Manuel Carrión: Diagrammatica Question 2

[This is the second of five questions about the the Diagrammatica ProjectThe questions were provided by Manuel Antonio Carrión Lira of the Design School at the University of Valparaiso, Chile (  An abstract of Manuel's research and the full set of questions can be viewed here.]

  • Manuel Carrión: Do you think that there are some basic elements that make a diagram what it is? What makes a good diagram?

At the basic level, I see a diagram as a collection of language and lines.  The words present elements or fields of significance, while lines form the vectors or trails that connect them.  Lines posit relationships between words, and therefore do the work of traditional grammatical structures.  These diagrammatic relations unfold spatially and in multiple directions rather than sequentially in a single line of text. All writing is diagrammatic and even blocks of text present a highly specialized type of diagram.  Literacy, learning to read, teaches us how the line of connection runs (from word to word, from the end of one line to the next, from the top of the page to the bottom).  This linear direction is not visually represented because it is learned as a practice.  In other diagrams, however, the lines are represented.  Lines show the reader which elements are connected, which elements are included in the field of another element, and which elements are excluded from those fields.

Diagrammatica is an exploration of the multiple ways in which viewers can read these lines and spatial relations.  The design possibilities are infinite.  Lines can lead in any direction and the space can extend as far as necessary.  Nevertheless, I have a personal preference for a certain aesthetic. Among other traits, I appreciate the following:

  • Sparsity and simplicity of design.  Diagrams with fewer elements and less lines often produce a greater impact than diagrams filled with texts and complicated networks of connection.
  • A spatial layout that makes significant use of the 2D plane.  In some diagrams only the lines of connection signify meaningful relationships.  In others, placement of elements is also significant: whether an element is above or below another element, whether two elements are nearby or distant, whether one element is within a larger element, etc.
  • Innovative use of visual contrasts as coded differences.  Examples include bolded text vs. normal text; the use of single lines, double lines, and dotted lines to imply different types of relation; the assignment of elements to different shapes depending upon function; etc.

I also appreciate diagrams that played an important role in intellectual or cultural history, such as diagrams from an influential text or diagrams interpreting the work of a significant theorist.  The particular content or message of a diagram is not necessarily important.  If a diagram represents a new way to communicate spatially, it has offered a significant contribution.

Most importantly, a good diagram is a tool that we can think through.  The diagram is an abstract machine for the production of knowledge and insight.  Does the spatial layout of a diagram ask new questions?  Does it provide new answers?  The diagram displays a set of relations: element A relates to element B.  The production of knowledge occurs as a reader fills in the gaps: element A relates to element B because of XYZ.  Semiotically, meaning arises from the transformation of meaning.  And diagrams are a prompt for the transformation of meaning: they ask readers to translate their spatial structure into language.  Readers literally read between the lines; they describe and expand those lines with additional terms and concept.  The reader of a diagram posits that this line means this, or this location means that.  Traditional semiotics had been criticized for relying too heavily on language as a model.  So how can we discuss systems of sign relations (including language) without resorting to language?  Diagrams offer one such method.  They transform linguistics structures into spatial relationships and visual patterns