Early Modern Literary Studies
Hell and Hypertext Hath No Limits: Electronic Texts and the Crises in Criticism
Hilary J. Binda
Tufts University

Binda, Hilary J.  "Hell and Hypertext Hath No Limits: Electronic Texts and the Crises in Criticism." Early Modern Literary Studies 5.3 / Special Issue 4 (January, 2000): 8.1-29 <URL: http://purl.oclc.org/emls/05-3/bindmarl.html>.


  1. In the introductory essay to the recent collection Hyper/Text/Theory, George Landow cites Jean-Francois Lyotard: "the spectacular introduction of what are called the new technologies into the production, diffusion, distribution and consumption of cultural commodities . . . [is] in the process of transforming culture."[1] While many contemporary critics of hypertext take for granted Lyotard's implicit argument that a causal relationship exists between technological innovation and cultural transformation, Landow points to the ways in which cultural shifts simultaneously produce technological changes. Textual innovations like electronic hypertext -- multilayered documents that enable internal and external multidirectional linking -- might be read as a manifestation of cultural shifts that occur and recur long before their technological counterparts. This paper will consider a dramatic text from the early modern period, Marlowe's Doctor Faustus, as a representation of cultural anxieties analogous to, if not coterminous with, those anxieties more recently associated with the advent of electronic hypertext. Faustus' obsession with the written word and his desire for endless knowledge very likely register the spectre of infinite knowledge made possible (and, also, frustrated) during the Renaissance by the spread of the printing press and the increasing circulation of printed material. Rather than read this material history causally as the source of the anxieties surrounding language that this play concerns itself with so centrally, I want to suggest that in the early modern period as well as the twentieth century the condition of language itself functions as the source of anxieties expressed both by Marlowe and by many contemporary writers on the new technology.

  2. Distinguishing his reading from Lyotard's more admonitory perspective on the new technologies as potential systems of domination,[2] over the past two decades Landow has contributed enormously to a more optimistic and technologically engaged discussion on the possibilities of the new electronic medium. The credit and praise due to him for his serious and continuous efforts to theorize the conjunction of the new technologies and the field of literary criticism is so familiar to any who work in this field that it need not be repeated here. I want to begin by considering a particular line of thought apparent in the article by Landow cited above, a line of thought that is also evident in other analyses of hypertextuality. Focusing his discussion around hypertext in particular, Landow argues that "[N]etworked hypertext, which offers liberation, idiosyncrasy, and even anarchy, obviates the kind of control feared by Lyotard and other cultural prophets who confuse increasingly old-fashioned centralized mainframe computing with the new information technologies as a whole" (33). I am particularly interested here in the utopian, liberationist terms with which many theorists of hypertext are making sense of this new and increasingly widespread technology.

  3. Landow's description of hypertext as offering "liberation, idiosyncrasy and even anarchy" is fairly common in contemporary analyses of hypertext that claim it as a poststructuralist or "nonlinear" practice. In "Nonlinearity and Literary Theory," for instance, Espen Aarseth distinguishes hypertext from what he calls "linear" electronic texts, suggesting correctly that electronic texts can simply replicate the more linear structure of print and that the electronic medium alone is not enough to initiate the new mode of textual production, what Landow calls the "major paradigm shift" (35) marked by genuine, "nonlinear" hypertextuality (Aarseth 52). Aarseth's analysis is useful on a number of fronts, not the least of which is its close attention to the particular differences between distinct hypertextual electronic forms, such as the nonlinear "novel," the cybertext, and the MUD (Multi-User Dungeon). His aim in considering the juncture of electronic publication and literary theory is to work out a definition of textuality that has been tested against these relatively new forms. While Aarseth's project is an important one, his desire to distinguish between what he calls "text" and "script," where script represents the "visible words and spaces," seems to reinscribe the traditional content/form binarism (53). He calls for the study of "text as information," as a kind of Ur-text that exists "beyond form," despite his claim to perform a deconstruction of this model of transcendent textuality (58). Although I do not have Aarseth's great breadth of knowledge about electronic forms, I want to propose here a challenge to certain unspecified assumptions in his and Landow's analyses -- first through a discussion of my own experience creating or editing a hypertextual edition of the complete works of Christopher Marlowe, developed and maintained at the Perseus Project at Tufts University, and second through a brief reading of Marlowe's Doctor Faustus.

  4. In "Nonlinearity and Literary Theory," Aarseth enriches Landow's provocative, Barthesian formulation of the hypertext reader-as-author: accordingly, the reader-author implicitly imposes some degree of mastery over a text and over the process of reading-become-writing. Such a model posits the reader as being somewhat in control of the text he or she produces through the selection of individual pathways or links that constitutes the "writing" of the text. Aarseth addresses Landow's additional claim that the quantity of text available in an electronic hypertext troubles textual mastery: rather than simply produce in the reader-author a sense of control, this new process of reading, Aarseth suggests, also creates the feeling that he or she is missing something, possibly something important. I want to suggest that this paranoia-inducing instability inherent in the reading/writing of hypertext is not finally so different from the process of reading what Aarseth calls a "linear" book-form text.

  5. I will argue here that the theoretical production of a distinction between linearity and nonlinearity, apparent in numerous scholarly and popular culture representations of hypertext, reflects an attempt to hold apart two characteristics of textuality that, in fact, exist in tension rather than as two separate phenomena; and, I will argue, this attempt is necessarily vexed. Without discounting the genuine transformations in modes of reading and writing in which the increasingly widespread incorporation of hypertext undeniably participates, and without returning to a Lyotardian conservatism (a concern with the "dangers" of the new technologies), I will now explore what appears at times as an almost messianic strain figured as "nonlinearity" in some contemporary theories of hypertextuality.


  6. What gets called the "crisis in literary theory" by George Landow, Espen Aarseth and others responding to the conjunction of the electronic medium and literary analysis, turns on the notion that critics need the same text to perform their work. As readers become writers and texts become the effects of a reader's individual selection of hypertextual pathways, hypertext erases textual reproducibility. Landow argues further,

    In the nonreproducible text, critics find themselves in a situation analogous to the pre-print world, in which scribal drift insured that one could never be sure that another reader had read precisely the same text. Students of oral and preliterate culture have argued that criticism as we understand the enterprise derives directly from print culture, specifically from printed text's fixity and multiplicity - qualities that permit the critic to write for other readers who can consult the "same" text. What then will the critic do? (35)

    While drawing on the provocative analogy between textuality before the printing press and after the Internet, Landow is making a logical and straightforward point about the threat the electronic medium may pose to the field of literary analysis. I want to examine, perhaps more closely than Landow intended, another way that this formulation signifies and perhaps runs against the grain of his own progressive argument.

  7. Specifically, this theorization positions the printed text in opposition to hypertext; where hypertext allows readers to pursue endless pathways such that, as Landow claims, "Quantity removes mastery and authority"; the printed text is implicitly defined as having "fixity" (35). Certainly Landow himself would not endorse the notion that a printed text produces fixed meaning, and nor would he, I think, intentionally establish this dichotomy. The logic of his argument, however, depends on it: one implication of this analysis is that the printed text could in fact be mastered or completely understood since, according to this formulation, it is both finite and the "same" in each reading, its meaning therefore determinable by the critic. Despite the radical claims of many contemporary theories of hypertext, setting hypertextuality in opposition to what is implicitly defined as the inherently linear qualities of the book seems to perform an ultimately anti-hypertextual move. Under the guise of a post-structuralist unsettling of meaning as we know it, the deployment of the trope of hypertextuality may ironically effect, or claim to effect, the stabilization of the word. Certainly a level of comfort, a purging of paranoia, is afforded by this particular deployment of hypertextuality; the reader/user is, after all, covertly and ironically assured of the potential stability of the text "beneath" the hypertext. The notion of a transcendental Ur-text that Aarseth challenges returns in this revolutionary rhetoric that accrues around the concept of hypertext.

  8. Not only is the word (and, perhaps, textuality more broadly) produced as a stable function through its theoretical opposition to hypertextuality, but the trope of hypertextuality becomes a means of "grounding," or making familiar the unfamiliar. In the essay by Aarseth noted above, he uses a sentence of Derrida's as an epigram: "The future can only be anticipated in the form of an absolute danger." The implication in this textual conjoining, of course, is that "nonlinear" hypertext occupies this space of the future as absolute danger. Despite this apocalyptic claim, the theoretical assertion of hypertext's absolute nonlinearity has the paradoxical effect of securing a linear mode of analysis: within this particular logic, familiarity with hypertext may be read as that which would provide us with access to this telos of future-as-absolute-danger. This utopian gesture --  hypertext's promise of "liberation, idiosyncrasy and even anarchy" -- effectively promises to reveal the unknown, the structureless realm of nonlinearity, endlessness or limitlessness. And yet hypertextuality can't possibly fulfill this promise since, as Derrida has shown in so much of his work, the future-as-absolute-danger must remain in place as the condition of textuality, the very condition of our teleologically organized lives and deaths. Ironically then, this notion of hypertextual apocalypse guarantees its own failure to deliver on its promise of liberation. My argument is not one for a more rigorous theorization of nonlinearity with regards to hypertextuality, but rather for an analysis of this particular trope as necessarily inscribed within and thus simultaneous with the discourse of linearity and limitation, of signifying chains and the inevitability of their ends in meaning.


  9. I will now situate my discussion of the production of an electronic hypertextual edition of Marlowe's works within a schematic history of relevant debates within the field of textual criticism. Next, I will consider some of the ways that our electronic version of the complete works of Christopher Marlowe begins to address both what Jerome McGann calls the crisis in textual criticism and George Landow's question: in the age of hypertext, what will the critic do?

  10. In 1983, Jerome McGann identified a crisis in textual criticism that had been brought to a head by the enormously influential Fredson Bowers and the New Bibliography. Central to this mid-century editorial philosophy was the concept of the "eclectic text": when an author's manuscript no longer existed, as is so often the case with early modern texts, the job of the editor was to develop methods for eclectically reconstituting the lost original document from the various extant editions. As McGann explains, in the case of the 1608 Quarto and the 1623 Folio of King Lear, according to the New Bibliographers "the 'eclectic' text of the play . . . will be, presumably, closer to 'what Shakespeare originally wrote' than either of the early printed texts" (4). One of McGann's important interventions was to suggest that "if . . . these are not two relatively corrupted texts of a pure (but now lost) original, but two relatively reliable texts of two different versions of the play (as we now think) -- then our general method for dealing with such texts is called into serious question" (4). McGann's argument is synthesized by D.C. Greetham in the book's Forward: "there is no 'text itself,' . . . the search for the single unitary utterance characteristic of the modernist well-wrought urn is a chimera" (xv). In place of the "eclectic text," McGann proposes instead "versions, historically mediated texts, texts in infinite regress from their moment of composition, materialist texts that achieve meaning only through the continued negotiation with the institutions of their reception and transmission."

  11. This crisis in textual criticism -- according to which "editing is no longer assumed to be merely an empirical procedure, ungrounded in ontological and epistemological attitudes and beliefs" -- is intrinsic, I believe, to what Landow and others have identified more recently as the crisis in literary theory precipitated, or at least registered, by the advent of electronic hypertext (McGann xi). As we have seen, Landow argues that reading electronic hypertext often does away with an "original" or reproducible text as the transcendental eclectically reproduced original text is effectively fractured in McGann's analysis into versions.

  12. Recent editors of Marlowe's Doctor Faustus, a play that presents extremely complex textual and editorial issues, have followed McGann's lead. David Bevington and Eric Rasmussen, most notably, analyze the textual history of this play's multiple versions, the 1604 A-text and the 1616 B-text.[3] Tracing the changes in editorial analyses of the two versions provides the basis for an interesting historical study of the shifting attitudes toward the process of editing early modern texts that Bevington and Rasmussen summarize. In short, the play's earliest editors, Dilke in 1814 and Oxberry in 1818, for instance, use the longer and more comic and theatrical B-text as their copy text. Nineteenth and early twentieth century editors, however, prefer the A-text because it does not appear to contain the B-text's Birde-Rowley additions and revisions. The A-text is thus selected on the basis of its being less adulterated. As Bevington and Rasmussen further delineate this editorial history, the mid-twentieth century witnesses a return to favor of the B-Text, taken to be the original text rather than a poor memorial reconstruction, as was believed to be the case of the A-text. Most recently, and largely due to the tenacious scholarly sleuthing of Eric Rasmussen, the repressed A-text has returned to favor again. This time, no longer as a memorial reconstruction but as an authorial manuscript with scenes interleaved from two dramatists.[4]

  13. I summarize this history here by way of supporting Rasmussen and Bevington's claim that "[c]ritics and editors need to be keenly aware of differences between the two texts" (47). They argue further,

    Faustus' use of verbal magic, so carefully dramatized in the A-text as he moves from a gloating sense of his own near omnipotence to helplessness and despair, is replaced in the B-text by an incoherent delight in stage trickery and a demonstration of control and mastery when Faustus ought to be seen as increasingly out of control. Critical differences such as these between the two texts offer confirming evidence that the A-text is close to the Marlovian original and that the B-text trivializes the very nature of Faustus' tragic experience by its endless appetite for stage contrivance. (47)

    By restoring the full textual autonomy of each text, we can avoid the traditional practice of debasing one in favor of the other. Rasmussen and Bevington's approach to the text clearly requires revisiting textual variants. Leah Marcus has now famously named this approach to the early modern material text "Unediting the Renaissance," a process that Marcus suggests "will allow for the perception of elements of consistency and for a recognition of difference" (65). As Bevington and Rasmussen conclude, and as Marcus later demonstrates, "Both texts of Doctor Faustus continue to deserve our divided attention" (Bevington and Rasmussen, eds., 77). It is precisely this divided attention that the Perseus Project's electronic edition of Doctor Faustus aims to facilitate.

  14. The electronic Marlowe hypertext housed by the Perseus Project allows for the "simultaneous" reading of the A and B versions, while identifying the links between them, thereby obviating the editorial dilemma of choosing the copytext. In creating this site, I used W. W. Greg's 1950 version of the A and B texts as a model for linking lines, passages and scenes between the two texts. Greg's parallel text edition retains traces of the New Bibliography's unifying impulse: the textual gaps and blank pages in Greg's edition, for instance, give the reader the sense that things are missing from the shorter A-text. This was, of course, consistent with Greg's own analysis of the B-text as closer to the original and thus more authentic than the A-text. Our electronic edition does not induce the same sense of what Marcus describes as the "textual paradise lost" of Greg's parallel edition; the differences in organization and display in the electronic edition preserve a greater degree of textual autonomy. The electronically linked A and B texts offer the reader the choice of viewing the original spelling and punctuation or our modernized version of either text, while also enabling the simultaneous display of the similarities and differences between the two texts.

  15. The Marlowe site begins to exploit the potential of the new medium to do more than simply reproduce the book. It points to the capacity of the electronic text to expand textual boundaries, in a sense, through a redefinition of textuality as inclusive of a set of links between texts. In our site, these links also include those developed between the play and its primary "source," P. F.'s 1592 prose narrative entitled The History of the Damnable Life and Deserved Death of Doctor John Faustus (also known as the English Faust Book). A reader may choose to read this text in its entirety, to do so with a view of links to the A and B texts, or to do so from the perspective of either version of the play as a set of links between the relevant chapters of the Faust Book and segments of both dramatic texts. Rasmussen uses this comparative approach to the play and its source in a particularly fruitful way: determining that the A-text is closer to the Faust Book than the B-text significantly bolsters his argument against the A-text as a memorial reconstruction.[5] One of our hopes is that by making this particular nexus of texts more available to non-specialists, who would not previously have had access to them, the Marlowe site will facilitate this type of scholarship that involves close textual analysis and close reading.

  16. This process of electronically "unbinding" the book not only dismantles the traditional boundaries of the text but clearly undermines what McGann calls the Romantic ideology of the solitary originator. The kind of textual analysis that McGann, Bevington and Rasmussen, and Marcus make a case for, in Marcus' words, "invests the 'author function' with a potential for dynamism and profligacy which it does not have in the Foucauldian paradigm or in the usual assumptions that guide editorial practice," the assumption that a single author produced a single text (65). It is precisely this dynamism of the author function that an electronic hypertextual resource can enact. Reading the "source" as a comparative version of the Faustus story, as Rasmussen and Bevington so fruitfully compare the A- and the B-texts, may very well produce new and surprising readings of any of the texts involved in such a study. It is perhaps this element of disjunction or surprise, as Barbara Johnson has suggested with regard to deconstructive readings, that characterizes, above all else, a genuinely hypertextual reading.[6]

  17. Along these lines, Ian Lancashire has argued for the reshaping of electronic texts into corpora: "corpus creation," he states, "is one of the newer objectives of scholarly electronic editing . . . . [And o]ne of the first corpus-making tools, of course, is an SGML [Standard Generalized Mark-up Language] document-type definition, which transforms texts and tags into an interpreted structure" (140-1). This corpus-formation is precisely what we have tried to produce by encoding texts in our site in SGML. An electronic textual corpus might be understood as a kind of textual intercourse where databases or "books" talk to one another and even, perhaps autoerotically, to themselves. I have described the way that different versions of a text, the A- and B-texts and the Faust Book, for instance, each stored as a distinct text-file, can be combined in their visual display. But single text-file electronic texts can also recombine and multiply prolifically on their own. For instance, each of the texts attributed to Marlowe and collected in our site (with the exception of the more complex Doctor Faustus) is maintained as a single file, but each text's variants, accidentals and emendations have been tagged according to the particular subset of SGML tags identified and standardized for texts by the guidelines known as the TEI (Text Encoding Initiative) -- to the extent that these accommodate such a complex encoding scheme.[7] As a result of this standardized tagging, a reader may select from a pop-up menu whichever version of the text he or she wants to read. For instance, one might choose the 1594 Octavo of Edward the Second or any of the later historical collations, such as Dodsley's 1744 edition. The selected version is then created on the fly, the product of a series of Tcl/Tk program executions. Each version displays textual variants in a different color with a link to the right that identifies and provides access to the versions that differ from the selected text.

  18. This capacity for textual multiplicity and recombination is certainly one of the defining traits of an SGML encoded electronic text. To argue, however, that such an electronic text is the manifestation of the post-structuralist explosion of linearity and narrative would be, not surprisingly given my previous analysis, at least partially misguided. Certainly, this argument about hypertext is often made with a different kind of electronic text in mind, one more like Michael Joyce's now famous hypertextual novel, Afternoon, a story. My practical experience working on the Marlowe site, however, illuminates challenges to this claim that are certainly different from, but also surprisingly similar to, those I have presented above. Specifically, in spite of SGML's capacity to produce nonlinear interface or display options, the underlying structure of an electronic text is exceedingly linear. SGML cannot very gracefully handle multiple hierarchies, and as a result an electronic text's tagging scheme must abide by an extremely rigid, single-hierarchy organization. While encoding a text using SGML is not the only way to create a hypertext, a rigorous underlying structure is necessary in other cases as well. This seemingly "technical" issue in many ways cuts to the quick of electronic nonlinearity. While critics often describe hypertextuality as a wild associative environment, this nonlinearity apparent at the front-end, in a text's interface, is in fact undergirded by the back-end's relentlessly linear structure. More to the point, it is precisely this rigid linearity that allows for and even guarantees the nonlinear capacity of many electronic texts.

  19. An on-line edition like ours certainly facilitates a kind of reading that the printed text perhaps more demonstrably resists, though as I will suggest in a brief reading of Doctor Faustus, this resistance is far from successful. Undoubtedly, hypertext such as this has the potential to de-privilege the author as the primary organizing principle of textuality. To claim, however, that a hypertextual site is therefore unorganized or "anarchic" would be not only inaccurate but, as we have seen, impossible. The links between the sources certainly facilitate hypertextual reading, but this does not mean that a reader proceeds in an entirely nonlinear fashion, unaffected by the so-called limitations and editorial decisions inherent in the structure of the printed book. Our links between texts, after all, must also qualify as editorial productions and certainly function to guide, narrativize or in Faustian terms limit one's reading in various ways.

  20. And finally, this limitation is also imposed at the level of encoding an electronic document. As I have said, producing an SGML textual database requires rigid linear back-end structuring. I do not, however, cite this as evidence of the inevitability of the textual cohabitation of linearity and nonlinearity. Rather, it provides a convenient analogy for envisioning this pairing at the level of technological engagement. I leave the question of its necessity to the computer programmer-philosopher. More simply, in structuring a text-database, we make decisions about what aspects of a text we tag and thus what information we retain. And at the next level, we shape an electronic text through the decisions we make about what information gets displayed and how we choose to display it. As a result, various versions and various meanings, though multiple and potentially shifting, are inevitable. In this sense a hypertextual site like ours reproduces the oscillation that I have argued is inherent in any hypertextual reading (electronic or not) between a free-for-all and a guided tour, between exploratory and directed learning, and finally between nonlinearity and linearity, meaning and its deconstruction, where each pole must be understood as forming and not simply informing the other. This cohabitation is precisely what Marlowe's Doctor Faustus both resists and understands.


  21. I will now turn briefly to Marlowe's play, first as a way of reading in the printed word the disruption of linearity that has recently been identified as the domain of hypertextuality, and second as a means of exploring the imbrication of linearity and nonlinearity. In their discussion of the question of Marlowe's orthodoxy or heterodoxy, Bevington and Rasmussen write, "Cohesion is perhaps a will o' the wisp as an interpretive goal; unsettling contradictions and ambiguities may give us a truer glimpse of the play's restless uncertainty about the ultimate authority of divine truth" (24). In making a case for a poststructuralist reading of Doctor Faustus, they continue: "This poststructuralist reading will perhaps not win many adherents among those who are out of sympathy with recent criticism, but it has the great advantage of addressing a distinctively contradictory quality in Marlowe's dramatic writing." While Bevington and Rasmussen are far from the first to locate a "contradictory" and unsettling quality in Marlowe, they suggest here that Marlowe's work has many of the qualities that we now associate with poststructuralism. Marlowe thus provides an excellent, if anachronistic, lens through which to read the phenomenon of hypertextuality.

  22. In his article "Marlowe's Doctor Faustus and the Ends of Desire," Edward Snow maps Faustus' own desire for limits, despite what appears to be a simultaneous wish for limitless imperial powers. Often considered one of the play's central problems, Faustus does not bargain for immortality but, rather, for a pointedly circumscribed twenty four years of limitless possibility. As the number twenty four appears to reflect the cycle of a day and functions as a microcosmic figure for the cycle of a lifetime, Faustus' request of only twenty four years of power might be interpreted as his pleading for the assurance of his own death. Such a gesture implies that this limit may be integral to the experience of life; he demands that a limit be placed upon the limitlessness that is all too available to him and thereby suggests that the guarantee of an end is the very condition of life or temporality.

  23. What Mephistophilis and, at least in moments, Faustus himself understands, of course, is not only that "Hell hath no limits, nor is circumscribed / in one self place," but that "where we are is hell, / And where hell is we must ever be" (A.II.i.124-26).[8] As a figure for radical discontinuity, "hell" names the principle of disorder that continually threatens Faustus' life and is repeatedly figured as a threat to the coherence of his body. Christopher Ricks' argument that the Bubonic plague is a subtext in this play supports this notion of hell as the structure of endlessness: "[T]he plague was hell on earth; it was both more hideously hell than this earth had seen and yet -- just because it was on earth -- it fell hideously short of that eternity of suffering" (112).

  24. The undecidability of the definition and location of hell in the play certainly resonates with numerous Renaissance debates on religious doctrine that include Calvin's insistence that hell was both metaphoric, a state of mind, and material, a place of torment. The magician's retort to Mephistophilis, "Come, I think hell's a fable" (A.II.i.130), suggests not simply his momentary disbelief in a nonmetaphorical locus of torment, but also perhaps his desire to master the principle of disorder that hell signifies: by discursively transforming "hell" into a "fable," story, or narrative, Faustus tries to invest nonlinearity with a linear structure. And yet, "fable" signifies simultaneously as a fiction or a lie. Faustus' attempt at mastery through the narrativization of "hell" is thereby inherently subverted by the instability of language. While attempting to resolve himself of all ambiguity, he perhaps inadvertently signals the necessarily fabulous structure of narrative linearity or meaning. Faustus asks, "Is to dispute well logic's chiefest end?" and as Bevington and Rasmussen suggestively posit, the possibility that "end" is only ever "disputation," that meaning never settles, must be confronted (A.I.i.8).[9]

  25. Hell is not, however, singularly figured as the limitless realm of nonlinearity that Faustus inhabits. Rather, it functions simultaneously as telos, as what Derrida, in the sentence cited earlier, calls the future. Hell is not simply "where we are"; it is also where Faustus ends, his limbs issuing from hell's "smoky mouths". The trope of hell names Faustus' eventual death, and by guaranteeing an end, structures teleologically the narrative of his life. Like hypertextuality in the essays I consider above, hell operates in this text doubly as the trope of nonlinearity that disrupts the linear or narrative structure of Faustus' life and simultaneously as the trope of the future, Faustus' "absolute danger," that reinscribes linearity.

  26. As Faustus agonizes over his eventual death -- "hell, ah, hell for ever!" -- he links the complex and double function of hell with the act of reading where he laments, "O would I had never seen Wittenberg, never read book" (A.V.ii.25-26, 20-21). And, of course, his last words, calling up Tamburlaine's own book burning, function similarly: "I'll burn my books - ah, Mephistophilis!" (A.V.ii.123). In the magician's futile attempt to redeem himself, Doctor Faustus suggests the depth of the genuine threat posed by the book.[10] Far from a guarantee of fixity, hierarchy, and linearity, the book has enabled Faustus' disruptive theatrical productions and more generally has disorganized the narrative and hierarchical structures that keep lives, bodies and reality intact. Textuality in Doctor Faustus, as Marjorie Garber has shown, reveals in fact its own instability. Signing is characterized by Faustus as an attempt to "make an end immediately" (A.II.i.72), and yet this is precisely what the printed word won't allow him to do. After signing the deed with the devil in his own blood, and following Faustus' claim of absolute certainty and finality as he repeats Christ's last words, "Consummatum est," he discovers an admonitory inscription on his own arm ("Homo, Fuge!") that instantaneously disappears (A.II.i.74). He responds to this event that reveals both the textual condition of his own body and the instability of that body/text with the now-famous words: "here's nothing writ" (A.II.i.80). Bevington and Rasmussen point out the paranomasiac function of the word "deed" that signifies both text and action, a pun that underscores the uncertainty structuring not only the text but the realm of acts or experience.[11] The structure of the pun itself, of course, figures this general textual ambiguity that Marlowe is perhaps all too aware of, but more specifically it also undermines the distinction between the realms of experience and language, deeds and deeds, and thus gestures to the textual character of bodily experience. The productive capacity of language to multiply significance endlessly and to thereby undermine the coherence of meaning and bodily integrity as a transcendent referent resonates in late twentieth century theories of hypertextuality. Marlowe's relationship to the condition of textuality that (dis)organizes experience is complex and ambiguous, imbued with both anxiety and excitement.

  27. Reading the phenomenon of twentieth century hypertext through Marlowe contributes, I urge, to an understanding of our own conflicted relationship to textuality and complicates a naive celebration of deconstructive hypertextual nonlinearity. Marlowe suggests that such nonlinearity is always inscribed within a realm of limitation -- in other words, structured as a narrative, albeit a complex one. Paul de Man anatomizes this structure of textuality in his work on irony where irony names the disruptive principle that registers meaning's capacity for infinite regression that can only be -- and, crucially for this analysis, always is -- stopped, stabilized, or interpreted through the temporal, narrative structure of allegory. This logic of simultaneous de Manian irony and allegory, structuring the relation between what I am calling nonlinearity and linearity, is further apparent in the remarkable capacity of Faustus' body to reconstitute itself following various episodes of fragmentation.

  28. In the B-text, after Martino and Benvolio, for instance, seem to have beheaded him and threaten to "put out his eyes, [so that] they shall serve for buttons to his lips to keep his tongue from catching cold," they ask, "having divided him, what shall the body do?" (B.IV.ii.63-64). Frederick states, "Give him his head for God's sake," and Faustus answers:

    Nay keep it: Faustus will have heads and hands,
    Ay all your hearts to recompense this deed.
    Knew you not, traitors, I was limited
    For four and twenty years to breathe on earth?
    And had you cut my body with your swords,
    Or hew'd this flesh and bones as small as sand,
    Yet in a minute had my spirit returned,
    And I had breathed a man made free from harm. (B.IV.ii.69-76)

    In this passage Faustus seems at least momentarily to both have and not have, need and not need his head. As such, fragmentation appears simultaneous with the return of form. Though he begins by acknowledging his headless state ("Nay keep it"), he continues, "Knew you not, traitors, I was limited," asserting his status as a bounded body. This passage marks the oscillation between these two-states-as-one by linking bodily coherence ("Yet in a minute had my spirit returned, / And I had breathed a man made free from harm") to the very fact of being "limited," of being linear. Furthermore, Benvolio's amazing formulation of Faustus' eyes made buttons to the lips reinforces the reflexive relation between language and bodily coherence, between the silencing and dismembering of Faustus. Just as silencing Faustus is analogous to dismembering him, alternately language may be understood as that which guarantees or at least restores coherence. Such a formulation thus reveals the correlation between the status of Faustus' body as a text and the inevitability -- fabulous as it may be -- of meaning, narrative closure or bodily containment.

  29. Like hypertextuality, Doctor Faustus is not, we are seeing, simply about the way that things fall apart. Marlowe's play is not singularly or even primarily, as Huston Diehl suggests, a Protestant text that rehearses the trauma of iconoclasm, of image-breaking and incoherence. Rather, it is a text that insists relentlessly, in a characteristically Marlovian fashion, upon the double movement of the dialectic of coherence and incoherence, linearity and nonlinearity, a dialectic that also operates in the double operation of the trope of "hell." I have tried to suggest here that theories of hypertextuality acknowledge more readily this capacity of the non-electronic word to perform hypertextually. Whereas Landow and Aarseth argue rightly that an electronic text is not necessarily a hypertext, I am suggesting not only that a printed text or a book more generally may in fact be a hypertext, as I'm sure they would agree, but, as Marlowe shows us, that it is necessarily such. Marlowe can help us to theorize hypertext in such a way that we attend both to the impossibility of conceptualizing nonlinearity exclusively through this trope and to the troubling effects inherent in such a utopian gesture, specifically, the attribution of stability to the printed word. Finally, the "material text" of Doctor Faustus, its complex editorial history, perhaps even more palpably than this reading of the play itself, subverts a utopian reading of the opposition between nonlinear hypertext and the printed word.

Works Cited


1.  See Lyotard (34; ctd. Landow 32).  I wish to thank Paul Werstine for pointing me in fruitful directions for research on this project and others. I would also like to thank Raymond Siemens for his continuous support.

2.  Lyotard's warning resonates with those of several Frankfurt School philosophers. See, in particular, Herbert Marcuse.

3.  I am deeply indebted to David Bevington and Eric Rasmussen's thorough and informative introduction to the Revels Plays edition of Doctor Faustus and to Rasmussen's A Textual Companion to Doctor Faustus, in which he lays out extensive textual and historical evidence for his very persuasive arguments about the textual history of the play.

4.  See Rasmussen, where this argument is laid out in great detail.  Marcus also assumes the authorial status of the A-text, but produces a different reading of its distinction from the B-text. In the introduction of the Revels Plays edition, Bevington and Rasmussen summarize and critique Marcus on this point (using an earlier publication of her essay): "Leah Marcus' argument . . . that the B-text revisers are proimperial, internationalist, and Anglican in support of James I's foreign policy of reversing hostility towards Spain and the Empire, overstates the ideological differences between the two texts; even in the B-text we find popular ideals of Protestant national self-determination that would have appealed to Londoners who were wary of James' new foreign policy and toleration of Catholicism. The 1602 additions predate, in any case, James' accession in 1603. Still, Marcus rightly calls attention to the importance of seeing the B-text revision in the context of a changing political environment" (n160).

5.  Rasmussen (Ch. 1; 8-10, in particular).

6.  See "An Interview with Barbara Johnson" in her book The Wake of Deconstruction. Johnson states ". . . what the system of knowledge may not see as unknown is the thing it is blind to in its very construction. So, I wouldn't say that I absolutely think that there is no such thing as an unknown that lies beyond where knowledge can reach, but that that model for the unknown is not surprising; whereas the model of the unknown that says the unknown is right here in the very center of the known is, in some ways, that which enables a sense of knowing to occur, is always a surprise. And I think that literature... is about that" (84).

7.  The head programmer at the Perseus Project, David Smith, has proposed expansions to the TEI in the areas of transposition, variation of meta-data, and insertion of incomplete structures, problems that we confronted in developing the Marlowe site in particular.

8.  All citations are from the Revels Plays Edition (David Bevington and Eric Rasmussen, eds.).

9.  See Bevington and Rasmussen (eds., 39): "Faustus wilfully misinterprets Ramus' idea that disputing well is 'logic's chiefest end' (I.i.8) to mean that disputation should be an end in itself rather than a means to salvation."

10.  While the books Faustus considers burning may in fact refer to grimoires, manuscript texts on magic, the anxiety in this play that circulates around books and writing was very likely at least fueled by the explosion in printed material resulting from the spread of the printing press during Marlowe's lifetime. Further exploration of this hypothesis lies beyond the scope of this paper.

11.  See Bevington and Rasmussen (eds., 40): "A deed is both something Faustus longs to accomplish and a document by which all will be undone."

Responses to this piece intended for the Readers' Forum may be sent to the Editor at L.M.Hopkins@shu.ac.uk.

1999-, R.G. Siemens and Lisa Hopkins (Editor, EMLS).

(HB, RGS, 24 September 1999)