Kolkata Urban Archive: A Semiotics for the Archive

By S Bharat  |  December 16, 2019

This article is the third part of a three part series reflecting on the archiving process undertaken to create the Kolkata Urban Archive. Access the second and first parts. 

One of the most common questions I ask of people, as an archivist, is “what is this document”, usually while holding up a piece of paper. Typically this question elicits information about what the document contains—“it’s about forest rights” or “it’s about a court case” or “it’s about an eviction in the city”. To which I respond, “but what is it?” 

At work here are two different ways of interpreting the question, and two different ways of framing the document’s ontology. The customary method, that we employ all the time in our daily lives, is to describe a document by what it says – its content. This is what we’ve all learnt to do, this is what we typically read a text for. For an archivist, this way of describing a text often surrenders its primacy to another based on the text’s context, purpose and provenance. I only became aware of this subtle shift in orientation through several such exchanges, particularly when I had to ‘teach’ my research assistants how to catalogue a document. 

It is not as if our customary ways of reading a text are completely divorced from the concerns of the archivist. Like many other students and avid readers in Delhi, I visited the Sunday Book Market at Daryaganj quite often. I quickly realised that reading the title and the blurb of every book on display was an impossible task given the sheer scale of the market. A more efficient method of sorting books had to be devised. Soon I was able to identify the publishers who bring out the books that interest me, and more importantly, the colour scheme, layout and design of their books. This enabled me to scan volumes quickly at a glance by just looking at the colour, font and layout of the spine. We all rely on such elements when we browse shelves at the library, at the bookstore or at the homes of friends and relatives, when we consult catalogues, when we browse the contents of a book. But the elements of a text that offer us this information do not always register in our minds as signs. Unless compelled by circumstance, we interpret them without necessarily recognising that we do so. 

The literary theorist Gerard Genette proposed the term ‘paratext’ to describe all the “accompaniments” of a text that do not belong to the text but nonetheless structure and govern the text’s presentation and reception. Such accompaniments include titles, prefaces, book covers and frontispieces, chapter headings and sub-headings, headers and footers, but also extend to things like advertisements, reviews, interviews, and facts about the author’s life itself. It is through the paratext that a text takes on a recognisable form and makes itself available to a reader. We all rely on such paratextual elements all the time, in fact they are deeply embedded in our habits  of reading, which is possibly why we are not always conscious of them. It is only when paratextual elements are missing that we notice them. 

Genette, with Borges’ ‘The Library of Babel’ on his mind, likens the paratext to a vestibule which “offers to anyone and everyone the possibility either of entering or of turning back”. As an archivist, it is in such vestibules that I dwell the most. What is a catalogue, after all, but a whole structure composed of such vestibules? The archivist turns his or her attention away from the text itself to look at an entirely different set of signs emitted by the document. Who wrote it? When? For what purpose? This turns the archivist’s gaze towards titles, towards prefaces and acknowledgements, towards the fine-print and the corners of the text, towards its front and back. It also compels the archivist to look at the text as a material object produced by material processes in time and space. One develops a familiarity with handwriting styles, with layout and formats preferred by a person or an institution, with the colour, texture and dimensions of stationery. Documents dwell in many places over the course of their lives and pass through different hands. They acquire the traces of those movements, of lived reality, so to speak. A tear in a corner, or the shape of a stain can become powerful signs that enable the archivist to locate a piece of paper within its proper context. A stain shared by different pieces of paper is a good indication that they belong together. These are signs that move beyond the content of a text. The archivist produces categories and structures out of such signs, out of this texture and materiality. Where the paratext is missing or ambiguous, categories weaken, become messy, ill-defined. Is this is a magazine or a newsletter? What differentiates a pamphlet from a brochure? Aakhir yeh kagaz hai kya cheez? (What kind of creature is this piece of paper?)

In the course of answering these questions the archive produces its own paratext, in the form of categories, tags, metadata and annotations. It extracts this paratext from the document’s existing paratext but also produces it out of the document’s content. As such it would perhaps be inaccurate to say that the archivist privileges the paratext over the text. Instead, we may say that the archivist reads a text with the specific intent of extracting paratextual elements out of it. The archival catalogue is, then, a compendium of these paratextual elements, which is in turn adorned by its own paratext. In a Borgesian vein, it becomes possible to think of catalogues of catalogues ad infinitum, each turning paratext into text in a ceaselessly recursive system. In that sense it is only fitting to think of the catalogue as a structure built entirely out of such vestibules since the archive is precisely that structure of openings that allows different meanings, different ways of entering its corpus, to permeate it.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.