What can typology explain that configuration can not?

Paper in proceeding, 2015

This paper aims to contribute to a better understanding of the relation between space syntax and the adjacent field of urban morphology. We believe that this can benefit both fields in their further development and more specifically, this paper will show how typical approaches in urban morphology can be helpful in explaining variations in correlations between space syntax measures and pedestrian movement. That these correlations vary is shown by various scholars and the reoccurring argument is missing data input such as, amongst others, density, land use and public transport. We also see a problem in space syntax analysis in that there seems to be little consistency in exactly how pedestrian movement is best captured, that is, with what measure and at which radius. Hillier and Iida (2005) show for instance in their study of four London areas that the ‘best radius’ can be found with a radius of analysis varying from 12 to 102 segments. This is troublesome, especially if we are not able to explain why this is the case. In this paper we propose to use two typo-morphological approaches to explain such variations: the classification system for street morphologies developed by Marshall (2005) and the integrated density approach ‘Spacemate’ developed by Berghauser Pont and Haupt (2009; 2010). The results presented in this paper show that different neighbourhood types, in terms of density and street morphology, indeed have different patterns driving pedestrian behaviour and following that, ask for tailored spatial analysis. It is shown that in denser and more ‘griddy’ street patterns, the betweenness centrality measure is able to capture pedestrian behaviour, but in other neighbourhood types pedestrian behaviour is better captured when also closeness centrality and the distribution of attractions is included. Further, it is shown that what may be called the ’scale of operation’ of each neighbourhood plays a crucial role which needs to be considered when choosing the radius of analysis. This paper shows further that a first indication of pedestrian intensity and pedestrian distribution can be arrived at by using two relative simple spatial measures: ‘accessible density’ and ‘attraction betweenness’ respectively. Although this study is just a first tentative exploration in combining urban morphology with space syntax, we suggest that we based on these preliminary results can see many advantages in pursuing research in this direction.