Flat ontology and the deconstruction of scale: a response to Marston, Jones and Woodward

To identify the subject matter of human geography it is generally necessary to draw spatial boundaries at some level of abstraction. Whether the focus is upon patriarchalism or the new imperialism, areal differentiation or economic integration, the determination of such boundaries will usually be required at some stage in the proceedings. Spaces do not however occur in the singular: each locality, for example, is defined not only by contrast with other localities, but also with non-local territories of different sizes (such as bodies or regions). Indeed, it is by codifying this system, projecting a world that is divided not only into a 'horizontal' structure (in which similar activities are organized at similar scales in different places) but also a 'vertical' structure (in which different activities are organized at different scales covering the same places), that scale analysis acquires its conceptual power. The framework of nested scales was introduced during the 1950s and 1960s as a categorical device for describing spatial patterns at different levels of aggregation. From the early 1980s it was however argued that scales reflect real differences in the territorial organization of society, and it is on this basis that scale analysis (including perhaps the body, home, locality, region, nation, supranational and global levels) has extended its influence: 'integral to the production of space, capital produces certain distinct spatial scales of social organisation' (Smith 1984, 87; see also Taylor 1982; Kurtz 2003; Gough 2004; Uitermark 2005; for a useful review see Sheppard and McMaster 2004). But whether it is composed of nominal categories or real territories, the scale analytic cannot be segregated from the rest of traditional human geography but is symptomatic of this, and of the spatial structuralism with which it is generally imbued. Over broadly the same period, however, the writings of Lacan and Foucault, Deleuze and Derrida, Cixous and Irigaray have informed a post-structural critique of presence and identity that challenges the coherence of abstract structures. Feminists have argued, for example, that the spaces of structural geography, the spaces that can be exhaustively analysed scale by scale, express the territorial logic of patriarchy (Rose 1993, 149; 1996, 62). The masculine desire to stabilize meaning leads therefore to the drawing of boundaries around territories: 'envelopes are another solid then; they depend on a certain kind of space to constitute the masculine subject and his feminine (m)other' (Rose 1996, 71). Indeed, in her deconstruction of these boundaries Rose pursues the language of a 'paradoxical space', a space that lurks beneath the bounded space of geography, a space of flows and melding that (for example) undermines the distinction between the real and the metaphorical: 'It is to write as if the mirrors were not solid but permeable, as if the tain could move ...' (Rose 1996, 72; 1993, 140-1). But perhaps the first deconstruction of spatial structuralism after that of Derrida himself was provided by actor-network theory, which acknowledges the reality of macrostructures (such as nested scales)