What is a discourse? It’s a way to arrange facts, ideas, expression of feelings to give them order and consistency, to make them resonate with each other, or to inspire in those it is addressing a particular purpose. Different societies have developed during their history, different types of discourse – religious, political, philosophical, scientific, commercial, sentimental, each of them having its own rules, its own history. Do we get to the end of this history, or more generally to the end of the discourse itself? This is what one might think at first sight by measuring the impact that the digital revolution seems to impose this particular type of intellectual production. Roger Chartier announced a few years ago a “disturbance of the order of discourse” by the Internet and new media mixing everything: books, journals, personal letters, diaries and newspapers in the digital large funnel. At the “Web 3″ era, should one speak of a disappearance of the order of discourse? It seems so, as this new creed shows a clear focus and fascination for what lies in the speech, which is its foundation and primary material, without order or construction, namely the ” raw data “[1].
This radical change of perspective can be seen at work in several areas, starting at a political level with the open data movement. What is it? Cast as often form the United States, the open data movement aims at encouraging or compelling governments and their agencies to make public the data they collect or produce in all areas of their work: census data, of course, but also those related to crime, health and economic activity [2]. The purpose is twofold: on a political ground, it gives a better transparency and thus a better control for citizens on the activities of their governments and the effectiveness of public policies they implement [3]. Related to the economic context of the “knowledge economy”, it must allow civil society to benefit from a real source of information, enabling it to develop new services that will benefit the population [4]. In the U.S., the Obama administration wanted to set an example with the initiative data.gov where you can access and download many databases in all areas. Britain launched its own program [5], soon followed by many countries. In France [6], the central gouvernement is not as much active in this area as the local communities. Several major French cities including Paris [7] have begun to open their databases to the public.
Launched as a voluntary movement from government, open data, however, tends to go beyond their control. The sensational revelations came from Wikileaks which shown that the transparency of government could be exercised at several levels. More profoundly, Wikileaks is acting as the developer of the open data movement: it reveals a general trend resulting in disqualification of political discourse essentially considered as a mystification. The movement is by far more radical than what have undertaken critical theory. The point is no more to deconstruct the discourse to reveal it as deceptive, but to put it aside in order to focus on data alone, open to all, allowing everyone to build his own speech on this basis [8].
One might object that Wikileaks is not open to the data itself because what is made available through it are documents. The objection is probably valid if one thinks in terms of resource. Yet it would miss the essence of what makes the value of Wikileaks, whose modus operandi is not to make public some particularly embarrassing documents, but to give to the public a mass of documents which, because of the amount of information involved, behaves like a database in which everyone is called to dig according to their own interests [9]. This point clearly explains why collaboration was needed between Wikileaks and various newspapers and magazines (New York Times, Le Monde, El Pais) : the organization gives access to information and the press gives it editorial value.
In science, the same movement is on the agenda. There is a general movement prompting researchers to provide open access to the data on which they build their argumentation in scientific publications. If this idea seems obvious at first sight and in theory, its implementation raises a host of practical problems in many disciplines. The open data movement is going further and reverses the perspective: now, data tend to produce their own theory. It’s quite the meaning of the initiative Culturomics [10] that provides researchers and the public a basic lexicometric interface on the entire corpus of books scanned in Google books. The article published in Science by several researchers [11] about this new service is quite significant: it bears no strong theory, any proposal or specific assumptions: It shows several types of interactions with a database that delivers statistical answers to a set of spontaneous questions. As in the case of open data in the political and journalistic areas, for the scientific one, the equivalent movement gives a secondary importance to discourse and theory and places mere data at the center of the picture.
It is certainly striking to see how this triumph of data and the dissolution of the discourse that can be seen in these initiatives, comes at the same time than its exact opposite: the triumph of the discourse in its narrative mode. The story telling phenomenon that touches all sectors too, is the exact opposite of open data. Well documented by the famous book written by Christian Salmon [12], the technique of story telling aims at causing the adhesion of one for whom it is deployed and at the same time anesthetizing his critical sense. Seduction rather than accuracy is here set up as a cardinal virtue.
Between the open data on one side and the story telling on the other, it is ultimately a certain rhetorical tradition that seems doomed to extinction by quartering: built on a strong collusion and an almost inseparable mixture of factual elements (the inventio) on one hand, and the layout that organize them (the dispositio) on the other, the rhetoric was a way to build a shared representation, socialized of the world. Its weakening means a radical disjunction between the “reality” on one side, placed on the side of its independent factual representation or perception, and social relations on the other firmly placed under the sign of seductive and hypnotic artifice. This trend can be seen as congruent with the weakening of institutions and all forms of mediation, a phenomenon quite characteristic of our time.
It is also possible that the dissolution is only a time of transition and the fall of a rhetorical order linked to an outdated mode of organization of society and not the disappearance of every possible rhetoric. Three examples can be cited to illustrate how a “renaissance of rhetoric” may be trying to break, pushing with the reconstruction of a new discourse built on new foundations.
The first example is the work that is emerging on the concept of data visualization. While reference books are starting to be published on the subject, that great blogs are the latest research in this area [13], or as digital artists to explore the many possibilities [14], many observers are beginning to evoke the concept of “data journalism” whose job is largely to develop both modes of representation, data visualization, as well as interfaces to access data that are satisfactory to the public [15]. From one point of view, all these works that are now together under the banner of “data curation” can be seen as building a new kind of rhetoric.
Secondly, the issue of data representation in science shows similar recent developments. The triumph of the data at the expense of interpretive discourse could be similar in the field of humanities and social sciences to the domination of social “naturalist”, “objectifying” science over the humanities gathered under the banner of hermeneutics. It is precisely from a consideration of the representation of data that the contribution of humanities and their singular concern of the subjective representation of the actors is reintroduced in the new context. This was brilliantly demonstrated by Johanna Drucker in a conference organized by the MIT [16] when she showed how, in history, one can design modes of representation of data that do justice to the subjective perception that players could have of the situation described. The triumph of the data may be accompanied by a somewhat paradoxical revival of semiotics, and finally, hermeneutics, via the issue of representation which is far from obvious. From another point of view, the sociologist Dominique Cardon poses at the open data movement as a whole a similar question [17]: the availability of data does not resolve the question of the nature of the data that will be exploited, and ultimately the focus that will be favored by the analysis: the individual or aggregate? Drucker and Cardon said almost the same thing: data depends on the perspective for which they are constructed, aggregated, represented.
Finally, we can conclude this overview by reporting a new service, launched in recent days in public beta. Significantly called Storify [18], this online service allows any user to gather and organize various documents from the Internet and social networking platforms in order to build a narrative “story”. The reader is a new rhapsode, he is encouraged to “sew” the data together to produce a text frame from the myriad of available documents. The entire Web in its various components, is mobilized as a database, but it is in the prospect of widespread production of discourse open to all and carried on under conditions very different from what was known far. The mere fact that such an offer exists is perhaps the premonitory sign of a return to discourse after a short period of reduction to the data. Time will tell whether Storify meets its purpose.
Anyway, the open data movement cannot escape the question faced by all innovations: thequestion of its adoption by the majority [19]. In other words, the question if data are fed through the activity of symbolic production users, if they make sense to them and help them articulate a discursive activity without which there is no possible society.
Notes
[1] Cavazza, Fred. “Du contenu roi aux données reines.” FredCavazza.net, Juillet 19, 2010. http://www.fredcavazza.net/2010/07/….
[2] O’Reilly, Tim. “Government As a Platform.” Dans Open Governement. Lathrope, Daniel et Ruma Laurel (éds.). O’Reilly Media, Inc., 2010. http://opengovernment.labs.oreilly.com/.
[3] Cardon, Dominique. “En finir avec le culte du secret et de la raison d’Etat.” Le Monde, Décembre 3, 2010. http://www.lemonde.fr/idees/article….
[4] LiberTIC. « Pourquoi la réutilisation des données publiques à des fins commerciales doit-elle être gratuite ? » InternetActu.net, mars 9, 2011. http://www.internetactu.net/2011/03…
[5] « L’open data n’est plus une chimère » : entretien avec Nigel Shadbolt,” RSLN, Février 24, 2011, http://www.rslnmag.fr/blog/2011/2/2….
[6] « La France de l’Open Data : où en est-on ? » Regards sur le numérique, mars 21, 2011. http://www.rslnmag.fr/blog/2011/3/2…
[7] Thierry Noisette, “Open Data : les données publiques de Paris ouvertes sous licence libre by-sa,” L’esprit libre, Décembre 19, 2010, http://www.zdnet.fr/blogs/l-esprit-….
[8] Fabrice Epelboin, “Bastille Day à #AmnDawla en Egypte : la révolution par l’open data radical,” ReadWriteWeb French edition, Mars 6, 2011, http://fr.readwriteweb.com/2011/03/…
[9] Kayser-Brill, Nicolas. “StateLogs : Wikileaks begins to reveal 250,000 diplomatic files.” OWNI, Digital Journalism, Novembre 27, 2010. http://owni.fr/2010/11/27/wikileaks….
[10] Guillaud, Hubert. “Culturomics : Comprendre les “lois” de la culture - La Feuille – Blog LeMonde.fr.” La Feuille, Décembre 17, 2010. http://lafeuille.blog.lemonde.fr/20….
[11] Michel, Jean-Baptiste, Yuan K Shen, Aviva P Aiden, Adrian Veres, Matthew K Gray, The Google Books Team, Joseph P Pickett, et al. “Quantitative Analysis of Culture Using Millions of Digitized Books.” Science (Décembre 2010). http://dx.doi.org/10.1126/science.1….
[12] Salmon, Christian. Storytelling : La machine à fabriquer des histoires et à formater les esprits. Editions La Découverte, 2008.
[13] Andrew Vande Moere. « Information Aesthetics », s. d. http://infosthetics.com/.
[14] Harris, Jonathan, et Sep Kamvar. We Feel Fine : An Almanac of Human Emotion. 1er éd. Scribner Book Company, 2009.
[15] Bradshaw, Paul. « How to be a data journalist ». The Guardian, octobre 1, 2010. http://www.guardian.co.uk/news/data….
[16] “Johanna Drucker : Humanistic Approaches to the Graphical Expression of Interpretation”. MIT World, 2010. http://18.9.60.136/video/796.
[17] Dominique Cardon, “Zoomer ou dézoomer ? Les enjeux politiques des données ouvertes » Article »,” OWNI, Digital Journalism, Février 21, 2011, http://owni.fr/2011/02/21/zoomer-ou….
[18] Gahran, Amy. « Storify launches public beta : Curation is a core news skill ». Knight Digital Media Center, avril 28, 2011. http://www.knightdigitalmediacenter…
[19] Michael Gurstein, “Open data : Empowering the empowered or effective data use for everyone ?,” First Monday 16, no. 2 (Février 7, 2011), http://www.uic.edu/htbin/cgiwrap/bi….
