Distributed and Conditional Documents: Conceptualizing Bibliographical Alterities Johanna Drucker University of California, Los Angeles

To conceptualize a future history of the book we have to recognize that our understanding of the bibliographical object of the past is challenged by the ontologically unbound, distributed, digital, and networked conditions of the present. As we draw on rich intellectual traditions, we must keep in view the need to let go of the object-centered approach that is at the heart of book history. For this reason, this seems an apt moment to emphasize scholarship of books in the period of colonial expansion and cross-cultural encounter (particularly, though not exclusively, that of the 17th and 18th centuries). In part because of the character of the works they study, scholars working in this area put ideas of performative materialities into the context of networked environments as a basic framework for their analysis. We know that classic studies in bibliography have emphasized the understanding that textual artifacts are frequently constituted through complex processes and received through the parallax of varying cultural, social, or historical perspectives. Taken to their logical extension, such approaches suggest that cultural artifacts are constituted within cycles of circulation where lines between production and reception blur [1]. But in book history, an object-centered approach persists, even, as we shall see, in revisionist models of the field. We have to shift outside its modern or western frames to grasp an alternative conception—in which a book is conceived as a distributed object, not a thing, but a set of intersecting events, material conditions, and activities. Books, documents, textual artifacts can no longer be thought of as autonomous objects that circulate in a context, but must be reconceputalized as event spaces within an ecology of changing conditions.

My argument begins, therefore, with a few assertions. First, that we have much to learn from the scholarship on Old and New World contact that touches on bibliography, document studies, and book history for formulating a non-object centered conception of what a book is. Why begin with scholarship focused on artifacts composed three or four hundred years ago? Because contact experiences unsettled every certainty that had been in place in the Renaissance, shaking the foundations of historical, religious, geographical, and philosophical knowledge across nearly every domain. The importance of these exchanges has registered in theoretical and critical writings with increasing frequency in recent decades. Second, that the insights from these studies can be usefully combined with a theory of the “conditional” document to develop the model of the kinds of distributed artifacts we encounter on a daily basis in the networked conditions of current practices [2]. Finally, I would suggest that this model provides a different conception of artifacts (books, documents, works of textual or graphic art), one in which reception is production and therefore all materiality is subject to performative engagement within varied, and specific, conditions of encounter.

Twenty years ago, in his groundbreaking book, The Darker Side of the Renaissance, Walter Mignolo argued that European colonization in the Renaissance was replicated by scholarly practices up into the present because they assessed “other” cultures’ textual practices from a western perspective. One of his examples was standard account of writing systems, derived from the work of such well-respected scholars as Ignace Gelb and David Diringer [3]. Mignolo’s argument was that these performed their “colonization” by normalizing the history of inscriptions on the basis of Western alphabetic scripts. In their accounts, writing systems “developed” through a series of “progressive” stages from “proto-writing” in pictures and signs to an advanced “true” alphabetic script, which was taken to be the highest level of achievement in this technological matrix.

We should keep in mind, pace Mignolo, that Diringer and Gelb, among others in the early and mid-20th century, were still piecing together the archaeological evidence on which such a master narrative could be constructed. Well into the 19th century, a figure like the British cleric and scholar, Charles Forster, was still tracking the “one primeval” language and attributing the invention of writing to a divine origin[4]. Chronologies of human history were based on biblical accounting, and only went back 4-5,000 years until geological evidence to the contrary made its way into scientific and historical perception. Thus the “modern” formulation of progress has to be seen in its own historical frame. Still, as Mignolo points out, the typology of the Diringer/Gelb approach, which underpins current studies of the history of writing and the alphabet, enforced a binaristic hierarchy in which the writing systems of the New World, in particular, were always subject to a prejudicial judgment and characterized as inferior, inadequate, or undeveloped: writing systems in Mayan glyphic inscriptions and Aztec codices never “reached” the alphabetic stage.

Not only is it impossible to fit these non-western materials into a standard model of textual production and bibliographical studies, but, in the larger point Mignolo makes, a confrontation between that standard model of writing, literacy, and books can be staged on the basis of a fresh encounter with these indigenous materials and their conditions of production and use. In essence, Mignolo is launching an attack on the fundamental coloniality of knowledge in the realm of bibliographical studies and suggesting that it be rethought (Mignolo, 1995).

If we take this seriously, the challenge is to think about what a future history of the book would look like if it began its formulation with New World examples of writing included from the outset. Rather than add (or try to add) indigenous glyphs, signs, quipu, and wampum as anomalies or exceptions to a “normative” bibliography, we would formulate a broader, more inclusive field of practices and works on which bibliographical studies could be constructed. Similar sentiments and impulses can be found in the small but growing literature that scholar Jesse Erickson designates with the term “ethno-bibliography” and that Jason Hewitt also called to my attention within his study of “fundamental semiosis,” which examines the emergence of signs within human cognition and culture [5]. Inspired by their comments and the work of Mignolo, my argument takes up studies by Robert Fraser, Birgit Rasmussen, Betty Booth Donahue, D.F. McKenzie, Jared Diamond, Jerome McGann, and Phillip Round, and others, to make a general proposal about how to put this changed concept of “the book” into dialogue with the prevailing/current models of book history and to think through the implications of this for our collective approach to pedagogy in the field [6].

In his work on encounters between old and new world cultures, Jared Diamond makes the point that “guns, germs, and steel” and “alpha-numeric notation” were not “superior technologies” to those found among the indigenous people, but they were embedded in a technological system that allowed “instrumentalization of control” in a way that shifted and skewed power relations from the outset (Diamond, 1998). In other words, a techno-ecology, not technology, is what we have to examine if we are to understand the contact encounters—and more important, learn from them. The imprint of the “technology” model—the core of which is what Mignolo is pointing to in his analysis of the “progressive” version of writing systems “advancing” towards the alphabetic—is still so present and prevalent that we barely see it. The naturalization of colonial power in knowledge production—ours—successfully conceals its workings. How to undo this?

Before I answer that question by turning to the work of some of the authors mentioned above, let me pause to situate the argument I will make within the intellectual traditions in which book historians have developed a well-articulated  series of successive paradigms for the field. Each builds on and extends (sometimes contests) the other: from the bibliographical attention to descriptions of objects; the reconstruction of their production; authentication of their intellectual content; study of their impact and effects; analysis of their lineages and genealogies [7]. “Histories of the book” usually map the development of writing, early codes for recording speech or language acts, and the sequence of technologies from sticks to clay to brushes, papyrus, leather, vellum, parchment, paper, and print (and recently, electronic formats and digital files). From wall and monument to tablet and scroll to codex and screen, the technological developments march along and with them a well-marked history of milestones in publication methods, major figures, important works, and shifts in the controls over intellectual property, production means, and distribution networks.

The narrative version of the “history of” is complemented by a statistical, sociological methodology associated with the French Annales school [8]. Not content with the description of physical artifacts, knowledge of their makers, or conditions of production, the Annales historians added considerable breadth by extending the field to studies of commerce, politics, economics, and other aspects of book history that would not be immediately extractable from the object, but required analysis of account books, documents and records, and other historical materials. The very act of periodization, such as that performed by Roger Chartier in his attention to the “break” between scroll and codex, and then manuscript and print, for all its benefits and virtues, reinforces certain assumptions that are readily undone when points of continuity, rather than over-determined notions of difference, are brought into play (Chartier, 1992). A historical approach based on changes in technology (manuscript to print) does not necessarily map onto shifts in practices (e.g. publishing), for instance, while the study of numbers of readers, book sellers, copies in circulation offers yet other insights.

In addition to adopting techniques from the broader field of social history, book historians have created specific models for analysis in their domains. These successive models have built a series of useful intellectual frameworks for analysis in the field, beginning with Robert Darnton’s “What is the History of Books?” published in 1982. Darnton’s “communications circuit” emphasized the interconnection of the many agents (author, printer, binder, bookseller etc.) in the lifecycle of a book—but the book is an autonomous object moving through this circuit. In 1993, Nicolas Barker and Thomas Adams proposed an alternative version emphasizing the dynamic “events” in that lifecycle (publication, distribution etc.), stressing social processes over individual human agents as the crucial elements of a book’s existence (Barker and Adams, 1993). A decade later, Michael Suarez’s thoughtful “Historiographical Problems and Possibilities,” published in Studies in Bibliography, laid out the many complexities that continue to plague the development of the field of book history, including those of periodization, gaps in knowledge, the multiple dimensions of the sociology of bibliography (Suarez, 2003-2004). The bibliographical context of his publication is significant, as it signals an effort to keep critical issues from that field in dialogue with book history. Suarez’s analysis exposes the difficulties of periodization in particular, and demonstrates the need for more complex and less reductive approaches. A subtle and exemplary set of studies by Adrian Johns offers a useful demonstration of historical methods, much of it dispensing earlier myths about the impact of printing technology (Johns, 1998). In The Nature of the Book, Johns put forward detailed individual cases that show the extent to which exceptions to generalized rules further complicate any “models” we create. The summary effect of these and other contributions to the field is to provide a highly useful set of analytic approaches that reveal different facets and aspects of objects under investigation. 

Each, however, assumes the existence of a book as an object, a priori. But for works outside the western tradition (or even within it, I will argue) the object constituted by the historical and theoretical inquiry may be an event space. There may not be an object in play at all, only a distributed condition of literacy and/or semiotic communication across physical traces and inscriptional or productive apparatus. Rather than relying on a forensic, descriptive, object-based approach for their analysis, such works may have to be conceived from a performative approach. Even where actual books are part of this alternative legacy, they call for reading of the polysemous field of their composition and conception and its performative dimensions, rather than assuming its literal, physical, or textual self-identity. Production, in other words, may not always result in an object, but even where it does, reception produces a performatively constituted event in response. Thus the cultural parallax described in D.F. McKenzie’s still dazzling study of the “Treaty of Waitangi” has to be expanded beyond the discussion of two crossed gazes, each from a different cultural perspective, misunderstanding each other’s foundations and assumptions about the symbolic and literal value of an object, a treaty. It needs to be expanded into a model in which recognition of constitutive processes replaces the assumption of an a priori object that is misread (McKenzie, 1986).

Marking, making, inscribing, reading, are all aspects of a system of social and cultural production. A semiotic object does not sit inside it, like a gem in a setting, in a context-based model of object and conditions. Instead, the object is constituted, like an organism in a medium, as an effect of the very conditions that bring it into being. In the same way that cell walls and chemical/physical/biological processes create the conditions of semi-autonomy that define a living organism in an ecological system, the semiotic “object” is an effect of constitutive conditions in the culture of which it is an integral part. Its reception is a secondary act, provoked by the material traces of production, but reception is a primary act in so far as it constitutes a text or artifact as an event, a performative reading or engagement.

If we return to the contact studies, we can see how such an approach is required. When Mignolo describes the cultural politics of encounter between Mayans and Spanish, he points to the asymmetry present from initial contact (Mignolo, 1995). The 16th century Jesuit José de Acosta “ranked writing systems according to their proximity to the alphabet,” in spite of the recognition that the indigenous people had a highly developed literate culture (4). This included vocabulary designating Incan men of letters, “quipu camoyan,” scribes, “tlacuilo,” and surfaces for painted narratives “amoxtli” (75).  Mignolo insists that we move beyond cultural relativism, particularly the sort based on comparative approaches privileging old world norms and conventions as standards on which terms of comparison are established. With rare exceptions, Mayan literacy has always been conceived from the European perspective (76). Among the exceptions, the aforementioned Acosta, who observed of the quipu that “in every bundle of these, as many greater and lesser knots and tied string […] in short, as many differences as we have” (83).

Acosta recognizes difference as the basis of signs. His recognition of the fundamental non-equivalence of these semiotic systems is equally striking. He knows that the bibliographic practices based in alphabetic literacy are inadequate for addressing literacy conceived in a fundamentally different mode. Each sign systems may be as complicated as the other, but they cannot be put into a relation of reciprocity. In Nahutl, emphasis is placed on the connection between spoken words and an agent, Mignolo continues, and the Mexicans “had a set of concepts to outline their semiotic interactions”(103). If their “Sages of the Word,” were resident in the “amoxtli” or surfaces, learning was located in the body of elders, transmitted orally. The Christian philosophy of the word, conceived in connections between the archetypal book (of God) and the metagraphic book (of communication), was embedded in the Franciscan view of writing and book (106).  Mignolo makes clear that this distinction doesn’t transfer to Nahuatl practices. None of these indigenous inscriptions is self-evident, each has to be read within the cultural ecology of signs, practices, event spaces, and knowledge technologies. How is this different from alphabetic writing, really? Is the semiotic code of alphanumeric writing any more self-evident than quipu knottings? Any less dependent on the act of reading for its productive of significance? The differences reside in their specifics at a more fundamental level.

Perceived asymmetries and cultural obstacles to equivalence have been recognized for decades. But the implications of these contact moments of the 16th and 17th century are still present at the deeper level—in the still unarticulated recognition of the ways cultural semiotic systems emerge, organize the cultural world, and then pass themselves off as natural, erasing the process by which semiotic conception occurs. In other words, Mignolo’s argument is not that we need better “translations” across sign systems, but that we need a way to understand difference and specificity at the level of original semiosis—in attending to the emergence and structuring effects of the formation of sign systems. The ways signs and literacy are thought, conceived, and acted are distinct in these contact zones, and the bibliographic requirements for this alternative ecology of signs can’t be developed—or turned into a critical or pedagogical method—as a simple appendix or corrective.

When Mignolo discusses later developments in the 17th and 18th century exchanges and the philosophical foundations of their attitudes towards signs, writing, and history, he shows how the cross currents of belief in the “universal history” of humankind were—and are—at odds with contact experience and exchanges. Boturini Benaducci, the 18th century ethnographer, for example, in his study of quipu, undercut the idea that the alphabet was the sole authority for the historical record (151-161). Mignolo emphasizes the paramount importance of attending to description and discourse as well as objects—because the objects are constructed by the discourses of inquiry and scholarly attention precisely in so far as they align with the conceptual principles on which the discourse itself operates. In classic post-structuralist parlance, the object of knowledge is constituted, not perceived, by the discourse.

As long as difference is construed as otherness, the asymmetry of these colonializing discourse persists. To move beyond this dilemma in book history, we can rely on a few concrete examples in scholarship of the last two decades to show the way.

Elizabeth Hill Boone, whose edited volume, Writing Without Words was published in 1994 (also 20 years ago), was aware that she was working after two decades in which post-structuralism and deconstruction had shaken up the authority of text and power relations. Jacques Derrida’s reformulation of the primacy of “writing” over the authority of “voice” was at odds with the literacy studies formulated by Jack Goody, Walter Ong, and the Canadian media theorists around Marshall McLuhan beginning in the 1960s [9]. But theoretical ambitions had a difficult time getting traction on material realities. Bibliography remained book-based, antiquarian in its attention to physical facts of collation, misprint, wrong-font and crooked sheets with overprints and recycled dingbats, cuts, or initial letters even if “grammatology” reformulated attitudes towards inscription. Bibliography met critical textual studies in the work of Jerome McGann (1993, 2001) and Dennis Tedlock (1983). Their performative concept of the text and the book, had a strong emphasis on the codependence of conditions of production and circumstances of use [10]. From these, as well as the other strains of intellectual thought already mentioned, we can begin to see both the limits of traditional bibliographical models for an encounter with “alterity” and to sketch an approach that is not “post-colonial”—i.e. a task of corrective recovery and retrospective inclusion of new examples to an old paradigm—but “de-colonizing,” to use Mignolo’s term, a project of rethinking the fundamental frameworks that constitute the object of inquiry at the center of our field. On what foundations, then, do we conceive of the “book” that comes to figure on such grounds? What, in fact, is a “book” in this shifted frame?

The contact zones of the 16th and 17th century are places in which the assumptions underpinning western bibliography are exposed and their limitations revealed because many of the textual and inscriptional practices are distributed in character and/or highly contingent, dependent on circumstances of use in ways that make it impossible to ignore qualities that pass with less notice in the bibliographical traditions that take the book object for granted. Literal, forensic, formal materialities, so crucial to bibliography, have to be extended by a performative understanding of materiality that engages bibliographical objects in terms of what they do, how they work, not just what they are (Kirschenbaum, 2008). This approach to performativity, this doing and working, is constitutive, and asserts that an object emerges from the co-dependent conditions in which it appears (Drucker, 2009, 2013). These codependencies occur at many levels—within the composition of the text, the structure of the object, its embedded condition within social practices, and across activities of editing and translation.

In her study of William Bradford’s 17th century Of Plimoth Plantation, Betty Booth Donahue shows the extent to which the book is a record of the “indianization” of the colonists (Donahue, 2011). To cite Donahue, “In American Indian epistemology the earth is First Text, and the study of its features constitutes textual exegesis” (20). Within the frameworks of this alternative semiology, Donahue tracks Bradford’s absorption of spatial constructs and directions, cosmology, and knowledge of natural history as they are encoded in Indian systems of language, work, and ceremony. Bradford absorbed the structuring principles of native cosmologies into the language in the text. The work is constituted as a border zone which embodies the native tribal leaders realization that they were preparing the land for a new narrative (5-18, 19-38). The outcome was not inevitable at the outset, and though its course is marked by fatal asymmetries, this re-reading and rethinking allow an alternate bibliography to take root as one aspect of a de-colonization of current epistemology.

Phillip Round opens his book on printing in “Indian Country,” Removable Type, with a study of the volume commonly known as the “John Eliot Bible.” He says, “In their stubborn materiality and monumental presentation, however, books were […] useful signs of the ‘visible civility’ Eliot demanded from his Native parishioners” (25). He goes on to paraphrase the work of Matthew Brown, a scholar whose work emphasizes the ways “the culture of the book in Puritan New England provides us with ample opportunities to explore Euro-American settlers’ representations of imagined Native peoples,” and all the asymmetries that implies. But, as Round goes on to say, Brown, like many scholars, refuses to view “the books in the Indian Library as ‘ethnographic facts drawn from the contact zone or as neutral sources of Algonkian expression.’” (25). Round asserts, instead, that the Indian Library “actually grew out of a fundamentally unstable bicultural communicative field.” Round takes apart each step of the composition of the Indian Bible, demonstrating that its translation, orthography, composition, and design function as a “crucial mediating semiotic in New England’s colonial middle ground” Eliot was dependent on collaboration with Christian Indians “to work up a syllabic orthography of the Massachusett language” (26). James Printer, the “Nipmuck convert,” and Job Nesuton worked closely with Eliot to produce the Bible: “The physical properties of the 1663 Mamusse wunneetupanatamwe up biblum God […] reveal the collaborative, bicultural social horizon from which the Native print vernacular emerged” (27). Round goes on to note all of the details in layout, typography, and design that differentiate the Algonquian Bible from the English one, stressing the impossibility of translation: “the Algonquian vernacular cannot stretch to accommodate many of the underlying ideological principles of either Protestant doctrine or book culture that inform the Bible’s production.” And, “In the Algonquian edition, the concept of ‘book’ itself is untranslatable.” Thus the pages are peppered with a kind of hybrid Algon-ish, with “words ‘Booke,’ ‘Bibleut,’ ‘Chaptersash,’ ‘Bookut,’ and ‘Bookash’.” (29).

Contact encounters erased the literacies and practices of indigenous people. We know this, but revisiting the way these encounters have been written and assessed forces a reconceptualization of bibliographical studies. This is only becoming apparent in more recent work. In Queequeg’s Coffin, for instance, Birgit Rasmussen recounts debates about relations between knowledge, recording practices, and sign systems in the literate cultures that existed in the New World at the time of contact (2012: 2). Her argument focuses on ways that the concept of “literacy” is a colonizing discourse that has to be dismantled and rebuilt if the full inventory of non-western notational frameworks are to factor into it. Among other indigenous forms of literacy, for instance, she discusses the practices by Indian warriors of putting public postings along their routes, in waterproof ink, as a distributed information system across the landscape (3). Native languages included terms for writing and grammar. Wampum was its own system of encoded information, never meant to be separated from the context in which it was used, and served as the foundation of oral recitation and performance. Such artifacts have to be approached through a revised bibliographical mode, not as static objects under examination, but as transactional objects whose very identity is constituted through exchange. The erasure of these practices has been systematic, Rasmussen demonstrates over and over again, through the repeated assertion that native peoples lacked writing—or lacked “real” writing. The painful history of the Mayan and Aztec codices is too familiar to need repetition, but rethinking the still extant and remarkable documents produced by Bernardino de Sahagún, with his native scribes, along with that of Guaman Poma and his Nueva corónica and buen gubierno,” the Popul Vuh narratives of the Guatemalan highlands, the Chilam Balam (Mayan works from the 17th and 18th centuries) as an “inter-animated” semiotic exchange, she offers a way to think through a “decolonizing” scholarship in and through bibliographic practice (28-29).

In our current moment, this has tremendous relevance for the ways we will apprehend digital artifacts. Their identity is dependent on a complex of conditions. They are, in essence, contingently configured in the dynamic flux of multiple co-dependencies in ways print artifacts only hint at, but which the distributed character of landscape signs, wampum performance, and quipu knowledge approach. Coming to terms with conditional texts and ephemeral documents whose “conditionality” is always shifting within the lifecycle of their production and use brings us right up against the recognition that these “artifacts” are not entities but events, not things “discovered” by an inquiry but objects constituted by it. With such an insight, we realize that we need not only to have the skills and techniques for practicing bibliography, but for reading bibliographically, taking the traces of material into their indexical and contingent relations and situated-ness, and then producing them through a reading across these distributed factors. This is what Armando Petrucci suggested as a method in Public Lettering, when he argued that the “spaces” of Rome are the effect of signage and written traces, not merely surfaces and sites whose identity is anterior to the inscriptional acts (Petrucci, 1993).  

Contact zones characterize 16th and 17th century encounters between the old world of the European “west” and the New World cultures (whose established communication and semiotic systems were so radically different from those of the colonizers that they disturbed their epistemological belief systems—and were thus distorted, rejected, ignored, or subject to repression or eradication). These exchanges, so important to the philosophical formulations of the late Renaissance and early Enlightenment, where the questions about peoples, identities, universal history, language, religion, civilization, and humanity all came up for question, are particularly useful as the start point for thinking about bibliographical work now and for the future. Why? Because the systems-ecological approach to the semiotics of biblio-literacy exposed in those encounters have implications that have been engaged only somewhat to date in the field of book history and bibliography.

As we engage with the pedagogical challenge of formulating future histories of the book, we have to move beyond connoisseurship and antiquarianism, into this realm of meta-bibliographical description and performative, constitutive practices. We need to defamiliarize our own practices, which take for granted their forensic attention to production histories and reception histories, and attend instead to the assumptions on which they work. By shifting our frameworks from western-based conceptions of bibliography to ones grounded in an ethnographic alterity we can reformulate the foundations on which we work.

So far my argument has drawn on approaches to this question informed by the study of books, literacy, writing, and inscriptional forms of knowledge and communication that broaden the traditions of bibliographical study. The future of book history will be altered substantially by including such works in a bibliographical approach that starts with these diverse forms as part of the field, rather than adding them as “other” to its “mainstream” traditions. Such a shift has wider implications for ways diversity is understood within intellectual and historical frameworks. Instead of registering “otherness” in relation to a normative “sameness,” we can construe all forms of identity as alterities. Thus our conception of a book shifts from that of an autonomous object that “contains knowledge” and to the notion it is part of a “knowledge ecology” and exists in a co-dependent relation to the cultural systems of production/reception in which it functions. The point is not merely to extent bibliographical or historical frameworks to include previously little studied or marginalized works, but to reconsider the foundations on which such frameworks established their own “colonizing” approaches to bibliographical knowledge, and to undo them in a way that takes up the call for “de-colonization” in other intellectual realms. This is the decolonialization called for by Mignolo, which cannot be accomplished by extending the existing epistemologies to include “other” objects. It has to begin with dismantling the foundations of these epistemologies and rebuilding them with the full field of objects/practices in view.

What makes this so timely is that a bibliographical approach grounded in this alterity proves to be highly relevant to understanding documents produced in networked and digital environments.  The ideas of conditional texts and ephemeral documents have their roots in a wide array of communities, including the creative realms of poetics and printing with their direct understanding of composition and production, academic structuralist and post-structuralist discourses, and encounters between digital humanities and theories of critical editing [11]. Textual scholars and bibliographers have long struggled with the difficulties of establishing the authority of texts, extracting versions from material witnesses, seeking any one of several elusive objects, each of which might be elusive differently—the intention of an author, or in the case of sacred or canonical texts, the “first,” most complete, or least corrupted version. Like an asymptote that never reaches the limit of the original, these vectors of inquiry expose the impossibility of certainty, and the conditional character of textuality within the larger problems of bibliographical study.

In the editing community the notion of a fluid text, with its apparatus of critical editing comprised of codes and elaborately governed rules of application, produces its own specialized language (and debates). Attention goes to the “lemma,” that sequestered figment of text arrayed with all of its variants, hints, and whispers, pulled out of its place in the weave of prose or the tightly made stanza, so that a phrase once breezily skimmed within a passage or verse is now tied with as many small strings and stakes of explanation and filiation as Gulliver in the hands of the Lilliputians.

The task of the scholarly apparatus is to expose the complex produced-ness of any single textual artifact. All words and phrases, it turns out, have a tendency to licentious errancy and promiscuous use that must be accounted for through revision and review. The result is not so much an ordering as a scattering, refraction of any text into a myriad of facets so that no text ever appears as a single, intact, defined phenomena, but always a result of combinatoric circumstances and happenstances. An author’s ability to muster vocabulary or references, to conjure phrases as if from a store laid by or snatched from language heard passing in the street or unearthed from the rapid streams of conversation or culled from reading and gleaning published sources through a diligent porousness of mind, shows that texts are indeed textura weavings of threads twisted and plied, their appearance of wholistic integrity an illusion produced by the tightness with which the many borrowings and offerings come together.

Therefore temporality of textual production is not so much a question as a given. Alluded to by dates on manuscripts, the identification of a hand or condition of the writing links to a particular phase of an author’s life, or even the use of a medium (that typewriter, piece of letterhead, or postmark on an envelope used for scrap writing). The times of these documents are used to piece together the production history of a text. Milestone dates of publication or review, galley corrections or editorial communication, second editions or later printings each and all bear their testimony in some more or less discernible, material way. However fluid the text, the documents have their own kinds of persistence and permanence, however palimpsestic and complex their relations to the streams of production in which they participate may be.

The documents remain, or appear to, and their ephemerality is no more or less poignant than our own. We therefore mark and measure the lifespan of a work by similar metrics—ones that measure the relation of relativity static things (documents) to fleeting ones (events). If we subscribe to the notion that a reading a makes a text anew in every instance, that still does not undermine our commonsense commitment to the existence of the thing/document as a thing—paper and print, ink and substrate, loose sheets or bound books or any other physical evidence.

But the conditionality of artifacts changes in networked environments, and here the tools of archival studies, informatics, digital scholarship and data curation, have to be drawn on to assist the bibliographer and textual editor. The analysis of tracking logs, click trails and data mined results create new challenges to our conception of a document, as do text messages and mobile communications. The archivists’ vocabulary of respect des fonds, and diplomatics, include longstanding and constantly shifting concepts of order, classification, and organization. Recently the notion of the records continuum has been described by the Society of American Archivists as “de-emphasizing the time-bound stages of the lifecycle model, recognizing that records do not ‘die’ but are reused and reanimated in various communities at different times.” Frank Upward, among others, has been discussing “post-custodial” approach to archival work, which, though it does not disregard the care/custodianship of physical artifacts nonetheless also takes into account the need to think about the “continual transactionality” of documents (Upward, 1996). These are “fluid” to a new degree and in novel ways.

The possibilities of “new” activity are not just re-workings of the activities of quiet shelves, files in vaults, and instruments of textual accounting. The editing techniques descended from the scholars of biblical, sacred, and classical texts within erudite communities of rabbinic Hebrew, orthodox Greek, Sanskrit, and classical Latin scholars, among others, were used assiduously to assemble as complete a knowledge base as they could. These are essential, but not sufficient, to current tasks. We have not only to deal with fragmentary evidence, but with fleeting, fugitive compositions. The bright surfaces and pixel arrays of screens and devices display twitter streams and postings according to the browser’s whimsical responses. Documents are configured momentarily, in passing, on the fly. Are these audio-visual-textual offerings “documents” in any traditional sense? No WayBack machines in the world can reconfigure today’s search results the same way tomorrow. The composition is dependent on what I do in the course of the day, what I have searched and will search a few moments later, but also, on the always shifting field of networked texts within range of the search engine algorithm. The “document” that is displayed as the result of a search for “topic modelling” contrasts with that of “topic modeling” (spelling variance), and each is different from the results displayed a day earlier.

This category of contingent and ephemeral documents can’t be “captured” except with screen grabs, and the same is true of a host of other temporary screen configurations that are based on a set of protocols meeting query conditions and making a display. For the document specialists, these aberrations have to find their place among the envelopes and antelopes, the kitted scarves and knotted skeins, the furtively scribbled prayers thrown on a flame and the never-revealed ciphers of occult messages that have all come to stand before the dock and be deemed worthy or not to be included among the documentary elect. Other specimens of “conditional” documents exist and they also alter or trouble our understanding of, for instance, the “temporality” of a document as a feature of its ontological/inherent existence.

A conditional document is not a speculative one, not imaginative or imagined, but is produced by protocols and processes that use structured conditions as a way to run, operate, select information, and display it. It is configured as the outcome of specified process varyingly specific constraints (a filtered search in a closed data based vs. a search by Google across the WWWeb). We could argue that the protocols, processes, and constraints built into the structure and organization of the database are the documents. But that would only dodge the question of how we think about search results pages, for instance, as a document. Documents and means of their production are related, but not the same. The conditions are a means of conjuring that remain distinct from their creations, so merely saving the search or query protocols does not preserve the documents that arise as their results.

Revisiting other examples of conditional texts and their odd temporalities expands the range on which this argument builds: the results of data mining in a faceted search, a display of structured texts in filtered search, the results of natural language processing on a corpus, and any engagement with the ever-changing networked domain of the Web. Seeking to define a “document” in this circumstance, we can try to default to the entirety of the system, and suggest that the document depends on all the code/codes, networks, systems software/hardware, and contingencies.  But then the whole ecology of the Web has to come into play, as it should, but this does not help establish limits on our already spiraling-out-of bounds definition of a document. Faceted searching in a closed system will yield repeatable results whose circularity returns us to the system itself as the “full” document. But, again, in an environment like the WWWeb, this is meaningless. The web is too changeable, organically so, too whimsical in its shifts of information mood and weight, its zones of access and obscurity, its constant flicker of appearances and disappearances. The implications for scholarship, policy, or any activities that depend upon documentary evidence are profound. Imagine that you create a Google n-gram for a term that has had currency in a very short and recent time frame (e.g. “metadata” just had a vogue moment in popular press). If these results are used to support an argument, then what is the record of what is displayed on my screen? A print out is a surrogate that cannot satisfy the need to show the choices and results or the field from which the document is drawn. A screen shot suffices as a snapshot, but loses the interaction with the living web. Even working within a more controlled or close environment, one faces the challenge of how to cite and display query results as a facet of a larger database. 

Another good example of text documents with unstable ontological identities are those produced by Matthew Hurst’s Hapax Legomenon, a natural language processor that distills of language culled from the Web between a particular start and stop date. These have no stability as enduring documents, and in fact, exist only as long as they are on view. We can conditionally configure documents, dependent on our fickle attention for their very existence, but they can vanish along the errant whimsy of our lines of attention intersecting our computational realms.

If ever the principles of a Heraclitan flux were embodied in the very ontology/phenomenology of an artifact, it is here, now, in the fleeting immediacies through which a document composes itself for our eyes only and for an instant’s disregard and then vanishes. Siblings and cousins and shades of resemblance may reassemble, so like the original we mistake them, momentarily, for that earlier temporary object brought into being but our attention, but then, with regret, relief, and other realization of the subtle but significant difference between the initial document and this “new” one, we realize the perils of our connection to refresh rates. No corpses remain. The past history of the documentary field is only in the cache, very easily emptied, more difficult to preserve. Gone, not in the same way as Ozymandis’s past glory, crumbling into ruins, over which we may wax romantically mournful. No, this is a profoundly new form of vanishing—without an inscription (forensics and their orthodox positivism aside). They leave no trace because there is no ground on which to register it, no way to  preserve or recover the phantasm whose materiality is dependent on so many contingencies and co-dependencies of distributed hardware and related software, networks and clock speeds, protocols and display capacities. Ground and figure were/are co-dependent, and not just on each other, as smoke rings are on the density of air, but on the temporally configured conditions that produce these ephemeralities and make them available for a moment—or more—of cognition. “Documents” have never been more material—and yet—they push against new limits of ephemerality, stretching the temporal spectrum of existence to nano-thresholds below perception and above its limits.

Questions of preservation and access follow, but the ones of ontological and phenomenal existence come first. The event spaces collapse into fleeting ephemerality possessed of a mad momentum. When I need to reference the results of a Search by referring to a temporally configured “document” that appeared on account of the precise conditions that existed at the moment of production I realize that the human universe resembles, more than before, those universes we produce in observation of quantum phenomena as well as those of the cosmos at its opposite and expanding scale.

Temporality defines documentality in these conditional circumstances, it does not merely enact specific variants of a document at different rates or speeds, its being as a document is an expression of momentary–highly fleeting—conditions whose relation to each other is temporal. The “being” of a document is always in a condition of “becoming,” just that now that act is unmoored from a substrate with memory.

As we consider the relations of documents to their temporalities, we see this is an issue of the very identity in/as production of documents. Rather than the history of a document, and/or its place in a temporal continuum, now, a conditional document is a contingent configuration, a fleeting document, which is produced across a span of time. This changes the identity and status of documents, and is not just a matter of degree, but of ontology—a difference in kind and character of how a document is in and of the cultural and material world.

Bibliography and book history, in their future formulations, will have to contend with these changed conditions. The distributed character of the document includes distribution across space(s) and temporalities. The artifact has no singular autonomy under such circumstances (if it ever did, it now loses that illusion). A constructivist epistemology, one that takes the conditional of objects into account, as well as taking seriously the constitutive acts of engagement, can still track its allegiance to the traditions and conventions of bibliography and document studies. The requirements pressed on us for reconceptualization within the networked environments of document and text production cause us to reflect retrospectively, not just redefining our understanding of the past, but mining it for insights we were not able to have in advance of our current moment.