This dream informs Merchant in another instance of the conjectural letter—or conjectural letters—and a putative relationship to presence. The critical textual moment for the Salerio/Salarino/Solanio issue occurs within the play's most important staging of the scene of reading—in 3.2, the moment (discussed above) just after Bassanio has made the correct choice of the lead casket, and a character arrives carrying Antonio's letter. It is the identity of this character that has caused considerable debate. The 1987 New Cambridge Shakespeare edition, edited by M. M. Mahood, identifies the three characters in its "List of Characters" thus:

Solanio
    gentlemen of Venice, and companions
    with Bassanio
Salarino

Salerio, a messenger from Venice

The entry for Salarino is noted at the bottom of the page: "He may very probably be the same character as 'Salerio'," and we are asked to consult the "Textual Analysis" that supplements the text.³⁰ In the pages of the "Textual Analysis" devoted to a discussion of these characters, Mahood offers a careful review of the parameters of this textual problem and the solutions to it offered by various editors:

Earlier editors of the play were reluctant to believe that Shakespeare, after naming two characters "Salarino" and "Solanio" … would have made confusion worse confounded by bringing on a third character called "Salerio." To have created so superfluous a character would have violated "dramatic propriety," put the actors to unnecessary expense, and shown a singular lack of inventiveness in the choice of names. … In the New Shakespeare edition of 1926, [John Dover] Wilson concurred with Capell in making Salarino and Salerio one and the same person but decided that Shakespeare's name for him must be "Salerio" since this occurs five times in the dialogue. He therefore substituted "Salerio" for "Salarino" or its variants in all previous stage directions and speech headings. All subsequent editors have followed Wilson in this, and Salarino has not put in an appearance for the past sixty years. On a number of grounds, I have restored him to the text of this edition. (M, 179)

Mahood argues there is "no prima facie case against Shakespeare having had three different personages in mind. On the other hand, the positive evidence in favour of three characters is admittedly slight" (M, 179). After a lengthy discussion of the various arguments both for and against the eliding of Salarino and Salerio, Mahood decides to maintain the distinction between these characters within the text, while noting in the textual apparatus the possibility that this decision may be untenable. This decision is underwritten, however, not by an argument for one character over the other, but is instead guaranteed by an appeal to a reputed authorial intention or the (lost, conjectural or—at the very least—the specular) authorial script:

It is always open to the director to identify Salarino with Salerio, thereby economising on minor parts and very probably fulfilling Shakespeare's final intention into the bargain. But the printed text must, I believe, retain three Venetian gentlemen with similar names because, whatever his intentions, Salarino, Solanio, and Salerio all figured in the manuscript that Shakespeare actually gave to his actors as The Merchant of Venice. (M, 183)

Embedded within this final comment are a number of crucial issues. To begin with, Mahood accepts a fundamental distinction between the play as it is performed and the play as a text: in the first instance, the textual stand taken vis-à-vis Salarino/Salerio simply doesn't signify; in the latter, the textual becomes occasion for taking a stand. In other words, this textual matter finally doesn't matter if the play is imagined in performance—as spoken language—, but matters a good deal more if it is instead imagined as a text—as written language.

This constitutes a performative version of the logocentrism described by Derrida: spoken language is imagined as prior to and more immediate than the written, with the consequences in this particular instance being that in production the play is substantially different in such a way as to allow an editorial emendation that in print would be inadmissable. At the same time, Mahood suggests that whatever the decision in performance, in print the three characters must nevertheless still appear. The performed play, then, enacts yet another splitting, reifies the posited distinction between performed and textual play, as an actor may be—in performance—Salerio while in print he may (still) be Salarino.³¹

There is another issue at stake in Mahood's double-vision of a single version of the play, and it is an issue relevant to our understanding of Merchant more generally. In the above paragraph Mahood identifies the three characters as "three Venetian gentlemen," while in the "Textual Analysis" she suggests that their status as "gentlemen" is perhaps open to some question, and that, moreover, Salerio may not be a gentleman at all, as his nomination "a messenger from Venice" may well suggest:

a messenger from Venice (3.2.218 SD) could imply that not only is Salerio not to be confused with the two men-about-town, but that his social status is rather different. Gratiano's "My old Venetian friend Salerio " (218) need not imply equality; it can be a condescending form of address and also an explanatory phrase such as the audience would not need if it had met Salerio four times already. Salerio … can be seen as a kind of state functionary … . This would accord with his role in the trial, where he is a kind of gentleman usher. (M, 181-82)

But Salerio's social status is not the only one at stake: while the "social nuances of four hundred years ago are not … something on which we can speak with confidence today," Mahood suggests, "it would be quite easy to make out a case, in the play's first scene, for a social difference between Solanio and Salarino on the one hand and Bassanio's more immediate group of friends on the other" (M, 182). Mahood clearly brings certain notions of class and class distinction to the play, and just as clearly suspects Shakespeare to have done so as well.

While Mahood's decision to retain Salarino seems to depend in part upon his presumed class-based differences from Salerio, it is in fact underwritten by an unquestioned adherence to the tenets of traditional textual criticism—particularly the faith in authorial intentionality. In this regard, then, Mahood's inclusion of Salarino is effectively no different from Wilson's exclusion of him.

Wilson's discussion of what he calls "the muddle of the three Sallies" (W, 100) is a careful analysis of this textual problem and has stood as the almost universal "resolution" reproduced by every editor of the play until Mahood. Wilson's argument—that "Salarino" is a repeatedly misrecognized or misprinted version of "Salerio"—is heavily indebted to a complex textual genealogical argument in which the copy-text for the 1600 Hayes Quarto is believed to have been pieced together not from prompt-books or manuscript (the latter is the argument favored by recent editors), but from what Wilson calls "secondary theatrical manuscripts" (W, 105). Wilson finds corroborating evidence for this conclusion in a number of the play's more striking textual characteristics: the evident addition of texts into the play—specifically, the letters read aloud in 3.2 and 4.1 and the three scrolls of the casket scenes—, the play's stage directions, the related matter of the "three Sallies," and what Wilson deems the evident playhouse additions to the play.

Wilson notes the curious textual features associated with the letters and the scrolls—that they are bibliographically "marked off" within the Quarto, and that for each a speech heading is missing. This bibliographical distinctness, Wilson claims, is "a textual fact of capital importance":

For the absences of prefixes before the letters and the duplication of prefixes in the speeches afford clear evidence that both letters and scrolls are, bibliographically speaking, textually distinct from the rest of the copy, or in other words, insertions … . Any text, therefore, in which letters, songs or scrolls are seemingly insertions, is to be suspected of being derived, not from the original "book," but from some secondary theatrical source, composed of players' parts. (W, 97-98)

The "frequent vagueness" of entry directions ("Enter Bassanio with a follower or two" [2.2.109], one of whom later turns out to be Leonardo; the entry for the "man of Portia's" whom we later learn is Balthazar, and the "Messenger" [5.1.24] who "is discovered four lines later to be Stephano, one of Portia's household") prove that, as Wilson had argued earlier in the New Shakespeare edition of The Comedy of Errors, "the dialogue had been copied out (from the players' parts) by one scribe and the stage-directions supplied by another … who possessed very vague ideas of the text he was working on" (W, 100). It is precisely this "scribe responsible for the stage-directions" whom Wilson holds accountable both for the "muddle of the three Sallies" and for the general textual state of the entire Quarto.

In his argument for resolving the Salerio/Salarino crux, Wilson lays the responsibility for the problem entirely at the hands of the scribe, reconstructing, based upon his sense of evidence, what must have happened in the scribe's production of the text:

Whence then came this curious "Salarino"? If we assume, as we have already found ourselves entitled to assume, that the text before us was made up of players' parts strung together, transcribed and then worked over by a scribe who supplied the stage-directions, the reply is not difficult. … This scribe had before him at the outset, we must suppose, a transcript from the parts containing only the bare dialogue and the abbreviated prefixes, so that he would be obliged to rely upon his memory of the play upon the stage for the full names of those characters which were not mentioned in the dialogue itself. Now the form "Salarino" is found, apart from the stage-directions, nowhere in the dialogue and in only one prefix, which occurs at 1.1.8. The prefix "Salari" (which is of course a variant spelling of "Saleri") is, on the other hand, fairly frequent. The beginning of all the muddle, we suggest, was that the scribe found the prefix "Salari" in his text at 1.1.8, took it as a contraction for "Salarino," added "no" to it, and framed his entry-direction accordingly. It accords with this theory that the only time we get the erroneous "Salanio" in the prefixes is at 1.1.15 … . Clearly, we think, the meddling scribe made the two changes at the same time. (W, 103-104)

From this description of an imagined scene of scribal intrusion and disruption of the Shakespearean text, Wilson constructs an entire narrative of the scribe's work and his absolute consistency in his erroneous and meddling ways:

"Salarino" (or "Salerino") marches happily along in the stage-directions hand in hand with "Salanio" (or "Solanio") up to the end of 3.1, by which time the former name had become such a habit with the scribe that when he comes upon "Salerio" in the dialogue of 3.2 he quite fails to recognize his identity and puts him down as "a messenger from Venice." (W, 104)

The final evidence for Wilson's theory of the "assembled text" is what he identifies as the playhouse additions to the play itself, arguing that "texts derived from secondary theatrical manuscripts are likely to preserve traces of actors', or at least of playhouse, additions." Wilson identifies an early section of 5.1 as such a trace—a "piece of 'fat,' as the modern actor would call it, [that] has clearly been inserted in the text": the prose lines introduced by Lancelot's repeated "sola's" and concluded with what Wilson conjectures is the misassignment of "sweete soule" (W, 105). In his analysis of the significance of the textual irregularities he finds in this brief passage, Wilson has recourse to the assistance of W. W. Greg ["whose authority on matters of this kind is unrivalled" (W, 106)]; when asked by Wilson what he made of the "sweete soule" matter, Greg theorized a version of the assembled text argument:

I think it is pretty clear that the preceding passage was an insertion in the margin, or more probably on a slip, ending up, as was usual, with a repetition of the following words to show where it was to come. The sense shows that the insertion must have begun with the Messenger's words: "I pray you is my Maister yet returnd?" I suppose that the printer finding the words repeated in the MS, omitted the second occurrence. The compositor would not be very likely to do this, but a proof-reader might—or there may have been an intermediate transcript. (qtd. in W, 106)

Authorized by Greg's words, Wilson continues his argument by wondering why there should be this addition at all—especially as "the passage … might be omitted without any injury to the context." The answer, Wilson declares, is simple: "to give the clown who played Lancelot an opportunity of making the theatre ring with his 'sola!'." "Evidently," Wilson concludes, "the clown in Shakespeare's company, Will Kempe presumably, was fond of caterwauling tricks" (W, 106-7).

Let us for a moment consider the rhetoric of this derisive passage which manifests a certain ideological bias brought to bear not only on the passages under review, but to the editing of the entire play, and, moreover, to that play's meaning.³² In this passage Wilson makes the small but serious mistake of referring to Will Kempe not as the comedian of Shakespeare's company, but as its clown. To confuse or conflate the two is to eradicate any distinction between actor and the part an actor might play upon the stage; the consequences of this confusion are significant. In Wilson's rhetoric, Kempe literally is a clown, and as such occupies the same position in the space of the social world that a clown does in the space of the theater. So Kempe's addition here—his "piece of 'fat'"—is pure clowning, but clowning with serious ramifications. For Wilson, Kempe's addition represents nothing less than the eruption of chaos and disorder into the otherwise decorous and high-aesthetic world of the Shakespearean play. Kempe becomes the sign of both social and aesthetic disruption and literal (Ietteral) textual corruption.

Wilson's vision of Kempe as the figure of radical instability does not end here, however, for as Wilson says, "if an addition was made to this 'assembled' prompt-book at one place, why not at others?" (W, 107); the text stands hopelessly vulnerable to the pernicious effects of Kempe as the socially and aesthetically disenfranchised figure of instability and subversion. Wilson identifies a second "prose-patch, this time of a ribald nature," in 3.2:

It is pretty certainly a textual addition, and we suspect that it was made by the same hand as wrote the 'sola' slip. Indeed, we are inclined to go even further and to attribute a whole scene to this unknown scribe. (W, 107)

The passage under review here—the opening 59 lines of prose—includes Lorenzo's famously obscure charge, "the Moor is with child by you Lancelot!" (3.5.35-6), and ends when Lancelot exits to prepare dinner. Wilson argues that not only is this so-called prosepatch an addition, but that the entire scene was (again) instigated by Kempe:

It is the verse with which the scene closes that seems to provide the clue we are seeking. The first five and a half lines of this verse are a tribute to Lancelot, or rather to the actor who played him, while the reference to "A many fools that stand in better place" is obviously intended as a hit at some successful rival. In a word, we suggest that Shakespeare had no hand whatever in the composition of 3.5, which might be omitted altogether without loss to the play; that it was added to the "assembled" prompt-book at the same time as the insertions at 5.1.39-49 and 3.2.214-18; and that while 3.5.60-5 was written by some second-rate poet as a compliment to William Kempe, Kempe himself may have been responsible for the very dull fifty-nine lines of prose with which the scene opens. (W, 108)

Wilson concludes his discussion of the copy for the Hayes Quarto by suggesting that Kempe not only presumed to write in Shakespeare's hand, but, also that it was he who was the "unknown" and "meddling" scribe Wilson's theory of the text had posited:

To sum up, our contention is that the manuscript used as copy by Roberts' compositors in 1600 contained not a line of Shakespeare's handwriting, but was some kind of prompt-book made up from players' parts, to which a theatrical scribe (maybe Kempe himself) had added stage-directions and additions of his own devising. (W, 108)

For Wilson, Kempe's intrusive and radically disruptive acts of destabilizing self-promotion are complete, but at a material cost to the integrity of the Shakespearean hand and text. Wilson's theory of the production of Merchant attributes virtually everything that is of uncertain authority and authorship—and therefore everything that is deemed aesthetically bankrupt—fully to the hands of Will Kempe.³³

These suspicions of Kempe's destabilizing presence in Merchant betray Wilson's fundamental distrust—not to say fear—of the lower class of which Kempe is made to stand as the embodiment. Wilson's "aristocratic" position, in turn, stands in steadfast opposition to such a disruption, as it seeks to guarantee the "sovereignty" of the Shakespearean texts against dissent, disruption or subversion "from below." This is precisely the sort of political and critical conservatism Terence Hawkes has so brilliantly analyzed in Dover Wilson's career as a "social" writer on Russia and its revolution, and as a literary critic.³⁴ Hawkes describes Wilson's conservatism (like Tillyard's) as "a version of what, by the time of the second world war, had become a standard British response to national crisis: the construction of longpast, green, alternative worlds of percipient peasants, organic communities, festivals, folk art, and absolute monarchy to set against present chaos."³⁵ Such a vision imposes a radical reconstruction of "peasant" and "folk" culture as happily acquiescent to the absolute monarch. This is an Edenic vision of folk culture that fails to see in it any potential source of subversive energy, any potential for misrule.

But this vision is not imagined, however, as necessarily natural. In fact, it takes the very deliberate and careful intervention on the part of people such as Wilson (and their appropriation of figures such as Shakespeare) to produce it, to identify potentially disruptive people such as Will Kempe, and recreate them as docile (royal) subjects. This is achieved, in Wilson's view, through both a well-regulated and maintained aesthetic and nationalistic education.³⁶

Wilson's political conservatism (like his critical and editorial conservatism) is dedicated to the preservation of so-called traditional values: Nation, high-aesthetic value, and the sovereign individual—whether that individual is Shakespeare or Tsar Nicholas II. And these transcendental values are themselves underwritten by a Hegelian philosophy and historiography that understands human activity not merely as diachronic, but as processive and, finally, teleological.

It is precisely against this teleological or exclusively linear model that Hawkes offers Telmah. For Hawkes, Hamlet is structured on the model of recursivity: events, words and phrases appear and then are replayed again. Hawkes warns us, however, not to be deceived by this recursivity and its symmetries:

It would be wrong to make too much of "symmetries" of this sort, and I mention them only because, once recognized, they help however slightly to undermine our inherited notion of Hamlet as a structure that runs a satisfactorily linear, sequential course from a firmly established and well defined beginning, through a clearly placed and signaled middle, to a causally related and logically determined end which, planted in the beginning, develops, or grows out of it.

Like all symmetries, the ones I have pointed to suggest, not linearity, but circularity: a cyclical and recursive movement wholly at odds with the progressive, incremental ordering that a society, dominated perhaps by a pervasive metaphor of the production line, tends to think of as appropriate to art as to everything else.³⁷

The metaphor of the production line bespeaks a deeprooted notion of (historical) progress and it is this philosophy of progress that authorizes and determines Wilson's editorial practices and produces his version of Merchant. Moreover, this philosophy of progress and the epistemology of presence together have powered traditional textual criticism, regardless of local responses to textual problems. Mahood's decision, for instance, to retain or restore Salarino to the play is a good one, though I disagree with her traditionally-determined reasons for doing so. Our current understandings of (Shakespearean) textuality no longer require or endorse the appeal to authorial intention or authorial script. My argument is more concerned with the untenable nature of traditional editorial practice typified by Wilson than with evidential weight behind retaining Salarino. In fact, it seems to me not much to matter how there came to be three characters with such names in the Hayes quarto, but simply that there came to be these three "letteral" configurations we have decided to call characters. The matter of the three Sallies is important here not because it stands as yet another site for our intervention in the attempt to solve a textual crux, but rather precisely because it marks the eruption—inexplicable and yet undeniable—of the accidental.

In the anticipated aftermath of the collapse of traditional textual criticism, can we theorize a textual practice and a theory of textuality not determined by a Hegelian processive philosophy?