A critique of John Bryant’s “Liquid Text”

And what rough beast, its hour come round at last…

Hardly anyone will argue that a paradigm shift is upon us following the advent of digital editions. Perhaps one of the most vaunted changes is the seemingly unlimited space that computers provide, so that the old restrictions of the print edition (ex., one book, one text), disappear when we make the transition to the cloud. Now that we have these possibilities, it is difficult to look back at the proliferation of critical editions in the 20th century, which aimed to establish an authoritative text that would fit the bill, without questioning all the theoretical industry that went into rationalizing them.

Has genetic criticism not also been caught in this carnal sin? One of the cornerstones of genetic criticism, for example, continues to be the conceptual demarcation of pre- and post-publication texts. This jealously policed border severing the chronology of a work coincides with the need during the early years of genetic criticism to carve out a niche in the world of editorial projects where manuscripts (and their study) could be disseminated in a way deserving of the power of the press. At the time, geneticists were under heavy attack from those zealous souls who revered authors’ final wishes. While critical editors of the past were happy to borrow from working papers to bolster their claims about an author’s wishes —even if they kept only glimpses of the manuscripts tucked away in their apparati— genetic critics stopped cross-border traffic all together. After all, the early genetic critics were not medieval Irish monks, and the development of more efficient photographic techniques in print, together with increased access to modern archives, had made manuscript dissemination feasible once more —territorial claims had to be made.

Now a new material reality and reader practice bears upon scholars. As a result, we are starting to see a push in many different quarters for a marriage between genetic and textual criticism, a blur between the epochal pre- and post-publication divide—yes, simply because we believe we can have our cake and eat it too. Several grand online editions have sprouted all over the digital landscape with a diverse set of editorial philosophies and uneven degrees of quality. Perhaps the famous Rosetti Archive’s proof-of-concept still provides one of the best examples of what has been done so far to take advantage of an ample digital space. The Archive not only brings print versions in dialogue with Rosetti’s manuscripts, it also brings both of these in dialogue with Rosetti’s visual art. Jerome McGann, the lead investigator of the RA, would be the first to admit the archive was a “failure,” but only in so far as it points our lamps to the dark road ahead for digital editing.

Not surprisingly, alongside this growing number of editorial projects, we are seeing a new batch of theories that justify the translation of objects which erstwhile enjoyed a different ontological status into a common medium. We are, I would say, in the midst of a theory rush. Though we are all sinners here, moved to think by our machines, as we relocate to this brave new world, those with critical, editorial and digital masteries should strive for theoretical work that does not come as gratuitous alibi for our new toys. Even more important perhaps, we should endeavor to protect the hard-earned lessons of the past that connect editorial practices to larger questions about literary texts. I would like to begin some of my own observations in this direction via a strong critique of one of the most visible textual theories to come out of American textual studies in recent years: John Bryant’s theory of “The Fluid Text.” Bryant provides an excellent starting point for several reasons: He only arrives to digital editing after flirting with print editions; his claims about texts transcend the pre- and post-publication divide; he tries to reconcile the eclectic, social-text and genetic camps; and, his digital edition of Herman Melville’s “Typee” has many features which make it a model work of electronic editing.

All texts that show variance, whether pre- or post-publication, are fluid-texts, argues Bryant, and the task of the fluid-text’s historian/editor is to focus

on the interpretation of private and public pasts in order to make the evidence of literary versions, revision, and adaptation accessible to readers for critical and cultural analysis. (61)

So far, so good. Such a spirit of inclusiveness and public generosity effectively allows for these grand editions stepping only on the toes of those invested in definitive editions. In order to arrive at this exhortation, though, he manages to undermine a few crucial distinctions along the way. He begins his book, The Fluid Text: A Theory of Revision and Editing for Book and Screen, by pitting G.T. Tanselle’s “intentionalism” against a straw-man of Jerome McGann’s “social-text.” In order to carve out a space for his own fluid-text, he misreads McGann to be only concerned with the post-publication life of texts. In his haste, he misses one of McGann’s most important insights: what makes a text social is already rigged into the textual condition itself in as much as all texts, at any given stage of their career, participate in an “editorial horizon (the horizon of their production and reproduction).” (McGann 21)

The misreading becomes important when we try to understand why Bryant felt he had to add a new category to McGann’s distinction between a linguistic code and a bibliographic code. For a digital edition which could take in pages in the order of gygabytes, which showcased a hypertextual apparatus plus side-by-side comparisons, he needed a category which could be operative anywhere the author made a revision, (manuscript, print, proof, a napkin, you name it), and he sort of found it there in a pocket of his own sewing. He calls this new category the revision code, “a bibliographic encoding of an authorially intended revision strategy.” (Bryant 55) This unfortunate hybrid comes from the failure to acknowledge that McGann’s insistence on the bibliographic code already implicated the documentary realm. In other words, a revision in a manuscript can be easily understood within that already strange marriage between the linguistic and the extra-linguistic (read bibliographic) which characterizes the textual condition, without recourse to a mystifying third. The difference, if any is to be made, lies in the distinction between public and private. That working papers were not meant to be distributed widely, does not mean their non-linguistic codes are not imbricated with the graphemes in meaning-production.

With the “revision code” in his toolkit, in a chapter with an aptly pop-Buddhist title, “Work as Energy,” Bryant finally arrives at his “ontology of process.” This he associates with genetic criticism, but because he must transcend the pre-publication camp as well, he will define texts as the “manifestations of a culture over time.” (62) No longer the author, nor particular social contexts, but a flow of energy over “a culture” unbinds the literary work and in turn justifies the fluid-text edition. But how can we speak of a culture in these terms? In Césaire’s case, we know his texts operated in many different cultural spaces (read here editorial horizons): surrealist, anti-colonialist, Marxist, etc. Without looking it up, I bet the same could be said for Melville.

Finally, the need to corner the market on the revision code, his foundation for the grand digital edition that a dash of PHP and CSS stirred in a pot of HTML makes possible, requires him to oppose revision to genesis (defined by him as non-revised text) —territorial claims are being made. This move leads Bryant to offer a very troubling statement: “What can one say about genesis except that it is everything and nothing to speak of? Words happen. They seem to come from nowhere. Ab nihilo. Boom. What can you say?” (94) My friends, if indeed one cannot say anything about non-revised text or genesis, our business model is in trouble; so, before we have to declare Chapter 11 on our intellectual inheritance, let me answer Bryant’s question and get down to business:

There are only additions, deletions and transpositions. Everything else is superstition.


Works Cited

Bryant, John. The Fluid Text: A Theory of Revision and Editing for Book and Screen. Ann Arbor: University of Michigan Press, 2002. Print.

McGann, Jerome J. The Textual Condition. Princeton, N.J.: Princeton University Press, 1991. Print.