In her text “Design for Information”, Isabel Meirelles begins her survey of design paradigms commonly used in the visual presentation of information with the arborescent model of hierarchical trees (Chapter 1; Hierarchical Structures: Trees). In relation to such structures, we are told, in the section titled “Representation”:
“Looking at hierarchical structures over time, it becomes apparent that ordered datasets are represented visually in two basic graphical forms, which sometimes are also combined: stacked and nested schemas [Meirelles, pg. 18].”
Gilles Deleuze and Felix Guattari, in their critique of hierarchical structuralism, contrast the hierarchical structure of the arborescent model with a contrasting, non-hierarchical organizational principle: that of the rhizome.
Thus, the first question that comes to my mind at the very beginning of “Design for Information’ is: Does a rhizome-based model even have a place in information design? If so, what would this place be? The answers to these questions may come quickly, by following what Meirelles has next to say about visual representations:
“In visual representations, the use of space is always schematic, independent of whether depictions of elements are direct or metaphorical. Spatial encoding is central to how we construct visualization, in that the geometric properties and spatial relations in the representation — the topology — will stand for properties and relationships in the source domain [Meirelles, pg. 20].”
In relation to the hierarchical nature of arborescent structures, Deleuze and Guattari contend:
“The tree and root inspire a sad image of thought that is forever imitating the multiple on the basis of a centered or segmented higher unity… Arborescent systems are hierarchical systems with centers of signifiance and subjectification, central automata like organized memories. In the corresponding models, an element only receives information from a higher unit, and only receives a subjective affection along preestablished paths. This is evident in current problems in information science and computer science, which still cling to the oldest modes of thought in that they grant all power to a memory or central organ. Pierre Rosenstiehl and Jean Petitot, in a fine article denouncing "the imagery of command trees" (centered systems or hierarchical structures), note that "accepting the primacy of hierarchical structures amounts to giving arborescent structures privileged status.... The arborescent form admits of topological explanation.... In a hierarchical system, an individual has only one active neighbor, his or her hierarchical superior....The channels of transmission are preestablished: the arborescent system preexists the individual, who is integrated into it at an allotted place" (signifiance and subjectification).”
“To these centered systems, the authors contrast acentered systems, finite networks of automata in which communication runs from any neighbor to any other, the stems or channels do not preexist, and all individuals are interchangeable, defined only by their state at a given moment — such that the local operations are coordinated and the final, global result synchronized without a central agency. Transduction of intensive states replaces topology, and "the graph regulating the circulation of information is in a way the opposite of the hierarchical graph....There is no reason for the graph to be a tree" (we have been calling this kind of graph a map) [Deleuze & Guattari, pgs. 15–16].”
Even in the 1960s, then, it appears that questions were being raised regarding the universal applicability of the arborescent model to computer-based organizational principles. Whether or not a rhizome-based model would have some form of applicability to information design, however, remains to be established. In contrast to the arborescent model of hierarchical tress, Deleuze and Guattari advance the model of the rhizome:
“Unlike the tree, the rhizome is not the object of reproduction: neither external reproduction as image-tree nor internal reproduction as tree-structure. The rhizome is an antigenealogy. It is a short-term memory, or antimemory. The rhizome operates by variation, expansion, conquest, capture, offshoots. Unlike the graphic arts, drawing, or photography, unlike tracings, the rhizome pertains to a map that must be produced, constructed, a map that is always detachable, connectable, reversible, modifiable, and has multiple entryways and exits and its own lines of flight. It is tracings that must be put on the map, not the opposite. In contrast to centered (even polycentric) systems with hierarchical modes of communication and preestablished paths, the rhizome is an acentered, nonhierarchical, nonsignifying system without a General and without an organizing memory or central automaton, defined solely by a circulation of states [Deleuze & Guattari, pg. 20].”
Interestingly, Meirelles does note the exceptional nature of information design when maps are involved:
“In cartography and geo-informatics, data are divided into spatial phenomena (geography) and nonspatial phenomena, called thematic data [Meirelles, pg. 124].”
“Different from all other visualizations, thematic maps are not concerned with conceptualizing the topological structure, which is provided by the geographic information in the form of the base map. All other visualizations in this book require the crucial step of deciding on the most appropriate topological structure, especially with regard to visual representations of abstract data [Meirelles, pg. 128].”
As we shall see, the idea of thematic data will again gain prominence in Meirelles’ analysis — under the designation of “nominal data” — but for the moment let us again consider Meirelles’ reference to schemata, and do so within the context of space and time; for fully three chapters in “Design for Information” deal with such matters: Chapter 3 (Temporal Structures: Timelines and Flows), Chapter 4 (Spatial Structures: Maps), and Chapter 5 (Spatio-Temporal Structures).
Schema hold a very special place in the philosophy of Immanuel Kant:
“The synthesis and the schema are always the forming of a correspondence between, on the one hand conceptual determinations, and on the other spatio-temporal determinations. What defines the synthesis as distinct from the schema? The synthesis is an act of the imagination which operates here and now; there is no synthesis if it is not an operation of your imagination that you do here and now. For example, here and now, you see a diversity; or else here and now you see an organization of space and time. You will recall that this space and this time are not yet determined: there is something in space and time. A synthesis must yet be effected which will give you a certain space and time, in such a way that you carry out a sort of isolation: if you say "that is a table", you have carried out a synthesis of space and time in conformity with a concept.
“So, in the synthesis, I have indeed effected a correspondence between a determination of space and time and a conceptual determination, the determination of space and time being carried out by the synthesis of apprehension and reproduction, and the conceptual determination referring to the form of the any-object-whatever in so far as this form of object will be determined by the diversity upon which I effect the synthesis.
“The schema. Put yourself in the reverse situation.
“The schema: you have a concept, and the problem is to determine the spatio-temporal relation which corresponds to this concept. The synthesis is just the opposite, it's this: you carry out a spatio-temporal operation and you specify the concept according to this determination. So the operation of the synthesis, valid here and now, will correspond with, in the other direction, the determination of the schema, valid at all times. There you have a concept and you are looking for the spatio-temporal determination which is likely to correspond to it.
“Given a concept, how can I produce it in intuition? Which is to say in space and in time, an object conforming to the concept. Producing in space and time, that is the operation of the schema. In other words, the schema does not refer to a rule of recognition, but refers to a rule of production. The synthesis of a house is the rule of recognition according to which I say "it's a house". You say "it's a house" in front of very different things. You effect a synthesis of the given such that you relate them to the any-object-whatever "it's a house". The schema of the house is very different, it is not a rule of recognition over random diversities. The schema of the house is a rule of production, namely that you can give yourself a concept of house. For example I can take a functional definition: house = apparatus made for sheltering men, this doesn't yet give us a rule of production. The schema of the house is what allows you to produce it in experience, in space and in time, something, objects conforming to the concept. But that definition does not get out of the concept; you can turn the concept around all you like in all directions, apparatus made for sheltering men, you will not draw rules of production from it, the rules of construction of the house. If you have the rule of production you have a schema.
“Here you can see in what respect the productive imagination is more profound than the reproductive imagination. The reproductive imagination is when you can imagine circles, concrete circles; you can imagine a circle drawn on a blackboard with red chalk, you can imagine a plate... all that is the reproductive imagination. But the circumference that allows you to make rounds, which allows you to round things, which is to say to produce in experience something conforming to the concept of circle, that doesn't depend on the concept of circle, that doesn't flow from the concept of circle, it's a schema, and that is the act of productive imagination.
“You can see why Kant feels the need to discover a domain of the productive imagination distinct from the simply empirical or reproductive imagination. You can see the difference between a schema and a synthesis, if you have understood that I have finished with my first point: what the difference was between the two fundamental acts, within the context of knowledge: the schematism and the synthesis.
“The schematism is not a case of reflective judgement, it is a dimension of determining judgement [Deleuze, COURS VINCENNES, 04/04/1978].”
All of which is not to point out any contradiction with the information that Meirelles presents; but rather, to tease out some important correlations that might otherwise be glossed over: that schema are productive in nature (Meirelles: “In visual representations, the use of space is always schematic, independent of whether depictions of elements are direct or metaphorical. Spatial encoding is central to how we construct visualization, in that the geometric properties and spatial relations in the representation — the topology — will stand for properties and relationships in the source domain”); and that non-hierarchical approaches are not topological in nature but instead are more closely related to maps (Meirelles: “Different from all other visualizations, thematic maps are not concerned with conceptualizing the topological structure, which is provided by the geographic information in the form of the base map. All other visualizations in this book require the crucial step of deciding on the most appropriate topological structure, especially with regard to visual representations of abstract data.”). However, it must be noted that between these two positions lies an area of prominent concern which needs to be explored more fully: the temporal.
When we first encounter temporality in “Design for Information” it quite quickly reverts to the context of philosophy:
“Time is an abstract concept and, thus, not inherently visual. Much of the terminology we use for time is based on our concrete experience of space and of the physical environment [Meirelles, pg. 83].”
“And here, Augustine pulls off an audacious coup de théâtre: He locates his nonmetric measure in our memory. The true measure of time is an inner measure. Centuries later, Henri Bergson would also contrast metric time with the time of our consciousness or ‘inner durée.’ [Meirelles, pg. 85].”
I’m not going to say that Meirelles is wrong in this, only that some very important considerations are being left out of the conversation here. Yes, Henri Bergson (author of such texts as “Matter and Memory”) did indeed say that we apprehend the temporal through memory; but the point he was making wasn’t that time is a product of memory — as might be inferred from the position Meirelles appears to present — but rather that we can only distinguish between different kinds of things because, through memory, we have access to the lived histories that distinguish each as different from any other. In this context, Bergson had some very insightful observations to make concerning the nature of the temporal:
"The important thing here is that the decomposition of the composite reveals to us two types of multiplicity. One is represented by space... It is a multiplicity of exteriority, of simultaneity, of juxtaposition, of order, of quantitative differentiation, of difference in degree; it is a numerical multiplicity, discontinuous and actual. The other type of multiplicity appears in pure duration: It is an internal multiplicity of succession, of fusion, of organization, of heterogeneity, of qualitative discrimination, or of difference in kind; it is a virtual and continuous multiplicity that cannot be reduced to numbers... Everything is actual in a numerical multiplicity; everything is not "realized," but everything there is actual. There are no relationships other than those between actuals, and no differences other than those in degree. On the other hand, a nonnumerical multiplicity by which duration or subjectivity is defined, plunges into another dimension, which is no longer spatial and is purely temporal: It moves from the virtual to its actualization, it actualizes itself by creating lines of differentiation that correspond to its differences in kind. A multiplicity of this kind has, essentially, the three properties of continuity, heterogeneity, and simplicity [Deleuze, "Bergsonism", pgs 38 - 43].”
Whatever point I am trying to make here might at first seem very far away from the ideas presented by Isabel Meirelles in her text “Design for Information”; but in fact, by returning to Bergson we are picking up exactly where Meirelles leaves off in her final chapter, “Textual Structures”:
“Objects, names, and concepts are examples of nominal data. We distinguish nominal datum on the basis of quality: A is different from B. The questions we ask about nominal data are what and where. Nominal data have no implicit quantitative relationship or inherent ordering, and questions such as how much don't apply.”
“Nominal data are considered qualitative and are rarely visualized without correlating to other kinds of data. [Meirelles, pg. 187].”
It is quite obvious that the distinct Meirelles makes between quantitative data and nominal data is in fact the distinction between spatial and temporal phenomena. Meirelles also makes a point of including concepts within the field of nominal data; and this is very interesting in that through Jean-Paul Sartre we find a direct connection between concepts and visual schema:
"It is characteristic of the schema that it is intermediate between the image and the sign. Its matter demands to be deciphered. It aims only to present relations. By itself it is nothing."
"Through these black lines we aim not just at a silhouette, we aim at a complete man, we concentrate in them all his qualities without differentiation: the schema is full to bursting. To tell the truth, these qualities are not represented; in the proper sense, the black features do not represent anything but some relations of structure and attitude. But it is enough of a rudiment of representation for all the knowledge to be weighted down there, thus giving a kind of depth to this flat figure."
"The majority of schematic drawings are read in a definite sense. Eye movements organize the perception, carve out the spatial environment, determine the fields of force, transform the lines into vectors [Sartre, pgs. 29–30]."
"But it is very evident that the comprehension is realized in and by the construction. The structure of the concept to be comprehended serves as a rule for the elaboration of the schema and one becomes conscious of this rule by the very fact of applying it. So that, once the schema is constructed, there remains nothing more to comprehend."
“Comprehension is not pure reproduction of the signification. It is an act. This act aims at making present a certain object and this object is, in general, a truth of judgment or a conceptual structure.”
"Next, and especially, it is enough to produce in oneself one of these schemas and as observer to note that they do not at all have this role of sign and representative. Without doubt, there is in the schema a representative: it is the affective-motor analogon through which we apprehend the shape and its color. But the schema itself is an analogon no more: it is itself an object having a sense... We reach here the true sense of the symbolic schema: this schema is the object of our thought giving itself to our consciousness. Thus the function of the schema as such is not at all to aid comprehension; it functions neither as expression nor as support nor as exemplification. I willingly say, using an indispensable neologism, that the role of the schema is as presentifier [Sartre, pgs. 103–105].”
Again, we are not at odds with the gist of the thesis that Meirelles presents to us, for throughout “Design for Information” we find an emphasis upon aligning design principles with what is known of the functional neurology which underlies vision. However, with Kant, and Sartre, and Deleuze, we find an emphasis placed upon the PRODUCTIVE nature nature of visual schema; and we see, through Bergson, that in this we are dealing with temporal differences-in-kind.
“A rule of production is solely a determination of space or of time conforming to the concept. Take another case. You make yourself a concept of a lion; you can define it by genus and specific difference. You can define it in this way: big animal, mammal, with a mane, growling. You make a concept. You can also make yourself lion images: a small lion, a big lion, a desert lion, a mountain lion; you have your lion images. What would the schema of a lion be? I would say in this case, not in all cases, that the concept is the determination of the species, or its the determination by genus and specific differences.
“We speak both of an animal's territory and an animal's domain, with its paths, with the traces that it leaves in its domain, with the times that it uses a particular path, all that is a spatio-temporal dynamism that you will not draw from the concept. I am not going to draw from the concept of a lion the way it inhabits space and time. From one tooth you can draw something of a mode of living: this is a carnivore. But really the spatio-temporal dynamism of an animal, that is really - I can't say its rule of production - but it's something productive, it's the way in which it produces a spatio-temporal domain in experience in conformity with its own concept.
“The lion is Kantian, all the animals are Kantian[Deleuze, COURS VINCENNES, 04/04/1978].”
We can arrive at concepts through schema, because schema are productive — schema make present for us things not otherwise there, rather than simply representing them: schema determine things, rather than represent them by way of reflection. In this, we find that schema present a functional aspect of information design — as with a mapping which reveals what exists (as the “what and where” of when lions roam, as nominal data), rather than a topology that assigns hierarchical meaning as relative height (or some other ordinal variation).
Where, then, does this leave us with the initial questions we framed concerning rhizomes? What aspects of information design might be considered rhizome-based?
In the real world, dealing with literally thousands of data points that needed to be compiled for their sales teams, the information design team at Conde Nast found themselves throwing out the structural hierarchy that came with the data sets they were given to compile; and instead, they created their own categorizations based upon “what” and “behavior”:
In this, they were searching for a way to make the data they had on hand sensible for their sales teams: they needed to give a functionality to that data, such that it could inform their sales teams in ways which would make them more productive. To achieve this, they needed to create schema for this data that would allow their sales teams to determine how best to proceed, rather than simply reflect upon situations already past. The iterative processes through which these productive schema were created were themselves rhizome-based; and it is here, I would suggest, that a rhizome inspired methodology would best be applied to information design — as a mechanism for moving from data-as-maps to thematic schema of qualitative, nominal categorizations that allow new concepts to present themselves — since we can get concepts from schema, but, schema do not come out of concepts. Keynote to this is determining HOW schema might be used to produce specific differences-in-kind from the data sets that might act as underlying maps that could show what is acting how, and where that happens as distinctly distinguishing from other behaviors.
In short, we find that rhizome-based information design appears integral to interactive technologies that are productive in nature, rather than simply being reproductions of things already recognized. In this, a more basic concept of what constitutes a schema is necessary — as is a distinction between the synthetic nature of concepts, and the productive nature of schema.
Below: schematic drawing for JavaScript expressions [Crockford, pg. 127].
References Cited:
Crockford, Douglas. JavaScript: The Good Parts. Sebastopol, Calif. USA, 2008.
Deleuze, Gilles, and Guattari Félix. A Thousand Plateaus. Univ. of Minnesota P., 1994.
Deleuze, Gilles. Bergsonism. New York, NY: Urzone Books, Inc., 1988.
Deleuze, Gilles. Lectures on Kant; COURS VINCENNES, 04/04/1978. Translated by Melissa McMahon. IMHO, Editions. Webdeleuze, https://www.webdeleuze.com/sommaire.
Meirelles, Isabel. Design for Information : an Introduction to the Histories, Theories, and Best Practices Behind Effective Information Visualizations . Beverly, MA: Rockport Publishers, 2013. Print.
Sartre, Jean-Paul. The Imaginary: A phenomenological psychology of the imagination. Translated by Jonathan Webber; Translation copyright 2004 by Routledge; reprinted in 2006. Published by Routledge New York NY
