1. “If we produced any thing [in Latin verse] worthy of notice before the elegies of Milton, it was perhaps Alabaster’s Roxana.” So opined Dr. Samuel Johnson in his Life of Milton, NOTE 1 and his high valuation can be matched by others. In The Worthies of England Thomas Fuller called Alabaster “as most rare poet as any our age or nation hath produced.” NOTE 2 More recently, in his 1847 essay Quæstio Quamobrem Poetæ Latini Recentiores Minus Legantur Walter Savage Landor named him, along with Thomas Campion, first in a list of England’s outstanding Latin poets. Further and similar notices could easily be quoted. Of all the approximately 150 plays in the repertoire of Renaissance academic drama written and produced at Cambridge and Oxford, the tragedy Roxana by William Alabaster of Trinity College, Cambridge [1568 - 1640], NOTE 3 possibly enjoys the greatest reputation.
2. The cynical may wonder how many of those who lavishly praised Roxana had actually read the play (which has not appeared in print since 1632), which has paid for its earlier high reputation by subsequent general neglect. Roxana has been studied by only a handful of scholars, and wider interest has probably been stifled by the dismissive and grossly unfair treatment the play received at the hands of Frederick C. Boas in his survey of Elizabethan academic drama. NOTE 4 Boas may have further increased Roxana’s obscurity by discouraging potential editors. All that is presently available is a photographic reproduction of the original printing, NOTE 5 and the text of a contemporary English version has never appeared in print.
3. Boas ridiculed Roxana as little more than a shortened adaptation of Luigi Groto’s 1572 revenge play La Dalida and came dangerously close to calling Alabaster a plagiarist. But various writers who have taken the trouble to compare the two plays in detail have remarked on the artistry with which Alabaster converted Groto’s interminable and hopelessly verbose play into an efficient and effective one. NOTE 6 Esther Kaplan, for example, wrote (p. 312):

 La Dalida, which runs to over four thousand lines, is reduced by more than half in Alabaster’s adaptation. Groto’s diffuse characterizations, illogical sequences, and addiction to the relentless reiteration of numerous themes make his drama tedious at best. Alabaster reworks this source with considerable care. He corrects these defects to a great degree and exhibits dramaturgical skills not clearly evident without the gauge of the Italian text.

To a large extent, he achieved the feat of turning a bad play into a good one by recasting it in the mold of Senecan tragedy, replacing Groto’s flaccid turgidity with neoclassical economy and urgency. In so doing, he paid close attention to the Senecan model upon which La Dalida is loosely based, the Thyestes. This achievement by itself would deserve considerable respect, and would count as a form of originality Boas was unwilling to acknowledge. But Alabaster was in fact more than an adroit play doctor, and we shall see that in Roxana he introduced his own idiosyncratic political outlook, and, in so doing, managed to transform the meaning of La Dalida’s absurdly gross horrifics. Roxana is a play of considerable originality, and Alabaster is rare among playwrights of his time for imitating not only the rhetoric and theatrics of Seneca’s tragedies, but also their political import.
4. Roxana was performed at Trinity College, Cambridge, at some point in the 1590’s. Its author, a Fellow of Trinity College, was one of the more interesting religious adventurers of the age (it is not entirely clear whether we should regard him as a spiritual quester or an eccentric malcontent). NOTE 7 The observation has been made that “even a compressed account of his life reads like a melodrama.” NOTE 8 His sundry religious defections drove him across the face of Europe as an exile, got him equally in trouble with Anglican and Catholic authorities, so that spells of imprisonment, punctuated by occasional jailbreaks, became a recurring theme in his life.
5. Roxana emanates from the period before his religious troubles began. He had already distinguished himself by attempting to write a Great National Epic, the Elisais (he only got as far as Book I, praised by Spenser at Colin Clouts Come Home Again 400ff.). NOTE 9 His literary reputation and family connections (his uncle, John Still, was an Anglican bishop and sometime Master of Trinity College) marked him for success, and in summer 1596 he served as chaplain to the Earl of Essex on the Cadiz expedition. The further ecclesiastical advancement that would doubtless have followed was abruptly cut short later in 1596, when Alabaster abruptly converted to Catholicism and became involved in a harebrained scheme to convert Essex in exchange for an offer of Spanish support for gaining the crown of England. Alabaster’s accounts of this episodes reveal complete unawareness that this was high treason, and that it could serve to put the Earl (as well as himself) in considerable danger, but the government authorities were quick to place him under arrest. In 1598 he escaped from the Clink and made his way to Rome, but returned to England in the following year to continue his attempts to convert Essex. He was again arrested and imprisoned. The recipient of a royal pardon soon after the accession of James in 1603, he offered to return to the Continent and spy on the activities of Catholic émigrés. Then he passed several years at Douai and Brussels until the publication of his theological tract Apparatus in Revelationem Iesu Christi (Antwerp, 1607), over the objections of Robert Persons, Rector of the English College at Rome. This got him in trouble with the Holy Office because of its interest in esoteric interpretation of Scripture of the Cabbalistic kind, so he was imprisoned by the Inquisition, and his book was placed on the Index. He managed to escape and get back to England, suffering another imprisonment at Amsterdam along the way. Taken in English custody in 1611, he recanted and was received back into the Anglican church. Gradually he returned to favor: in 1614 he was created a Doctor of Divinity at Cambridge by royal command and given the living of Therfield, Herts. In 1618 he was admitted to Gray’s Inn and was described in the Admission Register as a Chaplain to the King. At about this time he married Catherine Fludd (thus acquiring the alchemist Robert Fludd as his stepson), and received the firstfruits of the parish of Little Shelford, Cambs. He lived out the remainder of his life as a sedate Anglican vicar, devoting himself to Semitic philology combined with his peculiar brand of mystic theology, having learned one of the great truths about the Anglican religion: for those willing to subscribe to the Thirty-Nine Articles, it is a very latitudinarian denomination indeed. He died in 1640.
6. All of these adventures and tergiversions of course lay in the future. It is nevertheless worth bearing them in mind, for Roxana reveals signs of doubt, disaffection, and radical political thought that become more comprehensible in the light of the playwright’s subsequent career.
7. The play is preserved in two versions. The first, that of the original performance at Trinity College in th0e 1590’s, is represented by five manuscripts. These are:

• Cambridge University Library Ff.2.9, fols 1 - 16, art. 1
• Emmanuel College (Cambridge) ms. 185, art. 4
• Lambeth Palace ms. 838, art. 3
• Trinity College (Cambridge) ms. R.17.10 fols. 40 - 53 (verso)
• Yale University Library (Beinecke Rare Book and Library, Osborne Collection) ms. Vault/Shelves/Plays, item 5.

The text preserved by these manuscripts is substantially the same of that of the first printed edition, which appeared, as we are told in a notice on the last page, on March 31, 1632 (New Style). The contents of the title page are Roxana Tragaedia, olim Cantabrigiae, Acta in Col. Trin. Nunc primum in lucem edita, summaque cum diligentia ad castigatissimum exemplar comparata. Cui accesserunt etiam Argumenta. Londini, Excudebat R. Badgerus, impensis Andreae Crook, ad signum nigri Ursi in Caemiterio Paulino, 1631. NOTE 10
8. Alabaster took vigorous exception to the appearance of this volume, and responded by publishing Roxana tragaedia a plagiarii unguibus vindicata, aucta & agnita ab authore Gulielmo Alabastro. Londoni, Excudebat Gulielmus Jones. 1632. Since theater historians often uncritically take this statement at face value, it is worthwhile to consider his professed grounds for issuing his own version.
9. In the first place, he claimed to be the victim of plagiarism. The title page contains the subscript tragoedia a plagiarii unguibus vindicata (“a tragedy rescued from the claws of a plagiarist”), and in the dedicatory epistle to Sir Ralph Freeman he wrote of plagiarius quidam de crucis bene merito nuncupatus (“a plagiarist justly named after the cross”). The accusation is impossible to accept: Andrew Crooke identified himself as the publisher, but scarcely claimed to have written the play. Alabaster may have had two more legitimate complaints, that his name was omitted from the title page, and that he had given no authorization for Roxana’s publication. Let us examine the situation more closely.
10. In some respects, Crooke’s publication of Roxana finds a parallel in that of George Ruggle’s comedy Ignoramus. Here too, a printer (John Spencer) wished to capitalize on the reputation of a well known Trinity College play, and likewise procured a copy manuscript to provide his text. Both publications were equally unauthorized. There were also, one admits, differences between these two cases: Ruggle is given due credit for having written Ignoramus on the title page, and he, unlike Alabaster, was already dead at the time of publication. But it is less than wholly clear what importance ought to be ascribed to these differences. It may have been the case that Crook had no desire to conceal the play’s authorship, and failed to include Alabaster’s name for no more felonious reason than that it was not preserved in the manuscript at his disposal, a common enough occurrence with copy manuscripts. We have no way of ascertaining whether John Spencer knew or cared whether George Ruggle was still living in 1630. Then too, we have to wonder what exactly Alabaster meant by the word “plagiarism,“ since in his time plagiarism was neither a crime (authors were not paid royalties, so theft was not involved) nor considered a serious breach of literary ethics. One might mention, for example, that the description of the Armada’s defeat in Camden’s Annales for the year 1588 contains a straightforward Latin translation of the description in Hakluyt’s Principal Navigations. By modern standards, this would count as a flagrant case of plagiarism, but by contemporary ones it was unexceptional. Alabaster might more reasonably have accused Crooke of piracy than of plagiarism, but it may be less than wholly certain, by the standards of the time, that such a charge could be made to stick. Nobody, as far as I know, has ever seen fit to denounce Spencer’s edition of Ignoramus as a pirated book. In the present case, it may be worth noting that in 1648 Crooke issued a volume containing John Hacket’s Loiola, Edmund Stubbes’ Fraus Honesta, and two other comedies, typis R. C., sumptibus Andr. Crooke, likewise without naming the authors of three of these on their respectives title pages, and that texts of Loiola and Fraus Honesta, among others, are preserved in two manuscript play collections, Trinity College, Cambridge mss. R.17.9 and R.17.10 (which has only a truncated version of Loiola), the latter of which also contains Roxana. A distinctive feature of both manuscripts is that their authors’ names are unrecorded (although in the former of these mss. Loiola is bound together with a letter to Hacket). Whatever the precise relationship of these manuscripts, if any, may have been, clearly this kind of collection constituted a special subgenre of Cambridge play manuscript, and Crooke’s 1648 book is based on a third, similar one. It is even possible that this same manuscript also contained Roxana, which Crooke issued independently in 1631, and that in 1648 he returned to it and issued a edition of the other four. However this may be, his 1648 volume provides extra insight into Crooke’s working method, the sources of this text, and his willingness to work from manuscripts lacking authors’ names. All in all, the case for convicting him of plagiarism seems exceedingly weak.
11. Alabaster’s second complaint is that the text printed by Crooke is bad. It was founded on a corrupt manuscript: corruptum poematis huius exemplar nactus (“[Crooke] got his hands on a corrupt copy of this poem”), and worsened by the printer’s own mistakes: addidit de suo tot menda, ut certasse mecum videatur, uter crebrius peccaret (“he added so many blemishes of his own, that he seems to have entered into a competition with myself to see which of us could sin the more”). It is true that the first edition is marred by a fair number of misprints, although Alabaster was considerably better served by Crooke than Ruggle had been by Spencer, whose volume was so badly botched that he was obliged to issue a corrected one later in the same year.
12. Besides the understandable desire to gain proper credit for having written Roxana, Alabaster’s words point to a more pressing reason to issue his own edition: a frank admission of responsibility for some of the faults of the original version. Issuing a revised version led him to correct the errors of Crooke’s text, correct and improve his own Latin, and add a few new passages (although new material is scarcely extensive — the second edition is only fifteen lines longer than the first). Collation of the two texts reveals that the new one differs from its predecessor in over four hundred places. Regarded cumulatively, the changes might be considered substantial, but the great majority of them are minor alterations: corrections of mistakes in Crook’s edition; improvements in punctuation; and grammatical, metrical, and stylistic improvements achieved by rearranged word order, changing the mood of a verb, or by substituting words, phrases, single lines, or small groups of lines that normally retain the original idea or something quite close to it. Often the replacement passages are appreciably better; NOTE 11 occasionally the grounds for alteration are so unclear that one is tempted to regard them as mere tinkering. No significant material is eliminated, and only three new passages have been added: a satirical passage about physicians (53 - 61), a whimsical allusion to the Gunpowder Plot (221 - 226), and an exclamation about the havoc wreaked by king-waged wars (1331 - 63). But in no way is the fundamental conception of the original play altered, any of its scenes recast, or any of its characters redrawn.
13. It has been suggested that some of the alterations made by Alabaster were made for political reasons: NOTE 12

…it is easy to see why he would have been chagrined at the publication of the play as it stood in 1632, when he himself had enjoyed the fruits of Kingly patronage — had, as it were, joined the ranks of those whom he had earlier maligned. The observations, the fears, and the judgments regarding the Tudor reign that found expression in Roxana were no longer relevant; but the unspecified political allusions of 1591, particularly those condemning tyranny and royal prerogative, could apply all to easily to other times, including those of Charles I. Hence Alabaster’s emphasis in his preface on the youthful composition of the play. And in the text, in passages concerning the illegitimate wielding of power, we find the repeated substitution of ius - the rule of law, for Rex — the King himself, which softens speeches that would otherwise have professed a political bias now quite foreign to Alabaster.

Here is an account with a plausible ring. One could readily suppose that in 1632 Alabaster, long since reconciled to Anglicanism and enjoying a comfortable ecclesiastical living in his old age, could have been thrown into a panic by the unwelcome appearance of inflammatory material written in his youth. Fearing that these hotheaded effusions would jeopardize his present situation — plenty of people with long memories would hold him responsible for Roxana even if his name did not appear on the title page — he did everything in his power to discredit Crook’s edition and replace it in the public eye with a toned-down version.
14. But this fairy-tale is entirely unsupported by the facts. Collation reveals few if any places where any statements of a political nature have been eliminated or mollified. The above-quoted author wrote of the repeated substitution of ius for rex, but this occurs only in a single line (682), ubi quisque regnum, nemo quod iure obtinet (“To winne the crowne, that none can clayme by right”) which in its original form ended quod rex obtinet (“which no king can claim”). This one instance must be weighed in the balance against the many instances where adverse observations about kings and tyrants are allowed to stand unchanged. It is notable that one of the three passages added to the revised play (1331ff.) is a fresh complaint about kings. And the one line in the original text that might seem most objectionable, because it deals with the sensitive subject of kings and religion (110), usus tyrannis deos et metus facit (“Tyrants make gods, and unmake as occasion / Or feare constraines”), has been rewritten to deos tyrannis usus, haud pietas facit (“Custom, not piety, manufactures gods for the benefit of tyrants”). The latter line may be superior, but the sentiment is scarcely ameliorated.
15. Furthermore, the appraisal quoted above assumes that the play’s observations about tyranny or arbitrary royal rule, which are couched in very generalized terms, would strike Alabaster’s contemporaries as objectionable. This is perhaps open to question. NOTE 13 Roxana is firmly based on Seneca’s Thyestes. Both the Thyestes and the pseudo-Senecan Octavia have scenes in which an advisor unsuccessfully attempts to dissuade a monarch from tyrannical and lawless behavior. The debate between Oromasdes and Arsaces (III.i), which contains most of the statements that might be deemed politically sensitive, is highly imitative of these Senecan scenes and, as such, is, or at least could be claimed to be, a routinized topos. This scene indeed is written with an eye on a modern political situation and this passage, if anything in the play, might seem to have had the potential for getting him in trouble, yet it is striking that in 1632 Alabaster did nothing to recast it in the light of altered political or personal circumstances.
16. So let us consider this scene. Henry VIII’s marital dilemma is part and parcel of the play’s situation. In essence, the plot is about a king afflicted and alienated by his consort’s barrenness. The Henrican parallels are brought to the fore in Oromasdes’ dialogue with Arsaces (who we are told at 651 is the sole Bactrian who opposes him). I have just noted that this scene is modeled on interviews between autocrats and their advisors in Seneca’s Thyestes and the pseudo-Senecan Octavia, in which the advisor unsuccessfully tries to warn the ruler away from tyrannical behavior. But at the same time, III.1 is calculated to remind one of Henry’s dilemma. This is particularly so because, although the characters in the play are ostensibly pagan, the objections raised by Arsaces are much more appropriate to the sovereign of a modern Christian state than to a monarch of pagan antiquity, notoriously tolerant about the subject of divorce. Some of the religious and legal anti-divorce arguments used by Arsaces, to be sure, were also spoken by the Consigliere to Candaule in La Dalida III.i, but in the Italian play these are diluted by being mixed in with a number of other considerations not relevant to this issue. Alabaster judiciously pruned away such extraneous arguments and retained only those pertinent to Henry’s situation, adding many new ones of his own.
17. Part of Oromasdes’ argument is that, in divorcing Atossa, he would be acting in the best interest of Bactria. His fundamental theme is ius est, salutem quicquid auget publicam (“That which procures the kingdomes good, is right”). Specifically, the lack of an heir will sooner or later plunge the nation into war and anarchy (676 - 85):

Sic Bactriani decidet regni decus?
Nec per nepotum sceptra transmittam manus,
Privata vel si tangerent animum minus,
At vestra flectant; quanta bellorum seges
Hinc pullulabit, regis ignoti metu?
Quid factiosae principum manus agent?
Ubi quisque regnum, nemo quod iure obtinet,
Suum esse credit, quantus error partium
Manabit inde? Bella cum plures simul
Reges creabunt, unicum victoria.

Shall then the honour of the Bactrian race
Perish in us for want of lawfull heyres?
I speake not for myselfe, that moves me not,
But for your sakes. O what a world of woes
Will hence arise! What wast of guiltlesse blood!
When every factious peere will strive by might
To winne the crowne, that none can clayme by right.
How will the land be into parts divided
When bloudy warrs will many kings proclaime
And insolent victory will but suffer one?

 He insists that his choice is a matter of royal right (666f.):

Honor regalis in vili est situs,
Cui iusque fasque plebe cum sua par est.

The honour of that king is vile and base
That hath no privelege above his people.

Arsaces, on the other hand, takes the position that a sovereign has the responsibility to set the moral tone for his people (665f.) :

Rex plebis est mensura, pietatem docet
Qui facit.

The people take example by their King
He allwayes teacheth best, that liveth best.

Oromasdes points out that the Bactrian senatus approves his plan, but Arsaces responds by intimating that this senatus has been cowed into submission (670f.):

ORO. Reliquus senatus ordo consentit mihi.
ARS. Hoc est timendum nemo quod regi negat.

ORO. All our counsel els have sett to their hands.
ARS. More cause have kings to feare which none withstands.

(Arsaces’ reply might better be translated “And this is a source of fear, that nobody refuses to be dominated.”) Likewise, in his later soliloquy Arsaces repeats the complaint that the senatus is slow in asserting its own right (900f.):

Auget hoc regis malum
Facilis senatus, lentus assertor sui.

Whatsoe’re the kings desire’s his other lords,
Be’t right or wrong, they soone doe yeild consent.

(A more accurate rendition would be “A compliant senatus, slow to assert its right, increases this evil of the king” — the meaning of senatus is discussed below). In a similar exchange, Oromastes points out that a barren consort is injurious to his dynasty, and Arsaces rejoins that royal injustice is more harmful (674f.):

ORO. Regina sterilis impedit regni vices.
ARS. Iniusta facta regna regesque eruunt.

ORO. The barren queene denies our crownes succession.
ARS. Both crownes and Kings are ruined by oppression.

When an exasperated Oromasdes asks what precisely obstructs his divorce, Arsaces answers very comprehensibly and, significantly, adds a religious component to his reply (735f.):

ORO. Quae iura laedit coniugis repudium?
ARS. Laedit parentum, gentium, regni, deum.

ORO. What Lawes forbid us to divource our Queene?
ARS. All Law of nature, nations, realmes, and Gods.

18. I have quoted so liberally from this passage because from a political, if not a dramatic, point of view, III.1 is the plays’ most important scene. Reading it, one senses that Alabaster is really writing a thinly disguised debate between Henry and some contemporary advisor (probably, in view of his predominant interest in the subject of legality, Sir Thomas More), since the moral, legal, and religious issues and arguments raised by Oromasdes’ situation are the same as those raised by the Henrican divorce. It might be excessive to identify this scene as entirely pro-Catholic: Oromasdes’ arguments are cogently framed, his concern for the welfare of his reign is reasonable, and in this scene he does not attempt to bully Arsaces or threaten him with undue reprisal. Indeed, although in a later scene Arsaces complains of having lapsed from royal favor for his outspokenness (902f.), in the end it is Arsaces to whom the dying Oromasdes bequeaths the crown of Bactria (1514). A spectator, on the other hand, could scarcely fail to note that Oromasdes’ wrongheaded response to his predicament provokes the string of unmitigated disasters that occupy the remainder of the play. His characterization of Arsaces as sensible, and moderate is probably an accurate index of where Alabaster’s sympathies lay. Then too, the considerations that Oromasdes is an impetuous and ambitious autocrat who had dealt brutally with Moleon in order to gain the throne, and also a sexual predator, have the effect of tilting one’s sympathies further towards Arsaces. All in all, although this scene may not be a full-blown sectarian denunciation of the Henrican divorce, it is at the very least a questioning examination of that event that invites an equally thoughtful appraisal by the spectator.
19. The discovery of this element of political disaffection in Roxana, incidentally, requires a brief consideration of the play’s date. In an autobiographical document entitled (by someone else) Alabaster’s Conversion, preserved at the English College at Rome, our poet furnishes a chronology for his conversion: he dates its inception to Michaelmas 1596 (4.1), at which time he was back at Cambridge after having served as chaplain to Essex during the Cadiz expedition. The process continued during an unspecified time when he lingered at Court angling for a suitable parish (4.3); then he returned to Cambridge until his eventual arrest in 1597 (4.9, 7.2). Although a religious conversion is often preceded by a period marked by such things as doubt, disaffection, intellectual confusion, and inner turmoil, in his autobiographical account Alabaster mentions no such experiences. Possibly he is being less then candid in the interest of making his sudden conversion seem more dramatic. But if his testimony is to be taken at face value, then the standard dating of Roxana to the early 1590’s, based on the imprecise statement in this dedicatory epistle that he wrote Roxana ante quadraginta plus minus annos (“more or less forty years ago”), may require reevaluation. NOTE 14 Book I of the Elisaeis was finished in time for Spenser to have read it prior to writing Colin Clouts Come Home Again, and the consensus of Spenserian authorities is that work was written in 1592, although not published until three years later. It would appear that the Elisaeis was written before Roxana, and that its Anglican orthodoxy accurately reflects Alabaster’s religious convictions at the time he wrote it. This is turn suggests that Roxana was written when his beliefs were at least beginning to waver. Then too, it seems intrinsicaly unlikely that Roxana was written prior to Kyd’s The Spanish Tragedy and it may well also reflect the influence of Titus Andronicus (printed in 1594). A dating nearer to 1595 than to 1590 seems most plausible.
20. To return to Roxana’s political orientation, the further point needs to be made that the play’s politically significant passages are scarcely limited to III.1. Let us consider the play’s orientation more generally. We have already seen that Alabaster recast La Dalida into a play constructed very much in the Senecan manner. The finished product contains all of the ingredients of Seneca-imitating tragedy, such as its efficient neoclassical concision, the specifically Senecan rant of its poetics (as can be seen from commentary notes, much of the play’s phraseology is appropriated from Seneca), the conception of its characters, the presence of the supernatural, and a very liberal helping of atrocitas. Indeed, this last element is so prominent that in reading the play one might be tempted to disdain it as an amoral exercise in theatrical sensationalism. I say “amoral” because there is no overt criticism or moral evaluation of Atossa’s disproportionately savage revenge on adulterous Oromasdes, a revenge that involves the hideous destruction of at least two innocent victims (the children), and perhaps a third (Roxana, if we are to regard her as essentially blameless for her predicament). A common expectation for revenge plays is the presence of a Christian perspective, condemning the revenger for usurping a right to punish moral transgression that properly belongs to God. But, save for a perfunctory Epilogue tacked on to the English version, which may or may not reproduce the Epilogue of the play as enacted at Trinity College — the point is discussed in a note on the passageRoxana is thoroughly devoid of this particular form of moral analysis, which admittedly may come as something of a surprise in the case of a play written by an Anglican clergyman.
21. Then is Roxana as amoral as it seems? If we are looking for an evaluating perspective and fail to find one, it deserves to be suggested, this is only because we are looking in the wrong direction. The play’s evaluative element is not Christian, or at least not explicitly so, but rather political.
22. In his introduction to the new Penguin Hamlet (p. 10) the late T. J. B. Spencer quoted an obiter dictum of François de Belleforest, made while recounting the Hamlet story in his French translation of Saxo Grammaticus:

…where the prince or country is interessed the desire of revenge cannot by any means (how smal soever) have the title of condemnation, but is commendable and worthy of praise.

Alabaster’s appraisal of his brutal royal revengers is the exact opposite. His play contains a great amount of explicit condemnation of royal prerogative, royal lawlessness, royal harmfulness, and downright royal tyranny, virtually none of which was eliminated or toned down in 1632. It would appear that all of this material is meant to guide the spectator to the conclusion that the spectacularly arbitrary, cruel, brutal, and all but senseless behavior of his pair of monstrous revengers is the way that royalty behaves.
23. We have seen that Coldeway appreciated this element and thought it a critique of Tudor tyranny, and, in view of the facts of Alabaster’s life, there is an obvious temptation to regard this as a specifically Catholic condemnation of the Tudor execrata domus. Such an evaluation is possibly right. But I am unconvinced that this is the correct diagnosis. Alabaster’s adverse observations about kings and tyranny are couched in such generalized terms that they appear to have a broader application, and suggest a kind of proto-republican discontent with the entire contemporary notion of royal prerogative. I have already quoted his obiter dictum about kings manipulating religion for their own advantage (110). Nothing indicates that this applies only to Protestant sovereigns. The same can be said about a number of his observations. At 222f., for example, he writes:

At regni ob usum multa reges perpetrant
Quae non decent.

But for royal advantage kings do much that does not befit them.

Did he think that that Catholic rulers were exempt from this criticism? Or consider one of the three passages added in the 1632 version (1332 - 35):

Quid bella faciunt aliud, ubi tota oppida
Etiam immerentum, principis causa sui
Ruina sorbet una? Et hoc gloria?
Regis vocatur, culpa privati est scelus.

What else do wars accomplish, when they plunge whole towns of innocent men in ruin for the sake of their ruler? And this is called royal glory, a private man’s wrongdoings are sin.

Nothing suggests that Catholic monarchs wage more humane wars. Or again, consider one further exchange in the debate between Oromasdes and Arsaces in III.1 (668f.):

ORO. Reliquus senatus ordo consentit mihi.
ARS. Hoc est timendum nemo quod regi negat

ORO. All our counsel els have sett to their hands.
ARS. More cause have kings to feare which none withstands.

Here senatus is translated as “counsel,” but it is worth remembering that in Latin written by Englishmen senatus is the word regularly used to designate Parliament, and so it may have been understood by Alabaster’s contemporaries. In the same vein, in his soliloquy in III.4 Arsaces complains (896 - 900):

Rex Oromasdes angitur quenquam sibi
Obstare, demi credit a regno suo,
Quicquid negatur. Auget hoc regis malum
Facilis senatus, lentus assertor sui
Certusque regi si quid urget obsequi.

Great Oromasdes king of Bactria,
My prince and soveraigne whome I reverence,
Doth greive that any crosse his purposes.
Whatsoe’re the kings desire’s his other lords,
Be’t right or wrong, they soone doe yeild consent.

And at one point (705, a phrase perhaps significantly omitted from the translation) Alabaster employs the incendiary words Populusque censor principum, as if the people have the right to judge their sovereign (as the Scottish Humanist George Buchanan had recently scandalized all English loyalists by proclaiming, in his 1579 De Iure Regni apud Soctos Dialogus), a far cry from the doctrine of the Divine Right of Kings. Likewise, his statement at 110 about kings cynically manipulating religion is followed by a line to the effect that piety is only to be found among the common people:

At sola simplex turba veneratur deos.

…and but the vulgar serve them [the gods].

One cannot help wondering whether a contemporary spectator or reader might interpret these statements as a criticism of Parliament for not displaying greater independence. If so, did Alabaster seriously imagine that a more self-asserting Parliament would benefit Catholics? Or did he believe that a shift of the locus of power from sovereign to national representative body was desirable in its own right, because the people is the ultimate arbiter of its rulers? Such interesting and important questions deserve to be raised, even if they are incapable of being answered.
23. After a lengthy period of neglect and disparagement by all but a few (including, most conspicuously, T. S. Elliot), Senecan tragedy is enjoying something of a modern vogue. It has recently been pointed out that one of the reasons for this is that these plays, written during the reigns of Claudius and/or Nero, often explore the problems posed by tyranny, and so resonate with the issues of tyranny and totalitarianism that have so dominated the twentieth century. NOTE 15 For the Elizabethans, Senecan tragedy was paradigmatic and so was closely imitated. Perhaps this principle is most dramatically illustrated by Thomas Watson’s Latin translation of the Antigone prinrted in 1581, in which Watson added a Prologue and imposed a five-act structure on Sophocles’ play in order to make it conform to the obligatory Senecan paradigm. I have already mentioned some Senecan elements routinely imitated in academic tragedy, both academic and popular: these include such things as incursions from the realms of the supernatural and the Underworld, a specific kind of ranting rhetoric, and physical horrifics. Sometimes, as in Roxana, such imitatio involves the use of a Senecan plot. It also frequently involves a special kind of character delineation and psychology that works along the following lines: when a character is projected into a situation similar to that of one in a Senecan tragedy, he or she temporarily “becomes” that character, in terms both of psychology and of rhetoric. In Roxana, for example, as soon as she discovers Oromasdes’ unfaithfulness, jealous and aggrieved Atossa “becomes” Seneca’s Medea, just as Oromasdes is imbued with many of the features of the Senecan Atreus.
24. One could profitably spend a great deal of time exploring in detail the nature and workings of academic imitatio of Senecan tragedy. But, if one may be so bold as to anticipate the conclusion of a study not yet undertaken, the results of such a survey would in all probability be that the Elizabethans responded with great intensity to the dramatic and rhetorical elements of Senecan tragedy, but with no similar enthusiasm to the political contents. Groto’s La Dalida ostensibly imitates Seneca’s Thyestes. La Dalida is not only a bad play in the sense that is poorly crafted and almost absurdly long-winded. It also deserves censure for its sensationalistic wallowing in atrocitas. The reader of Roxana may be surprised to be informed that Alabaster has actually muted the horrifics of his Italian model. For example, he mercifully omits to translate Groto’s passage describing Queen Berenice, compared to an industrious anatomist, dissecting and mincing her victims, then consigning their flesh to pot and to pan (3473ff.). Even worse, Groto’s atrocitas appears to be intended to titillate the audience, but is otherwise pointless. The political elements in the play contributed by Alabaster, which have no equivalent at all in Groto, have a visibly transformative effect, giving the physical horrors of Roxana a startlingly new meaning. Whether written about the Tudors or about kings more generally, Roxana is used as a tool to probe the ethical and psychological nature of tyranny in much the same way that Seneca employed plays like the Thyestes to anatomize Julio-Claudian tyranny. Atossa and Oromasdes are genuine monsters. The savagery of the injuries they work on each other and on others is almost absurdly excessive, as are their melodramatic reactions and overblown rhetoric. As such, they invite viewing as gross and maliciously unfriendly caricatures of royalty. Confronted by the atrocities they commit, the audience seems invited to draw the conclusion “this is the way royalty behaves,” and a plausible case could be made that the atrocitas in Seneca’s Thyestes serves a similar purpose. Similarly, Bessus is a caricature of the scheming and unprincipled underlings who flock to royal courts. Played off against these characters is sensible, wise, pious, and unambitious Arsaces. He is both an exemplar of the statesmanlike traits Alabaster deemed praiseworthy, and a baseline against which royalty and its henchmen can be measured. And Alabaster has transformed the Consigliere’s monologue about the danger of flatterers in La Dalida (III.v) into a diatribe against the corruption and corruptiveness of a royal court (III.iv). All in all, the play contains much evidence that attacking royal tyranny is Alabaster’s chief programmatic interest.


25. The translation employed here is an anonymous contemporary one preserved in Folger Library ms. V. b. 222 (fols 29 - 37v). This has never appeared in print, although it was the subject of a 1980 Harvard dissertation by Ethel Rosenberg Kaplan. NOTE 6 There are two reasons for employing the Folger translation here. The first is that, after comparing it with the Latin, Kaplan’s verdict was (p. 26) “The English Roxana is a competent rendering of a Latin text. To the degree that it interprets the text in the light of seventeenth-century values, it provides a dimension unaffordable in a modern translation.”
26. But there is another and more compelling reason for using this translation. At the end of the translation is the subscription I. B. Scriptore. The same initials appear on the outer boards of the volume. The ms. is indisputably a copy text rather than a holograph. Besides a number of routine transcriptional mistakes, there are two places where the copyist was aware that his exemplar was missing some material and wrote desunt nonnulla, and at other points material is either omitted or imperfectly preserved. Clearly, therefore, I. B. (or J. B.) must have been the original owner of the manuscript and also its copyist, but scarcely the author of the translation. Although Kaplan came to the same conclusion, without attempting to identify the translator, I am strongly inclined to think the translator was Alabaster himself. On the basis of stylistic resemblances to Alabaster’s other English verse, the least that can be said is that, if he had produced an English translation of Roxana, the result would have been very much like the Folger one. Also, as we shall see, a working hypothesis that it was Alabaster’s work is helpful for making sense of what would, by any other understanding, be the considerable problems posed by the translation.
27. Huntington Library ms. HM 39464 is evidently incomplete at the beginning (if nothing else, it lacks a title page), and in its present condition consists of 37 bound and numbered pages. This volume contains eleven Latin poems with accompanying English translations. With the exception of the popular poem XVI, the items in this collection consist of poetic elaborations on familiar ethical themes. Some of these are elaborated to the point that they deserve to be classified as sermones, or satires in the manner of Horace. The anonymous contributor to the Huntington Library card catalogue tentatively ascribed this collection to Alabaster. One of its items (poem XVI), on the two Rainolds brothers, reappears in several other sources ascribed to Alabaster, and it is notable that the other poems in the anthology contain the same kind of stylistic defects and peculiarities that have been observed in Alabaster’s sonnets. On the strength of these features, I have included these unsigned items as authentic items of Alabaster’s occasional verse. NOTE 16 Stylistic features similar to those of Alabaster’s sonnets and the Huntington English poems can be observed in the Folger Roxana translation.
28. Alabaster was considerably more competent at writing verse in Latin than in English. In their Introduction, the editors of his sonnets observed that some of their lines contain only four or four and a half feet, and that it is not always clear whether such defective lines are to be blamed on copyists’ mistakes: in correcting them, “it is possible that the editor has here done for the author what he should have done for himself.” NOTE 3 The Roxana translation contains a number of similar lines, many of which can be fixed by adding an obvious word or two (as I have done, enclosed in angled brackets). Others cannot. These include 163 Of my childe and my nephewes sonne, 174 Much for your selfe, more for your children, 296 Yet why it should doubt, my brest knowes no cause, 411 Sin’s best safetie is to sin againe, 492 ATO. To thee it shines not. BES. Starrs who can hide? (possibly we are to read Yet starrs who can hide? or even Who can hide the starrs?), and 808 Ile ne’re goe to our house againe.
29. More important for diagnostic purposes is another observation made by the sonnet editors:

In many cases it is Alabaster’s incompetence which leaves the reader in doubt where the stress on a word should fall and uncertain whether to elide a syllable or not. It seemed better not to overload the commentary by suggestions as to how lines might be read, and to leave the reader to decide whether Alabaster’s frequent failures to ‘keep accent’ are due to a bad ear or to deliberate experimentation in handling the decasyllabic line.

The latter would appear to be the correct answer, and that the Latin-oriented Alabaster allowed himself to insert metrically equivalent substitute and resolved feet in the Latin manner: i. e. he reckoned an iambic foot to consist of three minims and freely substituted other feet which contained the same number. Thus a trochee can be be substituted for an iamb (a substitution which would never occur in Latin iambics). Take a line from the Huntington anthology (poem *XLIII.12) Proud purse-leaches, Harpies of Westminster. A reader unaware of Alabaster’s system would think that Westminster is receiving a very unnatural stress on the third syllable. For the same reason, a considerable number of lines in the Folger translation appear to contain similarly weird accentuations. Here only a few specimens: dearer in 77 Reft me of life, and throne dearer than life; Tython in 195 Aurora loved Tython an older knave, children in 209 And beare children . Poore soule what could she doe?, matter in 345 Whilst I determine some matter of weight, and 409 Jealousy will yield, when Atossa will not. Examples of this phenomenon could be multiplied indefinitely.
30. Another stylistic peculiarity of Alabaster’s sonnets is a proclivity for scanning the suffix - ion disyllabically, particularly in words appearing at the end of the line. This too is not uncommon in the Folger translation. A few illustrative examples are Argument 32, Was straight on fire with loves affection, 197 I am betraid; what means these questions?, 226 But every light suspition let flie, 499, Make me to scorne of base affection, 519 Her fathers death, and his succession, and 914 As rageth youths affection. Occasionally other suffixes are handled in a similar way: compare the scansion of such lines as 173 If fathers looke makes me not partiall, 878 Backbiteing envy, feighned dissembling, and 899 Gods will be done, and give me patience. (It must be noted that such scansion is not unique to Alabaster; it appears, for example, in line 21 of Abraham Cowley’s English translation of Alabaster’s poem XXI, And wisht a reparation to see.)
31. The sonnet editors also complained about occasional unintelligibility, usually resulting from elliptic writing, contorted syntax, or unnatural word order:

Faced with poems of such unequal accomplishment and a poet so inexact in his use of language, a commentator has at times to be content to suggest a general sense and acknowledge that often another reader’s guess may be as good as his own. To discuss some of the more obscure passages at any length would be ‘to waste criticism upon unresisting imbecility.’

Such offenses against intelligibility are mercifully rare in the Folger translation of Roxana. But (if we assume the text is right, for a line containing the main verb of the sentence must have dropped out), at least one line invites comparison with similar ones in the sonnets and the Huntington collection. This is 137, quoted here in context:

Ther fetch in hell from out some secret roome
Where hid from all, and feare knowne to herselfe,
[. . . . . . . . . . . . . . . . . . . . . . . .]

Presumably this is elliptic for Where hid from all, and [also hid from] feare knowne to herselfe, but in a manner quite familiar in Alabaster’s English poetry the reader is placed in the position of extracting meaning from a line that is scarcely self-explanatory. The same complaint can be lodged against 329 Silence must cure, which will encrease reveal’d. Again assuming the text is sound — one is tempted to substitute would for will — this is an excessively elliptic way of saying Silence must cure the evil, which will encrease if it is revealed. Once more, intelligibility is strained nearly to the breaking-point.
32. In general the English translations in the Huntington collection are in the same rhymed pentameter couplets favored by the Folger translator. But one Huntington item, *XLIV translates a Latin poem in elegiac couplets on the shortness of life, into the same four-beat rhymed verse that predominates in Roxana’s choruses:

Like as the Damask rose you see,
Or like the blossoms on the Tree,
Or like the daintie flower of May,
Or like the morneing of the day,
Or like the sunne or like the shade,
Or like the gourd
which Jonas had:
Even such is man, whose thread is spunne,
Drawne out and cut, and so is don.

33. We must ask what version of the Latin play this translation is based upon. In her dissertation, Kaplan collated the two print versions and four of the five manuscripts of the Latin Roxana — although a Yale graduate student, she seems to have been unaware of the existence of the Beinecke Library one — in an attempt to understand the history of the text, and then raised this question. This plunged her into a maze of difficulties. By her reckoning (pp. 53f), the English version reproduces about half of the innovations made in Alabaster’s printed text. In other respects it translates features of what Alabaster originally wrote, as printed by Crook. If two versions of the text existed, one derived from the Trinity College performance and the other consisting of Alabaster’s 1632 revision, how could a translation be based on a hybrid of the two?
34. How to solve this conundrum (and also, in Kaplan’s mind, how to account for the relatively minor variations between the manuscripts, which in fact are attributable to mistakes in the copying process)? Evidently (p. 9) “…analysis of all seven texts reveals that, contrary to his public disavowals, Alabaster did indeed revise his play frequently prior to issuing his own edition in 1632,” with the result that the Folger Roxana translates some intermediate version in which some but not all of the changes printed in 1632 had been made. Since this theory by itself would not serve to explain how Alabaster’s revisions came to circulate in copy texts, Kaplan was obliged to postulate one or even several revival performances in the 1620’s and 1630’s (p .13). Such mental gymnastics and the deus ex machina of unattested revivals can be eliminated by assuming Alabaster himself to be the author of the Roxana translation. It is likely enough that the changes and innovations in the Latin text printed in 1632 were introduced over time, and it no less likely that the Folger Roxana translates the text in an intermediate state, when some but not all of them had been made. But if we are to seek the location of this intermediate text, we need not search for it in entirely hypothetical revivals. We should rather imagine that this evolving Latin text existed in the most likely of all places, on Alabaster’s own desk.
35. The great majority of the differences between the English translation and the Latin text (in either extant version) are of kinds that are readily understandable in the translation process. Latin is a more terse and lapidary language than English, and the translator is often more interested in spelling out the ideas stated or implied in the Latin than in reproducing what it says with exactitude. This translation strategy sometimes leads to verbosity. English translations of Latin are almost invariably longer than the original, and it is chiefly for this reason why the English text is slightly over two hundred lines longer than its Latin equivalent. At other times the translation is simplified by the elimination of such things as mythological allusions. In commentary notes on the English version I have noted the significant differences between it and the Latin, and provided my own translations of what the Latin actually says. Here I will mention the most important discrepancies.
36. I have already noted the three passages added to the text by Alabaster for the version he published in 1632: 53 - 61, 221 - 226, and 1331 - 63. These are not translated. Similarly, it renders the original Latin text replaced in 1632 by lines 1526 - 31. It is reasonable to think that the translation is based on a version of the text as it existed prior to the insertion of these passages. Other important omissions are of lines 80 - 85, 208, 229f., 245, 517f., 1139 - 45, and 1278 - 1281, and this is hardly an exhaustive list. Some of these untranslated lines and passages may be instances of deliberate omission or simplification, but the copyist of the Folger manuscript was well aware that he was working from a defective exemplar (in the cases of both 1139 - 45 and 1278 - 81 he made a notation to this effect), and it is likely that at least some of the other missing material has been dropped out of the text by careless copyists. The translation’s attribution of English lines 1719 - 28 to the Chorus is presumably another result of faulty textual transmission.
37. The English text, on the other hand, contains some material unparalleled in the Latin. The most conspicuous such instances are the extra lines at the end of the Act III chorus (945 - 58), the extra ones at 1644 - 56, most of Atossa’s concluding speech (1737 - 56), and the Epilogue. These additions and alterations would be very difficult to understand according to the assumption that someone other than Alabaster himself was responsible for the English version. Why, one should have to ask, would a usually conscientious translator suddenly indulge in bursts of creativity or attempts to “improve” upon his Latin original? No similar problem confronts us if we identify Alabaster as the translator. Some of the abovementioned passages may indeed be cases of rewriting the Latin, but, obviously, when an author rewrites his own works no comparable ethical questions arise. It is also possible that some of these erratic passages represent points in the specific Latin text upon which the translation is based, where Alabaster made experimental alterations, but subsequently changed his mind and reverted to what he had originally written. The Epilogue may represent a special case: see the note on the passage 


38. The present edition is based on the text published by Alabaster in 1632, incorporating the author’s revisions and improvements. At some points Crooke’s text serves as a corrective for typographical mistakes in the authorized edition. The Latin text printed here employs modern conventions of orthography and punctuation. The edited text of the Folger English translation presented here is based on my own transcription of the manuscript, subsequently checked against that of Kaplan. The commentary notes on the Latin include the results of a collation of Alabaster’s 1632 text against the one printed by Crooke.