Environment and Planning A 2008, volume 40, pages 342 ^ 361

DOI:10.1068/a38255

Cartographic anxiety and the search for regionality

Joe Painter
Department of Geography and Centre for the Study of Cities and Regions, University of
Durham, South Road, Durham DH1 3LE, England; e-mail: [email protected]
Received 1 July 2005; in revised form 17 March 2006; published online 3 July 2007

Abstract. Despite the rise of relational and antiessentialist approaches to regional theory, many
accounts of regionality continue to work with territorial conceptions of regions as bounded wholes
or totalities. The author suggests that this tendency can be explained in part by the continuing effect
of cartographic anxiety and Eurocentrism on dominant understandings of regionality. The paper
examines the relationships between regional theory, different forms of totality and the cartographic
impulse, and discusses possible reasons for the Eurocentric cast of some regional research. It
concludes with a consideration of how regional theory might respond to cartographic anxiety and
Eurocentrism.

Introduction
Much of the expanding body of research on regions and regionalism has come (or been
brought) together under the signs of `new regionalism' and/or `new regional geography'.
This makes some sense. It highlights interactions between regional change and our
conceptualizations of what regions are and how they should be studied. It is, however,
a move I want to resist, or defer, as it risks conflating quite diverse literatures and
distinct empirical shifts.
In many countries the 20th century saw a reduction in (though never an eradication
of ) regional economic, social, and cultural differences. Regional loyalties and attach-
ment to the specificities of place often came to be seen as backward looking and as
potential threats to economic development and the territorial integration of nation-
states. During the long period of growth in industrialized capitalist economies after the
Second World War there was a widespread assumption that regional differences would
gradually disappear. During the 1950s and 1960s parties seeking regional autonomy or
independence had little electoral support. In many countries regionalist political ideol-
ogies were actively suppressed and/or came to be associated with the fringes of the
political spectrum.
It was not until the resurgence of regionalist politics in several countries in the
mid ^ late 1970s and the move of regionalist parties into the electoral mainstream that
geographers began to show much interest in the subject (eg Agnew, 1981; Knight, 1982;
Kofman, 1985; Smith, 1985). This early work was by political geographers, who treated
regionalism principally as a political movement for greater territorial autonomy. It had
clear affinities with cognate work in political science.
Keating (1998) argues that the upsurge in regionalist politics in the latter half of the
1970s did not represent a return to the antimodern `old provincialism' of earlier
decades. Rather, this was the beginning of a `new' regionalism that saw the region
not as a bastion of tradition and the antithesis of modernity, but as the vehicle for new
forms of development and enhanced economic opportunities. The idea of `new region-
alism' [which dates back at least to 1981 (Gerdes, 1981)] thus has a somewhat longer
history than is sometimes implied by its current uses.
Cartographic anxiety and the search for regionality 343

Keating's background is in political science, where substate regions are usually

understood as being administrative or governmental units. Keating's own book on
The New Regionalism in Western Europe acknowledges the variety of regions (cultural,
administrative, economic, governmental, etc), and draws connections between eco-
nomic, cultural, and political changes, but is concerned principally with regions
as actual or potential political units. Related work in international political economy
and political geography focuses on the formation of cross-border regions, with institu-
tions straddling international boundaries (Fall, 2005; Grundy-Warr, 2002; Ha«kli, 1998;
Kramsch, 2002; Sparke, 2002).
Studies of multilevel governance emphasize the development of a tier of regional
institutions between nation-state and municipality (Bache and Flinders, 2004; Hooghe
and Marks, 2001; 2003). A connected, but distinct, current of work considers federal-
ism as one strategy for managing regional differences in which a regional institutional
tier becomes constitutionalized and is accorded a formal role in a division of govern-
mental labour (Loughlin, 1996; Smith, 1995). Federalist ideas include a romantic
strand, linked to anarchist thought, that emphasizes local autonomy, self-government,
small administrative units, the rejection of the nation-state, and the political impor-
tance of place. This tradition seems to have been the source of the concept of a `Europe
of the regions', which dates back at least to the 1960s (Centre Europëen de la Culture,
1967/1968; Hëraud, 1963; Rhodes, 1974; Saint-Ouen, 1998).
The importance of place has also been central to debates in contemporary geog-
raphy. In their accounts of `new regional geography', Allan Pred (1984), Nigel Thrift
(1990; 1991), and Anssi Paasi (1986; 1996) examined the ways in which regions and
other kinds of places could be understood as both resources for, and the outcomes of,
human action (see also Lee, 1985; Sayer, 1989). Work in this vein sought to provide a
nonessentialist account of place with an emphasis on contingency and becoming, while
retaining the notion that places have a degree of integration and coherence.
More recent writing on place has drawn on poststructuralist and feminist ideas, actor-
network theory, and theories of practice, performance, and social complexity (eg Amin,
2004; Gibson, 2001; Massey, 2005). These writers challenge the notion that places should
be seen as bounded wholes and stress the (often quite extended) translocal connections
through which regions and other places are brought into being (Allen et al, 1998; Massey,
1991). Places are understood to be defined as much by their `constitutive outsides' as by
their internal dynamics. Moreover, places and regions are thought of as the effects of
myriad social and material networks that have complex geographies, disrupting the idea
that different kinds of places (neighbourhoods, cities, regions, nation-states) can be framed
as, or located at, distinct spatial scales (Marston et al, 2005).
Notwithstanding these critiques, scale remains a widely used concept in geographi-
cal research on regions. In fact, there are at least four more distinct understandings of
`new regionalism' in which ideas about scale are important, although the first three are
frequently conflated. The first is associated with the economic geography of Allen Scott,
Michael Storper, and others (Scott, 1998; Storper, 1995; 1997). Scott's and Storper's
theories derive from their analyses of the economics of industrial agglomeration, based
on untraded interdependencies, clustering, and the formation of regional industrial
milieus. This approach has led to the controversial claim that the regional scale has
become the key basis for economic competitiveness.
The second focuses on the `soft' institutional supports that supposedly promote the
economic developments analysed by Scott and Storper (Amin, 1999; Amin and Thrift,
1994; Cooke and Morgan, 1998; MacLeod, 2001a). This work was partly prompted by
Robert Putnam's (1993) study of the importance of civic traditions, trust, and associa-
tional networks in promoting good governance and economic development in Italy.
344 J Painter

Although it has often been suggested that there is a causal link between soft institutions
and regional economic success, the nature of the relationship remains opaque and
empirical research is inconclusive. We should therefore be wary of conflating these
two sets of regionalized processes into a generic `new regionalism'.
A similar caveat applies to the third `new regionalism', which focuses on `hard'
rather than `soft' institutions, and in particular on the `rescaling' of the state. Theories
of state rescaling suggest that globalization and the crises of the Fordist welfare state
have resulted in the `hollowing out' of the nation-state, with key state functions moved
`up' to international or supranational bodies such as the European Union and `down'
to regional and local scales (Brenner, 2004; Brenner et al, 2003; Jones, 2001; MacLeod,
1999; 2001b; MacLeod and Goodwin, 1999). It is suggested that this has been accom-
panied by a shift from government to governance and a move to networked forms of
social and economic coordination based on partnerships and an enhanced role for
nongovernmental organization and private firms (Goodwin and Painter, 1996; Jessop,
1995; 2000; Swyngedouw, 2000). The difficulty with assimilating state rescaling to other
understandings of new regionalism is that the administrative and governmental regions
on which it focuses correspond only weakly, if at all, either to the functional economic
spaces analyzed by Scott and Storper or to the networks of soft institutional relation-
ships that supposedly underpin them. Moreover, the regional structures of state
bureaucracies are often organized without regard to the preexisting geographies of
regional identity, cultural practice, or civil society, and sometimes are designed precisely
to cut across them to reduce the scope for regionalist mobilization.
Finally, a distinct discourse on new regionalism has developed in North America,
where it refers to metropolitan city regions (Barlow, 1997; Brenner, 2002; Frisken and
Norris, 2001; Norris, 2001; Sancton, 2001; Swanstrom, 2001). This literature includes
calls for new forms of regional governance, organized around functional urban regions.
This is a different territorial logic from that involved in regionalizing tendencies in
much of Europe, although support for city regions is growing there too. Whereas the
formation of a regional tier of government in Europe is partly a response to the idea
that the nation-state is too large and inflexible to meet regional needs, arguments for
city regions imply that existing municipal governments are frequently too small to
govern sprawling urban agglomerations effectively.
The richness of these literatures certainly speaks to the significance of regions,
regionalization and regionalism for cultural, economic, and political life. But it also
counsels against too rapid a move to collapse the distinctions between them into a
general (and sometimes highly normative) claim that regions are becoming a funda-
mental organizing principle of social life. It also seems important to respect the diverse
and often mismatched geographies of different kinds of regions and regionalism.
In this paper I consider some of the implications of these points for our understanding
of regions, and examine two neglected explanations for the pull that regionality has on
our geographical imaginations: Eurocentrism and cartographic anxiety.

Cartographic anxiety
Let me begin with the suggestive term `cartographic anxiety'. In the geographical
literature it was first invoked by Derek Gregory in his critique of the objectivist
tradition in human geography (Gregory, 1994, pages 70 ^ 205). In the same year it
also appeared, apparently coincidentally, in the politics and international relations
literature in an article by Sankaran Krishna (1994) on nationalism and mapping in
postcolonial India.(1) Both Gregory and Krishna cite Richard Bernstein's notion of
`Cartesian anxiety' (1983) as inspiration for the concept.
(1) `Apparently coincidentally' since neither author cites the other.
Cartographic anxiety and the search for regionality 345

According to Descartes, the application of human reason was the basis of the
search for truth. Descartes was therefore concerned to exclude from his philosophy,
and from the practice of science, anything that did not accord with Reason. Moreover,
without a clear and absolute distinction between Reason and Unreason to provide a
``fixed foundation for our knowledge ... we cannot escape the forces of darkness that
envelop us with madness, with intellectual and moral chaos'' (Bernstein, 1983, page 18).
What Bernstein calls `Cartesian anxiety' is, therefore, nothing less than the fear that
there may be no such fixed foundation or clear distinction.
Drawing on Derrida's practice of deconstruction, Bernstein argues that it is not
possible to make the kind of sharp division between Reason and Unreason on which
Cartesian thought depends. Rather, as Gregory puts it, ``the strange, the alien, the other
are not massing outside the gates of Reason: They are already and ... constitutively
inside'' (1994, page 72, original emphasis).
Gregory suggests `cartographic anxiety' as an appropriate label for the expression
of this Cartesian anxiety in human geography (1994, pages 72 ^ 73). This is partly
because of the resonance that the spatial metaphors associated with the Cartesian
approach have in geographical thought: metaphors of inside/outside and familiar/
alien, for example. But it also involves cartography in a more literal sense, which
Gregory illustrates with reference to Gunnar Olsson's (1991) discussion of lines of
power and Brian Harley's (1992) critique of map-making.
Krishna's account of cartographic anxiety in India also refers to the politics of
map-making. He argues that ``in a postcolonial society such as India, the anxiety
surrounding questions of national identity and survival is particularly acute'' (Krishna,
1994, page 508) and that this is expressed precisely in concerns over the creation and
representation of national boundaries. In consequence, ``the physical map of India
gains ubiquity as an iconic representation of the body politic'' (1994, page 510) in
ways that disguise the violence associated with the drawing and policing of borders.
For Krishna, cartographic anxiety is particularly associated with postcoloniality
(page 517). Anxiety over borders is not confined to former colonies, however. Nor, I
want to suggest, is it concerned only with national borders and nation-states. In what
follows I shall examine how the notion of cartographic anxiety might help us to
understand the politics of subnational territories, such as regions, in a place, Europe,
which was, for the most part, on the other side of the colonial divide.
Gregory and Krishna attend to each of the two senses of `boundary' evoked by the
phrase `cartographic anxiety'öthe epistemic and the spatialöand both note how these
are interrelated. Descartes was concerned primarily with the epistemic boundary
between Reason and Unreason. In the history (and present) of European colonialism
this mapped all too conveniently onto a spatial boundary between the West and its
orientalized Others. However, there is another strand of Gregory's critique of the
production of geographical knowledge that could be incorporated into the concept of
cartographic anxiety, although in Geographical Imaginations the link is largely implicit.
This is the will to knowledge associated with modern European science.

The exhibitionary world and cartographic reason

Gregory cites Timothy Mitchell's notion of the `world-as-exhibition' to argue that ``by
the closing decades of the nineteenth century, it was characteristic of European ways of
knowing to render things as objects to be viewed'' (Gregory, 1994, page 34). Visuali-
zation and legibility helped to constitute a way of knowing that regarded itself as both
objective and, in principle, unlimited. `Seeing' too easily becomes `all-seeing' öDonna
Haraway's `God-trick' (Haraway, 1991). For Mitchell `the world-as-exhibition' means
``not an exhibition of the world but the world organized and grasped as though it
346 J Painter

were an exhibition'' (1992, page 296). `Grasping' here implies both transparency and
completeness. It is a way of knowing that seeks systematically to draw more and more
of the strange, the complex, the Other, and the presently unknown into the ambit of
Reason and asserts that nothing is in principle unknowable. The world-as-exhibition is
thus a therapeutic response to the Cartesian anxiety, because it appears to offer a way
of expanding the territory of Reason to encompass more and more of the unknown or,
failing that, of bringing elements of the unknown inside the fortress to be dissected and
exhibited, and thereby made knowable.
From this perspective, the notion of cartographic anxiety can be enlarged to refer
to the desire to make geographical space legible. A detailed discussion of these con-
nections is provided by John Pickles (2004). Pickles shows that, while the history
of cartography cannot be reduced in any simple sense to a tale of the triumph of
objectivism and the logic of the world-as-exhibition, the epistemic regimes critiqued by
Gregory and Mitchell have undoubtedly had a profound influence on the development of
what he calls `cartographic reason'.
Pickles is concerned with the diverse principles and practices underlying the pro-
duction of actual maps. But in doing so he uses a number of conceptsöcartographic
reason, cartographic impulse, and cartographic imaginationöthat we can relate, not
only to the drawing of maps themselves, but also to geographical knowledge and other
forms of knowledge more generally. In other words, a cartographic impulse may be at
work whether an actual map is produced or not. In particular, I want to suggest that
dominant understandings of concepts such as `territory' and `region' have been struc-
tured in important ways by cartographic reason, and therefore that related practices of
territorialization and regionalism are generated, in part, by cartographic anxiety.

Boundedness, territories, and networks

As we have seen, the literature on regionality is multistranded, draws on diverse
theoretical sources, and works with a wide range of different and sometimes incompat-
ible definitions of the terms region and regionalism. The view that regions are bounded
territories (sometimes with distinctive essential identities) still holds considerable sway
in popular and policy discourses: ``people have often characterised the locations in
which they live as closed, bounded regions, to which they have strong territorial
affiliation and attachment'' (Hudson, 2001, page 3). Within academic discourse this
traditional view of regions as relatively homogeneous bounded territorial entities has
certainly not disappeared, but it has been subjected to a growing critique from relational
and antiessentialist approaches to geographical theory (eg Allen et al, 1998; Amin, 2004;
Amin and Thrift, 2002; MacLeod, 1998; Massey, 1991; 1993; 1999; Paasi, 2002).
By highlighting the social, cultural, and economic heterogeneity of regions, and by
deconstructing essentialist understandings of regional identity, relational critiques have
been very effective at undermining claims that regions are homogeneous or that they
should be understood in essentialist terms. However, while recognizing the internal
heterogeneity of regions and the absence of essential regional identities, relational
regional theory often retains the sense that regions are `bounded wholes'.
Martin Jones and Gordon MacLeod (2004) draw a useful distinction between
`spaces of regionalism' and `regional spaces'. The former refers to processes of political
mobilization around an idea of regional difference; the latter refers to the production
of patterns of regional differentiation mainly through economic processes. Spaces of
regionalism involve the ``(re-)assertion of national and regional claims to citizenship,
insurgent forms of political mobilization and cultural expression and the formation
of new contours of territorial government'' (Jones and MacLeod, 2004, page 435).
Jones and MacLeod suggest that relational approaches that are critical of bounded
Cartographic anxiety and the search for regionality 347

definitions of `region' are at their most convincing when they focus on ``trans-regional
and trans-national economic flows and interchange'' (2004, page 437, original empha-
sis). When it comes to spaces of regionalism, however, relational accounts ``bend the
stick too far'' (page 437) because ``when performing their practical politics, agents
often imagine and identify a discrete, bounded space characterized by a shared under-
standing of the opportunities or problems which are motivating the ... political action''
(page 437). As a result,
``there may be certain circumstances in which, as an object of analysis, the region
should be taken as a practical and `prescientific' bounded territorial space that has
been institutionalized through particular struggles and become `identified' as such a
discrete territory in the spheres of economics, politics and culture'' (Jones and
MacLeod, 2004, page 437, original emphasis).
The use of the term `prescientific' implies that, for Jones and MacLeod, bounded-
ness is (or can be) a real feature of the spatiality of the world that preexists scientific
analysis and is not simply a construction placed upon that spatiality by theorists.
It is important here to be clear about the distinction between regional boundedness
and regional boundaries. Some kinds of regionsöparticularly administrative or govern-
mental onesöhave formal boundaries. However such boundaries do not, in themselves,
constitute a bounded region. Regional boundaries will have more or less effect, depend-
ing on the powers of the relevant administrative or governmental institutions and the
extent to which those powers are exercised to different ends in different regions.
Boundaries may or may not affect the extent and nature of flows (of things, people,
money, ideas, practices or whatever) between regions. Thus, the fact that boundaries
demarcate eight official regions in England (for example) does not in itself mean that
any of them necessarily constitutes a bounded region which, in Jones and MacLeod's
terms, can be identified `as a discrete territory in the spheres of economics, politics, and
culture'. Conversely, to posit that a region is a bounded whole does not imply necessarily
that it possesses formal boundaries.
Jones and MacLeod suggest that relational approaches apply most convincingly to
the spatiality of `economic flows and interchange', whereas in the sphere of `practical
politics' and `political action' it may be more appropriate to understand a region as
`discrete bounded space' or `territory' (2004, page 437). Similar views have been
expressed in other recent studies including those by Peter Dicken et al (2001) and
Erik Swyngedouw (2004). In relational vein, Dicken et al propose a ``methodology
for analysing the global economy'', based on the study of ``actors in networks, their
ongoing relations, and the structural outcome of these relations'' (2001, page 91). They
also emphasize the practices that produce networks, rather than formal analyses of
network relations, and they question conventional scalar thinking by suggesting that
``different scales of economic processes simply become links of various lengths in the
network'' (page 95). However, when it comes to political institutions such as the state,
relational and topological thinking has to give way to territoriality. According to
Dicken et al, adopting a network approach to analyzing the economy should not
lead us to ``denigrate the role of the territorial state in global economic processes''
because ``national regimes of accumulation continue to create a pattern of `bounded
regions' '' (page 96). Territories and networks interact, claim Dicken et al, in a ``mutu-
ally constitutive process'' (page 97), but the clear implication is that they operate
according to different kinds of logic and in relation to distinct domains: economic
flows in the case of networks and political institutions in the case of territories.
Swyngedouw takes a similar view, arguing that ``the molecular strategies of capital
as mobilized by a myriad of atomistic actors produce rhizomatic geographical map-
pings that consist of complex combinations and layers of nodes and linkages, which are
348 J Painter

interconnected in proliferating networks and flows of money, information, commodities

and people'' (2004, page 31). However, he goes on to suggest that the rhizomatic
network logic of the economy is in tension with the territorial and scalar logic of
governance (2004, page 33).
Despite their different empirical concerns and theoretical emphases, Jones and
MacLeod, Dicken et al, and Swyngedouw all work with the same broad assumption
that concepts such as relationality, networks, and rhizomes are suited to understanding
the spatiality of the economy, whereas the geographies of governance and politics are
more appropriately addressed using the vocabulary of territory, scale, and boundedness.
The assumption that territory is a qualitatively distinct form of spatial organization
runs with the grain of much geographical thinking (eg Sack, 1986; Taylor, 1994). For
many practical purposes this seems a reasonable position. However, it is also possible
to see state institutions and bounded territories as equally produced, sustained, and
constituted by relational processes, flows of matter and energy, and rhizomatic net-
works. Much of Doreen Massey's work on relational theories of place, for example,
has been concerned as much with politics as with economics. There is also a growing
body of work on state institutions and governance processes that emphasizes their
porous, performative, and networked character (eg Barry, 2001, pages 85 ^ 103; Castells,
1998, pages 331 ^ 332; Nordstrom, 2003).
Both territorially defined institutions and territoriality as a feature of social life can
be understood as the effects and the outcomes of relational networks. For example, an
apparently `territorial' institution, such as the regional office of a government depart-
ment, can be interpreted as the product of rhizomatic flows. Taxes, policy documents,
communication technologies, public officials on secondment, architects who design
public buildings, private consultants, knowledge-sharing and expertise-sharing exer-
cises with other regional offices, secretaries, lawyers, statistics, food, water, energy,
and much more, flow in and out. Just like the picture of firms in the globalized
networked economy painted by Dicken et al and Swyngedouw, state institutions can
be understood as the product of networked flows and relational processes.
Formal territorial boundaries, too, can quite reasonably be interpreted as the effects
of networked social and material practices. This is not dissimilar to recent suggestions
that boundaries should be seen as social processes (Paasi, 1999) or social constructions
(Allen et al, 1998, pages 54 ^ 55). Equally, when Jones and MacLeod discuss political
struggles over regional boundaries in southwest England, the processes they describe
seem perfectly amenable to interpretation from a network perspective, as the authors
note (Jones and MacLeod, 2004, page 448).
What all of this suggests, I think, is that the fact that some regions appear to
function as bounded territorial entities does not need to be seen as evidence that
territoriality is a distinct form of spatiality that is qualitatively different from net-
worked or relational geographies. One question that then arises is why the impulse to
treat regions as bounded wholes is so persistent. It is often suggested that it stems from
defensive or exclusionary attitudes to place, or from the existence of strong emotional
place attachments. These may be important factors. However, defensive and exclusion-
ary understandings of place seem to be linked more to claims of regional cultural
homogeneity (about which there is now much scepticism) than to boundedness per se.
I suggest that a more significant factor in the continuing tendency to treat regions as
bounded spaces is cartographic anxiety, expressed in what Pickles has called the carto-
graphic impulse with its ``goals of systematization, boundedness and totalization''
(2004, page 197, note 6).
Totalization here refers to the practice of conceptualizing and representing places as
wholes, that is, as totalities. Most understandings of totality are relational, so totalization
Cartographic anxiety and the search for regionality 349

cuts across the standard distinction between `relational' and `conventional' regional
theory. It also usefully detaches the question of boundedness from that of homogeneity;
totalities can be internally differentiated (and thus heterogeneous) while retaining their
boundedness or wholeness. A tendency towards totalization is implicated in the carto-
graphic impulse because the exercise of cartographic reason seeks to parcel the world
into knowable places, to make those places legible, and to ascribe distinctive identities
to them. The desire for mappability and the fear of unmappability (cartographic
anxiety) find expression in the propensity to treat regions as bounded wholes öas
totalities. However, there are several different kinds of totality and they are not
necessarily equally implicated in the cartographic impulse.

Varieties of totality
The concept of totality has a long history in philosophy, history, and social theory
(Jay, 1984). It was particularly important in the work of Spinoza, Hegel, Marx, Lukacs,
and Althusser. Several different uses of the term can be distinguished. The notion of
a simple totality refers to a whole that is lacking internal division or contradiction.
More commonly, though, `totality' refers to a more complex phenomenon with evident
internal differentiation; in particular to the relationship between the whole and its
parts, and to the predominance of the whole over those parts. What makes a totality
a totality is that the different elements are integrated into a whole. The integration of
the whole may be smoothly functional (an ecosystem in dynamic equilibrium, for
example), or it may be subject to contradictions (as in Marx's analysis of the capitalist
mode of production). In both cases, though, the notion of totality is clearly a relational
conceptöit is the relationships between the parts, and between the parts and the
whole, that define the totality.
Stephen Cullenberg (1999) identifies three broad approaches to the notion of
totality: the Cartesian, the Hegelian, and the decentred Marxist, each of which can
be related to different ways of conceptualizing regions.
In a Cartesian totality the parts exist independently of, and prior to, the whole.
This atomistic ontology underpins the individualism of neoclassical economics, in
which the whole is seen as no more than the sum of the effects of the parts. In a
Hegelian totality, by contrast, the whole takes precedence over the parts:
``The whole is presumed to exist prior to and independent from its parts. The parts
express in and through their existence the inner essence of the totality. The totality
is Hegelian in the sense that its inner essence is dynamic and contradictory.
The contradictory nature of the inner essence means that the essence unfolds in
a dialectical manner guided by a telos, or end'' (Cullenberg, 1999, page 802).
The third type of totality, which Cullenberg refers to as a decentred Marxist totality,
is elaborated in Louis Althusser's (1969) discussion of overdetermination. As Cullenberg
puts it,
``Althusser's Marxist totality neither reduced the parts to an expression of the whole,
as does the Hegelian totality, nor the whole to the aggregation of its independently
constituted parts, as does the Cartesian totality. Instead, the Marxist totality is
conceived to be thoroughly nonessentialist as the parts (contradictions for Althusser)
mutually constitute one another'' (Cullenberg, 1999, page 805).
Althusser distinguished sharply between Hegelian and Marxist conceptions of
totality. He argued that, although Hegel appears to allow for ``an infinity of concrete
determinations'' (Althusser, 1969, page 102) in the constitution of historical societies,
together these diverse influences ``constitute an original, organic totality'' (page 102)
and are but different expressions of a ``unique internal principle, which is the truth of all
those concrete determinations'' (page 102, original emphasis). The Marxist totality,
350 J Painter

or `social whole' on the other hand, is an overdetermined `structure in dominance'

(Althusser, 1969, pages 209 ^ 10; 1971, page 134), in which the unity of the whole
is the product of the complex multiple and mutual determinations of its internal
contradictions.
Following Cullenberg, it makes sense to identify four broad kinds of totality: a
simple totality (a unitary homogeneous whole); an atomistic totality (a whole equal to
the sum of its parts); a centred totality (a differentiated whole that nevertheless
expresses a single animating principle or essence); and a decentred totality (a differ-
entiated whole understood in nonessentialist terms as the interaction of multiple and
contradictory processes and determinations).

Regionality, totalities, and cartographic anxiety

Each type of `totality' can be related to a different way of conceptualizing regions.
Thus in neoclassical regional economics regions are atomistic totalities, conceived
as the aggregation of individual economic agents and their decisions. In this approach,
the distinctive character of each region is the product of a particular territorial
combination of factors of production, resource endowments, labour-market conditions,
information flows, and the decision making of investors, workers, and consumers.
At first sight, the incorporation of `regional multipliers', `agglomeration effects', and
`increasing returns' into regional economic models might seem to constitute a whole
greater than the sum of the parts, and thereby to go beyond a narrowly Cartesian
approach. However, as Peter Sunley (2000, page 191) and Eric Sheppard (2000,
page 179) suggest, such ideas are not fundamentally at odds with the methodological
individualism and rational choice assumptions of neoclassical economics.
By contrast, thinking of regions as centred totalities involves the identification of a
core principle of regionalization that animates a dynamic process of region formation.
The core principle could take a number of forms. The concept of bioregionalism takes
relations with nonhuman nature as its core (eg Thayer, 2003). For Paul Vidal de la
Blache the concept of genre de vie was the essence of regionality. Ethnic regionalisms
place ethnicity, often defined in exclusionary terms, at the centre of regional identity.
Some stress the primary role of dense social and civic networks in underpinning region
formation (eg Putnam, 1993). Others identify economic imperatives as the mainspring of
regional differentiation. Economic geographers such as Scott and Storper move beyond
Cartesian individualism because they insist on the importance of social, cultural, and
institutional relations among economic actors, as causal influences that contribute to
the constitution of the regional whole (and are not merely the epiphenomena of market
dynamics). At the same time, their approach retains the idea of an animating central
principle: the development of agglomeration economies and untraded interdependencies
(Scott, 1998; Storper, 1995).
David Harvey has argued that urban regions (defined on the basis of urban labour
markets) have a tendency towards `structured coherence', involving a particular tech-
nological and organizational mix and a dominant set of social relations that ``embraces
the standard of living, the qualities and style of life, work satisfactions (or lack
thereof ), social hierarchies ... and a whole set of sociological and psychological atti-
tudes toward working, living, enjoying, entertaining and the like'' (1985, page 140).
For some writers, this tendency towards structured coherence constitutes urban regions
as totalities, albeit dynamic and internally differentiated ones. For example, Kevin Cox
argues that ``structured coherences define a totality of internally related elements,
discursive, spatial, practical, institutional, cooperative'' (2004, page 191, emphasis added).
Here, the animating principle is not agglomeration economies, but the dynamics of the
wage relation under conditions of capitalist urbanization.
Cartographic anxiety and the search for regionality 351

These accounts make it clear that treating regions as centred totalities does not
mean seeing them as homogeneous. However it does suggest that they are bounded,
with the bounds set by the sociospatial scope of the central animating principle. In this
way Hudson uses Harvey's idea of `structured coherence' to support his use of a
bounded definition of region (Hudson, 2001, page 3).
These accounts are diverse as are the kinds of regions to which they refer. They
adopt different theoretical and political positions and analyze different kinds of phe-
nomena. However, they all seem to share an assumption that there is an inner logic to
the regional differentiation of the world, that can be unpacked and analyzed in terms
of the relation between the parts and the regional whole, animated by a core process or
relationship.
Drawing on Mitchell's concept of the `world-as-exhibition', I suggested above that
the notion of cartographic anxiety encompasses the desire to make geographical space
legible. Descartes's goal of excluding all claims to knowledge that could not be
grounded in Reason finds its geographical counterpart in the desire to render social
space objectively mappable, and thus visible and knowable. The conceptualization of
regions as centred totalities functions in precisely this way. Centred totalities allow us
to see the world in terms of ordered wholes, which can be linked to the exhibitionary
regime of visualization:
``The experience of the world as a picture set up before a subject is linked to the
unusual conception of the world as an enframed totality, something that forms a
structure or system'' (Mitchell, 1992, page 304, my emphasis).
This notion of an enframed totality that `forms a structure or system' captures very
well how regions come to be (re)presented as integrated social wholes, in the ways
discussed. Furthermore, there are, by definition, many regions. Thus, we can speak of a
process of enframing a series of ordered wholes öa regional or interregional system.
As Mitchell puts it in Colonizing Egypt, ``enframing is a method of dividing up and
containing, as in the construction of barracks or the rebuilding of villages [or, we
may add, the demarcation of regions], which operates by conjuring up a neutral
surface or volume called `space' '' (1991, page 44). In this way, cartographic anxiety is
expressed in the imposition of a spatial grid of regionsöa regional `tier'öin which
each regional whole is assumed to be composed of comparable but distinguishing
elements (institutions, regional symbols, social groupings, and cultural forms).
But what of decentred totalities? Is it possible to envisage territories that are
integrated, bounded social wholes, but that are not defined in essentialist terms by a
central animating principle? The most obvious candidates, of course, are nation-states.
Much contemporary state theory rejects essentialist or reductionist models that see state
power as the epiphenomenon of a single animating principle, whether that be class
relations, a unitary ethnic identity, or a moral purpose. Instead, territorial nation-states
are seen as the outcome of multiple determinations: economic, cultural, institutional,
and so on. Bob Jessop has used the concept of autopoiesis to capture the ``inter-
dependence of a plurality of self-organizing systems'' in capitalist states (2002, page 5).
In geography, Peter Taylor's (1994) argument about the territoriality of modern states notes
the accretion of power and functions by the state across numerous domains (economic,
military, cultural, social) all of which are interrelated, but none of which can easily be
reduced to a causal effect of one of the others.
Applying similar thinking to regions is problematic, however. To be sure, it would
be unrealistic to expect to find examples of regions that function as integrated wholes
to the same extent as nation-states are often thought to do. But there are strands
in some new regionalist thinking that suggest that the consolidation of regions as
352 J Painter

`mini-societies' (in which social, economic, and political processes cohere within a
regional space) is a trend that is observable, or desirable, or both.
Yet, as Michael Keating has been at pains to emphasize, the regional geographies of
economic, cultural, and political processes and institutions are not necessarily the same:
``these varying definitions of the region not only do not always coincide, but they may be
in conflict with each other'' (Keating, 1998, page 376; see also Keating, 2001). This point
is illustrated by Keating with two maps (Keating, 1998). One shows the administrative
regions of the European Union and the other shows the `cultural nations' of Europe;
the territorial matrices in the two maps are very different. The boundaries of economic,
governmental, cultural, and other kinds of regions rarely correspondötheir geographies
are noncongruent, or, we might perhaps say, `incongruous' (see figure 1).
It is important not to conflate the diverse regionalities of economic, cultural,
political, and governmental practices for four reasons. First, the empirical pattern
of regions and regionalism does not suggest any systematic link between cultural
distinctiveness, strong regional government, and economic dynamism. Second, even
if economic processes are becoming increasingly regionalized, as Storper (1997), Scott
(1998) and others have argued, it does not follow automatically that political institutions
should be tailored to the same territorial scale.
Third, conflating cultural, political, and economic regionalism tends to overempha-
size the coherence and integration of regions. It is easy to fall into the trap of endowing
regions with personality and consciousness and slip into talking of regions as if they
were active subjects. It is social actors (individuals and institutions) that have political
power, and represent diverse and often conflicting social interests, even within one
region. Moreover, if we assume that political institutions, economic relations, and
cultural identities all mesh together in a functionally integrated way, there is a risk
that regionalism will lead to cultural closure, introversion, and the development of
regional chauvinism.

Cultural geographies

Economic geographies

Governmental geographies

Incongruous geographies

Figure 1. Boundaries of cultural, economic, and governmental geographies, and their incongruousness.
Cartographic anxiety and the search for regionality 353

Fourth, the degree of spatial congruence between economic, political, and cultural
practices necessarily tends to be lower at smaller spatial scales. Taylor notes that even
nation-states may not be quite the integrated totalities they appear: the ``state as
[territorial] container'' is increasingly `leaky' (Taylor, 1994, pages 157 ^ 161). However,
as spatial scope is reduced, the likelihood of a given territory being able to function as
an effective container for several different spheres of social life declines simultaneously.
The idea that the regional scale is of growing importance across social, economic
and political life has become axiomatic in many policy and academic circles. This, I
think, is partly a product of cartographic anxiety. It is in the cartographic imaginary,
more than in the materiality of social life, that regions exist as integrated, territorial
wholes. It is almost as if we feel the need to grasp the region as a mappable entityöto
locate it visually. Doreen Massey refers to this as the assumption of ``an isomorphism
between space and society'' (2001, pages 10 ^ 11), the idea that the social must be
mappable and that there must be a one-to-one relationship between cartographic
regions and distinctively regionalized social relations. Networks, flows, and the
noncongruity of social, economic, and political life disrupt conventional modes of
cartographical representation. John Pickles approvingly cites the comment of Jean
Baudrillard (1983) that ``the map has come to precede the territory'' (Pickles, 2004,
pages 31, 95). The challenge then is how to devise an understanding of the territorially
adequate to the unmappable.

Beyond totalization
One possible way forward is offered by the increasingly influential work of Anssi Paasi.
Not that Paasi regards maps as unimportantöquite the reverse. However, his
approach provides a way of understanding regions that recognizes their complexity
and contingency and does not depend on a totalizing model. Paasi's approach is, in the
terms outlined earlier, `decentred', in that he does not assert a priori that there is an
underlying economic, ethnic, ecological, or other logic to the formation of regions:
``A region is comprehended as a concrete dynamic manifestation of social (natural,
cultural, economic, political, etc) processes that affect and are affected by changes
in spatial structures over time'' (1986, page 110).
Moreover, he explicitly recognizes the complex geographies of the processes
involved:
``Regions and localities are a complex synthesis or manifestation of objects, patterns,
processes, and social practices derived from simultaneous interaction between
different levels of social processes, operating on varying geographical and historical
scales'' (1991, page 242, emphasis added).
Paasi (1991, page 243) argues that the ``emergence or institutionalisation of regions''
takes place through a fourfold process involving ``the constitution of territorial shape,
symbolic shape, and institutions and, finally, establishment in the regional system and
social consciousness of society'' (see also Paasi, 1986). For Paasi, `institutionalization'
does not necessarily refer to the establishment of formal regional government or state
administrative organizations for the region. The term `institutionalization' is being used
here in its wider sense to mean `the establishing or production of a region', although
one part of the process he describes does involve the formation of institutions of
various kinds.
Paasi draws on structuration theory and related writings by geographers such
as Allan Pred and Nigel Thrift, that emphasize `becoming' and argues that regions
do not solidify into permanent unchanging presence, but are subject to perpetual trans-
formation and, perhaps, ultimate disappearance (Paasi, 1986, pages 120 ^ 121). His
account allows for a general theory of the emergence of regions, from those that
354 J Painter

are ``vague ideas'' (Paasi, 1986, page 125) to formal governmental regions. Moreover,
Paasi emphasizes that his approach provides for a general theory of the emergence of
regions at any spatial scale and he insists that it is not limited to the level between
national and local, to which the term `regional' usually refers:
``The areal extent of the idea of the region set out here is not confined to any specific
regional level. Hence a region can just as well be a part of a city, a municipality or a
county, as a province or a nation state'' (1986, page 121).
In Paasi's nonessentialist approach, regions do not come into being because of some
underlying, necessary driver of regionality, but emerge from the interplay of contin-
gencies, historical circumstances, and geographical conditions (see also Sayer, 1989).
Moreover, although in Paasi's account regions have boundaries (vague or formal)
that does not mean that they necessarily constitute bounded wholes (totalities) in a
conceptual sense, although they may come to do so. Regions that, in Paasi's framework,
are fully institutionalized would come close to being decentred totalities.
For Paasi regions are always becoming in a continual process of social, symbolic,
and institutional sedimentation. While this implies that the process of region formation
will never be complete (the region is always `to come'), the longer it continues, the
more the region takes on a set of qualities that, in combination, distinguish it from
other regions. As a result, the kinds of regions described by Paasi are socially and
culturally rich and institutionally thick. He writes ``regions ... are centres of collective
consciousness and sociospatial identities'' (1991, page 241); ``a region is perpetually
reproduced in ... the spheres of economics, politics, legislation, administration, culture,
etc'' (page 244). As part of this process, ``individuals are socialized into varying
regional community memberships'' (page 246), the region ``is mediated into daily life
and is produced and reproduced in multitudinous social practices through communi-
cation and symbols, which can be common to all individuals in a region'' (page 249).
Furthermore, ``structures of expectations form a frame that is bound to a specific
region ... is quite permanent and is represented in the form of time ^ space-specific,
region-bounded, institutionally embedded schemes of perception, conception, and action,
which can comprise real, imagined and mythical features of the region'' (page 249).
This nuanced picture of the fullest expression of regionality raises the intriguing
question of how prevalent such fully institutionalized regions might be in practice.
Since Paasi's formulation explicitly avoids restricting the definition of region to any
particular spatial scale, the most obvious candidates for full regionality according to
these criteria are, once more, nation-states. It is debatable whether more than a
handful of territorial units below the scale of the nation-state have the kind of intense
impact on the multiple aspects of social life mentioned in Paasi's account. Those that
come closest are those substate regions that are also the home territories of stateless
nations. Elsewhere, the intensity of regionality (the extent of institutionalization in
Paasi's term), exhibited by the mesolevel territorial units commonly labelled `regions',
varies greatly, but is typically much lower than in the stateless nations. We also come
back to the problem of the lack of territorial congruence between the different kinds of
process that might feed into regional becoming. In some cases there may be a tendency
for congruence to increase over time, as cultural, economic, and institutional processes
interact within a defined space (as suggested by Harvey's account of `structured
coherence') but, pace Harvey, it is not clear that this tendency is a necessary one.

Regionality and Eurocentrism

I have suggested that the recent tendency to assert the importance of regionality and
to see regions as bounded wholes stems in part from the influence of cartographic
anxiety on the geographical imagination. This is particularly marked in writing on
Cartographic anxiety and the search for regionality 355

Europe, and it is worth considering whether a degree of Eurocentrism is not also an

important factor. Eurocentrism refers here to the tendency to treat European social
forms as the norm and to generalize them unproblematically, either by applying them
unproblematically to other places, or by treating them as a set of universal abstract
principles.
One reason for this seems to be the importance of regions in the politics and
political economy of European integration. European Union (EU) political discourse,
policy making, and institution building have all served to valorize regions (Keating,
1995). The EU celebrates regional cultural diversity, targets `lagging' and `peripheral'
regions and those undergoing economic `adjustment' for special assistance, and
promotes the decentralization of political power to regional governments.
Perry Anderson argues that `the region' was invented as part of the process of
European integration (Anderson, 1994; see also Keating, 1997). This implies that
conceptualizing regions as purportedly all-purpose generic territorial entities may be
a peculiarly European activity, the product of postwar European politics. This con-
fluence of policy and politics means that in the European Union the generic figure of
`the region' is mobilized in numerous contexts and has taken on a rhetorical power and
political importance out of proportion to the rather restricted budgets of regional
development programmes and the fitful and somewhat limited progress towards polit-
ical devolution. In other words, Europe does not represent `the norm', from which we
may generalize to elsewhere, and towards which others will converge. Rather, Europe's
obsession with regions is actually quite peculiar and thus the spatial imaginary of
bounded regions may also be peculiarly European.
I am not suggesting that discourses of regionality exist nowhere else. In many
non-European contexts, though, the term `region' has a far more specific set of
connotations than is the case in Europe. For example, `regional Australia' refers to
areas of the country away from the major urban centres, but excludes the very sparsely
populated nonagricultural areas known as `remote Australia'. In the United States,
there is a discourse of regional planning in which regions are typically conceived as
functional economic regions, often encompassing major metropolitan centres. Grass
roots regionalist movements exist in many parts of the world (for an Indian example
see Mawdsley, 1997), but they typically mobilize around a strong sense of ethnic
distinctiveness. Of course, ethnic regionalism certainly exists in Europe, but as a basis
for regionality in the EU, at least, ethnicity is the exception rather than the rule öcivic
regionalism is far more prevalent, even in places such as Catalonia and Scotland that
have distinctive cultural identities.
The idea that regions are `mini societies' was implicit in several versions of the dis-
course of `a Europe of the Regions' that was popular in the early 1990s. The discourse
ceased to have much significance in European Union politics after the conclusion of
the Maastricht negotiations (Jeffery 1997). In its most developed form, the discourse
posited a possible federal future for the EU in which nation-states withered away in
favour of a federation of regions. The `Europe of the Regions' concept has quite a long
history (Centre Europëen de la Culture 1967/68; Loughlin 1996; Rhodes 1974). Its
origins include the romantic federalism of Denis de Rougemont (Saint-Ouen 1998)
and Guy Hëraud (Hëraud 1963). Michael Heffernan has suggested that it can also be
traced back to the utopian visions of a regional Europe propounded by Patrick Geddes
and others before and during the First World War (Heffernan 1998: 99 ^ 106). Although
the establishment of a federation of European regions has never been EU policy, for a
period in the 1980s and 1990s these older federalist discourses resonated with the EU's
official support for regional development and decentralization, and with academic
accounts of `new regionalism' (eg Keating 1998; Murray 1992).
356 J Painter

This confluence of separate discourses may have fuelled the tendency to understand
regions as bounded wholes [though Keating, for one, argues strongly against such
reification of regions as unified entities (1998, page 13)]. Scott, for instance, cites the
development of a ``new Europe of the regions'' (Scott, 1998, page 146) as an example of
the emergence of what he termed ``regional directorates''önot necessarily regional
government in a conventional sense, but ``a system of local control and coordination
made up of any combination of formal governmental agencies, civil associations and
organizations, and private-public partnership'' (page 144).
According to Scott, it ``can be said with confidence ... that, if something approxi-
mating [regional directorates] does indeed come empirically into existence, they are
most likely to be physically anchored in dense polarized clusters of economic activity
coinciding with one or more metropolitan areas and a variable stretch of hinterland''
(page 145). However, very few of the putative regional directorates that he listsöthe
German La«nder, the Italian Lega Nord, the Belgian regions, the Spanish autonomous
communities, and the French regional councilsöcorrespond territorially to these sorts
of economic regions.
In fact, the geography of regional government in Europe is, for the most part, quite
different from the geography of economic activity; the two geographies are contin-
gently, rather than necessarily, interlinked. This is hardly surprising, since, as was
suggested above, the processes that generate governmental regions are connected
only distantly, if at all, to the territorializing imperatives of the capitalist economy.
In many cases the structures of regional government long predate the contemporary
restructuring of economic space that Scott is describingöGermany is an obvious
example. Where devolution has occurred more recently, its proponents often cite
economic advantages in favour of governmental decentralization (though these are
much debated). In doing so, they may even use the kinds of arguments put forward
by Scott. However, the new territorial structures are not normally determined by
reference to the map of `regional economies', but by fundamentally political considera-
tions. These may include cultural and historical affiliations, administrative convenience,
party advantage, the legacy of past institutional forms, and so on. The kind of strongly
institutionalized regionality described by Paasi and others may develop in some places,
but it is difficult to discern a widespread generic regional condition, even in Europe,
where the notion of regions as bounded wholes seems most at home.

Conclusion
The conceptualization of regions as bounded wholes, and the identification of regionality
as an integral element of contemporary sociospatial change are associated, at least in part,
with two distinct European logics. One, an epistemological logic, links Cartesian ration-
alism to a particular visual regime (the world-as-exhibition). It is expressed in cartographic
anxiety and the desire to corral complex nomad spatialities into coherent and mappable
territorial configurations. The other, a political logic, is related to European integration
and brings together a diverse network of political actors: utopian federalists, European
Union officials, regionalist politicians, and regional development practitioners.
The confluence of these logics has had three effects. The first is the persistent
(though certainly not universal) tendency to conflate the incongruous geographies of
different kinds of regionality into a generic concept of `the' region. The second is the
propensity to elide the distinction between an abstract concept of regions as bounded
wholes (whether atomistic, centred, or decentred) and the territoriality of concrete
social and material processes. The third is the failure to acknowledge the European
origins and contemporary European specificities evident in many purportedly general
theories of regionality.
Cartographic anxiety and the search for regionality 357

I do not wish to imply that cartographic anxiety can be easily shaken off, either
through the psychotherapy of thicker, more situated description, or through the
Prozac of ever more sophisticated geographical information systems, though each
of these, no doubt, has its part to play. If, as Gregory implies, cartographic anxieties
are as much constitutive as they are symptomatic of modern ways of seeing, then the
dilemmas they raise are inescapable. On the other hand, cartographic reason is
neither static nor immutable. The development of what Pickles (2004, page 192) calls
`countermappings' will have a vital role to play in the production of nontotalizing
accounts of regionality.
Such accounts are likely to involve at least four elements. First, research on strong,
highly institutionalized regions needs to be balanced by studies of decentred, weak, or
absent regionality. Although there are difficulties in such counterfactual studies, they
are necessary to obtain a full picture of the variety of forms of regionality and to insure
against tendencies to overgeneralize from particular iconic cases. Second, comparative
research is essential to test whether it makes sense to think of regionality (at least in its
new regionalist forms) as a general condition, evident in all large and complex societies,
or whether it is something that is peculiarly European. Comparative research also helps
to assess the power of specific approaches to studying and conceptualizing regions.
Does regional theory travel well between cultural contexts? Studies of the production
of place and space in different contexts are central to the discipline of geography. Yet,
driven in part by cartographic anxiety, comparative research can easily become a search
for similarity and corroboration, rather than difference. Instead cases should perhaps be
selected explicitly to identify the limitsöconceptual and geographicalöof dominant
models of regionality.
Third, we need to be cautious about the regionalizing activities of both nation-
states and the European Union. Top-down regionalization does not, in itself, signal or
produce a more general strengthening of regionality. Moreover, it arguably feeds
cartographic anxiety by establishing formal, official regional boundaries, sometimes
without regard to preexisting social, cultural, and economic geographies. Most impor-
tantly, though, it should not be seen as vindicating geographers' and other regional
researchers' interest in regions and territoriality. The study of regionality should not
imply the endorsement of regionalism as a political project, yet there is often a warmly
supportive tone to academic writing on regions. In political geography Gearöid
Oè Tuathail (1996) has argued for `anti-geopolitics' to challenge the taken-for-granted
framings of the world often reproduced by policy makers, academics, and the media.
When regionalism also becomes such a taken-for-granted framing, a similarly sceptical
approach from researchers is called for, one that examines critically the political inter-
ests at stake in different forms of regionalization, its differential social effects, and the
variety of subjectivities it generates.
None of this means that cultural, political, and economic processes are without
regional geographies, and still less that regional geography is somehow redundant. On
the contrary, the diverse and distanciated networks, through which all such processes
are constituted, can generate multiple and highly complex patterns of regional differ-
entiation. It is the fact that those patterns so rarely cohere into regional totalities that
makes regional geography so difficultöand so important.
Acknowledgements. Parts of this paper are based on research funded by the Economic and Social
Research Council (award no. L213252031) and supported by research leave granted by the Uni-
versity of Durham. The final version of the paper was mostly written while I held a Visiting
Fellowship in the Department of Human Geography at the Australian National University in
Canberra. The arguments in the paper have benefited greatly from the constructive comments of
358 J Painter

three anonymous referees, and conversations with generous colleagues including Luiza Bialasiewicz,
Stuart Elden, Katherine Gibson, Paul Harrison, Ray Hudson, Deirdre McKay, Gordon MacLeod
and John Pickles. The usual disclaimers apply.
References
Agnew J, 1981, ``Structural and dialectical theories of political regionalism'', in Political Studies
from Spatial Perspectives (John Wiley, Chichester, Sussex) pp 275 ^ 289
Allen J, Massey D, Cochrane A, 1998 Rethinking the Region (Routledge, London)
Althusser L, 1969 For Marx (Allen Lane, London)
Althusser L, 1971, ``Ideology and ideological state apparatuses (notes towards an investigation)'',
in Lenin and Philosophy and Other Essays (Monthly Review Press, New York) pp 127 ^ 186
Amin A, 1999, ``An institutionalist perspective on regional economic development'' International
Journal of Urban and Regional Research 23 365 ^ 378
Amin A, 2004, ``Regions unbound: towards a new politics of place'' Geografiska Annaler, Series B
86 33 ^ 44
Amin A, Thrift N (Eds), 1994 Globalization, Institutions and Regional Development in Europe
(Oxford University Press, Oxford)
Amin A, Thrift N, 2002 Cities: Reimagining the Urban (Polity Press, Cambridge)
Anderson P, 1994, ``The invention of the region 1945 ^ 1990'', WP 94/2, European Institute ^
European Forum, Florence
Bache I, Flinders M (Eds), 2004 Multi-level Governance (Oxford University Press, Oxford)
Barlow M, 1997, ``Administrative systems and metropolitan regions'' Environment and Planning C:
Government and Policy 15 399 ^ 411
Barry A, 2001 Political Machines: Governing a Technological Society (Athlone, London)
Baudrillard J, 1983 Simulations (Semiotext(e), New York)
Bernstein R, 1983 Beyond Objectivism and Relativism: Science, Hermeneutics and Praxis (Blackwell,
Oxford)
Brenner N, 2002, ``Decoding the newest `metropolitan regionalism' in the USA: a critical overview''
Cities 19 3 ^ 21
Brenner N, 2004 New State Spaces: Urban Governance and the Rescaling of Statehood (Oxford
University Press, Oxford)
Brenner N, Jessop B, Jones M, MacLeod G, 2003 State/Space: A Reader (Blackwell, Oxford)
Castells M, 1998 The Information Age: Economy, Society and Culture.Volume 3. End of Millennium
(Blackwell, Oxford)
Centre Europëen de la Culture, 1967/1968 Naissance de l'Europe des Rëgions Centre Europëen de
la Culture, Geneva
Cooke P, Morgan K, 1998 The Associational Economy: Firms, Regions, and Innovation (Oxford
University Press, Oxford)
Cox K R, 2004, ``Globalization and the politics of local and regional development: the question
of convergence'' Transactions of the Institute of British Geographers, New Series 29 179 ^ 194
Cullenberg S, 1999, ``Overdetermination, totality, and institutions: a genealogy of a Marxist
institutional economics'' Journal of Economic Issues 33 801 ^ 815
Dicken P, Kelly P F, Olds K, Yeung H W-C, 2001, ``Chains and networks, territories and scales:
towards a relational framework for analysing the global economy'' Global Networks 1 89 ^ 112
Fall J J, 2005 Drawing the Line: Nature, Hybridity and Politics in Transboundary Spaces (Ashgate,
Aldershot, Hants)
Frisken F, Norris D F, 2001, ``Regionalism reconsidered'' Journal of Urban Affairs 23 467 ^ 478
Gerdes D, 1981, ``Dimensionen des neuen Regionalismus in Westeuropa'' [Dimensions of the new
regionalism in western Europe] Oësterreichische Zeitschrift fu«r Politikwissenschaft 10 305 ^ 318
Gibson K, 2001, ``Regional subjection and becoming'' Environment and Planning D: Society and
Space 19 639 ^ 667
Goodwin M, Painter J, 1996,``Local governance, the crises of Fordism and the changing geographies
of regulation'' Transactions of the Institute of British Geographers, New Series 21 635 ^ 648
Gregory D, 1994 Geographical Imaginations (Blackwell, Oxford)
Grundy-Warr C, 2002, ``Cross-border regionalism through a `south-east Asian' looking glass''
Space and Polity 6 215 ^ 625
Ha«kli J, 1998, ``Cross-border regionalisation in the `new Europe'ötheoretical reflection with two
illustrative examples'' Geopolitics 3 83 ^ 103
Haraway D, 1991 Simians, Cyborgs and Women: The Reinvention of Nature (Routledge, New York)
Cartographic anxiety and the search for regionality 359

Harley J B, 1992, ``Deconstructing the map'', in Writing Worlds: Discourse, Text and Metaphor in
the Representation of Landscape Eds T Barnes, J Duncan (Routledge, London) pp 231 ^ 247
Harvey D, 1985 The Urbanization of Capital (Blackwell, Oxford)
Heffernan M, 1998 The Meaning of Europe: Geography and Geopolitics (Arnold, London)
Hëraud G, 1963 Europe des Ethnies (Presses d'Europe, Paris)
Hooghe L, Marks G, 2001 Multi-level Governance and European Integration (Rowman and
Littlefield, Oxford)
Hooghe L, Marks G, 2003,``Unraveling the central state, but how? Types of multi-level governance''
American Political Science Review 97 233 ^ 243
Hudson R, 2001, ``Regional development, flows of value and governance processes in an enlarged
Europe'',WP-6-01, Regional Economic Performance, Governance and Cohesion in an Enlarged
Europe, Department of Geography, Durham University
Jay M, 1984 Marxism and Totality: The Adventures of a Concept from Lukäcs to Habermas
(University of California Press, Berkeley, CA)
Jeffery C, 1997, ``Farewell to the third level? The German La«nder and the European policy
process'', in The Regional Dimension of the European Union: Towards a Third Level in Europe
Ed. C Jeffery (Frank Cass, London) pp 56 ^ 75
Jessop B, 1995, ``The regulation approach, governance and post-Fordism: alternative perspectives
on economic and political change'' Economy and Society 24 307 ^ 333
Jessop B, 2000, ``Governance failure'', in The New Politics of British Local Governance Ed. G Stoker
(Macmillan, London) pp 11 ^ 32
Jessop B, 2002 The Future of the Capitalist State (Polity Press, Cambridge)
Jones M, 2001, ``The rise of the regional state in economic governance: `partnerships for prosperity'
or new scales of state power?'' Environment and Planning A 33 1185 ^ 1211
Jones M, MacLeod G, 2004, ``Regional spaces, spaces of regionalism: territory, insurgent politics
and the English question'' Transactions of the Institute of British Geographers, New Series
29 433 ^ 452
Keating M, 1995, ``Europeanism and regionalism'', in The European Union and the Regions
Eds B Jones, M Keating (Oxford University Press, Oxford) pp 1 ^ 22
Keating M, 1997, ``The invention of regions: political restructuring and territorial government in
Western Europe'' Environment and Planning C: Government and Policy 15 383 ^ 398
Keating M, 1998 The New Regionalism in Western Europe: Territorial Restructuring and Political
Change (Edward Elgar, Cheltenham, Glos)
Keating M, 2001, ``Governing cities and regions: territorial restructuring in a global age'', in
Global City Regions: Trends, Theory, Policy Ed. A J Scott (Oxford University Press, Oxford)
pp 371 ^ 390
Knight D B, 1982, ``Identity and territory: geographical perspectives on nationalism and
regionalism'' Annals of the Association of American Geographers 72 514 ^ 531
Kofman E, 1985, ``Regional autonomy and the one and indivisible French republic'' Environment
and Planning C: Government and Policy 3 11 ^ 25
Kramsch O T, 2002, ``Reimagining the scalar topologies of cross-border governance: Eu(ro)regions
in the post-colonial present'' Space and Polity 6 169 ^ 196
Krishna S, 1994, ``Cartographic anxiety: mapping the body politic in India''Alternatives 19 507 ^ 521
Lee R, 1985, ``The future of the region: regional geography as education for transformation'',
in Geographical Futures Ed. R King (The Geographical Association, Sheffield) pp 77 ^ 91
Loughlin J, 1996, `` `Europe of the regions' and the federalization of Europe'' Publius: The Journal
of Federalism 26 141 ^ 162
MacLeod G, 1998, ``In what sense a region? Place hybridity, symbolic shape, and institutional
formation in (post-) modern Scotland'' Political Geography 17 833 ^ 863
MacLeod G, 1999, ``Space, scale and state strategy: rethinking urban and regional governance''
Progress in Human Geography 23 503 ^ 527
MacLeod G, 2001a,``Beyond soft institutionalism: accumulation, regulation, and their geographical
fixes'' Environment and Planning A 33 1145 ^ 1167
MacLeod G, 2001b, ``New regionalism reconsidered: globalization and the remaking of political
economic space'' International Journal of Urban and Regional Research 25 804 ^ 829
MacLeod G, Goodwin M, 1999, ``Reconstructing an urban and regional political economy:
on the state, politics, scale, and explanation'' Political Geography 18 697 ^ 730
Marston S, Jones J P III, Woodward K, 2005, ``Human geography without scale'' Transactions
of the Institute of British Geographers, New Series 30 416 ^ 432
Massey D, 1991, ``A global sense of place'' Marxism Today June, 24 ^ 29
360 J Painter

Massey D, 1993, ``Power-geometry and a progressive sense of place'', in Mapping the Futures: Local
Cultures, Global Change Eds J Bird, B Curtis, T Putnam, G Robertson, L Tickner (Routledge,
London) pp 59 ^ 69
Massey D, 1999, ``Power-geometries and the politics of space ^ time: Hettner-Lecture 1998'',
Department of Geography, University of Heidelberg
Massey D, 2001, ``Geography on the agenda'' Progress in Human Geography 25 5 ^ 17
Massey D, 2005 For Space (Sage, London)
Mawdsley E, 1997, ``Nonsecessionist regionalism in India: the Uttarakhand separate State
movement'' Environment and Planning A 29 2217 ^ 2235
Mitchell T, 1991 Colonising Egypt (University of California Press, Berkeley, CA)
Mitchell T,1992,``Orientalism and the exhibitionary order'', in Colonialism and Culture Ed. N B Dirks
(University of Michigan Press, Ann Arbor, MI) pp 289 ^ 317
Murray R, 1992, ``Europe and the new regionalism'', in Cities and Regions in the New Europe
Eds M Dunford, G Kafkalas (Belhaven Press, London) pp 299 ^ 308
Nordstrom C, 2003, ``Shadows and sovereigns'' Theory, Culture and Society 17 35 ^ 54
Norris D F, 2001, ``Prospects for regional governance under the new regionalism: economic
imperatives versus political impediments'' Journal of Urban Affairs 23 557 ^ 571
Olsson G, 1991 Lines of Power/Limits of Language (University of Minnesota Press, Minneapolis,
MN)
Oè Tuathail G, 1996 Critical Geopolitics: The Politics of Writing Global Space (Routledge, London)
Paasi A, 1986, ``The institutionalization of regions: a theoretical framework for understanding
the emergence of regions and the constitution of regional identity'' Fennia 164 104 ^ 146
Paasi A, 1991, ``Deconstructing regions: notes on the scales of spatial life'' Environment and
Planning A 23 239 ^ 256
Paasi A, 1996 Territories, Boundaries and Consciousness: The Changing Geographies of the Finnish ^
Russian Border (John Wiley, Chicheser, Sussex)
Paasi A, 1999, ``Boundaries as social processes: territoriality in a world of flows'', in Boundaries,
Territory and Postmodernity Ed. D Newman (Frank Cass, London) pp 69 ^ 88
Paasi A, 2002, ``Bounded spaces in the mobile world: deconstructing `regional identity' '' Tijdschrift
voor Economische en Sociale Geografie 93 137 ^ 148
Pickles J, 2004 A History of Spaces: Cartographic Reason, Mapping and the Geo-coded World
(Routledge, London)
Pred A, 1984,``Place as historically contingent process: structuration theory and the time-geography
of becoming places'' Annals of the Association of American Geographers 74 279 ^ 297
Putnam R D, 1993 Making Democracy Work: Civic Traditions in Modern Italy (Princeton University
Press, Princeton, NJ)
Rhodes R A W, 1974, ``Regional policy and a `Europe of Regions': a critical assessment'' Regional
Studies 8 105 ^ 114
Sack R D, 1986 Human Territoriality (Cambridge University Press, Cambridge)
Saint-Ouen F, 1998 Denis de Rougemont et l'Europe des Rëgions Fondation Denis de Rougemont,
Geneva
Sancton A, 2001, ``Canadian cities and the new regionalism'' Journal of Urban Affairs 23 543 ^ 555
Sayer A, 1989, ``The `new' regional geography and problems of narrative'' Environment and
Planning D: Society and Space 7 253 ^ 276
Scott A J, 1998 Regions and the World Economy (Oxford University Press, Oxford)
Sheppard E, 2000, ``Competition in space and between places'', in A Companion to Economic
Geography Eds E Sheppard, T Barnes (Blackwell, Oxford) pp 169 ^ 186
Smith G, 1985, ``Nationalism, regionalism, and the state'' Environment and Planning C: Government
and Policy 3 3 ^ 9
Smith G E (Ed.), 1995 Federalism. The Multi-ethnic Challenge (Longman, Harlow, Essex)
Sparke M, 2002, ``Between post-colonialism and cross-border regionalism'' Space and Polity
6 203 ^ 213
Storper M, 1995, ``The resurgence of regional economies, ten years later: the region as a nexus
of untraded interdependencies'' European Urban and Regional Studies 2 191 ^ 221
Storper M, 1997 The Regional World: Territorial Development in a Global Economy (Guilford Press,
New York)
Sunley P, 2000, ``Urban and regional growth'', in A Companion to Economic Geography
Eds E Sheppard, T Barnes (Blackwell, Oxford) pp 187 ^ 201
Swanstrom T, 2001, ``What we argue about when we argue about regionalism'' Journal of Urban
Affairs 23 479 ^ 496
Cartographic anxiety and the search for regionality 361

Swyngedouw E, 2000,``Authoritarian governance, power, and the politics of rescaling'' Environment

and Planning D: Society and Space 18 63 ^ 76
Swyngedouw E, 2004, ``Globalisation or `glocalisation'? Networks, territories and rescaling''
Cambridge Review of International Affairs 17 25 ^ 48
Taylor P J, 1994, ``The state as container: territoriality in the modern world-system'' Progress in
Human Geography 18 151 ^ 162
Thayer R L, 2003 LifePlace: Bioregional Thought and Practice (University of California Press,
Berkeley, CA)
Thrift N, 1990, ``For a new regional geography 1'' Progress in Human Geography 14 272 ^ 279
Thrift N, 1991, ``For a new regional geography 2'' Progress in Human Geography 15 456 ^ 465

ß 2007 a Pion publication printed in Great Britain

Conditions of use. This article may be downloaded from the E&P website for personal research
by members of subscribing organisations. This PDF may not be placed on any website (or other
online distribution system) without permission of the publisher.