Professional Documents
Culture Documents
In many languages the word territory typically refers to a unit of contiguous space that is
used, organized and managed by a social group, individual person or institution to restrict
and control access to people and places. Though sometimes the word is used as
synonymous with place or space, territory has never been a term as primordial or as generic
as they are in the canons of geographical terminology (Agnew 2005a). The dominant usage
has always been either political, in the sense of necessarily involving the power to limit
access to certain places or regions, or ethological, in the sense of the dominance exercised
over a space by a given species or an individual organism. Increasingly, territory is coupled
with the concept of network to help understand the complex processes through which space
is managed and controlled by powerful organizations. In this light, territory is only one type
2
of spatiality, or way in which space is used, rather than the one monopolizing its
employment. From this perspective, territoriality is the strategic use of territory to attain
organizational goals. It is only one way of organizing space.
In this paper I begin by exploring how territory and territoriality operate as modes
of spatiality, or conceptions of the uses of space in the social sciences. I then argue that
territory has become fatefully tied to the modern state, particularly in English-language
understandings. Finally, I suggest that two further modes of spatiality, spatial interaction
and place-making, provide analytically important ways of thinking about space and society
beyond the limitations imposed by a geographical imagination limited by a singularly
territorial conception of spatiality.
Spatiality
Territory is particularly associated with the spatiality of the modern state with its
claim to absolute control over a population within carefully defined external borders
(Buchanan and Moore, 2003, p. 6; Agnew 2005b). Indeed, until Sack (1986) extended the
understanding of human territoriality as a strategy to individuals and organizations in
general, usage of the term territory was largely confined to the spatial organization of
states. In the social sciences such as economics, sociology and political science this is still
mainly the case, such that the challenge posed to territory by network forms of organization
(associated with globalization) is invariably characterized in totalistic terms as ‘the end of
geography.’ This signifies the extent to which territory has become the dominant
geographical term (and imagination) in the social sciences (Badie, 1995). It is then closely
allied to state sovereignty and, sometimes, to an entirely nested, scale-based territorial
conception of space (from the local and the urban through the national to the global). Thus,
as sovereignty is seen to ‘erode’ or ‘unbundle,’ so it seems goes territory (Agnew 1994).
From this viewpoint territory takes on an epistemological monopoly that is understood as
absolutely fundamental to modernity. As such, it can then be given an extended meaning to
refer to any socially constructed geographical space, not just that resulting from statehood,
and can be used as equivalent to the term place in many languages including French,
3
Spanish and Italian (Scivoletto, 1983; Bonnemaison, 1996; Storper 1997). Especially
popular with some French-language geographers, this usage often reflects the need to adopt
a term to distinguish the particular and the local from the more general global or national
‘space.’ It then signifies the ‘bottom-tier’ spatial context for identity and cultural difference
more than a simple ‘top-down’ connection between state and territory but still within an
encompassing territorialized conception of spatiality. In absolute counterpoint, some
proponents of a postmodern conception of space see that space as completely “flat” without
any sort of territorial division or hierarchies whatsoever (e.g. Marston et al. 2005) and thus
provide a totally opposite but equally singular view of spatiality, albeit this time of
localized sites in a networked spatial topology rather than of an absolute territorialized
space.
Territoriality in its broadest sense, then, is either the organization and exercise of
power, legitimate or otherwise, over blocs of space or the organization of people and things
into discrete areas through the use of boundaries. In studies of animal behavior spatial
division into territories is seen as an evolutionary principle, a way of fostering competition
so that those best matched to their territory will have more surviving offspring. With
human territoriality, however, spatial division is more typically thought of as a strategy
used by organizations and groups to manage social, economic and political activities. From
this viewpoint, space is partitioned into territorial cells or units that can be relatively
autonomous (as with the division of global space into territorial nation-states) or arranged
hierarchically from basic units in which work, administration, or surveillance is carried out
through intermediate levels at which managerial or supervisory functions are located to the
top-most level at which central control is concentrated. Alternative spatialities of political
and economic organization, particularly hierarchical networks (as in the world-city
network) or reticular networks (as with the Internet), can challenge or supplement the use
of territoriality.
At least four models of the spatiality of power can be identified. I draw here on the
work of the French geographers Marie-Françoise Durand, Denis Retaillé and Jacques Lévy
(e.g. Durand et al. 1992) who have used idealized models of economic and cultural patterns
4
and interaction to understand long-term shifts in world politics. Each of their models is
closely associated with sets of political-economic/technological conditions and associated
cultural understandings. The logic of the approach is that the dominant spatiality of power
will change as material conditions and associated modes of understanding of them change.
Such processes of change are not construed as entirely spontaneous. Rather, this approach
to the historicity of spatiality implies that both material forces and intellectual perspectives
or representations interact in a dominant set of practices or hegemony to produce the
spatiality of power predominant within a given historical era. But each spatial model also
has a synchronic validity in the sense that political power in any epoch can never be totally
reduced to any one of them. In a sense equivalent to Karl Polanyi’s discussion of market
society in terms of the emergence of market exchange at the expense of reciprocity and
redistribution as principles of economic integration, as one model comes to predominate
others are not so much eclipsed as placed into subordinate or emerging roles. The models
offer, then, not only a way of historicizing political power but also of accounting for the
complexity of the spatiality of power during any particular historical epoch (Figure 1).
5
(Figure 1)
dynasties and seasons replacing one another in natural sequence. Political power is largely-
internally oriented and directed towards dynastic maintenance and internal order. Its
spatiality rests on a strongly physical conception of space as distance to be overcome or
circulation to be managed.
In contrast, is the geopolitical model of states in a “field of forces.” It revolves
around rigidly defined territorial units in which each state can gain power only at the
expense of others and each has total control over its own territory. It is akin to a field of
forces in mechanics in which the states exert force on one another and the outcome of the
mechanical contest depends on the populations and resources each can bring to bear.
Success also depends on creating blocs of allies or clients and identifying spatial points of
weakness and vulnerability in the situation of one’s adversaries. All of the attributes of
politics, such as rights, representation, legitimacy and citizenship, are restricted to the
territories of individual states. The presumption is that the realm of geopolitics is beyond
such concerns. Force and the potential use of force rule supreme beyond state boundaries.
Time is ordered on a rational global basis so the trains can run on time, workers can get to
work on time and military forces can coordinate their activities. The dominant spatiality,
therefore, is that of state-territoriality, in which political boundaries provide the containers
for the majority of social, economic and political activities. Political elites are state elites
and they mimic one another’s discourse and practices.
Third on the list of models is that of the “hierarchical network.” This is the spatial
structure of a world-economy in which cores, peripheries and semi-peripheries are linked
together by flows of goods, people and investment. Transactions based largely on market
exchange produce patterns of uneven development as flows move wealth through networks
of trade and communication producing regional concentrations of relative wealth and
poverty. At the local scale, particularly that of urban centers, hinterlands are drawn into
connection with a larger world which has become progressively more planetary in
geographical scope over the past five-hundred years. Political power is a function of where
in the hierarchy of sites from global centers to rural peripheries a place is located. Time is
organized by the geographical scope and temporal rhythm of financial and economic
7
transactions. The spatiality is of spatial networks joining together a hierarchy of nodes and
areas which are connected by flows of people, goods, capital and information. Today, such
networks are particularly important in linking together the city-regions which constitute the
nodes around which the global economy is increasingly organized. In some circumstances,
networks can develop a reticular form in which there is no clear center or hierarchical
structure. This is the case, for example, with the networks implicit in some business
models, such as strategic alliances, in which partnership over space rather than
predominance between one node and the others prevails and, more notoriously, in some
global terrorist and criminal networks.
The fourth, and final, model is that of the “integrated world society.” This
conforms to the humanistic ideal of a world in which cultural community, political identity
and economic integration are all structured at a global scale. But it also reflects the
increased perception of common global problems (such as environmental ones) that do not
respect state borders, the futility of armed inter-state conflict in the presence of nuclear
weapons and the advantages of defense over offense in modern warfare, and the growth of
an international “public opinion.” This model privileges global scale communication based
on networks among multiple actors that are relatively unhierarchical or reticular and more
or less dense depending upon the volition of actors themselves. The sprout-like character
of these connections leads some to see them as (in a term popularized by Gilles Deleuze)
somewhat like the “rhizomes” of certain plants that spread by casting out shoots in multiple
but unpredictable directions. Time and space are both defined by the spontaneous and
reciprocal timing and spacing of human activities. Real and virtual spaces become
indistinguishable. This model obviously has a strong utopian element to it but does also
reflect some emergent properties of the more interconnected world that is presently in
construction.
In the contemporary world there is evidence for the effective co-presence of each of
these models with the former territorial models somewhat in eclipse and the latter network
models somewhat in resurgence after a one hundred-year period in which the field of forces
model was pre-eminent (if hardly exclusive). If the trend towards regional separatism
8
within existing states portends a fragmentation that can reinforce the field of forces model
as new states emerge, then economic globalization and global cultural unification work to
reinforce the hierarchical network and integrated world society models. At the same time
movement towards political-economic unification (as in the European Union) and the
development of cultural movements with a strong territorial element (as with Islamic
integralist movements) tend to create pressures for the reassertion of an ensemble of
worlds. Historically, however, there has been a movement from one to another model as a
hegemonic or directing element. In this spirit I would propose a theoretical scheme
drawing from the work of Durand et al. in which, first of all, the “ensemble of worlds”
model slowly gave way to the “field of forces” model around 1500 AD as the European
state system came into existence (Figure 2).
Figure 2
9
But since then the hegemonic influence of the different models has tended to vary
geographically, so that by the nineteenth century a balance-of-power hegemony between
territorial states was dominant in Europe. Imperial hegemonies, however, were uppermost
in much of the rest of the world save for the public goods hegemony exercised by Britain
through its roles as upholder of the gold standard and entrepôt in a multilateral trading
system that unified an emerging world economy. As this model was establishing its
dominance, the modern “hierarchical network” also began its rise in and around the
framework provided by the state system. Under European colonialism the part of the world
in which states recognized one another as legitimate actors (what is now often called the
Global North) was divorced from the regions in which such status was denied. With
Independence after the Second World War numerous new states, irrespective of their
relative political efficacy, spread to cover most of the world’s land area. But many of these
new states were either clients of the United States or the Soviet Union – within two sphere-
of -influence hegemonies – or located in violent zones of conflict between them. In the
field-of-forces, therefore, these were hardly equal forces. Since 1945 the hierarchical-
network model has become more and more central to the distribution of political power as a
result of the increased penetration of state territories by global trade, population and
investment flows under an increasingly unilateral US hegemony. This is now a truly
planetary hegemony – the first in history – both with respect to its potential geographical
scope and to the range of its functional influence, based on the tenets of marketplace
society, even as its primary agent, the United States, may itself become less central to it.
With the end of the Cold War, which had produced an important reinstatement of the field
of forces model among the most powerful states, the hierarchical network model is in the
ascendancy with signs of the beginning of a trend towards an “integrated world” society
model. But this is as yet very much in its infancy. This framework is, of course, only
suggestive of long-term tendencies. What it does provide is a sense of the historical
spatiality of political power, associated in different epochs with different dominant modes
of spatiality and the co-presence of others. Ideal-types are a way of thinking about the
world, not to be used as a substitute for its actual complexities at any moment in any place.
10
necessity for polities to be organized territorially. As Hendrik Spruyt (1994, 34) claims, “If
politics is about rule, the modern state is verily unique, for it claims sovereignty and
territoriality. It is sovereign in that it claims final authority and recognizes no higher source
of jurisdiction. It is territorial in that rule is defined as exclusive authority over a fixed
territorial space. The criterion for determining where claims to sovereign jurisdiction begin
or end is thus a purely geographic one. Mutually recognized borders delimit spheres of
jurisdiction.”
Territoriality, the use of territory for political, social, and economic ends, is in fact,
as I mentioned previously, a strategy that has developed more in some historical contexts
than in others. Thus, the territorial state as it is known to contemporary political theory
developed initially in early modern Europe with the retreat of non-territorial dynastic
systems of rule and the transfer of sovereignty from the personhood of monarchs to discrete
national populations. That modern state sovereignty as usually construed did not occur
overnight following the Peace of Westphalia in 1648 is now well established.
Territorialization of political authority was further enhanced by the development of
mercantilist economies and, later, by an industrial capitalism that emphasized capturing
powerful contiguous positive externalities from exponential distance-decay declines in
transportation costs and from the clustering of external economies (material mixes, social
relations, labor pools, etc.) within national-state boundaries.
Absent such conditions, sovereignty–in the sense of the socially constructed
practices of political authority--may be exercised non-territorially or in scattered pockets
connected by flows across space-spanning networks. From this viewpoint, sovereignty can
be practiced in networks across space with distributed nodes in places that are either
hierarchically arranged or reticular (without a central or directing node). In the former case,
authority is centralized, whereas in the latter, it is essentially shared across the network. All
forms of polity–from hunter-gatherer tribes through nomadic kinship structures to city-
states, territorial states, spheres of influence, alliances, trade pacts, seaborne empires--
therefore, occupy some sort of space. What is clear, however, if not widely recognized
14
within contemporary debates about state sovereignty, is that political authority is not
necessarily predicated on and defined by strict and fixed territorial boundaries.
Two issues are crucial here: that political authority is not restricted to states, and
that such authority is thereby not necessarily exclusively territorial. Authority is the
legitimate exercise of power. The foundation and attribution of legitimacy to different
entities has changed historically. By way of example, the legitimacy of rule by monarchs in
the medieval European order had a different meaning from that of later absolutist rulers and
that operating under more recent democratic justifications for state power. In no case,
however, has the authority of the state ever been complete. There have always been
competing sources of authority, from the church in the medieval context to international
organizations, social movements, businesses, and NGOs today. More specifically,
transparency, efficiency, expertise, accountability, and popularity are as much foundations
of legitimacy as are nationality and democratic process. Thus, even ostensibly private
entities and supranational governments are often accorded as great or even greater authority
than are states. Think, for example, of credit rating agencies, charitable organizations such
as Human Rights Watch and the European Union. Using two countries as examples, within
the United States there is widespread popular suspicion of the efficiency and accountability
of the federal government, not just since the military debacle in Iraq and the pathetic
response to Hurricane Katrina. This often leads to perhaps excessive faith in the virtue of
privatization through corporate networks of what are elsewhere seen as “public” services
such as health care. In Italy, much of the popular enthusiasm for the European Union is
driven by the hope that Brussels will increasingly supplant Rome as the seat of power most
effective in relation to people’s everyday lives not so much territorially as in relation to the
functional effects in particular places of European-wide initiatives.
b) Place-Making
My second point about needing to diminish the overall emphasis on territoriality as
if it referred to spatiality tout court involves a rather different focus. This is the
significance of the human experience of space reflected at least in English language usage
15
of the word ‘place.’ In this perspective, space is bracketed, or put to one side, because its
“abstractness discourages experiential explorations” (Casey 2001, 683). In his
philosophical rehabilitation of place, Edward Casey (1997, x) notes how “place has been
assimilated to space. … As a result, place came to be considered a mere ‘modification’ of
space (in Locke’s revealing term) - a modification that aptly can be called ‘site,’ that is,
leveled-down, monotonous space for building and other human enterprises” (author’s
emphasis). Casey’s goal is to argue for the crucial importance of place in much thinking
about community and the public sphere, even though the connections are often not made
explicit by the thinkers in question. He wants to make place different from site and space,
even though he acknowledges Michel Foucault’s point that the modern world is largely one
of Leibnizian sites and relations rather than Newtonian absolute spaces (Casey 1997, 298-
300). In rethinking space as place, his primary interest lies in phenomenologically or
experientially linking places to human selves (also see Entrikin 1991; 2001). The central
issue is that of “being in place differently” (Casey 1997, 337) conditioning the various
dimensions of selfhood, from the bodily to the psychological, institutional, and
architectural. So, though the “shape” of place has changed historically, it is now no mere
container but, rather, a taking place, its rediscovery and naming as such is long overdue.
Thus: “Despite the seduction of endless space (and the allure of serial time), place is
beginning to escape from its entombment in the cultural and philosophical underworld of
the modern West” (Casey 1997, 339).
Symptomatic of the conceptual separation of space and place are the three dominant
meanings that geographical place has acquired in writing that invokes either space or place
(Agnew 1987, 1989; 1993). Each meaning tends to assimilate place to one or the other end
of a continuum running from nomothetic (generalized) space at one end to idiographic
(particularistic) place at the other. The first is place as location or a site in space where an
activity or object is located and which relates to other sites or locations because of
interaction and movement between them. A city or other settlement is often thought of this
way. Somewhere in between, and second, is the view of place as locale or setting where
everyday-life activities take place. Here the location is no mere address but the where of
16
social life and environmental transformation. Examples would be such settings from
everyday life as workplaces, homes, shopping malls, churches, etc. The third is place as
sense of place or identification with a place as a unique community, landscape, and moral
order. In this construction, every place is particular and, thus, singular. A strong sense of
“belonging” to a place, either consciously or as shown through everyday behavior such as
participating in place-related affairs, would be indicative of “sense of place.”
Attempts at putting space and place together must necessarily try to bring at least
two of these various meanings of geographical place together. Currently, there are four
main ways in the Anglo-American and French literature in which this task has been
approached: the humanist or agency-based (e.g. Sack 1997), the neo-Marxist (e.g. Lefebvre
1991), the feminist (e.g. Massey 1994), and the contextualist-performative (e.g. Thrift
1999). Each of these rejects the either/or logic in relation to space and place that has
characterized most geographic and social thought from the seventeenth century to the
present (Agnew 2005a). For the first, and one with which I am most in sympathy, the focus
lies in relating location and locale to sense of place through the experiences of human
beings as agents. In one of the most sophisticated statements of this perspective, Robert
Sack (1997, 58) provides the essential thrust when he writes that his “framework draws on
the geographical experiences of place, space, home, and world which people use in their
lives to integrate forces, perspectives, and selves.” From this point of view:
Place implies space, and each home is a place in space. Space is a property of the natural world,
but it can be experienced. From the perspective of experience, place differs from space in terms of
familiarity and time. A place requires human agency, is something that may take time to know, and
a home especially so. As we move along the earth we pass from one place to another. But if we
move quickly the places blur; we lose track of their qualities, and they may coalesce into the sense
that we are moving through space. This can happen even in my own home. If I am hardly there and
do not attend to its contents, it may seem unfamiliar to me, more like a part of space than a place
(Sack 1997, 16).
17
In this frame of reference, cultural differences, for example, emerge because of place-based
experiences and human agency but also because places are never separate but always part
of larger sets of places across which differences are more or less pronounced depending on
the permeability of boundaries between places as people experience them. Places are
woven together through space by movement and the network ties that produce places as
changing constellations of human commitments, capacities, and strategies. Places are
invariably parts of spaces and spaces provide the resources and the frames of reference in
which places are made.
In a recent research project on Italian electoral politics since the late 1980s, I and
my colleague Michael Shin (2008) have made the case for contexts of “place and time” in
accounting for what has transpired nationally in terms of the rise and fall of the various
political groupings. We argue that these are are not best thought of as invariably regional,
local, or national although they frequently have elements of one, several, or all. Rather,
they are best considered as always located somewhere, with some contexts more stretched
over space (such as means of mass communication and the spatial division of labor) and
others more localized (school, workplace, and residential interactions). The balance of
influence on political choices between and among the stretched and more local contextual
processes can be expected to change over time, giving rise to subsequent shifts in political
outlooks and affiliations. So, for example, as foreign companies introduce branch plants,
trade unions must negotiate new work practices, which, in turn, erode long-accepted views
of the roles of managers and employees. In due course, this configuration of contextual
changes can give an opening to a new political party or a redefined old one that upsets
established political affiliations. But changes must always fit into existing cultural
templates that often show amazing resilience as well as adaptation. Doreen Massey (1999,
22) puts the overall point the best when she writes: “This is a notion of place where
specificity (local uniqueness, a sense of place) derives not from some mythical internal
roots nor from a history of isolation – now to be disrupted by globalization – but precisely
from the absolute particularity of the mixture of influences found together there.”
18
We have used the term place, therefore, to capture the mediating role of such
geographically located milieux. What we mean by this word are the settings in which
people find themselves on a regular basis in their daily lives where many contexts come
together and with which they may identify. Or, as I have made the point previously
(Agnew, 2002: 21): “places are the cultural settings where localized and geographically
wide-ranging socioeconomic processes that condition actions of one sort or another are
jointly mediated. Although there must be places, therefore, there need not be this particular
place.” So, if, in this case, individual persons are in the end the agents of politics, their
agency and the particular forms it takes flow from the social stimuli, political imaginations,
and yardsticks of judgment they acquire in the ever-evolving social webs in which they are
necessarily enmeshed and which intersect across space in particular places. Mair (2006,
44) suggests that as party affiliations have weakened over the past thirty years in most
European countries, voting behavior is “increasingly contingent.” From our perspective,
this means that geographical patterns of turnout and affiliation will become more unstable
even as they often still respond to place-based if evolving norms of participation and
differing relative attraction to the offerings of different parties. Maps of the results from the
proportional representation parts of the 2001 and 2006 elections to the Italian Chamber of
Deputies show something of this geographical dynamic (Figure 3).
19
Figure 3
Conclusion
Clearly, there are important cultural and historical dimensions to both practices and
theories of spatiality and territoriality. Churches and polities (states, empires, federations,
etc.) have been the most important users of territoriality. Some churches (such as the
Roman Catholic Church) and some states (such as the United States) have more complex
and formally hierarchical territorialities than do others. Today, transnational and global
businesses erect territorial hierarchies that cut across existing political ones. So, even as
some uses of territoriality attenuate or even fade away, others emerge. Though varying in
precise form and complexity, therefore, territoriality seems always to be with us as an
important strategy for organizing human activities even as it must be considered alongside
other types of spatiality, such as interaction across space and place-making, that both direct
20
and give agency to human social existence. But as the modes of analysis and empirical
examples from my recent publications I have introduced today suggest, we must reject the
confusion of territoriality with spatiality, or how space is defined and used socially, and be
much clearer in our use of spatial terminology such as territory, space and place.
References
Agnew, J. A. 1987: Place and Politics: The Geographical Mediation of State and
Society. London: Allen and Unwin.
Agnew, J. A. 1989: The devaluation of place in social science. In J. Agnew and J.
Duncan (eds.) The Power of Place: Bringing Together Geographical and
Sociological Imaginations. London: Unwin Hyman.
Agnew, J. A. 1993: Representing space: space, scale and culture in social science. In J.
Duncan and D. Ley (eds.) Place/Culture/Representation. London: Routledge.
Agnew, J.A. 1994: The territorial trap: the geographical assumptions of international
relations theory. Review of International Political Economy 1, pp. 53-80.
Agnew, J. A. 2002: Place and Politics in Modern Italy. Chicago: University of Chicago
Press.
Agnew, J.A. 2003: Territoriality and political identity in Europe. In M. Berezin and M.
Schain (eds.) Europe without borders: remapping territory, citizenship, and
identity in a transnational age. Baltimore: Johns Hopkins University Press.
Agnew, J.A. 2005a: Space-place. In P. Cloke and R. Johnston (eds.) Spaces of
Geographical Thought. London: Sage.
Agnew, J.A. 2005b: Sovereignty regimes: territoriality and state authority in
contemporary world politics. Annals of the Association of American
Geographers, 95/2, pp. 437-61.
Badie, B. 1995: La fin des territoires. Paris: Fayard.
Barth, F. 1969: Introduction. In F. Barth (ed.) Ethnic Groups and Boundaries. London:
21