CICADA - Cross Feasibility Account - Two-month PhD Secondment
CICADA Textures in text: Mathematics of periodicals in time and space This is a Feasibility Grant proposal to develop a collaboration between German Studies/ Languages, Linguistics and Cultures (Dr Matthew Philpotts) and CICADA and Mathematics (Prof. Paul Glendinning). We are asking to employ a mathematics/computer science PhD student or equivalent for two months to explore the conceptual and methodological implications of a number of different descriptions of periodicals over time. Although the project takes as its starting-point Philpotts’ 1930 Periodicals Project (see next page), the methodologies developed will have impact on any investigation of text and its structure that goes beyond the standard ‘text mining’ approaches. We would expect one or two larger grant proposals to be submitted as a consequence of this work (e.g. Philpotts bid to ERC Starting Researcher Scheme, November 2012).
Over many years periodicals develop their own identity and expectations for the reader. This is much more than a simple sets of subjects, but includes lengths of articles, order of articles, juxtapositions, frequency, font, format, advertising, authors, editors, and so on, all of which influence and are influenced by (multiple) images of the periodical. These different and overlapping parts that make up a periodical are characterised by considerable heterogeneity and variation over time, but this variation is always mediated through the recurrent patterns of individual and specific types of periodical and by the recognisable central identity embodied in their title. This tension between the ‘centrifugal’ and ‘centripetal’ forces at work in these serial publications has been recognised by a number of scholars in the emergent field of periodical studies, some of whom have tentatively explored the usefulness of aspects of chaos theory as a means to make sense of the distinctive properties of periodical publication. But these approaches do not go beyond chaos as a descriptive and partially understood metaphor for the complexity of periodical form. One of the aims of the Philpotts’ 1930 Periodical Project is to see whether this richness of texture in the production, content and evolution of periodicals can be more systematically conceptualised using ideas and tools developed in Mathematics and whether those ideas might in turn help to develop new methodological approaches. So, for example, how might we develop the metaphor of the periodical as fractal, where self-similar structure is revealed at different levels (whole run, annual volume, individual issue, contribution, page), not only to describe information and geometry (the fractal dimension of a specific periodical), but also to develop periodical typologies and to understand how some periodicals are more successful than others at reconciling the forces of chaos and order which shape them?
This proposal is to investigate several possible approaches to the mathematical and computer sciences aspect of this project. In each case we are looking for a description of the methodology, a sense of the information required to implement the methodology, and an estimate of the type of results that could be obtained and how they might contribute to a humanities based appraisal of the periodical. We would expect recent work of Dr Muldoon, Dr Shapiro and Dr Brooks in CICADA to be relevant to some of the approaches taken.
There are three areas we would hope to explore in detail during the two months of this part of the project, and we would hope to obtain sufficient detail that the results could be slotted into proposals for further funding.
- Methodology of fractals and image processing. There are relatively standard methods available to describe texture and structure in image processing (e.g. Fractal Geometry in Digital Imaging by Martin Turner, Jonathan Blackledge and Patrick Andrews, 1998). We would like to know how these could be implemented on data from periodicals and what the measures might tell us.
- Methodology of virus evolution/genetic algorithms. By viewing the periodical as having an array of properties, could we understand spatial and temporal evolution using the methods currently developed to describe genetic evolution. What would this tell us about the periodicals?
- Shapiro Twitter methodology: Shapiro and co-workers are looking at ways of monitoring and following themes and impact on twitter feeds. There may well be similar themes within periodicals that would make it possible to use these new ideas.
Please contact Paul Glendinning (Paul.Glendinning at manchester.ac.uk) if you are interested in this project.
Funding will be 25,854 GBP per annum pro rata.
The person appointed will need to be in post by 1 February 2012 at the very latest.
The last date for expressions of interest is Monday 9 January 2012