Probing different centralities in city regions: a space-syntactic approach

This paper presents preliminary findings from the first attempt at extending the spatialconfigurational analysis of individual cities’ street network to the city-regional scale street network, hoping to capture simultaneously the inter- and intra-settlement spatio-functional dynamics that may explain different centralities – historically evolved centres, planned centres, emergent metropolitan centres at various spatial scales – in rapidly expanding city regions. This involves the analysis of aggregate space-syntactic properties, multiple-radii integration and choice measures of the complete spatial network of two rapidly developing, geo-morphologically varied, Malaysian city regions – Penang Island and Johor Bahru District. It is demonstrated that spatial network analysis is an effective tool for studying different centralities in mostly planned, spatially non-contiguous city regions. The spatial network approach aptly encapsulates cityregional morphological variations; gives effective spatial accounts of centres of different hierarchies and sizes at various spatial scales; identifies the presence of global and intermediatescale spatial relations that may define centres’ global significance and regional strength; depicts the nesting of local centres within larger centres as well as overlapping centralities across spatial scales; and accounts for the “alternative” global-oriented location pattern of emergent metropolitan centres in city regions. More specifically, it is found that historically evolved centrality tends to be more intelligible and synergetic than planned centrality; that higher ratio of planned centrality in city regions may possibly have aggregate weakening effects on their overall intelligibility and synergy levels; and that modern metropolitan shopping/commercial centres are spatially emergent under a different spatio-functional logic that is best captured by choice analysis. Within the paper’s restricted analytical depth and specificity, it is asserted that spatial network analysis effectively describes inter-settlement centrality patterns and potentially complements economic geography and regional science’s spatial interaction modelling of regional/urban centrality, with the crucial advantage of not losing sight of centres’ internal spatial structure. However, more in-depth quantitative analyses and detailed micro-structure studies are necessary to substantiate this claim. The paper concludes by opening up more questions that need further addressing in future studies.