cultures and ‘natures’ – that articulate the
desires, fantasies and fears of their authors
and the grids of power between them and
their ‘Others’. The concept is not confined to
ostensibly fictional works. On the contrary,
there is an important sense in which all geog-
raphies are imaginative: even the most formal,
geometric lattices ofspatial scienceor the
most up-to-date and accurate mapsare at
once abstractions and cultural constructions,
and as such open to critical readings.
The concept was originally proposed by the
Palestinian/American literary critic Edward
Said (1935–2003) in his influential critique
of orientalism (Said, 2003 [1978]; see
Gregory, 1995a). His emphasis on power
(and in particular colonial power) was alien
to the concepts ofenvironmental percep-
tionandmental mapsthen current inbeha-
vioural geography, and underscored the
radical ‘non-innocence’ of representation.
In some measure, Said’s formulation antici-
pated ideas ofsituated knowledge: he was
concerned to disclose the privileges that
European and American authors typically
claimed when representing other cultures, to
chart the asymmetric grid of power within
which (as he put it) ‘the West’ watches, ‘the
East’ is watched, and hence to criticize the
partialities of their constructions.
Said’s emphasis on viewing, watching, look-
ing, observing – onvision and visuality–
drew attention to the cultural construction of
the gaze. While Said’s critique of Orientalism
was shot through with visual metaphors, the
imaginative geographies with which he was
centrally concerned were textual. Human
geographers have been drawn to both the text-
ual and the visual image, however, including
art forms such as film and photography
(Schwartz and Ryan, 2003), and they have
drawn attention to viewing as an embodied,
vitally sensuous practice (Martins, 2000). Un-
like the other constructs of behavioural geog-
raphy, therefore, none of these imaginative
geographies are seen as the product of purely
cognitive operations. As cultural constructs,
their images are animated by fantasy, desire
and the unconscious, and indeed the very
idea of an ‘imagination’ has been extended
through geographies indebted to various
forms of psychoanalytic theory (Pile,
1998). These images carry within them com-
parative valorizations – what Said called a
‘poetics of space’ – by means of which places
are endowed with ‘figurative value’. Such
constructions also involve a poetics (and a
politics) ofnature, and there has been con-
siderable interest in recovering imaginative
geographies of other ‘natures’ as well as other
cultures. At the limit, these distinguish a ‘nor-
mal’, temperate nature from other, intemper-
ate natures and install their own cultural
subtext about those who inhabit such ‘unnat-
ural natures’ (Gregory, 1995b; Sioh, 1998; see
alsotropicality).
Said claimed that these figurative values
enter not only into the production of ‘other-
ness’ but also into the identity-formation of
the viewing subject in a complexdialectic.
Imaginative geographies thus sustain images
of ‘home’ as well as ‘abroad’, ‘our space’ as
well as ‘their space’: ‘Imaginative geography
and history help the mind to intensify its own
sense of itself by dramatizing the distance and
difference between what is close to it and what
is far away’ (Said, 1978, p. 55). Hence Orford
(2003) showed that imaginative geographies
are regularly mobilized to separate the space
of ‘the international community’ from the
space of those facing security crises, so that
intervention is always after the event and con-
structed as selflessly humanitarian; she argues
that this separation is achieved through tactics
of localizing and distancing (‘their space’) that
work not only to legitimize the actions of the
global North as so many virtuous efforts to
find a solution, but also in many cases to ob-
scure the active involvement of its inter-
national actors in thegenerationof the crisis.
Similarly, Graham (2006) contrasts the im-
aginative geographies that have been used to
separatehomelandcities from ‘terror’, ‘tar-
get’ or Arab cities during the ‘war on terror’,
even as they are also being integrated through
techno-science ‘into a single, transnational
battlespace’. Studies like these show why
Gregory (2004b, p. 256) concludes that ‘im-
aginative geographies are doubled spaces of
articulation’ whose ‘inconstant topologies are
mappings of connective dissonance in which
connections are elaborated in some registers
even as they are disavowed in others’.
‘Dramatization’ is not the same as ‘falsifica-
tion’,however,and Said’s discussion under-
cuts the distinction between ‘real’ and
‘perceived’ worlds on which behavioural geog-
raphy depended. This is the most complicated
and contentious part of Said’s argument.
There are certainly passages where he con-
trasted what he called ‘positive geographies’
with imaginative geographies produced under
the sign of Orientalism. And yet, if imaginative
geographies are ‘fictions’ in the original Latin
sense offictio– something made, something
fabricated – this does not mean that they areGregory / The Dictionary of Human Geography 9781405132879_4_I Final Proof page 370 31.3.2009 7:05pmIMAGINATIVE GEOGRAPHIES