I'm giving a lecture on some of my recent digital research on Proust. The talk will cover methods in text annotation and visualization, with a view toward their theoretical implications for literary criticism. Along the way it will describe some of my experiments with text mining and social network analysis for generating and representing associative paths.

• Wednesday November 17, 5:00-7:00 pm
• Lucy Ellis Lounge, 1st floor Foreign Language Building
• University of Illinois, Urbana-Champaign

Just made a LOLCat to explain Proust and churches.

We can run Proust through social network analysis (SNA) software in order to generate network models among its information nodes. The visualization to the left shows the association (basically, an uncategorized tag I use to label church-related passages) of Venice as it is networked among passages (referenced by their ID numbers and pagination codes) and notes on narrative context. When manipulated in real time, the visualization highlights the links to other nodes and their related concepts or passages.  What this means for the study of Proust is that we can think of the novel as a network of nodes consisting of concepts, characters, narrative elements, and any other unit of meaning that might enhance exploration of its text.

We can even include various texts and external information for a genetic or contextual study of the novel. In a hypothetical archive containing digitized avant-texte and published variants of the Recherche, we could potentially see -- in strikingly visual terms -- the correspondence of, say, the impact of WWI on the development of different sections. This could provide new insights into Proust's writing process as this work continually ballooned and changed during and after the war. What kinds of associations got the most development during and immediately after the war, and in what points of the narrative did they occur? Which churches received the most attention and where are they located in both fictional and real space? There is a wealth of traditional scholarship addressing genetic and contextual issues like these. However, SNA presents an opportunity to view all of the information nodes simultaneously, a much more powerful (and accurate) tool for the study of a book than other print books are. In that way, we would see and move around in the Recherche as a writerly text that responds to its own inner needs in reference to the war. Tools like SNA also present new pastures for narratology. If the text is marked up appropriately, all instances of a particular narrative device or structure could be instantly recalled by a researcher and viewed in relation to any parameters desired. It would be an even more comprehensive supplement than Barthes and Genette. All of this goes to say that a more rigorous taxonomy of the Recherche would be necessary for a meaningful SNA application. At present, the interpretive apparatus of the Ecclesiastical Proust Archive consists solely of associations, which are uncategorized tags denoting concepts, themes, important details, architectural elements of the churches described, and so on. It would be far more meaningful to tag separately the characters, churches, architectural elements, themes, concepts, and plot elements that make up the rich density of this novel, as well as the images and other media added here to illustrate it, so that they become individual nodes within the information network. Then a far more rigorous and powerful visualization of the novel would be possible, and new discoveries will almost certainly be made. But so far, these notions pertain only to my particular study of the church motif. A far richer application would be made if we took an entire electronic text (or, better, all of the variants and translations) and allowed researchers to mark them up and add media by way of illustration. In that way, the Proust archive would become a collaborative, electronic research and editing environment that takes shape from individuals' own scholarly pursuits

Social network analysis (SNA) software combines a variety of methods commonly used in digital humanities research, such as text mining, visualization, and modeling. SNA software can pour over the data and metadata in the archive's XML files and generate a network of nodes. It could be trained to recognize and normalize names, or even pseudonyms, and the metadata, provided by readers, would tell it whether a given passage contained the idea of Venice, or the subject/object distinction, or jealousy (or all three). If it had a qualitative analysis component it could even recognize concepts. And of course there would need to be the capability for scholars to add and tag information about the documents in the archive. This is a daunting task, but eminently possible with the aid of text and data mining software. There are pitfalls, of course. The accuracy of any analytic tool depends on the quality of the data it operates upon. We must always be aware that tools like these, powerful and impressive though they are, always represent a state of the information realm. This is no different from traditional, print-based scholarship, but it bears consideration given the sometimes exaggerated hype of digital humanities at the time of this writing. So, now that AccessTEI has provided us an XML file with structural TEI markup, I'll be looking for ways to mine it with text analysis and SNA software. Stay tuned for more updates. As an aside, SNA has tremendous possibilities for the study of modernist magazine culture, which is an actual, publishing network. See my post from earlier today at the Magazine Modernisms blog.

In searching for suitable work in progress for the IPRH seminar, I came across this beginning to a meditative essay that I composed nearly eight years ago. I thought I would include it here as an example of the kind of writing I intend to do in correlation to this digital humanities project. The visualization and mapping tools could be used to enhance some of the issues that come up regarding the frequency of certain themes with particular places or place names.

---

In the final pages of In Search of Lost Time the narrator says of his intended book that he would “build it up like a church” (VI.507). He would also stitch it simply like a dress, regroup his forces like a general conducting an offensive, endure it like a medical regime (VI.507-9). But the book that he has written over the six volumes leading to this point features the church motif more prominently than the others mentioned in these self reflexive pages. In fact, the connection between self reflection and churches is prefigured on page one, where the narrator begins the story by explaining his nightly shift between waking and sleep.

And half an hour later the thought that it was time to look for sleep would awaken me; I would make as if to put away the book which I imagined was still in my hands, and to blow out the light; I had gone on thinking, while I was asleep, about what I had just been reading, but these thoughts had taken a rather peculiar turn; it seemed to me that I myself was the immediate subject of my book: a church, a quartet, the rivalry between François I and Charles V. (I.1)

The apparent blending of subject and object “did not offend [his] reason” (I.1) and, we later learn, became the major motivation for his search of lost time. One critical aspect of this passage is that it establishes the conflict between [illusory?] perception and imperceptible reality, of which the narrator’s other conflicts are types (for example the questions of Albertine’s sexuality and fidelity). The church and quartet are significant here in that they embody and document time itself. Time and harmonics are the essential elements of the art of music, the bringing to life of a continual emotional present that can be re-performed but never duplicated. An old church brings to life the presence of the past and is a supreme exemplar of the general in the particular. It also features many arts—architecture, sculpture, stained glass, painting, tapestry, music, narrative, fashion, even food and drink—that bring together their special effects to express the whole of human experience. It is therefore not surprising that the narrator concludes that books of the magnitude he will undertake are never complete: “How many great cathedrals remain unfinished!” (VI.508).

Defined briefly, a motif is a recurring thematic or structural element in any art, but especially in music, architecture, painting, and literature. In addition to manifestation through plot and character, motifs can be embodied in place or time. Churches in In Search of Lost Time comprise one motif among many. The recurring elements that unify Proust’s novel—the Madeleine, the Japanese paper game, the little phrase, the parish church at Combray, to name a few—allow for the modulation of themes and function like motifs in a piece of music, but also like the synaesthetic motifs of a church. Churches fulfill several functions in the novel ranging from an element of setting to the object of discourse. Churches are therefore highly appropriate as a motif because they embody both place and time. Considering the narrator’s comparison of his book to a great unfinished cathedral, the notion of a church motif, and possibly the motif itself, might be incomplete.

This paper will perform a meditation on the church motif of In Search of Lost Time, focusing on significant as well as seemingly insignificant moments. The goal is to define the import and function of the church motif, volume by volume, as a mean to developing a theory of narrative and further to illuminate Proust’s work. Questions raised along the way will be varied and speculative because this piece will attempt to discover the ground upon which a more formal study will be based. For that reason, the paper might seem in places to be disjointed, incomplete, incoherent, or sketchy. Since my orals lists will focus on modernism, realism/naturalism, and narrative theory, I will also attempt to begin fleshing out questions and issues to address while reading for the exam. This paper will also make use of and address issues pertaining to the Ecclesiastical Proust Archive, a beta version of which is at http://web.gc.cuny.edu/provost/apit/itech/proustarchive/search.asp.

I

Volume I features a lengthy meditation on the parish church of Combray along with numerous references to it and other churches. The meditation on Saint-Hilaire constitutes twelve pages of exquisite description of the parts of the church, its content, and what it meant to the narrator as a young boy. Saint-Hilaire is a shaping influence on Combray and its citizens’ activities. It is also an icon that represents their place, their essence. For the narrator, it is a point of origin and guidance in his geographical, temporal, and biological (biographical?) movement as well as in his vocation; it is later in this volume the setting for several formative events, notably his first observation of the Duchesse de Guermantes at the wedding of Dr. Percepied’s daughter. Other churches, visited or imagined, figure prominently in his walks along the Méséglise and Guermantes ways, which form the “deepest layer of [his] mental soil” (I.260). And Saint-Hilaire always manages to show itself from afar as he returns to Combray.

The passage quoted in the initial paragraph of this essay (I/I.1) underscores the extent to which the narrator is as much a reader as the writer of his book—both the book held in childhood and the one we ourselves read—as much its subject as all the other themes. Likewise, its readers will be furnished with “the means of reading what lay inside themselves” (VI.508). One question raised here pertains to realism—whether reality lies in things themselves or in the experience or memory of them. The passage could possibly be linked to memory, for the narrator claims later that memories reside not in the mind but in things. However, since memory is not reality but a reproduction of it, the novel probably has more of a non-realist aesthetic despite the meticulous realism of external description. How can we relate these issues to modernist aesthetics such as surrealism? Whatever is happening aesthetically, Proust seems to be questioning traditional epistemologies based on the subject/object dichotomy, especially when the subject becomes its own object of inquiry. How does this fit with modernist trends? What are the relationships of reality and epistemology in (non-)realist aesthetics?

Epistemic and aesthetic issues also appear in the next church passage, which deals with layers of representation and authenticity. Similar to the first passage’s blending of reading and dreaming, this one occurs during bedtime reading, when the narrator’s grandmother gives him photographs of paintings to calm his nerves.

She attempted by a subterfuge, if not to eliminate altogether this commercial banality, at least to minimise it, to supplant it to a certain extent with what was still art, to introduce, as it were, several “thicknesses” of art: instead of photographs of Chartres Cathedral, of the Fountains of Saint-Cloud, or of Vesuvius, she would inquire of Swann whether some great painter had not depicted them, and preferred to give me photographs of “Chartres Cathedral” after Corot, of the “Fountains of Saint-Cloud” after Hubert Robert, and of “Vesuvius” after Turner, which were a stage higher in the scale of art. (I.I/53-4)

The narrator’s sardonic tone highlights the...

realism, different valuations of art object, layer of closeness not necessarily relevant to that of representation (photograph or painting of the original vs. photograph of a painting of the original). Proust archive images—digital simulacra of simulacra; what matter the color + lighting differences? For example the Corot Chartres Cathedral painting images: one is significantly darker than the other.

Does the file manipulator/editor’s hand or intent matter? Relationship of narrator to grandmother—irony is she wants to give him something of a higher or better aesthetic value, whereas photographs themselves can be valuable aesthetic objects (not to mention the aesthetic value of the cathedral itself, which is the ultimate referent here). To her mind painting is the higher art, though is mediating the boy further from the original beauty of the cathedral. However, if you consider the painting itself is beautiful and an art object, you are still only a step away from the painting, which puts you two steps away from the original. It all depends on where you wish to stop/stand, which is part of the point of the novel. The narrator later comes to realize that mediation is all there is and that the truth, or reality, is always in it (page ref?). This scene also prefigures the repeating mediation of experience and memory through memories of both, with the church figuring as the point of origin. Church is appropriate because it surrounds entirely when you’re in it, and is always present outside the self when in its area/town. Church is itself a memory vault, the memory of history, providing experience of external and internal memories in all their dimensions. Photographs of paintings of churches help to convey/expose this epistemic structure/cycle and prefigure what narrator’s primary mission will be throughout the novel.

In addition to my written summary, you can find the resulting action plans, poll data, and Twitter commentary here:

https://wiki.projectbamboo.org/display/BPUB/Workshop+4
http://search.twitter.com/search?q=%23pb4
OVERVIEW

I think Bamboo Workshop 4 went well overall and I was very glad to attend. To my surprise, pedagogy seemed to be one of the areas with the most energy and support behind it, and from various quarters including IT, Librarians, and -- to a *lesser* extent -- faculy. The meeting polled the constituents several different times on institutional interest and willingness to commit to working on the areas of the planning document. I marked us down as willing to be leaders in the Education and Curriculum section, and understand from George that we might be willing to lead on others.

The Bamboo organizers worked very hard to change the agenda on the fly
according to the needs and input from the meeting constituents.

My sense of the plan is mixed. I definitely think it's a good idea, and I'm enthusiastic about working on it, but budgetary restraints in these economic times are going to introduce difficulties for overstretched IT departments who might be called upon to allocate personnel or financial resources to Bamboo.

My other main concern was that Bamboo might replicate or rebuild some existing services (like MERLOT or SAKAI), but by the final reporting session on day 3 it was clear that much thinking and planning had been done to avoid that. The question is how to do it. This should be more clear below.

The meeting resulted in sets of principles, 1st year, and 2nd year action plans for all sub-sections that can be viewed on the wiki. The following page will give you a strong sense of the meeting's results, as it contains the poll numbers on the various sections, as well as the action plans, and notes on the discussions taken by Bamboo folks:

https://wiki.projectbamboo.org/display/BPUB/Workshop+4

You can also view the Twitter discussion (#pb4) here:

http://search.twitter.com/search?q=%23pb4

Following is my summary of the days' events.

DAY 1

Goal of the workshop is to generate a proposal for the 1st 3 years, as a series of one year proposals. Resources are starting to come into focus. They need to know who (institutionally) will take part, and what kinds of resources they can allocate to the project.

PB4 worked toward PB5, seeking our input on major program elements, the consortial model, etc. through non-binding polls and discussions. The following will happen immediately after PB4:

• Program staff mtg in Chicago Mon/Tue to debrief + digest
• Create straw man of first phase implementation proposal
• Schedule conversations with institutions and organizations to guage interest and commitment.
• Refine and edit Bamboo program
• Develop and refine the Implementation Proposal
• Have something up and operational in 12-month timefrime (what Mellon wants)
• Meet June 17-19 Washington DC, at UC Berkeley center there, right before digital humanities 2009.

After PB5, they wil:

• Finish + finalize Bamboo Program
• Finish + finalize Implementation Proposal
• Finalize conversations with institutions and organizations (what we can contribute and particpate in: a "sign on the dotted line" kind of moment; who's in, who's going to wait and see?)

Chad and David would like an email on Sunday from institutions about what kind of roles and commitments they might like to make in the project. I am thinking we could commit to the Education and Curriculum part of it (at least I'd like to be actively involved in this), and possibly others?

OVERVIEW OF DOCUMENT

1. Vision

2. Scope of work -- the idea of moving a big multifarious project forward but not as a single monolothic object -- has to be broken up some how with various semi-autonomous elements (the Cloud)

- How to make this happen at the technology layer (Cloud)

3. Major Activities

• The Forum - the are already mature efforts (SAKAI, Fluid) that are humanities led, so we need to be careful not merely to replicate these.
• Bamboo would take a step back and somehow connect some of these, but also lead where identified needs are lacking in resources
• Bamboo would be a kind of glue for a longer term infrastructure
• Social model where different people can be in leadership roles
• But not proposing own social networking environment like Facebook or SAKAI, rather a way to connect them and still allow you to find project related to your disciplinary work, etc. (learned societies).

4. The Cloud
Infrastructure for sharing services, gadgets that minimizes risk, is inherently redundant, low in cost, and introduces the potentional for broad adoption across institutions, organizations and georgraphical boundaries in a sustainable and reliable manner.

Constraints - financial? Ballpark figures have changed to around $1m per year for first three, could be more or less. I assume this is a diminished figure from what was initially offered by Mellon.

Scholarly Narratives working group report -- they tried to fill a gap between scholars' description of work in PB1 and Tech designs in PB2. Met with tech and scholarly people; idea was to provide connection between scholarship and design work. How do we identify and define the needs we're trying to address? How can we design tools and infrastructure to support that kind of work?

PB needs a map -- scholars' needs must be expressible in terms of tech capability. The tech must be describable in terms of the schoalrly activities they support. PB "tri-group" team envisions such a map.

Sch. Narrative --> Recipe /  Activity Definition --> Tools / Content / Services
scholarship <----------> technology

The Scholarly Narratives are often not described as overtly digital or technical; frequently overlap one with another; require deconstruction and analysis.

Recipes are one way to tease out relevant elements of narratives: facilitate processes for other institutions who might need insight in producing their own projects.

The Scholarly Network is ffor managing bibliographies, centralizing resources, seredipitously finding other scholars working on same subject, etc.

The Forum will advance the narrative-recipe-activity service process to become more streamlined and efficient, ie, allow PB to better sort wheat from chaff.

1st Poll assessing our institutional interest in the following sections of area 3.

2nd poll, assessing our institutional interest in the following sections of area 4.

Services Atlas
- Bamboo inviting contributions for user interface design. What will API look like? Etc.

Section 5 - Lab

I don't remember much about this. Sorry: my attention was failing by this point in the day.

DAY 2

This day began with an activity that broke down the constituency according to their institutional roles: (1) faculty, (2) administration, (3) technology / IT / Library. I was placed in group 3 as a technologist. During our small table discussion, teaching started as the main focus. Some of the librarians (these from the Open University in Britain) felt that expertise in pedagogy seems to have shifted to technology services. Also, teaching is important politically because funding and resource allocation for research projects (esp. in small liberal arts schools) needs to show relevance and impact on students.

I added that if a mission of Bamboo is to facilitate digital humanities research and bring up the next generation, much of that happens through teaching, so I would think PB would have a vested interest in best practices and maximizing learning.

Networking -- the IT executives in my discussion group felt that the networking tools proposed for Bamboo would be more of a drain on their staff than a help to their work. However, they also felt that it's good for technologists, faculty, and librarians to keep getting together like this to discuss it all. The IT execs, though open to the project, seemed to have a less idealistic notion of how the work of Bamboo would benefit their own departments, or even their universities as a whole.

Risk of Bamboo is to demonstrate a clear payoff to researchers (faculty) within a defined timeframe. If Bamboo can't communicate that, it will be difficult to make it work.

One IT person said that humanities scholarship is an individual pursuit and that the Forum would not take load off him.

The IT folks began to ask whether PB would just become another MERLOT. Facebook is successful because it reached a critical mass. I don't think MERLOT has the kind of institutional buy-in -- and therefore the motivation for continued commitment for development and improvement -- that PB is looking for. PB strikes me as being fundamentally a multi-*institution*-driven project rather than a social network of individual humanists.

Q&A re: small table discussions:
If digital humanities is going to flourish, then institutional and disciplinary Silos will have to break down and reform. Linear Recipes might be too narrow -- need feedback loop type structure (not sure what was meant by this). Scenarios might be better way to think of Recipes. Why use Bamboo over something like FB for networking? Can we leverage or interoperate with something like FB? Tools etc. are short lived relative to interoperability standards.

Following the Q&A of small table discussions, we took a poll on the level of institutional willingness to commit to the sections of PB in areas 3 & 4.
DAY 3

Small table discussions on what to do next in Bamboo. Establish an integrated core for the first year.

Long range plan has to be to reach beyond that initial group of humanists building Bamboo.

Neil Fraistat: keep humanities scholars engaged in shaping Bamboo at all stages, else the technical/humanities divide will remain.

Elli Mylonas: True partnerships have to go both ways.

Charter: Do we need different versions of it or different schematics to address the different constituent audiences in Bamboo?

While reading Franco Moretti's Graphs, Maps, Trees: Abstract Models for Literary History, it occurred to me that the Ecclesiastical Proust Archive should do the same within itself. The first step this Winter will be to complete a major information overhaul, marking up and encoding all passages and images rigorously in XML. Then, dynamic real-time visual tools can be used to illuminate the Recherche, narrative, and the manner in which archive users have been interacting with the novel.

Moretti's use of graphs to illustrate the publishing data about the novel in different times and locations throughout history shows fascinating patterns about its system of subgenres, its rises and falls, and the relationships it bears with politics and economics. Such models could be applied to the Ecclesiastical Proust Archive, but for purpose of illuminating its internal relationships. Graphs could be used to show various aspects of the church motif and how they are patterned throughout the work. Would, say, the rises and falls of particular associations tell us anything about the novel's exploration of memory or subjectivity, especially anything that might not be obvious in Proust scholarship hitherto? If so, what do these patterns tell us about narrative itself, and of the motif as an element of narrative?

The mapping techniques that Moretti applies to certain English novels reveal interesting patterns in their plot elements, such as the consistency in which certain types of plots form distinct rings around the geographical center of the action. What would we learn from maps of churches in the Recherche, and how they relate to its exploration of subjective memory, national memory, local memory, memorialization through architecture, archives, and narrative? Would the regions of France, their churches, and how the churches signify within the narrative tell us anything new about the Recherche?

Perhaps even trees dealing with associations, categories, or motifs could tell us a lot about the church motif and its operation within the whole narrative.

As well, these kinds of tools could be used to illuminate the relationships between critical discussion on the blog and the text and images in the archive. Which passages get the most attention? The least? Which associations and image properties are most or least discussed? How do these change over time?

A unified, dynamic, and interactive visualization section of the Ecclesiastical Proust Archive could potentially show so much about the Recherche and narrative that has not yet been seen. It could prove to be a new method of inquiry into the novel, Proust, narrative, literary scholarship, and more.

Here are some basic relationships that a visualization application could explore:

Graphs

• Association by pagination location in the novel.
• Association by chronological location in the narrative.
• Association by chronology of composition (would require extensive textual scholarship).
• Association by church.
• Blog categories by association, and/or by image property, and/or by church.
• Real, fictional, and hybrid churches by location in the novel, in the narrative, by image type, etc.

Maps

• Geographical locations of churches.
• Geographical locations of associations and churches.
• Geographical locations of associations by churches, broken down by real, fictional, and hybrid churches.
• Character by geographical location, church, associations, and image properties.
• Any of the above by critical categories in the blog.

Trees

• Breakdown schema of how real and hybrid churches are used in particular associations (say, romantic love or the subject/object distinction).
• Schema of how blog categories explore certain associations or image properties.
• Schema of an image property and the kinds of associations it tends to appear with.

As a note for a future post, in order to make the data more effective in visual applications, a rigorous categorization of the associations (which are non-categorized) will probably be necessary.

(Continued from this post and this post.)

This project began as a spreadsheet documenting the church passages for a term paper in a Proust seminar. It was subsequently compared by someone else in a textual scholarship seminar to Roland Barthes' S/Z. The comparison holds on two counts: the arranging of passages in a cross-referenced grid system and the inclusion of interpretive keys as paratexts.

Barthes' method in S/Z, a narratological analysis of Balzac's novella Sarrasine, breaks down the entire story into passages (which he calls "lexia"), beneath which appear his analyses according to five semiotic codes: the hermeneutic (HER), semantic (SEM), symbolic (SYM), proairetic [or actional] (ACT), and referential (REF). This enables him to perform a step-by-step reading that remains attentive to the plural of the text. This method,

through its very slowness and dispersion, avoids penetrating, reversing the tutor text, giving an internal image of it: it is never anything but the decomposition (in the cinematographic sense) of the work of reading: a slow motion, so to speak, neither wholly image nor wholly analysis; it is, finally, in the very writing of the commentary, a systematic use of digression (a form ill-accommodated by the discourse of knowledge) and thereby a way of observing the reversibility of the structures from which the text is woven; of course, the classic text is incompletely reversible (it is modestly plural): the reading of this text occurs within a necessary order, which the gradual analysis will make precisely its order of writing; but the step-by-step commentary is of necessity a renewal of the entrances to the text, it avoids structuring the text excessively, avoids giving it that additional structure which would come from a dissertation and would close it: it stars the text, instead of assembling it. (12-13; Barthes' emphases)

To a contemporary reader, Barthes' digressions take a form remarkably similar to blog posts with category tags and commentary. Each segment of the book is anywhere from one to about five pages in length and begins with a number, a title, a passage from Sarrasine, and then commentary that incorporates any of the five semiotic codes that might be present. And they progress rigidly in chronological order according to the tutor text -- as we sometimes say of the seriality of blogs under the "tyranny of the timestamp" (but which can be "adjusted"). He uses the codes as a system for both teasing out the plural of the tutor text in the act of reading and for referring to other passages containing the same types of signifiers, which is like tagging in Web 2.0.

One of the true innovations of Barthes' approach in S/Z is the simplicity of the overall structure. In using only the basic procedures of analysis, labeling, and cross-reference, without the inhibiting burden of organizing them around large themes or an articulation of the whole text, he is able to use the digressive episodes to mine each lexia for its plurality. Each segment becomes a self-contained discourse on the lexia it falls under, making connections as it pleases. Or as he says:

If we want to remain attentive to the plural of the text (however limited it may be), we must renounce structuring this text in large masses, as was done by classical rhetoric and by secondary-school education: no construction of the text: everything signifies ceaselessly and several times, but without being delegated to a great final ensemble, to an ultimate structure. Whence the idea, and so to speak the necessity, of a gradual analysis of a single text. Whence, it would seem, several implications and several advantages. The commentary on a single text is not a contingent activity, assigned the reassuring alibi of the "concrete": the single text is valid for all the texts of literature, not in that it represents them (abstracts and equalizes them), but in that literature itself is never anything but a single text: the one text is not an (inductive) access to a Model, but entrance into a network with a thousand entrances; to take this entrance is to aim, ultimately, not at a legal structure of norms and departures, a narrative or poetic Law, but at a perspective (of fragments, of voices from other texts, other codes), whose vanishing point is nonetheless ceaselsessly pushed back, mysteriously opened: each (single) text is the very theory (and not the mere example) of this vanishing, of this difference which indefinitely returns, insubmissive. (11-12; Barthes' emphasis)

The notion of Literature as a single hypertext of voices, "a network with a thousand entrances," is where the Ecclesiastical Proust Archive begins its own construction (though I hadn't studied S/Z until long after the search engine was built). It singles out one strain of the narrative in order to examine up close the multiple voices and "entrances" and "vanishing points" of the Recherche. The purpose is to read Proust in a way that hasn't been done before, and also to further the study of narrative by using new tools (search engine, blog, taxonomic and folksonomic organization, hypertext) that were foreshadowed but unavailable to narratologists during the 1970s:

to take up the structural analysis of narrative where it has been left till now: at the major structures; it is to assume the power (the time, the elbow room) of working back along the threads of meanins, of abandoning no site of the signifier without endeavoring to ascertain the code or codes of which this site is perhaps the starting point (or the goal).... (S/Z 12)

In more selectively culling its lexia but less selectively organizing its interpretive codes (the uncategorized associations), the Ecclesiastical Proust Archive highlights both the entrances and vanishings of the text -- where each instance of the church motif begins and ends and the voices and codes that weave therein, that channel them from other parts of the narrative but are amplified and cut off. (In much the same way, the church is for Proust's narrator both the origin and the end, the orienting post.) And for the sake of recalling these instances it uses the advantages of the digital medium to archive and reorganize the text(s), to build itself accretively on its voices, just as a church or a book embodies those voices (of history, of love, of war, of strife and hope).

In the last post I asked a question related to David Greetham's metaphor of membranous transmission between archives.

In conceiving of a text as an archive (of knowledge, voices, attitudes, values) consisting of inter-membranous citations, this text interrogates its tutor text, and also itself. How must Proust be read here through the collect of its church motif (citations) and through the heterogeneous images (also citations) that supplement it?

In the ensuing discussion I neglected to consider the obvious question of genre. What makes the membrane metaphor so rich is its basis in the notion of leaves -- of a book. The Proust passages constituting the church motif have "crossed several membranes (membranae or 'leaves' of a book) to interrogate the integrity of the archives from which they have been drawn" (Werner and Voss 1). They have, first, been translated and revised (Enright revision of the Moncrieff/Kilmartin translation) from an original (to them) printed version in French, itself an edited variant of whatever beginnings it had in manuscript; second, been singled out through my acts of reading and interepretation; third, been transcribed into a spreadsheet by myself and the woman whom I subsequently married; fourth, been imported into a database that operates upon them in response to searches of their words and phrases, as well as the paratexts (associations, context notes, image properties, pagination) that form relations with them.

Hence, each fragment of the collect constituting the core text of this archive has passed through several leaves or membranes before arriving in its place here. Only one of those leaves surviving in the present constellation is in print; the other three are digital. In that way, the digital archive-text provides several functions that allow for an interaction of digital and print membranes through its multi-layered memorializing of readings. The digital text is a deliberately partial trace of the whole print text, and its native ability to be reorganized allows for a non-sequential reading of its component parts. Thus the fascicles (OED -- "A bunch, bundle. Now only in scientific use. Formerly also fig."; "A part, number, ‘livraison’ (of a work published by instalments)" -- demarcate the points of loss in the original, allowing readers to reconstitute, to re-member the original narrative in meaningful ways by means of the pupil text.

Membrane -- OED -- "classical Latin membrna a membrane (in animal bodies), parchment membrum"

Memory -- OED -- "classical Latin memoria memor mindful, remembering (a reduplicated formation)"

Memory as the act of preservation through reduplication (of the original, through writing), of committing to archival parchment, to a node in the database. Re-membering -- collecting and reassembling the membranes, the planes of memory in the novel's signifiers and (here) signifieds, the pieces of a motif extrapolated from an organic text. Proust's churches as the archives of both personal and collective memory; his book as the same; this archive as... ?

Before addressing Barthes' S/Z, I felt it necessary to broach this subject of the membranous layers between print book and digital archive. S/Z deliberately fragments (or "stars") the text of Sarrasine in order to tease out the full ambiguity of its signifiers, to get as close as possible to the writerly text by operating methodically upon the minutiae of the readerly one. Barthes ultimately concludes that a full articulation of the text's signifying structures is impossible because the text itself is not a closed system. This archive begins with that conclusion as an assumption, limiting its selection of citations but using the mobility of the digital medium to approach the writerly text of a narrative strain running through the original. The digital medium is perfectly suited to interrogate the valences of the print text by spontaneously realigning its parts to match the reader's intent.

What can the digital archive see in the book from which it derives?

While looking over some materials from one of the courses that sparked this project, I came across some notes on archive theory that seem especially relevant. There is a strong connection between the poetics of the archive and the activity of archiving.

In The Poetics of the Archive, Marta Werner and Paul Voss remind us that recent theories shift aspects of physical archives onto the conceptualization of texts and discursive practices. The archive's dual function as a guardian of memory and a mechanism for controlling access to that memory make it indistinguishable from the process of knowledge production.

If the first archons originally conceived of the archive as a space of pure knowledge, then for those who came after, including oursleves, the archive has more often revealed itself as an ideologically-charged space. This space, inseparable from the ensemble of operations deployed within it, confers order on its contents and creates a system whereby an official record of the past may be preserved and transmitted instact. The archive may be, in effect, a political space, a genedered space, a memorial space. (ii)

"This space, inseparable from the complex of operations deployed within it": The Ecclesiastical Proust Archive is the search engine, blog, forum, image galleries and the operations readers use to access its records. What does it record? The entire collection of passages forming the church motif; my readings of those passages -- in the form of the associations and context notes that appear as search parameters (if selected) and as paratexts in the results (if selected); the images that contain (archive) my memories -- as well as those of hundreds of other people alive and dead -- of churches in France that are also archived in Proust's novel; potentially the readings of other researchers in the comments field and the forum; the many thousands of pharmacological and pornographic offerings of comment spam quarantined by a plugin.

In making the church motif of Proust's Recherche the controlling idea of this archive, I have, as archon, already imposed an order and a system on the rest of its content. In so doing, I have also preconditioned the readings that take place here, making the interpretive discourse both a result of the archival function and a part of that function. As David Greetham points out, via Derrida, in "'Who's In, Who's Out': The Cultural Poetics of Archival Exclusion," the exergue or collection of citations before the beginning of a discursive piece sets the tone, meaning, and form of what follows. The collection of passages in this archive therefore functions similarly to the miscellaneous citations that perform as epigraphs in Greetham's essay: "they have thus crossed several membranes (membranae or "leaves" of a book) to interrogate the integrity of the archives from which they have been drawn (and redrawn) and the one into which they are imported" (Werner and Voss 1).

In conceiving of a text as an archive (of knowledge, voices, attitudes, values) consisting of inter-membranous citations, this text interrogates its tutor text, and also itself. How must Proust be read here through the collect of its church motif (citations) and through the heterogeneous images (also citations) that supplement it? This is where the reading of Proust alongside the relational attitude of the juxtaposed images generates much complexity. Some images depict an actual church named in the text (e.g. Chartres for "Chartres") in a documentary attitude. Some depict a real church on which a fictional one was based (e.g. the église Saint-Jacques at Illiers-Combray for the "église Saint-Hilaire of Combray") in a sort of demistifying, "source identification" attitude. [The hyphenation of that town's name in honor of Proust is another interesting example of archiving.] Because of the archontic rule I set myself for including an image for every passage, some images depict a real church for a fictional one that has no basis (or no single source) in reality (e.g. my ghostly black and white photos of Chartres porches for passages in which the narrator "dreams of meeting his love on the porch of some Gothic cathedral"), in which case the relationship is based on an analogue of architectural elements and/or an emotional affect held in common. While there are more combinations in the image/text relationships (and many more yet to be teased out), the question naturally arises of their effects upon other readers.

As the progenitor and editor of this archive, my readings are memorialized -- inscribed in the very architecture -- in a way that must necessarily hold greater sway over those who perform readings here later.

The history of the archive, on the one hand a history of conservation, is, on the other hand, a history of loss. The archives of antiquity have long since vanished; we receive their contents as fragments of only as citations in later works. (Werner and Voss i)

Much of recent theory considers archives compiled by single authors/editors, of which the present one is still an example. But what happens when the archive becomes collaborative, when the fragments of the original novel-archive are brought into new relationships with images or other texts by the editorial/authorial voices of other readers? How will the external forces of time, cultural and ideological shifts, and scholarly contribution alter its content and its meaning?

The complex relationship between the archive and memory is subject not only to external, historical forces, but also to its own interior dynamics: “the archive’s dream of perfect order is disturbed by the nightmare of its random, heterogeneous, and often unruly contents†that make it “always only partially decodeable†(ii). Hence, The Ecclesiastical Proust Archive deliberately embodies recent theories that question the archive’s teleological function: it self-consciously collects violently decontextualized citations and external heterogeneous images for the purpose of closely reading, and re-membering, a novel.

Its operation is thereby similar to Roland Barthes' archive of Balzac's novella Sarrasine, which will be addressed in the next post.

It occurred to me as I lay awake last night that Drupal could actually do much of what came out in our discussions at if:book a year ago. Dan Visel suggested allowing users to add their own images and their own passages (permissions nightmares), or to comment on searches, which is interesting. Drupal wouldn't be able to do a search comment. However, entering each passage as a page or story would enable:

• More images to be attached to passages.
• Commenting on the passages and, perhaps with a module, the images.
• Integration of Proust passages and commentary with services like del.icio.us, twitter, technorati.
• Use of modules to serendipitously or randomly highlight passages, images, and critical content.

The Drupal search tool would recall all of these. However, the downsides would be:

• Less immediate access to the search results, since they'd show up as headlines and teasers instead of displaying all info in a neat table as at present.
• There would be no way to conduct a pagination search for in-depth study of a particular segment of the novel.

Again, as I wrote in the previous post, the archival structure of this site must be "respectful" of the organicity of the novel genre. A Drupal or Drupal-like integrated search engine and Web 2.0 tool would open up possibilities inherent in the digital archive genre, but might go too far in doing violence to the novel genre.

With Web 2.0 (user-produced content), institutional considerations would have to address the topical specificity of the archive, lest it become an encyclopedic, directionless, Proustian wiki. That could mean instituting an archive staff committed to study of the church motif and narrative, which would require a grant or some other financial backing. At the very least it would mean vetting the readers who are allowed to post content (i.e. students, faculty, researches demonstrably focusing on Proust, etc.). But that too is inseparable from what an archive is -- a container of information, whose information is controlled, selected, interpreted, and presented by the archon and both the intra- and inter-institutional politics of its time and place.