Bertin and organising Information

If you have been following these posts you may well have realised that I'm fascinated by the ability of the diagram to communicate information visually. A key book when it comes to an understanding of diagrams is the 1967 text 'The Semiology of Graphics' by Jacques Bertin, but because most of my readers will be English speaking the first and often only diagram theorist and text that people tend look at is Edward Tufte and his 'Visual Display of Quantitative Information', which is a wonderful book if you have never come across it, but Bertin was there first and his book set the standard for our thinking on how diagramming worked. 

Bertin: A separation of semantic elements 

Bertin had a particular formalist understanding of diagrams, and he began each analysis of a graphic by looking at two critical issues: 1. the visual mark or what he called 'implantation' (E.g. a dot on a scatter plot graph) and its position on the plane (where the dot is placed relative to the x-axis and y-axis that can be seen to bisect any flat image). 2. The idea that lay behind or unified that was being diagrammed, which he called the 'invariant'. Bertin called the variable features that constitute the invariant 'components' and the 'components' themselves were made up of different parts called 'elements'.

Bertin has some wonderful diagrams in his seminal text. For instance he sets out the rules for how to construct a graphic diagram. He states that every mark in the image is significant, as well as any absence of a mark. (This is rather like a musician pointing out that silence is what makes sound intelligible). He then uses a diagram to illustrate how the six visual variables that we can apply can work: size, value, pattern, colour, orientation, and shape. (Nb value in this context is tone, or tonal value).

The six visual variables

What I thought was interesting about his diagram was that it looked as if it still belonged to the 1960s. The 'touch' or 'feel' of the layout, the visual weight given to each element reminded me of many book covers and posters of the time.

Typical 1960s design

So was Bertin analysing how diagrams and graphic images worked, or was he simply reflecting the visual mood of the times? It was a time when in England and I suspect across Europe and the USA, many of the tenets of Bauhaus teaching were central to how design and fine art were taught. We are all whether we like it or not products of our time, even so, when Bertin highlights some important concepts that need to be born in mind when designing information graphics, we can still look at them and think about how they can be used. 

• Relative size is the most widely useful and easily perceived visual variable, no matter what information is visualised.
• Size and planar position are the only variables that can accurately communicate quantitative information.
• For communicating order, tone or value (from light to dark) is far superior to colour (from one colour to another).
• Colour, orientation, and shape function best when creating visual associations and building visual groups.
• Ideally, the graphic should balance white space with 5%-10% ink.
• Focusing on fewer components in a graph aids memorisation and allows freedom to use more exotic graphic forms.

Bertin then goes on to look at how you evaluate the image in terms of its ability to communicate.

• External: From all the information that exists in the world, what information is represented in the image? 
• Internal: What visual variables are used to represent the components in the graphic? 
• Relationships: How are the components related? What questions can I ask of this image? What can I learn? 

• He then describes the three types of questions he thinks you should be asking.

• Elementary: An elementary question focuses on a specific element of the graphic. For instance by using the two axes to pinpoint a specific thing, where a dot is for instance, or to focus on any annotation, or a work's title. 
• Intermediate: An intermediate question focuses on a group of elements and usually illuminates a trend, idea or implication. 
• Overall: The overall question seeks an answer to the general message of the image. 

Left to right: Elementary, Intermediate, and Overall Reading Levels

Bertin analysing how a diagram works

You could relate these stages of evaluation to those used by Edmund Burke Feldman in his model of art criticism. However I'm attempting to use Bertin's approach to critiquing a draft novel that I have been sent to appraise. It's an interesting task and one that needs some sort of structure to help build a critique, so I have decided that Bertin's approach might help. It is interesting to see how it translates, for instance, 'External': the novel could be about anything, but it isn't, it is a sort of pilgrimage and is set in the 1930s. 'Internal' the visual variables seem to be very much to do with setting, landscapes, and events such as floods etc. 'Relationships': These seem to be mainly about people and settings and how they interact. 'Elementary': seems to translate to individual words in sentences, are they appropriate, do they feel right and offer the right texture for the internal mouth?  'Intermediate' can I see where an idea is going? 'Overall', where does it all head to, what is the message? 

Every now and again it is useful to see how a particular way of evaluating things can be applied to something that it wasn't designed to evaluate. In science this would be seen as wilfully wrong, as measuring instruments tend to be very specific and to measure the force of magnetism by using a wind scale would be regarded as foolish. But art isn't science and as a way of thinking it can reconcile opposites and embrace paradox. However it is also for fine art students very useful to think about the tools of the graphic designer as they are often about clear communication and as such can be very useful to any fine artist worried about whether or not their idea is getting through to others. 

References

Few, S (2009) Now You See It: Simple Visualisation Techniques for Quantitative Analysis London: Analytics Press

WSJ Guide to Information Graphics by Dona M. Wong

Visualize This by Nathan Yau

The Visual Miscellaneum by David McCandless

From Visual Miscellaneum by David McCandless

See also:

Feldman's model of art criticism

Drawing as data visualisation

Jorinde Voigt: Drawing as abstraction and notation

Kurt Vonnegut diagrams the shapes of stories

Diagram as art form

Minjeong An