Shpuza; Urban Shapes and Urban Grids: A Comparative Study of Adriatic and Ionian Coastal Cities
extent to which grids have developed as constrained by bounding urban shapes.
Previous research on office buildings (Shpuza and Peponis 2005, Shpuza 2006) has shown that office floorplate shape affects internal layout integration according to the generative principles of layouts. This paper uses these findings as a departure point to inquire the effect of settlement shape on syntactic features of street networks.
Unlike buildings, where layouts are bound by the limitations posed by building shells, in cities the relationship between urban shapes and grids is mutual. Considering geographical constraints, urban shapes affect urban grids by exerting the limitations of where it is possible to extend, while grids affect urban shapes due to growth occurring along outlying streets. Therefore the study addresses the interaction between urban shapes and urban grids and is aimed at discovering robust generic principles that link shape properties of urban shapes to properties of street networks.
Space syntax studies that have dealt with comparative or typological classification of urban form have so far relied on small samples of cities or have considered limited representative fragments of urban grids. In contrast to building typologies, which are formulated on the basis of large samples, the typological classification of urban form is generally marred by the problem of dealing with large systems and the economy of surveying, observing, representing and computing. It is necessary to support the formulation of typological classifications of urban form with samples of sufficient size.
The search for generic affinities between urban shapes and urban grids is based on the morphological analysis of a large sample which enables statistical reliability. The sample includes 50 coastal cities and towns located in the Ionian and Adriatic seaside of Italy, Slovenia, Croatia, Montenegro, Albania and Greece. Citiesi are listed in a clockwise direction starting from the southeastern corner of Sicily, going up the Apennine coast, down the Western Balkans, and ending in Peloponnese (table 2). The study does not inquire into the historical factors that have influenced the development of the cities, while acknowledging the likelihood for similarity among them due to continuous cultural and economic exchange across the region.
For the purposes of the argument “urban boundary” is defined according to the contiguity of the built area by disregarding administrative boundaries; “urban shape hull” is defined according to the shape enclosed inside the outmost urban boundary of the settlement; “holes” inside the urban shape consist of undeveloped land caught inside urban development; “urban shape” is defined as the shape resulting from subtracting holes from the urban shape hull; the “urban grid” is defined as the street network located entirely inside the urban shape. Urban grids are represented with linear maps (Hillier and Hanson 1984, Peponis et al 1998), also known as axial maps, and are characterized accordingly.
The chosen cities represent a wide diversity of urban shapes which result from variations in the overall shape hulls and the size and distribution of holes (figures 1 and 3). Shape hulls are affected by: islands, littoral strips and peninsulas which limit the further extension of cities (e.g. Augusta, Chioggia Sottomarina, Venice, Lido, Grado); and crescent areas around bays (e.g. Umag, Dubrovnik, Krk, Kotor, Sarandë). The holes in shapes consist of: undeveloped farm land located between peripheral orbital roads and town neighborhoods (e.g. Termoli, Poreč, Biograd, Nafpaktos, Aigio, Patras); hilltops, cliffs, ravines and creeks (e.g. Taormina, Messina, Trieste); railroad tracks