view, he articulates a political geography of
dog-stealing characterized by class antagonisms
and exploitation of rich by poor, and deeply
ingrained practices of ‘domestication’ itself – of
both dogs and women confined by Victorian
ideals of femininity, obedienced to male authority
and middle-class domesticity.
Critiques of science provided another avenue
to rethinking the animal–human nexus, particu-
larly the question of animal subjectivity.
Scholars in science studies, as well as some
self-reflexive scientists (see Birke and
Hubbard, 1995), promoted a critical reading of
scientific results and emphasized how social
interactions shape their content. Haraway (1989)
and others argued that foundational primatology
texts were ethnocentric and androcentric, and
questioned whether knowledge of the ‘natural’
world and its non-human inhabitants could be
anything other than partial and subjective (Braun
and Castree, 1998; Demeritt, 1994; Willems-
Braun, 1997).
Such arguments about partial perspective and
subjectivity in our knowledge of the natural world,
the social construction of nature and
animals, and shifting networks linking humans,
nature and quasi-objects, prompted responses
ranging from cautionary to hostile among natural
scientists (and some social scientists: see
Murdoch, 1997). Natural scientists worried about
potential practical implications for animals and
nature more generally, arguing that science
critics addressed inadequately, or failed to address
altogether, the dilemma that they themselves
posed. If nature and human knowledge of it is
fragmented, partial and socially constructed,
could people take responsibility for those sen-
tient life forms historically relegated to the
‘natural’ realm – whose habitats were being so
deeply degraded by anthropogenic change? By
stripping scientists of any authority or special
knowledge, critics of science effectively denied a
powerful voice endeavouring to speak thor-
oughly, thoughtfully and with some accuracy for
non-human life. A real prospect was further mar-
ginalization and silencing of these already
oppressed animal ‘others’ (Cronon, 1995;
Shepard, 1995; Soule and Lease, 1995).
This response sparked a series of ‘science wars’
noted for vicious caricatures of both realism
and relativism/constructivism. Some in
science studies sought to move beyond this
binary (Rouse, 1992), by showing how knowl-
edge is neither just socially constructed nor
simply real, and developing new vocabularies to
incorporate non-human agency into social analy-
sis (Whatmore, 1999: 27). Within such perfor-
mative approaches, both subjectivity and agencyare relational effects generated by a network of
heterogeneous, interacting, materials rather than
inherently innate properties and powers. But
debates continue, clearly sensitizing animal
geographers to the centrality of knowledge
produced by the environmental sciences, whilst
leading them to understand the limits to such
knowledge. Especially for human geographers
trying to take animal subjectivity seriously,
simply ignoring scientific findings was not an
option. Rather, it became imperative to investi-
gate those scientific disciplines dedicated to
understanding animal behaviour, animal–habitat
relations, and human influences on animal life
chances – such as ethology, cognitive psychol-
ogy, conservation biology and landscape ecology
(Whatmore and Thorne, 1998: 451).
Actor network theory (ANT), derived from
poststructuralism, provided new analytics for
studying society, nature and technology together,
whilst avoiding the denial of differences
between, and within, such categories (Law,
1992). ANT proponents argued that analytically
there was no a prioridistinction to be made
between humans and non-humans, that dividing
lines between people, machines and animals are
subject to change and negotiation. Machines and
animals may gain or lose attributes, such as intel-
ligence or independence, while people too may
take on and lose attributes of machines and
animals (Law, 1992: 383). Since humans cannot be
disentangled from non-humans, non-humans –
including animals – are partly constitutive of
society. The analytic treatment of animal subjec-
tivity by animal geographers has thus flowed
both from scientific information about their
behaviour and cognition, and from considering
them as groups socially constructed as minorities
or ‘outsiders’. Whatmore and Thorne (1998), for
example, arguing for consideration of animal
agency, followed Ingold (1992) in suggesting
that animals were best seen as ‘strange persons’
to be treated analytically in the same way as
human groups. Similarly, Philo (1995) sought to
view animals rather as marginalized, socially
excluded people; he speculates on the terror of
cattle at the hands of drovers and the possibility
of their transgression of human-set limits, as they
jump through shop windows and engage in
‘beastly’ sexual conduct.
Neither Whatmore and Thorne nor Philo
explored animal thinking and behaviour to better
understand the subjectivity of the creatures they
were trying to understand. In contrast, Gullo
et al. (1998) excavated scientific literature on
cougar ecology and behaviour, to explicitly assess
lion attitudes toward people and how, if at all,
they might have changed with the encroachment192 CULTURENATURES3029-ch09.qxd 03-10-02 10:48 AM Page 192