Natural movement: or, configuration and attraction in urban pedestrian movement

Existing theories relating patterns of pedestrian and vehicular movement to urban form characterise the problem in terms of flows to and from 'attractor' land uses. This paper contains evidence in support of a new 'configurational' paradigm in which a primary property of the form of the urban grid is to privilege certain spaces over others for through movement. In this way it is suggested that the configuration of the urban grid itself is the main generator of patterns of movement. Retail land uses are then located to take advantage of the opportunities offered by the passing trade and may well act as multipliers on the basic pattern of 'natural movement' generated by the grid configuration. The configurational correlates of movement patterns are found to be measures of global properties of the grid with the 'space syntax' measure of 'integration' consistently found to be the most important. This has clear implications for urban design suggesting that if we wish to design for well used urban space, then it is not the local properties of a space that are important in the main but its configurational relations to the larger urban system. Introduction: attraction and configuration Quantitative methods for predicting pedestrian movement in urban space have conventionally been adaptations of the models employed in vehicular studies. The trip-generation potential of built forms (buildings or urban blocks) is seen as the key quantity, congestion as the most likely problem, and the scaling of local pedestrian space to match attraction as the main design aim (for example, Pushkarev and Zupan, 1975). We might call this the attraction theory of pedestrian movement: movement is seen as being to and from built forms with differing degrees of attraction, and design is seen as coping with the local consequences of that attraction. It follows that attraction theories say little about the spatial configuration of the urban grid, that is, about the way in which the spatial elements through which people move—streets, squares, alleys and so on—are linked together to form some kind of global pattern. But it is easy to show that, theoretically at least, configuration can have effects on movement which are independent of attractors. For example, in the simple layout shown in figure 1(a), all journeys from 'sidestreet' origins to