastically between 1590 and 1620. While in the 1590s, unsurprisingly, there were only eight reissues of history plays, 14 were reprinted in the 1600s, and 17 in the 1610s (fig. 2). The overall number of playbooks (as opposed to individual titles) drawn from recent, usually English, history thus rose slightly from 27 in the 1590s to 28 in the 1600s, and dropped rather less precipitously to 18 in the 1610s.[5] At the same time, all other plays were reprinted 15 times in the 1590s, 27 times in the 1600s, and 44 times in the 1610s; the total of non-historical playbooks climbed from 49 in the first decade to 108 in the 1600s, and then declined again to 75 in the 1610s (fig. 3). History plays thus accounted for 36 per cent of all playbooks published in the 1590s, 21 per cent in the 1600s, and 19 per cent in the 1610s.

Figure 2

Figure 3

From this perspective, the genre looks quite a bit less moribund than the received account would lead us to expect. Readers remained eager to buy plays on English history well beyond the 1610s; more remarkably, perhaps, the same texts continued to sell for decades. Unlike the market for other plays, which could rely on no more than a few enduringly popular and frequently reissued books and was driven by a steady stream of new texts of which only about a third were reprinted, the market for history plays was dominated by a comparatively small set of texts that saw far more editions than the average non-history play—almost 60 per cent of all chronicle dramas published by 1620 were reprinted.

These divergent situations should encourage us to question to what extent the trade in printed drama actually reflected developments in the theatres. If non-historical plays had less staying power than those based on the chronicles, and thus had to be supplied to the reading public in greater numbers and at a steadier rate, the trade in those plays resembled quite closely the cycling in and out of new offerings in playing companies’ repertoires. On the other hand, the less volatile market for history plays may not be as representative of theatrical practices: if stationers’ customers still bought new editions of plays such as Richard II years after they had gone stale on stage, we can learn less about the theatrical fortunes of the genre as a whole from its development as a genre of printed drama than we can learn from other playbooks about the evolution of repertories in the 1590s and early 1600s. New chronicle plays may well have been written and performed in those decades, but with a market saturated in still-popular older texts, booksellers may have had little incentive to invest in new offerings in the same genre.[6] The economic logic of a market partly driven by strong reprint rates does not accord well with the imperative for constant innovation that propelled the continued evolution of company repertories. In other words, the publication data suggest two things. First, that while the history play may have been a dead-or-dying genre of performance by 1603, it nevertheless remained attractive for readers well into the 1620s. Second, the very notion that playwrights ceased producing history plays after James I came to the throne may need to be re-examined, as it relies on the shaky assumption that what was published can be understood as broadly representative of what was staged. If I am right about the unusual divide between the enduring popularity of individual history plays with readers and their comparatively normal (and hence relatively short) lifespan in performance, the history of the genre constructed through publication data may only intermittently relate to the development of the genre in the theatre.

III

There is one other reason to suspect that the sample of history plays surviving in print is not in fact representative of what audiences could have seen in the theatres. That reason can be found in the conventions of the early modern book trade. If further works on English history were written and staged after the 1590s, those conventions may have made it impossible for stationers to market these new plays—and consequently, they may not have been seen as particularly promising investments.

As Peter Blayney (1997: 399) has argued, the owner of a particular title

had not only the exclusive right to reprint the text, but also the right to a fair chance to recover his costs. He could therefore seek the [Stationers’] Company’s protection if any book—not necessarily a reprint or plagiarism of his own copy—threatened his ability to dispose of unsold copies of an existing edition.

In other words, if a play on a particular subject had already been published, that fact alone made it more difficult for another stationer to publish a new play on the same subject.[7] Blayney’s primary example for how this rule would have been implemented in practice is a history play: Henry V, which competed with The Famous Victories of Henry the Fifth, the Queen’s Men’s play first printed in 1598. Before Thomas Millington and John Busby could publish Shakespeare’s version, they would have had to make arrangements with the owner of the older play, the printer-publisher Thomas Creede. Apparently they succeeded: Creede was hired to produce both the first and second quartos of Henry V, presumably the price for his consent. Richard Dutton (2009: 144n25) has recently expanded on Blayney’s hypothesis by suggesting that there might be a similar connection between the fact that Creede owned the copy for The True Tragedy of Richard the Third and was paid to print all but the first of the early quartos of Shakespeare’s Richard III for Andrew Wise.

If Blayney’s interpretation of the Stationers’ Company’s rules is correct, any publishers who wanted to issue a history play on the same or a closely related subject as one already in existence would have had to come to terms with the owner of the earlier text(s). The examples of Henry V and Richard III show that such arrangements were possible, but they may not be especially representative. That Creede was the owner of plays probably made negotiations easier, since the bulk of his business was printing — his activities as a publisher were less significant and less lucrative.[8] He would therefore probably have regarded the guaranteed income from a commission as beneficial; what is more, the fact that Creede owned a press made a deal possible in the first place. Most other stationers dealing in history plays, however, did not operate presses and could thus not be offered the kind of bargain Millington and Busby (and possibly Wise) could offer Creede. In particular, all of Shakespeare’s chronicle drama belonged to stationers who worked as booksellers, not as printers: John Busby, Thomas Millington, Thomas Pavier, and Andrew Wise. And from the more crucial perspective of subject matter, nearly all the histories spanning the entire sequence of monarchs from Edward II to Richard III (and after 1605, on through Elizabeth I) were controlled by stationers who did not print books and may well have simply blocked attempts to publish new dramatic works on the same reigns. The only exceptions were Peele’s Edward I, which belonged first to Abel Jeffes and after 1599, to William White, both of whom were printer-publishers; Jack Straw, which was initially John Danter’s (though by the time of the second quarto of 1604, it had become Thomas Pavier’s); and three plays registered to Creede, the two already mentioned and Greene’s Scottish History of James IV (1598). In other words, the entirety of the royal history covered in Edward Hall’s The Union of the two Noble and Illustre Families of Lancaster and York had already been spoken for by the end of the sixteenth century, and the reigns preceding and following Hall’s account were beginning to be filled in as well. This situation must have made it relatively difficult for stationers to negotiate the licensing and publishing of new works on the same reigns or subjects—and must have acted as a significant disincentive to purchase play manuscripts whose subject matter overlapped with what was already covered in previously printed playbooks.

In the theatre, no such restrictions applied.[9] Concrete examples are difficult to come by, since the archival record of what was actually being staged (and when) is extremely spotty. But we do know that Strange’s Men performed a play about Henry VI throughout 1592 and early 1593 which may or may not have been one of Shakespeare’s; the Admiral’s Men had a new play about Henry V in November 1595, probably a few years before the Chamberlain’s Men staged Shakespeare’s version; and they bought (and presumably produced) a two-part work on Henry Richmond by Robert Wilson in late 1599, the summary of which sounds very much like a regular chronicle play.[10] There is also no evidence that Shakespeare’s Richard III and the Queen’s Men’s play on the same monarch did not coexist peacefully, nor do we know whether the Queen’s Men did or did not continue to perform their plays on Henry IV, Henry V, and King John after the Chamberlain’s Men had begun to stage Shakespeare’s works on those reigns.[11] Finally, there is no concrete evidence whatsoever that any acting company ever objected to another company mounting a performance concerned with a historical figure or moment central to one of their own plays.[12]

IV

How does Shakespeare fit into the modified history of histories I am sketching here? For one thing, it is no exaggeration to say that the popularity of chronicle drama can be linked directly to the popularity of Shakespeare’s plays. But the flip-side of that argument, as we shall see, is that the uncommon popularity of Shakespeare in print itself is largely an effect of the success of his history plays.[13] Others wrote popular plays on historical subjects, too, of course. Thomas Heywood had two hits: the first and second part of Edward IV, always published in one volume, and printed five times between 1599 and 1619; and the two parts of If You Know Not Me, You Know Nobody, always published separately—the first volume six times between 1605 and 1613, the second three times between 1606 and 1623. Marlowe’s Edward II saw four editions between 1594 and 1622. And the Queen’s Men’s anonymous Troublesome Reign of King John was a late-blooming success story, with three quartos between 1591 and 1622. But no other author had as many history plays printed, reprinted, and remain attractive in the long run as Shakespeare. While 27 plays from the genre by other authors were first printed between 1590 and 1620, those only led to 21 reprints over the same 30 years—less than one per play. By contrast, Shakespeare’s seven chronicle plays were reissued 18 times—almost three times per title. The genre would still have commanded a presence in the market for playbooks without his influence, but the extraordinary, disproportionate, and enduring popularity of the history play is largely an effect of readers’ continued desire to buy Shakespeare’s histories despite their increasing age and apparent lack of theatrical currency.

If Shakespeare’s historical works dominated the market, they had an even more prominent place in what was beginning to take shape as his published oeuvre. In the 1590s, his chronicle plays outsold all his other dramatic writings by ten editions to four. In the following decade, history plays held steady at ten editions, while his other plays accounted for twelve editions; and in the 1610s, again almost as many reprints of Shakespeare’s history plays appeared as of all his other dramatic works (five vs. eight editions)—no new plays of his being published at all that decade (fig. 4 and 5). By 1620, the seven individual history plays had yielded 18 reprints, whereas the eleven other texts had resulted in a mere 13. Only one history play (2 Henry IV) was not reissued, while the other six were reprinted on average three times; by contrast, only four of the non-historical texts were reprinted in editions other than the 1619 Pavier quartos, also an average of three times each.[14]

Figure 4

Figure 5

These figures suggest that by the time of the Essex performance, most stationers—and possibly, most readers of dramatic texts—would have identified Shakespeare primarily as a poet and an author of chronicle drama. At that point, most of the first and all of the second tetralogy was in print, and all but Henry V and 2 Henry IV (both first published the year before) had already been reissued, in some cases twice. Titus Andronicus and Romeo and Juliet had been published in previous years and recently been reprinted, but the only other plays of Shakespeare’s available from booksellers, Love’s Labours Lost (1598) and the three comedies newly issued in 1600, The Merchant of Venice, A Midsummer Night’s Dream, and Much Ado About Nothing, were destined for an indifferent career, not going into second editions at all or not until 1619 (and then under puzzling circumstances).[15] Through the 1600s to the end of Shakespeare’s career in or around 1613 and to his death in 1616, little happened to change his print persona as predominantly a historical playwright. No new chronicle plays appeared, but the old ones kept being reissued, while with the exception of Hamlet and Pericles, none of his more recent works did notably well. Titus and Romeo and Juliet saw one more edition each. But the impression that Shakespeare was mostly a writer of history plays likely solidified rather than weakened over the two decades between the first publication of his dramatic works and his death. Richard II was a primary reason for that: reprinted five times within 17 years, its success was rivalled only by Richard III (five editions in 14 years) and 1 Henry IV (six editions in 14 years). And some stationers clearly saw an opportunity to tap into Shakespeare’s specific reputation. Of the six apocryphal plays published before the folio with his name or initials on the title page, three treat historical themes of one kind or another: Sir John Oldcastle (Q2, 1619), Thomas Lord Cromwell (1602), and, somewhat further afield, Locrine (1595). Similarly, the Queen’s Men’s Troublesome Reign of King John was newly advertised as “Written by W. Sh.” in the second quarto of 1611, and as “Written by W. Shakespeare” in the third quarto of 1622.

It is probably worth restating my point once more, since it is so clearly at odds with what “Shakespeare” would come to stand for after 1623. David Scott Kastan is right to observe, in an argument recently elaborated by Lukas Erne, that “in his own age,” Shakespeare was preeminent among “published dramatists,” since “more editions of his plays circulated than of any other contemporary playwright” (2002: 26). But this preeminence was founded largely on the popularity of his history plays. Other dramatists saw almost as many non-historical plays reprinted as Shakespeare: Heywood, Marlowe, and Francis Sharpham each wrote two such plays that were reprinted at least twice by 1623, and Dekker, Jonson, and Marston did as well if their various collaborative efforts are taken into account. Kastan’s Shakespeare—the most frequently printed dramatist “in his own age”—would look very peculiar to us, at least if we take “his own age” to mean “his lifetime.” It is a Shakespeare that would leave A. C. Bradley with little to say, one who only engaged with Rome in the ahistorical form of Titus Andronicus, and one who did not write a tragedy with popular appeal after Hamlet; a Shakespeare uninterested in the generic experimentation of the late plays; and a dramatist on par with Chapman and Marston in finding little of a readership for the handful of comedies to his name (few of which showed any interest in gender confusion).[16] The folio, of course, eventually preserved a different sort of author and constructed a different sort of canon, but it matters that these changes did not take place until the 1620s. During his lifetime and for years after his death, at least in the world of books, Shakespeare was first and foremost not a writer of great tragedies or of popular comedies, but a historical playwright. The most significant innovation of the folio, from the perspective of the pre-folio years, is that it gave aggressively equal billing to “Comedies, Histories, & Tragedies.” In fact, the title page deliberately buries the lede, with the author’s signature genre relegated to the least prominent position.