The background of The Shrew : Texts and intertexts

Dramatic and theatrical intertexts

The study of The Taming of the Shrew is complicated by the play's relationship with the anonymous 1594 quarto The Taming of a Shrew. As is well known, the quarto has been considered variously a source, an adaptation, and a piratical version of The Shrew. Recently, Roy Eriksen and Donna Murphy have even strongly claimed Christopher Marlowe's paternity. ¹ The Taming of a (the) Shrew was a popular play in the 1590s, as is shown by references to it in poems, letters, and treatises. In Antony Chute's Beawtie dishonoured written under the title of Shores Wife (1593), there is a possible allusion to the last scene of The Shrew (‘Come on and kiss me Kate’, 5.2.180): ‘He calls his Kate, and she must come and kisse him’. ² Sir John Harington recommends the reading of A Shrew: ‘For the shrewd wife, read the booke of “taming a shrew”, which hath made a number of vs so perfect, that nowe euery one can rule a shrew in our countrey, saue he that hath hir’. ³ The popularity of the play – or of the two plays/versions – seems attested by some of the running titles in the Folio versions. The Elizabethans did not seem to distinguish between the two plays, so that The Shrew was never entered in the Stationers’ Register and even the Second, Third, and Fourth Folios have, by mistake, some pages with the running title ‘The Taming of a Shrew’, ⁴ as if the change in the title did not affect the identity and identification of the play. It is difficult to say in which form (or forms) this play circulated, unless as a conjectural reconstruction.

If we accept the so-called revision theory, the analysis of the intertexts underlying A Shrew and The Shrew will throw light on the process of theatrical construction of Shakespeare's play, showing us how the ‘original’ Shakespearean text could be at a certain point of its dramatic and theatrical history. Studies on intertextuality lead us to consider a different approach to the question of sources: no longer as mere comparative criticism, but as a semiotic scrutiny of the modes of selection, transformation, and transcodification of the source texts. In the case of the dramatic text, the study of the sources should take also into account the passage from narrative plot to dramatic plot; the passing from ‘told story’ to telling story; and the mode of transcodification from narrative form to dramatic form. The dramatic intertext could affect directly or indirectly a dramatic text. Plutarch is a direct intertext for the composition of Julius Caesar, Antony and Cleopatra, and Coriolanus, whereas Plautus is an indirect source when filtered through the Italian Renaissance plays. As far as the theatrical intertext is concerned, this is a particular kind of text, which ‘cannot but bear the traces of other performances’. ⁵

Intertexts: A Shrew and The Shrew

The Taming of a Shrew was entered in the Stationers’ Register on 2 May 1594 and published in a quarto edition by Peter Short in the same year. On 11 June, Henslowe reports a performance, not marked as a new play, of ‘the tamynge of A shrowe’ at the old Newington Butts theatre, where the newly formed Admiral's and Chamberlain's companies were acting. ⁶ The play was reprinted in 1596 with a few minor modifications, mostly in spelling. A third edition, entered in the Stationers’ Register on 22 January, was printed in 1607 by Nicholas Ling. In the same year (9 November), the rights to A Shrew were transferred to John Smethwick, who, in 1631, published a quarto edition not of A Shrew, but of The Taming of the Shrew, ⁷ printed from the Folio. The three quarto editions of A Shrew differ mostly in spelling. ⁸ The numerous spelling variants are often unmotivated and due to the hands of different compositors. Both the syntax and the spelling of single words make it difficult sometimes to understand whole passages of the comedy, as is often the case with many ‘bad’ quartos. Unlike other quarto editions, considered ‘bad’, A Shrew presents different textual problems and, above all, an independent story and plot. The variants in the three quartos testify to the instability of the text and the poor ‘editorial procedures’, whereas its dramatic structure reveals a textual and theatrical autonomy interwoven with that of The Shrew. Only since the 1950s has Shakespearean criticism begun to consider A Shrew a particular kind of ‘bad’ quarto; and only at the end of the century has the single Shakespearean authorship that lies behind the two Shrew plays been claimed. ⁹

For many years, Shakespearean scholars considered the Folio version as deriving directly from Shakespeare's manuscript, paying little attention to the dramatic and logical instability caused by the absence of the continuation and conclusion of Christopher Sly's plot, present, on the contrary, in A Shrew. In the Cambridge edition of The Shrew, Ann Thompson – unlike G. R. Hibbard, Brian Morris, and J. H. Oliver ¹⁰ – underlines how, in fact, ‘the Folio text as it stands could not have been used in the theatre’, adding that ‘the Folio text is unusual: the number of errors is uncommonly high in The Shrew’. ¹¹ W. W. Greg argues that ‘the entrances and exits and the designation of characters are not what we should expect to find in any manuscript that had been properly prepared for use on the stage’; the Folio ‘is obviously a makeshift and is more likely to have originated in the printing house than on the stage’. ¹² Stanley Wells seems to agree when he writes that ‘the Folio act division seems too incompetent to reflect theatrical practice at any period and is probably of printing house origin’. ¹³ The numerous mistakes and inconsistencies are evidence, writes Thompson, that ‘omissions are more likely to be a characteristic of the copy’ and that ‘the Folio text is a cut version of a longer play’. ¹⁴ In her recent edition, Barbara Hodgdon agrees that ‘F's copy probably derives from a transcript’ and that ‘there is some evidence of theatrical annotation and/or of revision or alterations’ and adds that ‘the editorial story invites viewing The Shrew as a composite text – one which offers an intriguing example of the pleasure (and dangers) of collaboration’. ¹⁵ If The Shrew, as Thompson suggests, depends on a longer version, what is the relationship between the two printed versions and between these and the supposed uncut version? The hypotheses are three: A Shrew is one of the sources of The Shrew; The Shrew is the source of A Shrew; A Shrew and The Shrew have a common third source, a lost Shrew play (Ur-Shrew). ¹⁶

Alexander Pope, though admitting that in A Shrew ‘there is scarce a line of this the same’ as The Shrew, could not help considering the old play as a sort of first sketch, ¹⁷ reputing the anonymous comedy even better. ¹⁸ He thus collated and conflated the two texts, inserting one of Sly's interludes, with cuts and rearrangements of lines, adding the Folio interlude at the end of the first act (p. 301), and two passages from A Shrew (sigs A3v, 9–12 and A4r, 14–17) ¹⁹ in the Induction of The Shrew (pp. 279 and 283). In F3, which includes Pope's manuscript annotations (Folger Shakespeare Library S2914 no. 20), we can find Pope's first insertions from A Shrew and his bizarre act and scene division. Pope reshuffled the scenes, turning 4.1 in the Folio into 4.4 (p. 223), breaking up a scene of the fourth act, and turning it into 5.1 (p. 222). In his edition, Sly's interludes from A Shrew (sigs A4r, 11–15 and A4v, 1) are added at the end of Ind. 4 (p. 283) and inserted after Ind. 5.12 (p. 284); two passages (sigs C1v, 21–8 and F3r, 24–31) are inserted at the end of act two (pp. 315–16), with some minor corrections and the dropping of two lines, and at the end of act four with cuts (p. 341). In 5.6 (p. 351; 4.4.32–42 in Hodgdon's edition), after Katherine's four-line speech to Vincentio, Pope inserted two speeches from A Shrew (sig. F1r, 1–14), which he preferred to the F text (p. 351). He inserted one of Sly's interludes at the end of 5.1 (p. 341); likewise, he used the Epilogue from A Shrew, cutting the first speech of the Tapster and a few other lines (A Shrew G1r, 19–25, G1v, 1–8). Of the four Sly interludes, plus the Induction and Epilogue, Pope utilised two speeches from A Shrew in the Induction and two of Sly's commentaries.

Until Edmond Malone, editors followed Pope and produced a text of The Shrew that was a sort of conflation of the Sly material in A Shrew and in the Folio version. Malone, in his 1790 edition, rejected Pope's interpolations, making the first attempt to attribute the authorship of A Shrew to someone other than Shakespeare (‘I suspect that the anonymous Taming of a Shrew was written about the year 1590, either by George Peele or Robert Greene’). ²⁰ These attempts continued throughout the nineteenth and the first half of the twentieth century. F. S. Boas rejected Marlowe's, Thomas Kyd's, or Greene's paternity, considering its provenance enigmatic and insoluble; ²¹ R. Warwick Bond, editor of the first Arden Shrew, assigned the authorship of A Shrew to John Lyly, ²² while E. K. Chambers designated Samuel Rowley. ²³

The second hypothesis (The Shrew as source of A Shrew) is the one that twentieth-century commentary has focused on the most. The third hypothesis (an Ur-Shrew) was accepted as alternative to the first by A. H. Tolman ²⁴ and as alternative to the second by Hardin Craig. ²⁵ George Ian Duthie, setting out from his pioneering study of the first quarto of Hamlet (1941), supported the theory of memorial reconstruction not of The Shrew but of a lost Shrew play, of which The Shrew is a rewriting. ²⁶ Peter Alexander in 1926 claimed that A Shrew is not an independent text but a particular kind of ‘bad’ quarto, a corrupted memorial reconstruction of the Shakespearean original. ²⁷ This hypothesis, a variant of the second, was accepted by John Dover Wilson in his 1928 Cambridge edition, but not all scholars have embraced it. Though a supporter of Shakespeare as a reviser, E. A. J. Honigmann rejected the revisionist theory for A Shrew/The Shrew, claiming that ‘A Shrew was composed after The Shrew, when the text of The Shrew was not available’. ²⁸ In their editions, Hibbard (1968), Morris (1981), and Oliver (1982) reaffirm Alexander's ‘memorial reconstruction’ theory. Thompson (1984), even though she admits the Elizabethans’ non-discrimination between the two plays, claims that ‘A Shrew is of no direct use in establishing an authoritative text for The Shrew’. ²⁹ Stephen Miller's old-spelling edition throws new light on the play. ³⁰ Miller supports the theory of adaptation, claiming that A Shrew ‘is a viable comic text of its period even if it is, as it seems, memorially derived in large part from The Shrew’. ³¹ What Miller does not say is from which text of The Shrew is derived (i.e., a longer lost version or the F version). I agree with Miller, but it seems to me that A Shrew is a much more complex play than has been hitherto realised. A Shrew, as Miller puts it, gives ‘a mine of information about theatrical taste of 1594’ ³² ; as such, by starting from this information it may be possible to have a new look at this ‘wretched play’.

The transmission of A Shrew is typical of most Shakespearean and Elizabethan texts, particularly of the so-called bad quartos. Arguably, the interest in a script depended on its stage success. There would seem to be little reason why actors and/or writers should meet to reconstruct it other than exploiting the success of a script whose rights belonged to someone else. This is the case of such early quartos texts as Hamlet Q1 (1603) or Romeo and Juliet Q1 (1597), but, unlike Hamlet and Romeo and Juliet, there is not for A Shrew a good quarto that ‘denies’ the ‘bad’ quarto and deprives it of its authority. On the contrary, as we have seen, A Shrew had two other editions in 1596 and in 1607. What distinguishes A Shrew from all the other ‘bad’ quartos is that it looks to be more derivative than other texts that are typically classified as memorial reconstructions. ³³ The original ‘has been handled more freely and rewritten more extensively (especially in the subplot) than is usually the case with memorially reconstructed texts’. ³⁴ The ways of transmission and the condition of the original also contributed to the diversity and the rewriting. Shakespearean scholars have argued for decades about the transmission of Elizabethan theatrical piracy and how original scripts were stolen from theatre companies. The theory of the stenographic transcription of the ‘bad’ quartos clashes with primitive stenographic systems, like Timothy Bright's Characterie (1588), which were not able technically to reproduce an acting speed of 20 lines per minute. Moreover, the extraordinary variety of Shakespeare's language could not be contained in Bright's system, where there was no alphabet, but only a number of characters corresponding to certain words to be learnt by heart. Nor could the method laid out in Peter Bales's Brachygraphie (1590) handle it, being a mere plagiarism of Bright's system. The only system, based on sounds, that might be able to provide a stenographic transcription of a theatre performance, was that which had been invented by John Willis in 1602, ³⁵ too late for most of the Shakespearean ‘bad’ quartos. Adele Davidson has tried to reconsider the relationship between Shakespeare and stenography, suggesting a link ‘between stationers, shorthand, and the theatre’. ³⁶ Starting from the tangible evidence of the use of shorthand in printing sermons, Davidson suggests ‘shorthand as a possible element in the provenience of early Shakespearean quartos such as Romeo’ ³⁷ and claims that ‘some person may have copied Q Pericles and/or Q Lear using (at least a few) techniques of short-writing, and thus may have influenced the printer's copy for these plays’. ³⁸ However, Davidson takes for granted the diffusion of Willis's method since the late 1590s; worse, she considers Bright's Characterie as being adequate to transcribe a theatrical performance. Claiming the early quartos to be stenographically transcribed in the theatre, or later by scribes, is total conjecture. It is not enough to claim that the differences between the two Lear texts can be explained through the conventions of Elizabethan shorthand. Shorthand conventions can explain some variants, but how can they explain the different titles and the unique passages in the Folio and quarto versions of King Lear, or the different sequence of scenes in Hamlet Q1?

If the ‘bad’ quartos could not have been transcribed stenographically, we must go back to the hypothesis of memorial reconstruction. To accept this hypothesis does not mean considering the ‘bad’ quartos to be simple piratical works; rather, it means also accepting the hypothesis of the mobility of the Shakespearean dramatic text. All of the Shakespearean texts transmitted either in the form of ‘bad’ quartos or in the Folio are only clues to the ‘mythical’ original text, a text that Shakespeare probably never wrote in such a static form, a text, on the contrary, that Shakespeare revised and adapted according to theatrical demands; a text that could even be modified in the hands of single actors, when the script was given to the company. Even Greg was conscious of the uncertainty of having a correct text when he wrote that ‘in the case of Shakespeare […] we cannot hope to achieve a certainly correct text, not so much on account of the uncertainties of transmission […] as because the author may never have produced a definitive text for us to recover’. ³⁹ If we accept the ‘revision theory’, ⁴⁰ then the ‘bad’ quartos are inexhaustible sources of information about the process of theatrical construction of Shakespearean plays. They are signs of changed stage conditions and changed theatrical demands.

In the case of A Shrew, we face a particular text that has its own autonomy, both theatrical and narrative, compared to The Shrew transmitted through the First Folio. It is because the canonical text presents dramatic and theatrical inconsistencies, and structural deficiencies, that the quarto – even bungled, even reconstructed with Marlovian fragments and passages from contemporary poetical texts such as Guillaume de Salluste Du Bartas – is extremely important both for its logical dramatic coherence and its story and plot structure, which show how the ‘original’ Shakespearean text could be at a certain time of its dramatic and theatrical history.

The Induction: Dramatic intertexts

The role of the dreaming man, employed in the two Shrew plays, draws on a common figure in literature from the Middle Ages to the Renaissance. The most ancient version is that of The Arabian Nights, whose tales go back to the ninth-to-tenth centuries CE. The tales originated in different cultural and literary traditions: the Indo-Persian, the Egyptian, and that of Baghdad. Variations on the theme circulated in Europe since the early Renaissance, starting from the versions of Marco Polo's journeys, translated into English by John Frampton in 1579. The second English translation is contained in the third volume of Samuel Purchas's His Pilgrimage (1613), ⁴¹ which comes from Ramusio. ⁴² The popularity of the story is testified also by some analogues of the sixteenth and seventeenth centuries, such as Heuterus's Rerum Burgundicarum (1583), ⁴³ David Chytraeus's Chronicon Saxoniae (1593), ⁴⁴ and Ralph Carr's The Mahumetane or Turkish historie (1600), ⁴⁵ a collection of fragments of Middle Eastern history translated from French and Italian. Versions in Latin circulated in the Renaissance and had their common source in a story told by Juan Luis Vives to Duke Francisco Béjar. ⁴⁶ The Catalan humanist, in a letter to the Duke, told how the Duke of Burgundy, walking through the streets at night, met a drunkard. To demonstrate his theory on human vanity, he took him to the court, making him believe himself a nobleman. After he spent a whole night at court, listening to music and even watching a theatre performance, the drunkard, again asleep, was dressed with his ragged clothes and brought back in the street. The first version of the story in English is told by Richard Barckley in A discourse of the felicitie of man (1598), ⁴⁷ where the Duke of Burgundy becomes the Emperor Charles V and the city of Bruges becomes Ghent. Later versions can be found in the English translation of Simon Goulart's Histoires admirables et mémorables (1606) by Edward Grimstone (1607), ⁴⁸ and in Robert Burton's Anatomy of Melancholy (1621). ⁴⁹

Hence, the story told in the Induction was indeed popular. Shakespeare could probably rely on an oral version and, according to Thomas Warton, also on a written version contained in a jest book published around 1570. Warton writes that he saw among William Collins's books

a Collection of short stories in prose printed in the black letter under the year 1570, “set forth by maister Richard Edwardes” […] Among these tales was that of the Induction of the tinker in Shakespeare's Taming of the Shrew: and perhaps Edwardes's story-book was the immediate source from which Shakespeare, or rather the author of the old Taming of a Shrew, drew that diverting apologue. ⁵⁰

The Induction: Theatrical intertexts

Elizabethan drama used a variety of introductory scenes that can be freely defined inductions, provided that we distinguish their dramatic and theatrical functions. According to Alfred Harbage, before 1594, 21 plays with inductions were performed or simply entered in the Stationers’ Register. ⁵¹ Eight of these plays, such as George Gascoigne's Jocasta (1573) and George Peele's The Battle of Alcazar (1594), present a dumb show as the introductory scene; eight other plays have what Thelma Greenfield defines ‘occasional induction’, ⁵² such as that of The Spanish Tragedy. Only four plays have the induction as frame play: A Shrew, The Shrew, George Peele's The Old Wives Tale (1588–94), and Robert Greene's The Scottish History of James IV (1590–91). After 1594, the forms of induction registered by Harbage are 49: eight in dumb show (such as John Marston's Sophonisba or the Wonder of Women, 1605–06) and only three in the form of frame play: Francis Beaumont's The Knight of the Burning Pestle (1607–10); Beaumont, John Fletcher, and Nathan Field's Four Moral Representations (1608–13); and Thomas Randolph's The Muses Looking-Glass (1630). Seventeen inductions have mainly an introductory function, presenting the play, and sometimes appearing throughout the play: for instance, Thomas Dekker's If This Be Not a Good Play the Devil Is in It (1611–12), in which the characters in the induction open and close the play. This kind of induction is similar to the frame play, but it lacks its dramatic development – see, for instance, the induction to Thomas Middleton's Michaelmas Term (1604–06) and that to Laelia (1595), an adaptation in Latin of the French translation of Gli ingannati, Les abusez, performed at Queen's College, Cambridge, in 1595. The other inductions from 1594 to the Restoration have been classified as ‘critical inductions’, where ‘human spectators appear on stage to watch a play and criticize it. Actors and tiremen rush about it in a last-minute flurry before the play opens’. ⁵³

As I have tried to show, neither Shakespeare nor the compiler of A Shrew had many dramatic and theatrical exempla for their frame play. The induction as dramatic and theatrical device was a way to emphasise the gap between stage and audience, make-believe and actuality, aimed at making the audience accept dramatic illusion. The Induction could have been influenced by the Italian prologues and epilogues, even though in the Italian Renaissance plays, they were mainly used to disclose sources and plot in the tradition of Terence and Plautus.

It is well known that the wooing subplot and the intrigues in Kate's paternal house have their source in Ludovico Ariosto's I suppositi (1509), or in George Gascoigne's English version Supposes (1566). Ariosto's debt towards Plautus is evident in the prologue to I suppositi, where, after Terence, the sources are disclosed. Less evident, and still a matter of dispute, is Shakespeare's debt to Plautus’ Mostellaria. ⁵⁴ As is well known, Shakespeare uses Plautus’ and Terence's Latin comedies as sources for the dramatic construction of his comedies. Mostellaria, like most of Plautus’ comedies, was not translated in the Elizabethan period, but Latin comedies were performed in the original by university companies at Cambridge and Oxford. According to G. C. Moore Smith, ⁵⁵ between 1548 and 1583 in Cambridge colleges, there were 17 performances of 12 of Plautus’ comedies. Among these was a performance of Mostellaria (Trinity College, 1559–60). That Shakespeare knew Mostellaria is evident from the names of Grumio and Tranio in The Shrew (but not in A Shrew, in which their names are Sander and Valeria), servants to Philolaches in Plautus’ comedy. Clues to the influence of Mostellaria in A Shrew are the names of two characters, Simo Senex and, above all, Philematio. In A Shrew, the Lord gives his name as Simon (‘Simon and it please your honour’ [A4r, 14]), and Christopher Sly decides to call him simply Sim (‘Simon, thats as much to say Simion or Simon / Put foorth thy hand and fill the pot. / Giue me thy hand, Sim’ [A4r, 15–17]). In Mostellaria, the character of Simo Senex is merely the owner of the house that Tranio pretends was bought by Philolaches. The coincidence between the two names could be casual, or simply a reminiscence, even though Simo Senex's lines on his wife (‘Atque pol nescio ut moribus sient / uostrae: haec sat scio quam me habeat male / peius posthac fore quam fuit mihi’ [3.707–10]) ⁵⁶ resemble very closely lines both about Kate and about Christopher Sly's shrew wife (‘Tapster. I marry but you had best get you home. / For your wife will course you for dreming here to night’ [G1v, 1613–14]). Sly and Simo Senex, as William E. Harrold writes, have wives who do not allow them ‘to take a nap at home’ and force them to go out of the house, the former to the tavern, the latter to the Forum. ⁵⁷ Moreover, in the Italian translation of the Mostellaria (Mustellaria) made by Geronimo Berardo and published in Venice in 1530, the name of ‘Simo Senex’ is ‘Simon vecchio’, and the speech prefix is Sim. In the case of Philematio, the influence seems also to be structural. In A Shrew, Philema, Alfonso's second daughter – corresponding to Bianca in The Shrew – is the object of Aurelius’ desire and the cause of the intrigues in Alfonso's house. In Mostellaria, Philematio, a courtesan, the object of Aurelius’ desire, is freed by him; she is the cause of the dissoluteness and the non-stop carousal at Theopropides’ house. What is more, the eavesdropping scene in the first act of Mostellaria, where Philolaches overhears Philematio and falls into a rapture over her, reveals his admiration and love:

Venus uenusta,

haec illa est tempestas mea, mihi quae modestiam omnem

detexit, tectus qua fui, quom mihi Amor et Cupido

in pectus perpluit meum, nec iam umquam optegere possum (1.161–4) ⁵⁸

On the contrary, in I suppositi a Ferrarese does it for him, and then Filogono asks him to do it again:

In A Shrew, in which the scene of the knocking at Aurelius’ door is missing, Jerobel (corresponding to Vincentio), upon his arrival in Athens, first meets Ferando and Kate (E4v), then his son, his servant Valeria, the Pedant, Alphonsus, and his daughters (F1v–F2r). Finally, both Mostellaria and A Shrew are set in Athens. If this is logical for Mostellaria, whose characters have Greek names, it is not so for A Shrew. In A Shrew, whose characters have English and Italian names, the setting in Athens seems due to imitation.

William E. Harrold lists a few similarities between Mostellaria, The Shrew, and A Shrew: Mostellaria has four similarities with The Shrew (the two characters Grumio and Tranio, the servant-beating-servant motif, the knocking at the door by the father, and the banquet); three with A Shrew (the setting in Athens, the names of Simo and Philematio); and seven with both The Shrew and A Shrew (country-town distinctions, music girl, motif of begging pardon from punishment, Tranio or his equivalent as a prompter of vice, drunken man who calls for drink as he falls asleep, man who has a shrew for a wife, young men [wooers] who are friends). ⁶² All these similarities, regarding both plot and dramatic structure, raise the possibility of a double process of revision involving the two Shrew plays and a supposed third uncut version. It is possible that most of the similarities between Mostellaria, A Shrew, and The Shrew were all in the uncut version. The Folio version maintains the similarities except those connected with the Athens setting, the Greek names, and the reference to Simo. A Shrew does not keep the names of Grumio and Tranio, the servant-beating-servant theme, and the banquet. It seems that, if we accept the revision theory, Shakespeare wrote a version following the folk tradition (the three sisters’ theme), ⁶³ Plautus, and Ariosto; then, revising the play, he went back to Ariosto, deciding to keep only two sisters as in I suppositi, changing the setting from Athens to Padua, and changing the names of all the characters, except Kate, Grumio, and Tranio. It is impossible to say if he decided to drop Christopher Sly's interludes and the epilogue, or if the dropping was due to new theatrical demands in the 1600s. It seems improbable that both the changes in the setting and names and the adding of one sister are due to the compilers of A Shrew.

The intertexts underlying the composition of the two Shrew plays throw light on their relationship. As I have shown, evidence of possible revision of an uncut version of The Shrew can be found in the relationship both plays entertain with Plautus’ Mostellaria. Useful to this purpose are also other intertexts underlying the two subplots, The Journeys of Marco Polo, George Gascoigne's Supposes, and Ariosto's original I suppositi. From this perspective, the intertexts of The Shrew/A Shrew appear to be a palimpsest of quotations, revisions, and translations, rendered the more obscure by the absence of the lost ur-text.