owners, though not to appropriate their
property. By 1500, around half of England’s
surface area was either already enclosed or had
never been common. Perhaps 20–25 per cent
was enclosed during the seventeenth century
and a similar amount between 1750 and 1830
(Wordie, 1983).
The importance of enclosure in England has
beenstressed inthree majorhistoricalprocesses.
First, enclosure reshaped the landscape.
Second, enclosure was entangled with the
destruction of the English peasantry (seepeas-
ant) and the rise of agrariancapitalism,though
its importance remains a matter of debate
(Humphries, 1992; Neeson, 1993; Shaw-
Taylor, 2001). Third, enclosure played an
important role in raising agricultural product-
ivity during the agricultural revolution
(Overton, 1996). By allowing farmers to spe-
cialize in the most profitable crops, enclosure
facilitated regional specialization and thus
higher overall productivity, and enclosure
was also a precondition for certain techno-
logical changes in agriculture; for example,
installing under-drainage or the selective
breeding of animals.
While enclosure has been subject to long-
standinghistoricalscrutiny, more recent work
has also begun to reinterpret the concept of
enclosure as a means of exploring thecontem-
poraryforms and processes ofneo-liberalism
(see RETORT, 2005). Human geographers
have seized upon enclosure as an ongoing
feature of capital accumulation. Whereas
traditional historical accounts of enclosure
emphasized its relationship to ‘the commons’,
recent work has proposed a more complex
spatial formation operating across a number
ofscales, sites and practices, from special
economic zones to genetic modification
and biometrics (seebiopower). Three major
axes of investigation have been identified
(Vasudevan, McFarlane and Jeffrey, 2008).
The first focuses on the role of enclosure as
a technology of contemporaryneo-liberalism.
Narratives of enclosure help illuminate the
reconfiguration of political sovereignties,
modes of subjectification and neo-liberal eco-
nomic norms through a variety ofterritories
andnetworks. The second explores the role
oflawas a key instrument through which both
old and new forms of enclosure are legitim-
ized, regulated and policed. The third
addresses the significance of enclosure as a
key dimension of our colonial present. While
studies ofcolonialismunderline the import-
ance of land tenure as a precondition for par-
ticular forms of displacement, dispossession
and discipline, contemporary instances of
imperial enclosure have also mutated into
new forms of theenclave capitalism that
has accompanied the securitization of global
mineral extraction (see Ferguson, 2006). Such
an expansive re-conceptualization of enclosure
has highlighted not only a complex set of
logics of spatial inclusion and exclusion, but
increasingly urgent forms ofresistancethat
centre on a messy and highly conflicted
reclaiming of the ‘commons’. lst/avSuggested reading
De Moor, Shaw-Taylor and Warde (2002);
Mingay (1997); Vasudevan, McFarlane and
Jeffrey (2008).endogeneity The property that a variable is
determined within amodelor geographical
system, rather than being ‘exogenous’ or
determined outside the model and so taken
as fixed. In statistical models such asregres-
sionand thegeneralized linear model, the
right-hand side or independent variables are
assumed to be exogenous, but situations arise
where they are in fact endogenous, creating
problems for model estimation. One widely
applied example is in fiscal competition and
‘tax-mimicking’ between neighbouring local
or regional governments (such as states within
the USA): a state or local government’s
expenditure (or tax-rate) is determined by sev-
eral (exogenous) characteristics of the state
but also by the expenditure (or tax-rate)
behaviour of neighbouring states, so that there
is spatial endogeneity; such models are an
important topic in spatial econometrics.
A second example is where school pupil per-
formance levels are influenced by the average
income levels of families within the school
catchments: the family income levels are
assumed to be exogenous, but affluent families
may have moved to the catchments of high-
performance schools, and so not be exogenous
after all, the so-called ‘selection problem’ in
estimation. There are several methods for tak-
ing endogeneity into account in the estimation
process, includinginstrumental variables,
maximum likelihood andbayesian analysis,
all of which are employed in spatial economet-
rics. For studies employing individual-level
data (as in some pupil performance analyses),
there is a technique of ‘matching’ the charac-
teristics of individuals between areas to elim-
inate endogeneity bias. lwhSuggested reading
Manski (1995).Gregory / The Dictionary of Human Geography 9781405132879_4_E Final Proof page 192 1.4.2009 3:17pmENDOGENEITY