Critical reflection on a data viz example proposed on UCC module's lecture

The Flow toward Europe – a breathtaking visualization of Helsinki based agency Lucify deserves to be viewed closely. First, because of a hight level of a technical feat, second, because of the twisted effect this feat does.

About technical side first: it is done on Ruby programming language and Node 8.js (runtime environment which allows to use Javascript off line): being ignorant in both of them I can just express my “wow” in front of a such (well working) visual complexity.

There is a nice animation: once image loaded, we can observe the uncountable number of spermatozoids-like things irrigating, silently and with unshakable constancy, our “old Europe” (understood as ECC). If accelerate the speed, the likelihood with spermatozoids will disappear and it will look rather as a salt falling, upside down (following the South-North axis) from the small holes of large non-europeans spaces. The “salt capital” increases in the country of reception doing white shiny bars representing the increasing population of migrants in each country. The crisp contours of those bars create a kind of crispy evidence of the “amount”. We are explained, in a text below, that one white point corresponds to 25 people: along with an impression of incessant moving of those points, the impossible mental computation of each point by 25 returns an impression of an “invasion” of Europe by those white constantly arriving things.

The fond is dark: user's interaction can slightly change the shades so as to distinguish the countries where the people go to (green shade) and those, from where people flee (mauve shade). We can think this choice is dictated by legibility: the dark moving object on the lighter fond, confusing with the border's lines, would be probably not the best option. But this picture is seen as whole and it is difficult since to separate a legibility issue from the sense of the whole. And the impression of a dark unhospital space (both where the situation is “good” and “bad”) dominates.

We are far beyond the statistics: even thought the picture is “based on data published by the UN Refugee Agency (UNHCR)” as the legend informs us, it is “ designed to provide an intuitive grasp of the scale of the problem”. Let's put aside (for a moment) a problematic term “problem”. Let's just realize that the data participates here to create not an “exact picture” but foremost a picture, leaving our intuition to do what it usually does: along with our senses and believes to lead through aesthetic experience and form an impression (“grasp”) of a “problem”. And let's pay attention to a fact that a human tragedy (which I would definitely prefer to the word “problem”) enters in a kind of aesthetic conflict with a perfect formal execution. It is what Sean Cubitt appoint as a data viz's paradox: revolting the human’s catastrophes (climate, poverty, wars) through “beautiful evidence”¹.

It would be quite useless to predict how this picture could be interpreted by the people with different aesthetic and political sensibility but it is quite obvious that it could be grasped in a different way. There is not so question about authors' intention but rather in a fact that each artifact is a cultural one, born and living in a changing and different cultural environments with their particular combinations of aesthetic and political agreements and contradictions. The pretension of Data Viz to be mere truth for everybody is, in itself, the mark that it moved from the domain of facts to that, of aesthetic. The numbers never pretended to be “truth” as this notion is not about numbers but about shared sensibility of what is “true” and what is not.

The Information Visualization, originally, was a part of iterative process of business or scientific analysis of a concrete situation – never an outline of it (“truth” about it). It is still its role in a data mining's process where data viz is not an aim but a tool, free from pretending on an autonomous aesthetical existence. Outside of the boundaries of sciences and in large waters of humanities, data viz morphed into an aesthetic phenomenon participating, in the first line, in what Jaques Rancière calls distribution of sensible,². While this fact seems to be obvious, we continue, so often, to read and hear about data viz as a “truth”³.

The plural sense of the word “insight” (scaled from “grasp' to “idea”) helps somewhat this ambiguity of data viz' status. The (very) often quoted Ben Shneiderman's “the purpose of visualization is insight, not a picture” ⁴ is nothing but a rhetorical sample: there is an insight in a Vermeer's painting, in an X-ray picture of my neck, as well as in a graph representing the annual variation of a Z-company's financical activity.

Nerveless, the question is not about to “take back the control” over this tool escaping his “objective” role: it is about the awareness of this pretension to be merely objective while the only truth which data viz, as a perfect instance of a mass image reveals, is the fact “ that reality is never given (datum), but is always to-be-constructed”⁵.

Risking the redundancy, I would like to conclude that the understanding of data viz as a tool of an aesthetic construction of reality rather than a tool to “deal in truth”⁶ is, in my mind, an unavoidable and urgent step, especially, when we pretend to make a visual overview of the large human phenomenons (which some are empresses to name “problems”).

1. Sean Cubitt, Data Visualization and the Subject of Political Aesthetics pp. 179-190, in: Berry, D., Dieter, M., ed. Postdigital aesthetics, Palgrave Mcmillian, 2015;
2. Jaques Rancière, Le partage du sensible, éditions La fabrique, 1998
3. Alberto Cario, The Truthful Art: Data, Charts, and Maps for Communication, New Riders, 2016
4. Need to confess that even very respectful people fall under the charm of this rhetoric: https://medium.com/@mbostock/a-better-way-to-code-2b1d2876a3a0
5. Sean Cubitt, The uncertenity of the mass image: logistics and behaviors, 4/10/2017:
https://www.tandfonline.com/doi/abs/10.1080/14626268.2017.1378687?scroll=top&needAccess=true&journalCode=ndcr20
6. https://visualmatters.com/dangers-visual-data-manipulation/