MZ Sitewide

Digital narrative

Giorgio Buccellati – November 2007

Back to top: Digital narrative

Discourse and narrative

There is a special way in which disparate data are appropriated automatically into the system and integrated into a meaningful new whole. We may consider this process a digital discourse. Through a series of operations that are in large measure automated, it produces a construct that organizes data in the form of an argument. Its formal characteristics are those of a narrative that serves the same purpose as the traditional one, but with a “story line” that is construed differently.

Fragmentation is so characteristic of a browser edition that it is valid to ask whether we can properly speak of “discourse.” I give a positive aswer, although it may be more appropriate to speak of “discourses” in the plural. Arguments are proposed and followed in a text mode, with hyperlinks providing the equivalent of traditional crossreferences, footnotes or figures. But arguments can be constructed with a maximum of flexibility, by following at will a myriad different threads that offer themselves at random.

The great advantage of such an open-ended structure is the speed with which one can follow an unlimited quantity of paths that link the most unexpected observations. In so doing, randomness can become purposeful.

It is important, however, to caution against the disadvantages. A digital discourse can reduce to a minimum the time for reflection and the ability to absorb both arguments and data. Also, the perception of the whole is quite different between a browser edition and a book, and so are the modalities of use. Therefore one may easily loose track of the broader import and one may as a result drown in details. In that case, randomness may generate a purposeless daydreaming. Thus it may be useful to encourage the notion that one should in fact “read” and “study,” not just “surf,” a browser edition.

One of the keywords that seem to describe a digital presentation of the data is non-linear: a suggestion is advanced here that while the image is not improper, it is not as dramatically at variance with “linear” thought as one generally thinks.

Back to top: Digital narrative


I use the term “sequential” to refer to the logic of an argument. When an author presents an argument, a linear thread is followed, which the reader is supposed to be able to follow precisely in its sequentiality – otherwise one will not do justice to the intent of the author.

The browser edition approach may substantially obscure the ability of the reader to do so. While the sequential thread is generally present within a given page (such as this one), one is generally at a loss in following an intended argument across pages. To obviate this, different approaches may be followed as I do in this website and in the Urkesh Global Record (UGR) in particular.

  1. Intent. – Apart from formal techniques, it is important that an author be specifically mindful of the need to make explicit the sequential thread of his or her argument across pages. The ease with which segments may be written precisely as segments, and then just as easily assembled in a frame that provides an outward overarching scheme, may lessen the urge to follow a long range logical path, and to show it to the reader. But this is indispensable if one expects a long range argument to unfold in such a manner that the reader may in turn follow it.
  2. Hyperlinks. – While links may provide only the opportunity for jumping off in an endless, and potentially aimless, associative stream, they may more positively serve to open unexpected inquiry paths. What matters is the control one must exercise in tracing the path itself.
  3. Headers. – A difficulty in following a long-range argument is the time that normal “reading” requires. Headers within a page help to streamline the process, and to allow the reader to look at them as signposts of the path taken by the argument.
  4. Side bars. – By remaining visible at all times, the side bars serve as tables of contents that assist the reader in gaining and maintaining a unified perception of the whole while at the same time accessing the more minute details, always seen in their fuller context.

Back to top: Digital narrative

Historical background

It is enlightening to place the intellectual dimension of digital discourse in its broader historical and cultural context. Since the beginning of the twentieth century, a variety of approaches developed in our culture, which had as a common thread the fragmentation of perception. The recurrent theme was to project profound discontinuities in the natural sequence, with the expectation that a higher synthesis would be set in motion, capable to reconstitute the fragments into a plane of reality where a new continuity could emerge.

In painting, cubism gives the most explicit evidence of this trend. The greater role of dissonance may be seen as an example from modern music. The stream of consciousness approach to narrative (as in James Joyce’s work) is an apposite case from literature. And deconstruction may be viewed as providing the most explicit philosophical framework.

This was the intellectual humus that has found one more application in digital technology, in particular the one relating to browsers. While the latter has come about primarily as a result of the development of the medium, rather than as a conscious experiment in providing a new outlet for an intellectual trend, the full utilization of the system harks back to that substratum, and should be so viewed.

Back to top: Digital narrative

Types of narrative

The narrative is the concrete form taken by the discourse.

There are two major types of narrative within a proper digital context (see also under authorship). The first (primary narrative) is the one that is built up automatically. The second (secondary narrative) follows the standard mode of presentation that derives from the articulation of a logical argument. The qualifications “primary” and “secondary” refer to the relevance of the digital construct. In the former, the argument is developed by the programs on the basis of all the available pieces of information – hence it is properly digital, and global. In the latter the argument is developed by the writer who then inserts, selectively, the relevant pieces of information – hence it is digital only in a derivative (“secondary”) sort of way, and it is tendentially incomplete.

The primary narrative, then, starts with the tesserae, all of them, and ends with the mosaic. The secondary narrative, conversely, starts with the mosaic, with the whole argument as perceived by the author. The tesseare that are adduced explicitly are those that suit the argument and serve to buttress its conclusions. It is found on the upper left hand side of the browser page and is produced manually as in a standard book, and it brings together the writer’s reflections as they are developed linearly and sequentially.

Back to top: Digital narrative

Primary narrative

The coherent whole that emerges from the scattered, atomistic observations is embodied in three aspects of the Urkesh Global Record (UGR):

  1. The organization of the central portion of the page, which organizes the observations in an ordered sequential mode.
  2. The saturation of the text with hyperlinks, which allow the reader to follow ususpected inquiry paths.
  3. The availability of higher nodes (accessed through the lower left hand side of the page), which organize the data in a synthetic mode.

A good example of primary narrative is the one that pertains to the types of contact. Specific observations about the contact association of individual features and items are recorded at different points in time and by different individuals. Yet:

  1. they all coalesce into a cohesive and coherent sequence (see for example J3f148): this can be “read” as a meaningful text, because the data are organized according to the depositional sequence as inferred from the emplacement.
  2. The hyperlinks branch out to every single element in the relationship, providing the full panoply of details. But there is more.
  3. The same data, coordinated with those of all other features, build into a higher node that shows the complete stratigraphic history of the entire unit. All of this is generated automatically from input entries that are not only quite simple, but, potentially, quite apart from each other in time.

Back to top: Digital narrative

Database and narrative

Two basic models - a database organization of the data and a print-like presentation of an argument. The problem is that never do the twain actually meet.

On the one hand, a database, no matter how referential the underlying structure may be, is a juxtaposition of elements which is dynamic in terms of how data are organized, but remains essentially static in terms of the argument: it only provides information for a narrative that unfolds outside of it. This is why one only consults a database, one does not read it or study it.

The exact converse obtains with a print-like presentation, typically encased in a .PDF file. Here the argument is developed dynamically, in a standard linear manner. But access to the supporting data is static, because they are summoned in an ad hoc fashion in function of the sequential steps proposed in the argument.

Back to top: Digital narrative