Land Use Policy 43 (2015) 149160

Contents lists available at ScienceDirect

Land Use Policy

journal homepage: www.elsevier.com/locate/landusepol

Local history and landscape dynamics: A comparative study in rural

Brazil and rural France
Florent Kohler a, , Guillaume Marchand b , Marcelo Negro a
a
CREDA UMR 7227, Paris 3 CNRS, 28 rue Saint-Guillaume, 75007 Paris, France
b
Universidade Federal do Amazonas, Amazonas, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: In rural France as well as in rural Brazil, the years 19601970 were marked by profound socio-economic
Received 29 March 2014 and environmental changes. In France, these changes were due to the agricultural modernization policy,
Received in revised form in Brazil, they were caused mainly by the political integration of the Amazon to the rest of the country
25 September 2014
by infrastructure and agricultural colonization. The apparent irreversibility of the parallel phenomena of
Accepted 10 November 2014
settlement, in France, and deforestation, in Brazil, gives us a comparative ground that this paper wishes
to explore. We focused on four Brazilian protected areas and two French rural communes and studied the
Keywords:
local attachment of the people and their collective attitude toward the environment. To do so, we assessed
Land use/cover change
Rural France
the main proximate factors identied in the literature, those determining the attitude of a human group
Brazilian inhabited protected areas toward its environment, an attitude inuenced by structural drivers like legal issues and law enforcement.
Environmental impact Our results suggest that these multiple factors are often randomly interconnected and can hardly be
Hysteresis modeled. Land use and cover change may be an interesting way to understand social and environmental
change, if accompanied by qualitative research about environmental and social perceptions. Our main
conclusion is that collective and individual choices are eventually framed by local history: we use the
notion of hysteresis to suggest that ancient causes may have enduring effects.
2014 Elsevier Ltd. All rights reserved.

Introduction
On both sides of the Atlantic, parallel and apparently irreversible
dynamics unfolded over the last fty years: the conversion of
forests to croplands and pastures, in Brazil, and the conversion of
croplands to settlements,1 in France. In both cases, governments
acknowledge the problem without these drifts being altered. The
apparent irreversibility of these trends has drawn our attention
to the possibility of a comparative approach, to check what similar
logics are at work in these seemingly distinct phenomena. To do so,
we used an approach based on the land use/cover change science
(LUCC)2 which combines geography and sociology, among others,
Corresponding author. Tel.: +33 6 30532133.
E-mail address: orent.kohler@gmail.com (F. Kohler).
1
According to Meyer and Turner (1992, p. 47) the term settlement can denote a
form of land cover or a form of land use (. . .). The category of settlement as a land use
includes areas devoted to human habitation, transportation, and industry. As land
cover, it incorporates highly altered surface such as buildings and pavements, but
such cover represents only a portion of the total area that a land-use classication
might accord to settlement.
2
According to Meyer and Turner (1992, p. 39), land use denotes human employ-
ment of the land and land cover denotes the physical and biotic character of the land
surface.
to understand and predict land use and landscape evolution (Caldas
et al., 2007; Lambin, 2002). The advantage of this approach is that
it is founded on observable phenomena, in an attempt to identify
structural and/or proximate causes of land use and cover changes.
In a seminal paper published in 1992, however, Meyer and
Turner were pessimistic about the possibility of developing models
to anticipate future changes, in a social engineering perspec-
tive. In fact, all things being equal, the trajectories recorded from
one region to another can vary signicantly. However, in present
Land Change Science literature, the modeling approach prevails
(Wainwright and Mulligan, 2002; Turner et al., 2007; see for the
present state of our knowledge, Verburg et al., 2013). Our approach
to landscape changes, in a social science perspective, is inspired in
geographical and anthropological works (Bertrand and Bertrand,
2002; Alphandry, 2001) in the sense that we are interested in the
modications of physical places (land cover change) and how these
are carried through by local populations, on a fty years period of
time. Our comparison will be conducted in the light of the results of
two multidisciplinary research programs, CLEVERT3 on the French
3
Conditions socio-environnementales pour la rhabilitation de la biodiver-
sit ordinaire (20102013) program funded by French Ministre de lEcologie

http://dx.doi.org/10.1016/j.landusepol.2014.11.010
0264-8377/ 2014 Elsevier Ltd. All rights reserved.
150 F. Kohler et al. / Land Use Policy 43 (2015) 149160
side and USART4 on the Brazilian side. We will see that despite
seemingly very different contexts (agricultural abandonment in an
old industrialized country, expansion of pioneer fronts in an area
under development since the 1970s), common features may appear
in the underlying causes of irreversible changes.
After referring to the colonization vs. conservation dilemma
in Brazil, and to the ecological turn which followed the Green
Revolution in France, we will compare, on one side, four Brazil-
ian inhabited protected areas, subjected to internal and external
pressures on the environment (three in the Amazon and one in
Bahia State, in the remains of the Atlantic rainforest); on the
other, two French communes torn between augmenting the urban
sprawling and maintaining the farmlands crisscrossed by hedges
and groves (bocage). For the purpose of our analysis, we explore
the concept of hysteresis, which describes the persistence of a phe-
nomenon even after its causes ceased to exist (Bourdieu, 2002), a
generally underestimated factor.
Contextualization of French and Brazilian situations
In rural France as well as in rural Brazil, especially the Brazilian
Amazon, the years 19601970 were marked by profound socio-
economic and environmental changes. In France, these changes
are due to the policy of agricultural modernization (Green Revolu-
tion, land consolidation), whereas in Brazil they are caused mainly,
but not only, by the Amazon integration policy led by the Military
regime (19641985), interweaved with an agrarian reform based
on colonization, and the recognition of Indian Rights by the Statute
of the Indian (1973). This Statute was the rst step toward the
recognition of cultural diversity, ratied by the Brazilian Consti-
tution of 1988, followed by the legal acknowledgment of a new
category of citizens in 2007, the so-called Traditional People,
including Indians, Maroons, and riverine populations.5
In Brazil, inhabited protected areas as a counterbalance to
development plans
In contrast to the Atlantic Rainforest, which destruction began
with Portuguese colonization (see Map 1), the threats looming
on the Amazon Forest became reality in the 1960s. Plans for the
development and colonization of the Amazon are ideologically
inseparable from the military regime and its Operation Amazon
planned in the 1950s (Simmons, 2002). The Superintendence of
the Amazon (SUDAM) created during this decade was the rst
development agency specically dedicated to this vast territory.
The National Development Plan (19661970) promoted the estab-
lishment of agribusiness companies by building infrastructure and
applying signicantly low taxation. According to Margulis (2003),
16% of total deforestation recorded until in the 2000s is the result
of tax concessions granted between 1970 and 1987. This plan was
followed by the National Integration Plan (19701974), redirecting
the policy of occupation by fostering agricultural settlements, so
as to provide plots of this land without people to men without
land (one of the slogans of the operation). The opening of pioneer
fronts in the state of Rondnia and in southeastern Par provoked
a vast migration of landless peasants from the Northeast and South
of Brazil, as an alternative to a genuine agrarian reform.
The years 19701980 saw an annual average of 21,050 km2 of
primary forest burnt to ashes, whether in the context of agrarian
(Commissariat Gnral au Dveloppement Durable) and by French Caisse des Dpts
et Consignations.
4
Usages et transmission des savoirs et reprsentations du territoire en Amazonie
(20092013) program funded by French Agence Nationale de la Recherche, coordi-
nated by F.-M. Le Tourneau.
5
Decreto n 6.040, de 7 de fevereiro de 2007.
reform (small producers), or through incentives to install large
landowners, outlining the current arc of deforestation visible in
Map 2. The different phases of migration led to a number of land
conicts, between settlers or small farmers against big landown-
ers (Hecht, 1985) as well as between these large landowners and
traditional populations (Araujo et al., 2009).
After the assassination, in 1989, of Chico Mendes (unionist ght-
ing for the rights of rubber tappers in acre), a new policy was set up
to protect these traditional populations and their territories. Among
the mechanisms involved was the creation of inhabited protected
areas such as sustainable development reserves (RDS) or extractive
reserves (RESEX), as part of the national policy for environmental
protection (Sistema Nacional de Unidades of Conservaco, 2000,
PP-G7 and ARPA programs6 ). Other groups gained a perspective of
social and land protection as a compensatory measure for historic
debt (Indigenous and Quilombola7 Territories).
We explored the Indigenous aspect during a 5 years eld
research lead in Bahia State, in the remains of the Atlantic Rainfor-
est. This case study provides a useful perspective on the Amazonian
context, where we focused on Quilombola communities. By assign-
ing a special status to traditionally occupied territories and giving
their inhabitants access to public services (education, health,
energy), the Brazilian government hoped to curb environmental
degradation, and avoid the errors that led to the destruction of the
Atlantic Rainforest, with only 8.5% remaining, mostly under (rather
inefcient) protection (Map 1). As for the Amazon, over 36% of its
surface is currently covered by inhabited protected areas (Map 2).
Map 2 shows that protected areas are able to contain the expan-
sion of the pioneer areas of the Amazon, sometimes constituting
residual forest enclaves within territories largely converted to agro-
pastoralism. However, when it comes to consider national scale,
these apparent good results should not hide a series of problems
identied in the scientic literature. Most of the inhabited pro-
tected areas are the result of multiple trade-offs, especially in
regions of ancient colonization, such as Bahia State (Map 2). In
these regions (Nordeste and Sudeste), the difculties to indemnify
large landowners led the authorities to cut back on non-habited
protected areas, creating overlapping territories with differing
environmental regulations. Due to the inextricable problem of land
tenure in Brazil and different conicts with a great array of popu-
lation (traditional or not) the conservation units have been largely
dened in accordance of social criteria: in the Atlantic forest, Cullen
et al. (2013) emphasized that conservation units created to protect
primates were delimited in function of the previous human occu-
pation and are, in fact, too small to maintain a healthy population.
We will see that the concerned populations, despite their status of
environmental stewards, are subject to conicting issues, making
it difcult to predict their environmental impact or future orienta-
tions.
In France, a rural world torn between agricultural activities,
environmental protection and the progression of suburbanization
It was only after the Second World War (19391945) that suc-
cessive French governments undertook to transform the traditional
structures of the peasant world, based primarily on self-sufciency.
This process culminated with the Agricultural Orientation Act of
August 5, 1960, completed in 1962, which established the principle
6
Respectively Pilot Program for the Protection of Tropical Forests of the G-7 and
Protected Areas of the Amazon, funded in the rst instance by the G-7 with additional
support from the World Bank and WWF.
7
Territory concession to Maroons, descended from fugitive slaves. The term
Quilomborefers to the territory, quilombolarefers to the population that lives in
it.
 F. Kohler et al. / Land Use Policy 43 (2015) 149160
Map 1. Localization of the studied area in the Brazilian Atlantic Rainforest showing protected territories and deforested areas (G. Marchand).
of parity between agriculture and other economic activities, main-
taining however the family-based structure of the farms.
Farmers access to land ownership was encouraged, through
pre-emptive rights given to tenants and sharecroppers. The reor-
ganization of plots (land consolidation) permitted the use of
motorized vehicles; the creation of agricultural colleges facilitated
the organization of an innovative agro-alimentary sector, coor-
dinated by the National Institute of Agronomic Research (INRA),
established in 1946. Cultural methods were then market-oriented,
increasing national productivity. This reform was strengthened by
the establishment of the European Common Agricultural Policy in
1962, with a negative impact on traditional societies and envi-
ronmental health. An ecological turn was then adopted by the
Agriculture Orientation Act of 1999 which established the princi-
ple of multi-functional agriculture (economic, environmental and
social functions) as a major tool for regional planning (Alphandry,
2001). This Act transposed into French Law two major European
Directives: Birds (1979) and Habitats (1985), which promoted the
creation of a network of ecological corridors, the Green and Blue
Infrastructures (IVB).8
But population pressure combined with the regional economic
dynamism generated a rise in land value. Habitat improve-
ment, recreational or commercial settlements, accelerated the
phenomenon of suburbanization or urban sprawling. This
phenomenon, less spectacular perhaps than deforestation, is a
serious problem by its scope and its irreversibility. The SAFER
8
The implementation of these corridors is part of the European strategy for bio-
diversity conservation, derived from the Convention on Biological Diversity of 1992.
The Pan-European Ecological Network is expected to restore ecologically functional
systems, by linking the protected areas established by the Natura 2000 network.
151
(Societies for Land Planning and Rural Settlement, regional Agen-
cies of French Ministre de lAgriculture) issued a statement
summarizing the reasons for concern: in fty years, agricultural
labor force has fallen below the threshold of 3% of active popu-
lation. The French agricultural area decreased by 20% during the
same period, currently representing 28 million ha. 2.5 million ha
were lost to infrastructure and housing. Farmland transformed
into articialized surfaces represented 60,000 ha/year between
1990 and 2000 and 77,000 ha/year between 2000 and 2006. After
2006, the settlement phenomenon continues with an even faster
pace: until 2010, the areas concerned are about 78,000 ha/year, or
3000 km2 in four years (Ministre de lAgriculture, 2010; SAFER,
2012). Map 3, built from the Corine Land Cover data from 1990 to
2006, illustrates this situation.
This apparently irreversible expansion of settlement can be par-
alleled with the dynamics of deforestation in Brazilian Amazon and
Atlantic Rainforest: local dynamics come into contradiction with
the legal scaffolding built precisely for the opposite objective
the maintenance of forested areas in Brazil and agricultural land
in France. It is therefore necessary to focus on structural driving
forces before considering the micro-local level, which allows us to
understand the internal dynamics of change.
Structural factors: globalization, shifting public policies and legal
apparatus inconsistencies
As stated by Meyfroidt et al. (2013), globalization process is
a major land change driver as it inuences both public policies
at different scales and local economic opportunities, accelerating
migration trends. An example of this twofold process was given by
Brondizio (2008) about the Aca palm (Euterpe oleracea) in Maraj
152 F. Kohler et al. / Land Use Policy 43 (2015) 149160

Map 2. Localization of the studied areas in Brazilian Legal Amazon showing protected territories and deforested areas (G. Marchand).

Island. Brondizio shows that traditional communities are deeply
transformed by the global valorization of very specic products.
But underlying factors can also be found in the structure of judi-
cial arsenal. If the law provides a framework which is ignored or
undermined, it may be due to two main reasons:
(a) The successive public policy guidelines are too discontinuous
and do not allow time for adaptation.
(b) Legal regimes intertwine, or one cancels the effects of the other;
the frames established by law stumble over different scales of
implementation and decision levels.
 F. Kohler et al. / Land Use Policy 43 (2015) 149160 153

Map 3. Localization of the studied areas in France, showing land cover change between 1990 and 2006 (G. Marchand).

Instability or discontinuity of public planning policies results
in the eld by the superposition of conservation and development
projects. This overlap is due, on the one hand, to the slow imple-
mentation of new policies, and on the other, to the inertial force
perpetuating the practices related to former ones. As Simmons
(2002) puts it, the successive plans implemented in the Amazon
region have created land instability, with an exacerbated envi-
ronmental degradation due to changes in land use, a predatory
advance of pioneer fronts, and a phenomenon of land grabbing at
the expense of small settlers and traditional peoples. These pro-
cesses went on for decades thanks to shifting legal frames, and took
advantage of the political invisibility of traditional populations.
In France, the successive orientations of agricultural policy were
implemented slowly. These orientations cannot achieve their goals
unless all agricultural stakeholders (producers, distributors, food
industry, but also technicians and zoo-technicians executives, edu-
cators and students of Agricultural Colleges) are mobilized. Their
implementation is postponed, as their objectives can be achieved
only with the advent of a new generation of farmers aware of the
related issues. Farmers throughout their life have seen different
policies, and thus experienced different ideological frameworks;
they often lament the inconsistencies of their career. Our survey
showed that farmers who now are in charge of landscape stew-
ardship were those who cared less about leaving some space
154 F. Kohler et al. / Land Use Policy 43 (2015) 149160
to nature (Kohler et al., 2014). If farmers feel close to nature, it is
because they feel they are producing it: this explains the appar-
ent paradox of their mentioning a strong natural connection, due
to their outdoors activity, but seeing no reason to leave room for
wilderness (Ahnstrm et al., 2009).
Since the French decentralization laws of 1982, the decisions
for urban planning and development have been transferred to the
municipalities, while the main environmental decisions are taken
at national level, as a transposition of EU law. A combination of
factors come together to encourage local authorities to attract new
residents by multiplying transportation and recreational facilities.
Farm buildings and farmland are largely exempt from taxes, so
municipalities have interest in promoting housing projects sub-
jected to local taxation.
French farmers are then caught in the crossre: on one side, an
environmental legislation governing their activities, experienced as
one of the many constraints such as climatic or economic uncer-
tainty weighing on the viability of their farm. On the other side, the
progressive suburbanization, concentrated (house lots) or diffuse
(urban sprawl) multiplying use conicts because of the pollution
generated by agricultural activity. In La Gentouze the urbanized
area increased from 20.4 ha in 1958 to 101 ha in 2008, and the trend
rmly continues in 2013. This set of environmental, economic and
social pressures deters potential buyers, and encourages owners
of agricultural land to convert them into residential areas, thus
accelerating the process (Ahnstrm et al., 2009).
In Brazil, the legal apparatus opposes a very restrictive envi-
ronmental legislation (at least until 2012) to an extremely liberal
land tenure system on public lands, allowing an adjustment of indi-
vidual ownership after one year of effective occupation, effective
meaning the transformation of forest or savannah into cropland or
pasture.9 Meanwhile, the need of an agrarian reform has produced
a legal regime that allows the landless to claim for the expropria-
tion of unproductive farms10 which unproductive areas often
coincide with the legal reserve mentioned above, an intentional
confusion encouraging the invasions (GTA, 2008). Landowners
obviously prefer to pay nes for environmental crime (usually
amnestied) rather than seeing their property invaded and con-
scated. Land tenure insecurity is a powerful driving force for
deforestation, and it is encouraged by the land property regime
more than it is inhibited by environmental legislation (Fearnside,
2001; Araujo et al., 2009).
With 53% of native vegetation located in private properties, for-
est conservation is central in Brazilian, as well as international,
environmental debate. Accelerated deforestation process from late
1980s, led to the revision of Brazilian Forest Code, requiring from
land owners the conservation of a percentage of native vegetation,
the so-called Legal Reserve. In the Amazon biome this percent-
age is up to 80% of the property (50% in specic cases). In the
Atlantic Rainforest biome, this Legal Reserve is, surprisingly, only
20%. Furthermore, private properties are supposed to protect the
Permanent Protection Areas, i.e. riparian vegetation, top of the hills,
and all land submitted to erosion (Soares-Filho et al., 2014).
This severe restriction to deforestation, difcult to apply and
monitor, unleashed the rebellion of agribusiness sector. These reac-
tions led to another revision of the Code, approved by the end of
2012. This new Code maintained the percentages of Legal Reserves,
but granted an amnesty to deforestation occurred before 2008,
which encompassed 90% of rural properties. Furthermore, the Per-
manent Protected Areas were included into computing the Legal
Reserve, which reduced the environmental liabilities by 58%.
9 13
Lei No 6.383, de 7 de Dezembro de 1976. Art. 29.
10
Brazilian Constitution, 1988, Chapter III, Articles 184186.
Methodology
We chose our case studies in order to assess, empirically, how
local dynamics may interweave, so as to shed light on their envi-
ronmental impact. In Brazil, we focused on the transmission of land
uses and representations in traditional communities, and in France,
on the willingness of local authorities to implement, or not, poli-
cies promoting ecosystem rehabilitation or restoration. We chose
an interdisciplinary approach that combines anthropology, human
geography, ethno-ecology11 and ecology. In France, biodiversity
sampling conducted in the eld allowed to objectify the environ-
mental conditions, using the diversity of species inventoried as
biological indicators.
We used the methods of collecting individual life stories,
including migration patterns and current mobility systems, and
household surveys12 that allowed us to identify environmental
perception and representations. Qualitative data collected through
open-ended conversations completed these surveys. We also col-
lected mental maps, which consist in asking community members,
without other indication, to draw maps expressing the local sense
of place. These maps reveal the interdependence of spatial, tempo-
ral, and social representations. GIS maps were then used to compare
the subjective representations of space with the objective char-
acteristics of the territory, so as the land use change might be
visualized by the inhabitants.
These methods have been applied in four Brazilian and two
French eldworks, the Patax Indigenous Land having been sub-
ject to a classic ethnography (see footnote 12). On the Brazilian
side (Maps 1 and 2), we conducted two eldworks in the Middle
Amazon (Par) where land overlapping is common: the Abu and
Jarauac Quilombo Areas. The other two sites were used as com-
parators: the Quilombo of Cunani in northern Amap, landlocked in
the National Park of Cape Orange, and the Patax Indigenous Land
(state of Bahia), which overlaps since 2008 the whole area of Mount
Pascal National Park. We decided to introduce this non-Amazonian
case study as it allows us to highlight a very interesting example
of the outcomes of social relegation. On the French side, the two
municipalities studied are fairly representative of rural evolution
since the Second World War in hedge areas (Map 3).
To be concise, we will expose, for each eld, only the relevant
aspects for our analysis.
Proximate driving forces: behavioral consistencies and
inconsistencies
Statistic or stochastic approaches, spatial analysis about human
occupation or vegetal land cover (Guillaumet et al., 2009), multi-
agent models, General Circulation Models (Lambin, 2002) and
DPSIR framework13 seek a fairly high level of generality so as to be
policy-relevant. On the other hand, social sciences acknowledge
that individual psychology is transcended by a combination of col-
lective norms and values (Stern, 2000). This collective dimension is
considered by the Institutional Analysis Development framework
(Ostrom, 2011) which allows us to visualize multiscale regulation
systems. Case studies about rural/urban connections can also be
mobilized in this effort to understand Land Use Change (Padoch
et al., 2008; Eloy and Lasmar, 2012; Seto et al., 2012). However,
11
We used ethno-ecology in Brazil, as Brazilian law did not allow us to proceed to
biodiversity sampling.
12
Anthropological surveys were applied to all the households on three Brazilian
study areas (Abu, Cunani, Jarauac); and to 10% of the inhabitants in French com-
munes. The method applied in Patax Indigenous Land in Bahia was exclusively
qualitative, based on an ethnographic inquiry from 2001 to 2005 (Kohler, 2011).
DPSIR stands for Driving forces Pressure States Impacts Responses. It has
been adopted by the European Environment Agency.
 F. Kohler et al. / Land Use Policy 43 (2015) 149160
Table 1
Environmental impact drivers, according to scientic literature.
Severe environmental impact
Land property Land property instability
Social position Social relegation
Urban accessibility Urban proximity
Social capital Social heterogeneity
Soil productivity potential Soil conversion facilitated
Demography Demographic pressure
as we shall see in Section Urban/rural connections: analytical
limitations for the cases of Cunani (Amap) and Abu (Par), these
connections are not free from inconsistencies.
Scientic studies focusing on environmental issues generally
mention an array of positive or negative factors relevant for land
use and cover change. Among the most cited driving forces (Table 1)
are the land property situation (Araujo et al., 2009); the social cap-
ital (Brondizio et al., 2009); the degree of isolation (Droulers and
Le Tourneau, 2010); the demographic pressure (Wittemyer et al.,
2008; Meyer and Turner, 1992); the social position (social relega-
tion or low social status see Andrade and Rhodes, 2012) and the
potential of soil conversion (Netting, 1993; Nepstad et al., 2006).
We want to show that these factors have a limited explanatory
value as they combine randomly and do not take temporality into
consideration.
Consistency of case studies with explanatory models:
Saints-en-Puisaye (Yonne, France) and Jarauac (Oriximin,
Brazil)
Consistent with our model is the commune of Saints-en-Puisaye
(Yonne, 600 inhabitants, 2770 ha), located in Burgundy, where the
land consolidation was completed in 1962. This consolidation has
strongly affected hedgerow surfaces, but had no noticeable impact
on urban type settlement, which augmented only from 30 to 38 ha
in 50 years (Map 4).
This commune is divided into two soil types: in the East, a cal-
careous soil favoring cereal crops, and in the West a clay soil, where
the activity of mixed farming was maintained. This commune has
the particularity of being far from urban centers (Auxerre is 50 km
away, Joigny 60 km), thus limiting the phenomenon of suburban-
ization. This relative isolation has created a virtuous circle in favor
of organic farming (10 out of 25 farms); many neo-rural inhabitants
were converted to organic farming in a marked ecological activism
atmosphere. The healthy environment of the commune fosters, in
turn, the proudness of organic producers and allows social control
over agricultural practices. The interviews conducted in the eld-
work have demonstrated a strong commitment to the landscape as
markers of local identity, especially because a majority of the popu-
lation is native or comes from surrounding areas. The consolidation
affected landscapes but had little impact, however, on the nature of
land tenure: in Saints, the smallholders still predominate, creating
good conditions for social cohesion.
Equally consistent is the case of Jarauac Quilombo Area in the
Trombetas region (Par, Brazil), which provides a parallel between
collective management areas, where exists a relative social control,
and areas of individual settlements. This lake is located in the region
of Erepecuru River, where fugitive slaves concentrated before the
abolition of slavery (1888), a region colonized shortly after by white
traders. In 19801990, major social and land conicts opposed
Quilombolas populations to self-proclaimed owners of the Brazil-
ian nut forests. These populations successfully argued their prior
occupation, which led to the demarcation of Quilombola territories
of Trombetas and Erepecuru, as well as a land reform settlement,
Terra Preta II (TRAJAP). Differences are signicant in the evolution
of the landscape of the region between the people who chose the
155
Mild or neutral environmental impact
Land property stability
Social valorization
Isolation
Strong cohesion
Soil conversion discouraged
Demographic stability or decline
demarcation of Quilombola territories (collective ownership) and
those who preferred individual plots (personal property). Defor-
estation is greatest in areas of agrarian reform (250 ha between
1992 and 2011 in TRAJAP) than in Quilombola territories (5 ha for
the Quilombo of Trombetas) for a number of households roughly
equivalent (80 against 109). In 1992, there were 877 ha of open
areas around Lake Jarauac. The boundaries shown on the map
did not exist: they illustrate the current situation. The cleared
areas are distributed as follows: future quilombola area Trombe-
tas: 155.67 ha; future quilombola area Erepecuru, actually occupied
by individual: 357.57 ha; on the left bank of the Lake, future
land reform settlement TRAJAP: 377.96 ha (Map 5). Agrarian reform
areas are more accessible and therefore more prone to conver-
sion to pasture, but production ow facilities are almost identical
in areas under collective use. In the collective land, the decisive
factor would then be the social control that promotes the main-
tenance of traditional activities such as extractivism, horticulture
and slash-and-burn clearing.
Urban/rural connections: analytical limitations for the cases of
Cunani (Amap) and Abu (Par)
Land stability and social control exerted within a community,
however, do not guarantee the sustainability of local practices. The
issue of isolation or urban center accessibility is another factor reg-
ularly invoked to explain the environmental impact of a population
(Droulers and Le Tourneau, 2010). We analyze the cases of Abui (PA)
and Cunani (AP) Quilombo areas, both very isolated even by Ama-
zonian criteria.14 Both Quilombolas communities face the growing
gap between on the one hand, a dened territory associated with a
particular ethnic identity, and on the other, a socio-economic real-
ity rooted in cities and marked by the expansion of family networks
in nearby towns (Padoch et al., 2008; Eloy and Lasmar, 2012).
In Cunani, after the construction of the National Road (1975)
and the creation of the National Park of Cape Orange (1979), the
80s were marked by migrations, many being denitive, because
of the restrictions imposed by the Brazilian Institute of the Envi-
ronment (IBAMA, now Chico Mendes Institute) for the exploitation
of natural resources and the cost of freight. The expectation that
the Quilombo area be legally recognized was not enough to stop
the exodus. While it used to be an important commercial center at
a time when the State of Amap lacked roads, the village is now
reduced to fty permanent inhabitants, and the decline is accel-
erating with the announced closure of the primary school. Aca
plantations, which do not require a permanent residence, seem to
grow while traditional slash-and-burn practices are losing ground:
more than 5000 plants of Aca were planted around the village in the
past ve years. These plantations are no longer considered as being
open access. The owners of the plots recruit a seasonal workforce in
14
In the case of Cunani the advantage of proximity (as the crow ies) of the city is
neutralized by the state of the road, impracticable part of the year, which makes the
transport cost prohibitive. Abu is located 250 km (by the river) from the city of Orix-
imin, and the trip is slowed down by the checkpoints of the Brazilian Environmental
Institute.
156 F. Kohler et al. / Land Use Policy 43 (2015) 149160

Map 4. Land cover in Saints-en-Puisaye (Yonne) between 1957 and 2009 (G. Marchand).

the neighboring town Calcoene, accelerating the decline of commu- to the isolation does not seem to entail a reduced environmental
nity life, while planting and harvesting are accompanied by their impact.
usual impacts: predatory shing, poaching for bush meat, and pol- By contrast, in the Abu community, land security has gener-
lutions of all kind. Against all odds, the demographic decline linked ated new revenues in connection with city mobility (seasonal or
 F. Kohler et al. / Land Use Policy 43 (2015) 149160
Map 5. Evolution of the forested area around Lake Jarauac, in 1992 and 2011 (M. Negro).
permanent jobs, social subsidies and pensions, and sales oppor-
tunity for livestock). These alternatives have been explored after
the devaluation of forest products, including Brazil nuts (Bertholet-
tia excelsa). Despite the distance that separates Abu and Oriximin,
people have developed bi-residential strategies based on the inten-
sication of river transport. Among the most unexpected, given the
layout of the site and the distance to the city, we will mention the
introduction of cattle, encouraged by the urban elite that ensure
the renewal of herds and the commercialization of the meat. It has
led to a concentration of wealth in the hands of some families, and
the privatization de facto of communal lands, being cattle breeding
an exclusive activity.
In both cases described, the factors land stability and degree
of isolation do not, by themselves, account for the evolution of
practices, nor do they guarantee a reduced ecological footprint of
the populations concerned. Family and business networks, a result
of unpredictable circumstances, contribute signicantly to reori-
ent practices, which in both cases have a still reduced but rapidly
increasing impact where one would expect a decrease.
Land conict and social relegation
The impact of poverty, social exclusion or relegation on defor-
estation has been studied; it is generally admitted that the impact is
exacerbated when traditional systems of redistribution are shaken
or have disappeared (for Brazil: Andrade and Rhodes, 2012; for
India: Agrawal, 2005). But little has been said about what is likely
to occur after the situation (social relegation and/or land dispute)
has been settled. Two examples will illustrate this particular phe-
nomenon of enduring impact, called hysteresis.
In La Gentouze (Vende, 1700 inhabitants, 1310 ha), land con-
solidation took place relatively late (19681972) to put an end to
157
a tense land situation (the land being mostly in the hands of large
landowners living on their incomes) and modernize local agricul-
ture. The consolidation coincided with access to land ownership of
many farmers and sharecroppers. The hedge landscape then disap-
peared, which was seen, at that time, as a positive development by
the majority of farmers. The sense of social and technical backward-
ness helped intensify the destruction of traditional landscapes,
hedges being uprooted even where they should have remained
in place (roadsides and property limits). The familiar landscapes
having disappeared, the lampposts served as landmarks for sev-
eral years. The prevailing discourse among farmers who lived the
consolidation period is that they needed to evolve: the peasant
condition was perceived (according to our interviews) as a default
situation due to the lack of alternative or capacity to engage in
another activity. Agricultural modernization was therefore seen as
liberation from social and environmental constraints, which might
explain the radical transformation of ancient landscape (Map 6).
The most striking example, however, remains that of the Patax
Indigenous Land, Extreme South of Bahia, legalized in 1991. This
region falls within the Atlantic Rainforest biome, and the case stud-
ied is exemplary in many respects. Patax Indians were evangelized
and settled in 1861. They adopted the livelihoods of smallholder
peasants, until the lands they occupied were classied as National
Monument in 1941, then as National Park in 1961. They were sub-
jected to repeated humiliation, having been forbidden for thirty
years to access the forest, until their rights were gradually recog-
nized (Statute of the Indian enacted in 1973, Brazilian Constitution
of 1988). In this region associated with the discovery of Brazil, the
Patax, when confronted to the authorities or tourists, pride them-
selves on their ecological sensitivity and harmony with nature.
They obtained, in 1991, the recognition of 1/3 of the area of Mount
Pascal National Park, a Biosphere Reserve, as Indigenous Land,
158 F. Kohler et al. / Land Use Policy 43 (2015) 149160

Map 6. Land cover in la Gentouze (Vende) in 1957 and 2008 (G. Marchand).

and became co-managers of the remaining forest. But the ensuing Indigenous Land was cleared (see Map 7). However, on the basis
social regeneration was not accompanied by virtuous environmen- of the same ecological arguments, the Patax obtained in 2008 the
tal practices. On the contrary, the activities of wood handicraft took extension of their territory, now encompassing the entire National
an industrial scale: between 1991 and 2008, 80% of the area of the Park as well as some surrounding communities.
 F. Kohler et al. / Land Use Policy 43 (2015) 149160
Map 7. Mount Pascal area (Extreme South of Bahia State) before (1987) and after
(2008) the delimitation of Patax Indigenous Land, created in 1991, just before its
extension in 2008 (R. Folhes).
This dynamics is based on the investment in school and supe-
rior education of the young generations, almost exclusively at the
expense of the forest cover. The depletion of forest resources,
according to this logic, is widely compensated with the access of
the young Patax to salaried work, the sign of their social advance-
ment. The other relevant factor to explain this deforestation lays
in the association between environmental conservation and past
discriminatory policy against backward Indians (Kohler, 2011);
in the same way, in La Gentouze a sense of social backwardness
compounded the environmental impact of land consolidation.
Discussion and conclusion
Our research, conducted in two French communes and four
Brazilian protected areas, aimed at emphasizing the difculty of
predicting the environmental impact and behavior of a given pop-
ulation.
According to cultural geographys main hypothesis, population
stability, length of occupation, landmark knowledge and collective
memory are signs of human-landscape connection and therefore
favor conservation. This set of signs is called sense of place (Simmons
et al., 2007, p. 577). The example of La Gentouze and Patax
Indigenous Land showed us that ancient landscapes can be anni-
hilated in a specic social conguration; the structure of feeling
created and recreated through the experience of everyday life
can be, precisely, a feeling of social relegation which effects can
endure for decades. The economic opportunities do not explain
everything: the way these opportunities are exploited and the
159
logic which underlies this exploitation reveal the way the past
is channeled. A qualitative eld study offers a clue to under-
stand underlying factors, the long-lasting relationship as revealed
by phenomena such as hysteresis, that LUC models can hardly
consider.
Moreover, peoples attachment to the places they live, when
taken for granted, may bias the scientic approach. We have there-
fore exposed different perspectives and case studies, in order to
verify the relevance of the LCS and its assumptions about the
structural and proximate causes of change, on both sides of the
Atlantic: protected areas in Brazil, and French rural counties divided
between suburbanization and maintenance of farming activities.
Multiple factors are at work when a human community deter-
mines its attitude toward its environment and its own destiny.
Globalized interactions are the rst structural factor identied as
they impact local economic trends and opportunities. The law,
which establishes the framework in which the choice will be exer-
cised, is another structural factor that can compromise all type
of community-based sustainability programs (i.e. independence of
law regimes, or inconsistent legal apparatus favoring infrastructure
and facilities).
Proximate drivers are more problematic. By discussing the rel-
evant proximate factors identied in scientic literature (land
property stability, social valorization, isolation, strong social cap-
ital, low potential of soil conversion, demographic stability or
decline), factors which allegedly favor a mild or neutral environ-
mental impact, we showed that these factors cannot account for all
the cases in study.
Lambin (2002) highlighted the difculty of taking into consider-
ation the historical dimension so as to understand not only how
but also why these changes occur. Our conclusion is that collective
inertia should be taken into account, as generational transmission
and new educational or economic opportunities interweave to slow
down or accelerate LUC. In fact, many cases result from hysteresis,
meaning that we should consider local history as a major driver
of environmental impact. The causes of land cover and land use
change may actually be anchored in remote local history: only a
qualitative approach allows us to understand why, all other con-
ditions being equal, a human community engages, or not, in an
environment-friendly dynamics. By studying local history, and col-
lective representations (the Natives point of view), we can better
understand peoples attitude toward their environment, so as to
propose appropriate conservation policies.
The authors wish to thank Florencia Amorena, Ludivine Eloy,
Ricardo Folhes and Claire Couly for their kind contribution to this
article. The authors also wish to thank Isabelle Laudier (Caisse
des Dpts et Consignations) and French Commissariat Gnral au
Dveloppement Durable.
References
Agrawal, A., 2005. Environmentality Technologies of Government and the Making
of Subjects. Duke University Press, Durham.
Ahnstrm, J., Hckert, J., Bergea, H.L., Francis, C., Skelton, P., Hallgren, L., 2009. Farm-
ers and Nature Conservation: What is Known About Attitudes, Context Factors
and Actions Affecting Conservation? Agronomy & Horticulture Faculty Publi-
cations, pp. 3847 (paper 361).
Alphandry, P., 2001. Les campagnes francaises de lagriculture lenvironnement
(19452000). Politiques publiques, dynamiques sociales et enjeux territoriaux.
PhD thesis, Institut dEtudes Politiques, Paris.
Andrade, G.S.M., Rhodes, J.R., 2012. Protected areas and local communities: an
inevitable partnership toward successful conservation strategies? Ecol. Soc. 17
(4), 14, http://dx.doi.org/10.5751/ES-05216-170414.
Araujo, C., Bonjean, C.A., Combes, J.-L., Combes Motel, P., Reis, E.J., 2009. Property
rights and deforestation the Brazilian Amazon. Ecol. Econ. 68 (2), 461468.
Bertrand, C., Bertrand, G., 2002. Une gographie traversire, lenvironnement
travers territoires et temporalits. Arguments, Paris.
Bourdieu, P., 2002. Le Bal des clibataires. Crise de la socit paysanne en Barn.
Points Seuil, Paris.
160 F. Kohler et al. / Land Use Policy 43 (2015) 149160
Brondizio, E.S., 2008. The Amazonian Caboclo and the Aca palm: Forest Farmers in
the Global Market. New York Botanical Garden Press, New York, NY, USA.
Brondizio, E.S., Ostrom, E., Young, O.R., 2009. Connectivity and the gover-
nance of multilevel socialecological systems: the role of social capi-
tal. Annu. Rev. Environ. Resour. 34, 253278, http://dx.doi.org/10.1146/
annurev.environ.020708.100707.
Caldas, M., Walker, R., Arima, E., Perz, S., Aldrich, S., Simmons, C., 2007. Theorizing
land cover and land use change: the peasant economy of Amazonian Deforesta-
tion. Ann. Assoc. Am. Geogr. 97 (1), 86110.
Cullen Jr., L., Sana, D.A., Lima, F., Abreu de, K.C., Uezu, A., 2013. Selection
of habitat by the jaguar, Panthera onca (Carnivora: Felidae), in the upper
Paran River, Brazil. Zoologia 30 (4), 379387, http://dx.doi.org/10.1590/
S1984-46702013000400003.
Droulers, M., Le Tourneau, F.M., 2010. LAmazonie brsilienne et le dveloppement
durable. Belin, Paris.
Eloy, L., Lasmar, C., 2012. Urbanization and transformation of indigenous resource
management: the case of Upper Rio Negro (Brazil). Int. J. Sustain. Soc. 4,
372388.
Fearnside, P.M., 2001. Land-tenure issues as factors in environmental destruc-
tion in Brazilian Amazonia: the case of southern Para. World Dev. 29 (1),
361372.
GTA, 2008. O m da oresta? A devastaco das unidades de conservaco e terras
indgenas no estado de Rondnia. Rondnias GTA Report, 62 pp.
Guillaumet, J.-L., Laques, A.-E., Lna, P., de Robert, P., 2009. La spatialisation de la
biodiversit: un outil pour la gestion durable des territoires. Editions de lIRD,
Bondy.
Hecht, S.B., 1985. Environment, development and politics: capital, accumulation and
the livestock sector in eastern Amazonia. World Dev. 13 (6), 663684.
Kohler, F., 2011. Tombeau des Aymors: le Monde souterrain des Indiens Patax.
Peeters, Louvain.
Kohler, F., Thierry, C., Marchand, G., 2014. Multifunctional agriculture and farm-
ers attitude: two case studies in rural France. Hum. Ecol., http://dx.doi.org/
10.1007/s10745-014-9702-4.
Lambin, E.F., 2002. Modelling land-use change. In: Wainwright, J., Mulligan, M. (Eds.),
Environmental Modelling: Finding Simplicity in Complexity. John Wiley & Sons,
Oxford, pp. 245254.
Margulis, S., 2003. Causas de desmatamento da Amaznia brasileira. World Devel-
opment Report, 100 pp.
Meyfroidt, P., Lambin, E.F., Erb, K.-H., Hertel, T.W., 2013. Globalization of land use:
distant drivers of land change and geographic displacement of land use. Curr.
Opin. Environ. Sustain. 5 (5), 438444.
Meyer, W.B., Turner, B.L., 1992. Human population growth and global land-use/cover
change. Annu. Rev. Ecol. Syst. 23, 3961.
Ministre de lAgriculture, 2010. Mthodologie Teruti-Lucas, http://www.
agreste.agriculture.gouv.fr/enquetes/territoire-prix-des-terres/teruti-lucas-
utilisation-du/
Nepstad, D.C., Stickler, C.M., Almeida, O.T., 2006. Globalization of the Amazon soy and
beef industries: opportunities for conservation. Conserv. Biol. 20, 15951603,
http://dx.doi.org/10.1111/j. 1523-1739.2006.00510.x.
Netting, R.M., 1993. Smallholders, householders: farm families and the ecology of
intensive. In: Sustainable Agriculture. Stanford University Press, Stanford, CA.
Ostrom, E., 2011. Background on the institutional analysis and develop-
ment framework. Policy Stud. J. 39 (1), 727, http://dx.doi.org/10.1111/
j. 1541-0072.2010.00394.x.
Padoch, C., Brondizio, E.S., Costa, S., Pinedo-Vasquez, M., Sears, R., Siqueira,
A., 2008. Urban forest and rural cities: multi-sited households, consump-
tion patterns, and forest resources in Amazonia. Ecol. Soc. 13 (2), 2 (URL:
http://www.ecologyandsociety.org/vol13/iss2/art2/).
SAFER (Socits damnagement foncier et dtablissement rural), 2012. State-
ment, http://www.safer.fr/communique-diminution-surface-agricole-ferme-
france.asp
Seto, K.C., Reenberg, A., Boone, C.G., Fragkias, M., Haase, D., Langanke, T.,
Marcotullio, P., Munroe, D.K., Olah, B., Simon, D., 2012. Urban land telecon-
nections and sustainability. Proc. Natl. Acad. Sci. U.S.A. 109 (20), 76877692,
http://dx.doi.org/10.1073/pnas.1211658109.
Simmons, C.S., 2002. The local articulation of policy conict: land use, environment,
and amerindian rights in eastern Amazonia. Prof. Geogr. 54 (2), 241258.
Simmons, C.S., Walker, R.T., Arima, E.Y., Aldrich, S.P., Caldas, M.M., 2007. The Amazon
land war in the south of Par. Ann. Assoc. Am. Geogr. 97 (3), 567592.
Soares-Filho, B., Rajo, R., Macedo, M., Carneiro, A., Costa, W., Coe, M., Rodrigues, H.,
Alencar, A., 2014. Land use: cracking Brazils Forest Code. Science 344 (6182),
363364, http://dx.doi.org/10.1126/science.1246663.
Stern, P.C., 2000. Psychology and the science of humanenvironment interactions.
Am. Psych. 55, 523530.
Turner II, B.L., Lambin, E.F., Reenberg, A., 2007. The emergence of land change
science for global environmental change and sustainability. Proc. Natl. Acad.
Sci. U.S.A. 104 (52), 2066620671, http://dx.doi.org/10.1073/pnas.0704119104
(published online before print December 19, 2007).
Verburg, P.H., Mertz, O., Erb, K.-H., Espindola, G., 2013. Human settlements and
industrial systems. Curr. Opin. Environ. Sustain. 5 (1), 433534.
Wainwright, J., Mulligan, M. (Eds.), 2002. Environmental Modelling: Finding Sim-
plicity in Complexity. John Wiley & Sons, Oxford.
Wittemyer, G., Elsen, P., Bean, W.T., Coleman, A., Burton, O., Brashares, J.S., 2008.
Accelerated human population growth at protected area edges. Science 321
(5885), 123126.