Conlusion
Visual Grammar

[Contents] [Previous: Chapter 5 Grammatical Research]
[Next: Bibliography]

Images meet Noam Chomsky's conditions for language (correspondence between deep and surface structure, discrete infinity). This justifies an approach to visual images that can be called linguistic. This does not mean that the same grammatical rules apply to images and texts alike. Images make use of different means of articulation that allow for entirely different grammatical structures, although there are some similarities as well.

When looking at the human brain and the way it processes (visual) language, a few things come to fore that should be taken into account when constructing a visual grammar. Most important of these is modular representation. The human brain has a few areas (modules) dedicated to the processing of different aspects of images (articulation, syntax and lexicon) and visual grammar should be modelled accordingly. The information of all these modules is ultimately passed to the conceptual module where several 'threads' of interpretation converge or possibly collide. This leads to a final interpretation that can take into account any or all these threads.

The brain modules specialize in the recognition of several grammatical structures and I have tried to capture the most basic of these structure into grammatical rules. These grammatical rules can be used to analyse visual representations, their flexibility able to cope with many different and unique visual representations. As such I feel that the theoretical framework of the visual grammar is quite successful. Furthermore it circumvents the problem of Kress and Van Leeuwen's grammar: it does not describe fixed categories of structures, even though it regards some structures as more 'typical' than others.

I did not design visual grammar in order to model meaning. Visual grammar is a tool in the employ of the researcher in the domain of visual graphics. One of the fields of research that this grammar can be used for and one that I find particularly interesting is the social stratification of (visual) language. The social stratification of language follows from the theories of Mikhail Bakhtin and Julia Kristeva. They claim that different social groups and contexts call for the use of different languages or sub-languages. Kress and Van Leeuwen also contribute to this with their notion of visual modality. These visual sub-languages are assumed to stress different grammatical structures, and with part of the research presented in this thesis is intended to investigate social stratification of visual language.

To put this grammar to the test I have analysed 406 posters. I used the data to test for correspondence between the grammatical rules. With one exception the correspondences were not very high. Also, all the correspondences proved to be positive, bolstering the claim that grammatical structures can sometimes be contradicting be it by intention or mistake. This is a positive result that strengthens the theoretical framework of my visual grammar.

The second objective was to research the assumption that language is socially stratified and grammatical rules play an important role in this stratification. The result do not support this claim unquestionably. There are certain groups of posters that are indeed strongly stratified from the rest, most notably Communist posters and artistic posters. For other groups significant differences do exists but are not as strong. Still, I feel that result of the research is such that further research with more homogeneous groups of posters should prove interesting and can be use to bolster the notion of visual sub-languages. Also, it might be interesting to compare sets of posters that are grammatically distinct. Perhaps these sets can also be classified as genres on their own. In fact I did something similar when I separated Communist posters from political progressive posters intuitively.

All in all, further research into this matter is recommended. More data should be gathered to include more posters from different regions and timeframes.

The aim of this thesis is to be a contribution to the ongoing development of the theory of the image. I drew on many sources from various intellectual traditions and combined these into a single framework for visual grammar. I feel that this framework reflects contemporary developments of the fields of semiotics, linguistics and cognitive science. The grammar constructed within this framework has proven to be successful in the analysis of visual language even though the results of the research into the social stratification of visual language were not entirely satisfactory.

[Contents] [Previous: Chapter 5 Grammatical Research]
[Next: Bibliography]

http://www.jorisdormans.nl/article.php?ref=visualgrammarc