Changing Ruralities, Changing Landscapes:exploring Social Recomposition Using A Multi-Scale Approach

ARTICLE IN PRESS
Journal of Rural Studies 19 (2003) 425444
Changing ruralities, changing landscapes: exploring social

recomposition using a multi-scale approach
Sylvain Paquettea,*, Ge! rald Domonb
a
Chair in Landscape and Environmental Design, Universit!e de Montr!eal, C.P. 6128, Succursale Centre-ville, Montr!eal, Qu!ebec, Canada H3C 3J7
b
Facult!e de lam!enagement, Universit!e de Montr!eal, C.P. 6128, Succursale Centre-ville, Montr!eal, Qu!ebec, Canada H3C 3J7
Abstract
Changes in agricultural practices have led to declines in the farming population and signicant changes within the landscape.
Paradoxically, many rural areas are recording signicant demographic growth. Given the extent of rural areas experiencing these
changes, little attention has been devoted to the relationship between rural migration processes and landscape developments.
Therefore, the various rural dynamics resulting from these phenomena merit closer investigation. Moreover, we should explore these
complex processes at scales where they are most evident. Through the multi-scale research approach used in this study, these
processes are investigated in their regional, local, and domestic contexts.
As shown by regional typological outlines, the great diversity of the rural areas indicates an increasing dissociation between the
agricultural and socio-demographic trajectories. From our local-scale study, changes in lot occupation revealed new residential
settlement patterns induced by an inux of in-migrants. Specic landscape characteristics seem to be a determining force shaping
these population ows. At the same time, these population movements have dissimilar inuences on evolving landscape dynamics.
With the exception of isolated agricultural abandonment trajectories, migrant relocation does not seem to affect specic local-scale
land use development. However, it is signicantly associated with individual domestic practices. These practices are indicative of the
migrants identity in the rural landscape; further, they suggest specic values for the landscapes qualitative dimensions. These
emerging urban and non-farming interests in rural landscapes challenge planning policies to guide the landscapes evolution for the
benet of its producers and consumers.
r 2003 Elsevier Science Ltd. All rights reserved.
Keywords: Urban-to-rural migration; Landscape dynamic; Residential settlement; Rural development; Landscape planning; Southern Qu!ebec
1. Introduction
Rural areas experience continuous transformation in
their social and physico-spatial composition. Productivist to post-productivist (Halfacree and Boyle, 1998),
industrial to post-industrial (Jollivet, 1997) or even
rural to post-rural (Murdoch and Pratt, 1993) are some
of the many visions of the functional changes taking
place within the countryside. Many of these transformations are associated with an increasing inuence of
urban and non-farming interests in rural places and on
their lifestyles. With the migration of specic segments
of the urban population to the countryside, there
emerges increased social demand for an array of new
functions and services (e.g., environmental, recreational,
*Corresponding author.
E-mail addresses: sylvain.paquette@umontreal.ca (S. Paquette),
gerald.domon@umontreal.ca (G. Domon).
aesthetic, etc.). These new demands are primarily

concentrated within agricultural (Ilbery and Bowler,
1998; Marsden, 1998) and forested areas (Egan and
Luloff, 2000; Erickson et al., 2002; Mather, 2001).
Without fail, the inux of urban migrants into rural
areas involves new development. If signicant academic
research (Meeus et al., 1990; Baudry, 1993; Domon
et al., 1993; Poudevigne and Alard, 1997) has helped to
reveal spatially differentiated evidence of agricultural
landscape development, the outcome of landscape
dynamics associated with newly introduced social
aspirations and interests have received comparatively
little consideration. Given that some rural areas are
becoming places of residence for an increasing number
of in-migrants (Fuguitt, 1985; Kayser, 1990; Kontuly,
1998; Dahms and McComb, 1999) and that these
migration patterns seem closely associated with specic
idyllic representations of the countryside (Willits and
Luloff, 1995; Hervieu and Viard, 1996; Lowenthal,
0743-0167/03/$ - see front matter r 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0743-0167(03)00006-8
ARTICLE IN PRESS
426
S. Paquette, G. Domon / Journal of Rural Studies 19 (2003) 425444
1997; Halfacree and Boyle, 1998; Swafeld and Fairweather, 1998; ORourke, 1999), the dynamic relationships existing between residential behaviour, landscape,
and rurality merit closer investigation.
While some academics have acknowledged that
landscape aesthetics act as driving forces for rural
in-migration (Halliday and Coombes, 1995), it is
unknown how landscape attributes act on migrant
residential settlements, and how they induce a particular
rural socio-demographic evolution. Equally unresolved is how rural migration affects landscape
development itself. Finally, it is unresolved if the
inux of new rural migrants may accelerate rural
landscape transformations or, conversely, help to
maintain their current attributes. The importance of
the landscape as a pivotal element for rural changes
should be better assessed. In particular the way the
landscape modulates social recomposition processes as
well as being recongured by them should receive more
attention.
Many authors view rural repopulation trends as an
opportunity to revitalise rural economies (Newby, 1990;
Stockdale et al., 2000). These questions are all the more
critical given that this development, within the perspective of an ever-increasing incidence of rural consumption practices (Ilbery, 1998, p. 259), is largely associated
with the preservation and the improvement of the rural
landscape qualities. Before addressing the issues that
merit further investigation, some conceptual clarications are essential.
1.1. Rural landscape changes: a conceptual approach
The word rural no longer has a single meaning
(Murdoch and Pratt, 1993; Halfacree and Boyle, 1998).
The coexistence of distinct rural experiences reects the
growing diversity of human interactions within rural
places (Mougenot, 1986). As a result, the landscape
can be dened as a tangible imprint continuously
reshaped by changing social representations of the
space. This denition of landscape is derived from the
notion of empreinte-matrice developed by Berque
(1990, p. 44). It designates the in situ results of human
intervention within a given milieu, it also designates the
interpretation and perception of that milieu. This
landscape concept derives from an idea whereby people
represent a given milieu through the way they use it, and
conversely, manage this milieu following the way they
interpret it (Berque, 1990). In this way material
imprints or traces in the landscape, reveal individual
or collective aspirations that are more truthful than
most autobiographies because we are less self-conscious
about how we describe ourselves [through landscapes]
(Lewis, 1979, p. 12). These inscriptions are not merely
neutral facts. Rather, in the context of our investigation, landscape attributes embody the many expressions
of what individuals value in rural places and territories

(Dubot and Lizet, 1995; Schein, 1997). The way distinct
landscapes are shaped by everyday practices and
experiences of rural residents constitutes our starting
point.
Given the signicant repopulation of some rural
areas, it is understood that contemporary experiences of the countryside are increasingly inuenced by external viewpoints. In contrast, until a
few decades ago, the experience of the countryside
was characterised by close and long-lasting attachments
to the land (Lowenthal, 1997; ORourke, 1999).
As an example, the way rural land is currently settled
based on residential needs as well as farming or
forest production (Erickson et al., 2002) is
indicative of specic rural identities (ORourke, 1999;
Primdahl, 1999). An urbanite purchasing a rural
property with a view of the surrounding countryside,
or maintaining a vernacular house, also represents a
distinct manifestation of the individuals values associated with rural places. Rather than painting a static
portrait of these practices and uses, this paper emphasises their evolving nature. By way of this emphasis, we
attempt to show evidence of the diversity of the
landscapes dynamic directions, hereafter known as
landscape trajectories.
1.2. Study outline
Given the empirical gap that exists between social
recomposition and landscape development relationships, we need to explore the various rural dynamics
that result from these phenomena. Moreover, these
processes must be explored at the scale where they are
the most evident. This paper intends to reveal these
processes in their regional, local, and domestic manifestations by taking advantage of research results from a
multi-scale case study.
The following section describes three perspectives
(Fig. 1). Distinct typological outlines have created a
preliminary portrait of the trajectories and actual
proles of rural municipalities in Southern Que! bec at
a regional scale. This large-scale characterisation was
necessary to show the dominant tendencies of the
processes involved, as well as their diversity and their
specic spatial expression. Once the new residential
settlements were revealed, a smaller-scale evaluation
determining how landscape attributes shape the sociodemographic evolution of rural communities was
carried out. This was done through detailed analysis of
the settlement dynamics of a clearly delineated territory.
This case study also explored how socio-demographic
changes contribute to specic landscape transformations. Beyond local phenomena, the effort to characterise changes affecting rural landscapes through rural
recomposition processes needs to be complemented by
ARTICLE IN PRESS
427
2.1. The evolution of rural municipalities in southern

Que!bec: a regional overview
Social recomposition has led to changes in rural areas
in southern Que! bec, as elsewhere in western countries
(Fuguitt, 1985; Kayser, 1990; Robinson, 1990; Beesley,
1991; Jean, 1997). Focusing on agricultural and sociodemographic dynamics is helpful in order to better
understand the ongoing evolution of rural municipalities, and the extent and diversity of the phenomena
evolved. This focus also reveals the spatial patterns of
these phenomena. Documenting this evolution also
helps to identify local situations that potentially require
a more detailed examination, such as rural municipalities in transition that are characterised by uncertain
dynamics.
Fig. 1. Multiple-scale research design.
an investigation of residential practices. This domicile

level, or individual lot scale, investigation becomes
essential, given that socio-demographic changes must
be documented at a scale where their manifestation
is the most apparent. Following an overview of
the dominant trends related to each level of analysis,
the third section of this paper attempts to establish
relationships between the processes that express
themselves at regional, local, and domestic scales.
The implications of these results, namely trends
likely to shape rural areas in the future, are discussed
in the fourth and nal section. Specically, we
explore the trends related to demographic evolution
and landscape dynamics as well as rural community
changes. In conclusion, this paper outlines future
research directions.
2. Rural recomposition and landscape dynamics: a multiscale perspective

The methodological strategy of this research is
based on three complementary analyses derived from
as many differentiated scales, namely: regional, local
and domestic scales (Fig. 1). Based on a previous
research initiative (Paquette and Domon, 1999; 2001a,
b), this section species the methodology and provides
an overview of signicant results for each level of
analysis.
2.1.1. Study area and database development: a regionalscale study framework

Located extreme south of the province of Que! bec
(Canada), the study area (6300 km2) is composed of 108
municipalities within six Regional County Municipalities (RCM)1 (Fig. 2). Our interest in this region lies in
the diversity of the populations structure as well as its
biophysical characteristics that include the Saint Lawrence Lowlands and the Appalachian Plateau Uplands.
Primary agricultural data obtained from Canadian
census data of 1961 and 1991 (farm population, number
of farms, capital and farm income, average area per
farm, areas of cultivation, pasture, and cattle production, areas of corn (maize), hay, and alfalfa crop
production) was collected for each municipal subdivision. In addition, socio-demographic data derived from
the 1991 Canadian population census (population
density, age distribution, population according to the
level of schooling, number of workers per sector of
activity, average and median income, rate of activity,
and unemployment rate) were also considered, as well as
the property values (1991) for different land use
categories (residential, cottages, industrial, agriculture,
trades and services) provided by the ministe"re des
Affaires Municipales du Qu!ebec.
Based on this data, different typological portraits
were drawn using cluster analysis.2 This statistical
tool was of great importance in the development of
the subsequent regional typologies that enabled the
1
The study area consists of the RCMs: Haut-Saint-Laurent, Jardinsde-Napierville, Haut-Richelieu, Brome-Missisquoi, Memphr!emagog
and Coaticock.
2
Using SPAD software (v. 3.2), this analysis identies homogeneous
groups on the basis of the distance between each pair of municipalities.
The average of each of the variables is calculated for each group. With
the help of a statistical test, it is possible to recognise the averages that
diverge signicantly from overall averages for each variable and each
group. These measures also give an idea of the internal variability of
the groups.
ARTICLE IN PRESS
428
Fig. 2. Location of the study areas. Inset shows location in Quebec (Canada).
identication of agricultural trajectories3 (196191),

agricultural proles (1991), and current socio-demographic proles (1991). Accordingly, this methodological approach allows us to illustrate socio-spatial
differentiations from distinct viewpoints.
2.1.2. Regional-scale agricultural trajectories and sociodemographic profiles: the dominant trends
The extent of agricultural transformations, their
distinct spatial manifestations, and the great diversity
of socio-spatial situations are the most signicant
features in the recent transformation of southern
Que! becs rural areas (Paquette and Domon, 1999).
From the four agricultural trajectories (196191)
identied using cluster analysis (Fig. 3a), the most
signicant results are: an agricultural intensication
and an agricultural regression concentrated within welldened areas. Agricultural intensication trajectories
were revealed primarily by an increase in the proportion
of cultivated land conned to the Saint Lawrence
Lowlands. Agricultural extensication processes and
regressing agriculture (i.e. a decrease in the number of
farms; no increase in cultivated areas, particularly those
devoted to corn production) were concentrated within
the Appalachian Plateau Uplands. This spatial pattern is
accentuated by the resulting agricultural proles (1991)
of the associated municipalities (Fig. 3b). The average
farm capital, farm income, and area under cultivation
contributed signicantly to different spatial patterns
between the municipalities experiencing intensive production and those characterised by marginal farming
3
In this instance, the agricultural trajectory is a result based on the
rate of change in the agricultural data between 1961 and 1991.
activity. Given their dissimilar and crucial inuence on

agricultural potential, biophysical attributes (e.g., topography and geomorphological deposits) seem to be the
determining factor underlying this dual land use
dynamic in southern Que! bec (Domon et al., 1993;
Paquette and Domon, 1997; Pan et al., 1999) as
elsewhere in western countries (Meeus et al., 1990;
Simpson et al., 1994).
The extent of farming decline was also revealed by
this study. For the entire study area, the proportion of
the farming population fell by two-thirds during the last
three decades and corresponds to only 14% of the total
rural population in 1991. These results suggest that
traditional farming activities no longer predominate
within large parts of the territory. Consequently,
contemporary ruralities are not dened exclusively
through agricultural activities (Jollivet, 1988; Jean,
1989; Murdoch and Pratt, 1993; Marsden, 1995).
This situation is exemplied by the diversity of sociodemographic proles within southern Que! becs rural
municipalities (Paquette and Domon, 1999). From the
seven proles generated (Fig. 3c), three situations were
identied: the agricultural rural area, the periurban rural
area, and the rural amenity area.
*
Agricultural rural area. As previously stated, agricultural activities were spread out along well-dened
spatial patterns. Even for municipalities where
agriculture is predominant, its signicance has
greatly declined from a socio-demographic perspective. In fact, nearly 75% of the active population
within these municipalities has income from other
sources (Paquette and Domon, 1999, p. 291).
Although the farming imprint is still perceptible in
ARTICLE IN PRESS
429
Fig. 3. Regional typologies showing: (a) agricultural trajectories (19611991); (b) resulting agricultural proles (1991) and (c) socio-demographic
proles (1991) of southern Quebecs municipalities.
ARTICLE IN PRESS
430
land use, agricultural vocations are no longer

exclusive.
Periurban rural area. While traditionally conned
close to larger urban agglomerations, the current
periurban development now penetrates well within
the truly rural area (Lucy and Phillips, 1997). In
the study area, results suggest that the rapid
demographic growth associated with such development spreads around small regional urban centres
(Paquette and Domon, 1999). Although this periurban area presents a continuum of situations with
regard to population density, socio-professional
prole, and residential land value, it constitutes an
undeniable expression of the current social changes
experienced by rural communities in southern Que! bec.
Rural amenity area. Land occupation for the purpose
of seasonal recreational activities (e.g. second-home
development) is another recurring theme of many
rural municipalities. This prole characterises some
municipal entities that have no recreational appeal
(i.e. open water, ski resorts). Even more signicantly,
as shown by land values attributable to cottages, the
importance of second-home owners has become
signicant in traditional agricultural municipalities.
As observed elsewhere (Brunger et al., 1991), second
home development diffuses into larger rural areas,
including traditional farming territories.
Our examination of agricultural trajectories and their

socio-demographic proles suggests that new residential
settlement dynamics are emerging throughout southern
Que! becs rural areas as in other regions (Marie! and
Viard, 1988; Brunger et al., 1991; Theobald et al., 1996).
The increasing dissociation between agricultural and
socio-demographic trajectories clearly supports this
assertion (Fig. 3). The demographic growth of municipalities that experience simultaneous regression in
agriculture represents the most striking manifestation
of these dynamics (Paquette and Domon, 1999). For
those municipalities outside the intensive farming belt,
the complex relationships between rural recomposition,
residential settlement, and landscape dynamics merit
closer investigation.
2.2. Rural recomposition and landscape dynamics: some
local evidence
Given the population growth of municipalities
affected by a decline in farming and the extent of
second-home development throughout vast rural areas,
an exploration of the role of landscape attributes in
guiding new residential settlement patterns becomes
necessary. Many large-scale urban-to-rural migration
surveys have acknowledged scenery (Halliday and
Coombes, 1995) or scenic amenity (Bryant et al.,
1982; Coppack, 1988) as important factors of attraction

in explaining moves to rural destinations. Other
investigations have documented the spatially selective
nature of such processes (Gorton et al., 1998, p. 217).
Few studies describe the ner scale empirical evidence
(Riebsame et al., 1996; Theobald et al., 1996) of the
interaction between landscape characteristics and social
recomposition. Such information becomes essential
given the empirical gap that exists regarding the impact
of occupational changes associated with rural repopulation on the landscape (Primdahl, 1999).
Our methodological strategy attempts to shed light on
these complex relationships by using a detailed analysis
of settlement dynamics of each lot within a clearly
dened area. Based on in situ observations, analysis of
land use changes, as well as socio-demographic, and
residential history information surveys, the study
attempts to identify residential settlement patterns, and
local land use and built environment transformations
(hereafter referred to as landscape trajectories).
2.2.1. Study area and database development: a local-scale
study framework
Havelock Township (88 km2), located in the southernmost corner of the province of Que! bec (Canada) is
approximately an hours drive from downtown Montreal (Fig. 2). This area is representative of a larger
group of rural municipalities characterised by transitional phenomena of an agricultural as well as a sociodemographic nature (Fig. 3). This study area is also
representative of other settlement experiences along the
urban fringe in North America (Brunger et al., 1991;
Hart, 1998). Havelock Township shows a demographic
growth of 21.2% between 1961 and 1991, while the
farming population decreased from 63.4% to 29.8% of
the total population during this same period (Paquette
and Domon, 2001a). The landscape diversity of this
study area enables us to explore residential settlement
patterns in a wide range of situations (Fig. 4). Agroforested upland areas surround agricultural lowlands
with an elevation ranging from 100 to 340 m. Many
locations on the summits and the upper hillsides offer
panoramic views of the region.
Three types of data were collected to characterise
residential settlement patterns and landscape trajectories
at the lot scale: visual analysis, landscape trajectory, as
well as socio-demographic and residential history data.
All the 254 lots retained for the study4 were visited
between August and October 1998.
*
Visual analysis data. The following indicators, derived from Jacobs et al. (1986) were used for the
4
The resulting database comprises a large sample of the population
(254 of a total of 297 properties located outside hamlets). Residential
settings identied as suitable for study exclude vacant (31 lots) and
inaccessible lots (12 lots).
ARTICLE IN PRESS
431
Fig. 4. Havelock Townships land use (1997) and dominant topographical locations.
visual analysis: width of the visual eld, depth of the

visual eld (distance to the horizon), prevailing views,
viewshed (potential view), and visual accessibility
from other locations along the road network
(Table 1). With the exception of the visual access
from the road, all these indicators allow the view to
be documented from the residential site (e.g. from the
front door). Topographical entities were also identied for each home location based on the terminology
used by Gerardin and Lachance (1997). The different
combinations of visual elements revealed by the
cluster analysis are hereafter designated as landscape contexts.
Landscape trajectory data. For each lot, landscape
trajectories were identied on the basis of land use
change and built environment transformation
(Table 1). Land use was interpreted for two study
periods (1968 and 1997), using 1:15 000 aerial
photographs. Using the IDRISI geographic analysis
system (Clark Labs, 1997) and a digitalised cadastral
map of Havelock Township, we then evaluated
the extent of the ve most dominant trajectories
at the lot scale (Table 1). Built environment
transformation (home and farm building changes)
was assessed during observations in the eld
using 1:20 000 cadastral (19665) and topographical

maps (19926). Field evaluation was completed by
systematic verication of aerial photographs from
1965 and 1997.
Socio-demographic and residential history data.
Derived in part from the work of Kayser (1990),
data relative to place of birth, previous place
of residence, occupation, place of work, age group,
year of lot acquisition, and the presence of relatives were collected during face-to-face interviews
with Havelock residents using a standardised
questionnaire designed for this study (Table 1).
Information was successfully obtained for over
70% of the households (181/254). Informal discussions with residents also gave many insights into
residential histories, household situations (e.g.,
5
Minist"ere des terres et for#ets du Qu!ebec (1978), Carte cadastrale a"
1/20 000 de Ch#ateauguay (Feuille no. 31 H-4 S.O.), Qu!ebec. (Revised
from 1966 aerial photographs.)
6
Minist"ere des resources naturelles (1994), Carte topographique a" 1/
20 000 de Saint-Jean-Chrysostome (Feuille no. 31 H 04-200-0101) et de
Hemminford (Feuille no. 31 H 04-200-0102), Qu!ebec. (Revised from
1992 aerial photographs.)
ARTICLE IN PRESS
432
Table 1
Visual analysis, landscape trajectory as well as socio-demographic and residential history data used in the Havelock local-scale case study
Variables
Indicators
Empirical materials
Visual eld
Open, ltered, closed
In situ observation
Depth of the visual eld
o100 m, 100 m2 km, >2 km
In situ observation
Prevailing view
Elevated, horizon, closed
In situ observation
Topographical entity
Summit, upper/lower hillside, hillock,

boulder, river terrace, small valley, plain
In situ observation
Viewshed
Very small, small, intermediary, large,

very large
Digital terrain model

(topographical database)
Visual accessibility
High, intermediary, low, none
In situ observation
o1%, 1%50%, >50%
Aerial photographs
Areas remaining under cultivation
o1%, 1%25%, >25%
Aerial photographs
Areas remaining under pasturea
o1%, 1%15%, >15%
Aerial photographs
o1%, 1%20%, >20%
Aerial photographs
Pasture-to-foresta
o1%, 1%20%, >20%
Aerial photographs
Residential transformation
Constructed before 1965, after 1965

original demolished (after 1965)
In situ observation,
aerial photographs
Original farm building dynamics
No building before 1965, all original

buildings presents, one building demolished,
many buildings demolished
aerial photographs
New building dynamics (1965)
No new building, one new building,

new buildings
aerial photographs
Occupation
Full/part-time farmer, farm worker, small
Interviews
Areas remaining under forest covera

a
Abandoned land-to-forest
business, salaried worker, professional,

retired, without earned income
Place of work
Havelock, adjacent municipalities,

Mont!er!egie area, Montreal area, without
xed place of work, not applicable
Interviews
Place of birth

Mont!er!egie area, Montreal area, other
places
Interviews
Previous place of residence

Mont!er!egie area, Montreal area, other
places
Interviews
Residential occupation type
Permanent, weekly, seasonal,

occasional, transitional (toward
permanence), non-resident
Interviews
Age group
1824, 2544, 4564, 65 years and over
Interviews
Family acquisition
Present, absent
Interviews
Year of personal acquisition
Before 1960, 196069, 197079, 198089,

1990 and after
Interviews
Relatives
Present, absent
Interviews
Relative to the overall lot area.
ARTICLE IN PRESS
433
Fig. 5. Resulting landscape contextsHavelock township (N=181).
unstable economic conditions) and the reasons

for migrating7 to Havelock. This additional qualitative material enabled us to situate our results within a
broader perspective.
Given the complex nature of the database, and in
order to permit the differentiation between phenomena
at the scale of each lot, cluster analysis8 was used to
discern the dominant tendencies related to residential
settlement patterns and landscape trajectories.
First, signicant relationships between the visual
analysis data (active indicators) identied for each
lot and residents data (illustrative indicators) were
singled out. Using this data, we were able to evaluate
the potential attractiveness of landscape contexts
7
The distinction between local and migrant populations is based on
the place of birth and available information on the previous place of
residence.
8
Based on the distance between each pair of lots, clusters are
established from active indicators using SPAD software (v. 3.2).
Illustrative variables are then projected onto multi-dimensional axes in
order to reveal signicant correlations between the different clusters
and potential explanatory factors. For each variable of these clusters,
the percentages that diverge signicantly from the percentage of the
whole population were calculated using a statistical test.
on residential settlement. Cluster analysis also allowed

us to distinguish landscape trajectories at the lot
scale. In this case, land use change and built environment transformation act as active variables while visual
and socio-demographic indicators are considered illustrative ones.
2.2.2. Residential settlement patterns
From the four landscape contexts identied
(Woodlot-closed view; Upper hillside-panoramic
view; Agricultural lowland-limited view; Lower
hillside-potential view) for Havelock Township
(Fig. 5), our results showed that the population
does not spread uniformly throughout the different
landscape contexts. Some widely recognised sociodemographic characteristics associated with rural
migration movements (Kayser, 1990; Beesley, 1991;
Dahms and Hallman, 1991; Thomson and Mitchell,
1998; Walmsley et al., 1998) are signicantly correlated
to specic landscape contexts. Urban background,
professional occupation and age group (4564) are
highly associated with Woodlotclosed view and
Upper hillsidepanoramic view. Specic landscape
attributes sustain selective rural migration ows, and
ARTICLE IN PRESS
434
act distinctively on the overall recomposition process.

However, these signicant tendencies do not represent
the only way in which these rural migration phenomena
occur. For example, urban migrant relocation to
farming lowlands is another manifestation of rural
recomposition. The differentiated residential settlement patterns associated with urban-to-rural migrants suggest new residents relationship to space.
Our discussion with migrants reinforces this assertion.
As also observed by other researchers (Hervieu and
Viard, 1996; Ryan, 1998; ORourke, 1999), our informal
interviews suggest that lifestyle enhancement values
(tranquility, proximity to nature, privacy and rural
ambiance) represent some of the migrants motives for
living in the countryside.
2.2.3. Landscape trajectories
The ve landscape trajectories outlined (Signicant
persistence of agricultural activities; Moderate persistence of agricultural activities; Advanced agricultural
abandonment; Former agricultural lots in transition
and Residential development lots) identify distinct
ways in which individual lots evolve. Lot evolution
ranges from: the persistence of farming activities, to
transformations in housing development or hobby
farming, to agricultural abandonment. Although limited
to a specic area, only the Advanced agricultural
abandonment trajectory is signicantly associated with
urban migrants (Paquette and Domon, 2001a). While
half of these residents can be associated with the process
of agricultural abandonment (i.e. purchased their lots
between 1960 and 1979), the other half acquired their
lots as this abandonment was well under way (after
1980). Since it is conned to a limited area (16 lots), this
single result does not by itself suggest that a migrants
lot occupancy results in instantaneous abandonment of
farming. The migrant population takes part in several
landscape trajectories simultaneously, though with
varying intensity. The participation of many urban
migrants in the continuation of agricultural activities is
a clear illustration of the complex nature of these
landscape dynamics. For example, while some local
farmers cultivate migrant landowners lots through
rental agreements. The migrant landowners themselves
assume agricultural activities on other lots. Therefore,
our results prevent us from pointing out a clear
differentiation between the migrant and the local
population when considering their respective inuences
on these landscape trajectories.
A signicant part of this complexity may be related to
the time lags within which these processes have evolved
(i.e. socio-demographic changes and land use dynamics).
For instance, built elements are generally subjected to
more rapid alterations in comparison to the more
gradual changes associated with land use (e.g. land
abandonment). To better situate the impact of the
selective residential settlements previously observed on

rural landscapes, a closer investigation is required. This
investigation should be undertaken at a scale where
these processes are the most clearly expressed, namely,
at the domestic scale.
2.3. Rural landscape changes: a domestic-scale survey
Given the attractiveness of rural landscapes and the
magnitude of the in-migration movement, it is crucial to
show how these new rural populations affect the
qualities of the landscape itself (Paquette and Domon,
2001b). A characterisation of residential practices
becomes essential given that domestic space is one of
the only areas where individuals can exert control on the
landscape (Despre! s, 1991). Domestic space also reects
personal identity (Hummon, 1989; Abu-Ghazzeh, 2000),
and the ways in which residents interact with others and
their surroundings. Ultimately, our investigation at this
scale shows how domestic practices shape the evolution
of the rural landscape. It also reveals attitudes vis-a" -vis
the rurality that upholds them.
2.3.1. Study area and database development: a domesticscale study framework
Within Havelock Township, the domestic-scale study
constructs a detailed portrait of the diverse residential
practices (181 lots). These were observed using various
empirical methods (in situ observations, interviews,
analysis of local building permits, etc.). The indicators,
drawn from the work of Langlois et al. (1993), Phipps
et al. (1994) and Archambault (1995), are selected to
emphasise small-scale domestic landscape changes as
well as spatial appropriation (Table 2). They refer to the
homes characteristics, the use of exterior residential
space, and to land use at the lot scale.
*
Home characteristics. Observations have allowed the

evaluation of home conditions and exterior materials
as well as the homes orientation and its distance
from the road (Table 2). Additional information on
building renovation was obtained through local
building permits based on the method used by
Halseth and Rosenberg (1990).
Use of exterior space. These criteria correspond to the
presence of temporary (ower, vegetable garden,
property maintenance, visual links to the road) and
semi-permanent domestic space use and management
(lawn and shrub area; tree areas, their spatial
distribution and age; ornamental elements; type of
boundaries; garden sheds) (Table 2).
Land use. This term refers to the percentage of the
lots area occupied by the dominant land use types
(1997), as interpreted at the scale of each lot
using 1:15 000 black and white aerial photographs
(Table 2).
ARTICLE IN PRESS
435
Table 2
Residential practices (home characteristics, use of exterior space and land use) documented for the Havelock domestic-scale case study
Variables
Indicators
Empirical materials
Home condition
Excellent, good, poor
In situ observation
Exterior facing
Brick, synthetic, wood, stone, other
In situ observation
Major renovation
Present, absent
In situ observation
building permits
interviews
Home orientation
Facing road, facing view
In situ observation
Building renovation
Present, absent, no construction
In situ observation
building permits
interviews
Distance from the road
Less than 5 m, 5100 m, >100 m
In situ observation
Boundary type
Hedge, stone wall, wire fencing
In situ observation
Visual link to the road
direct, ltered, none
In situ observation
Lawn area
>80%, 5080%, o50%
In situ observation
Shrub areaa
1030%, o10%, none
In situ observation
Abundant, frequent, few, absent
In situ observation
Flowers
Flower bed area
1040%, o10%, none
In situ observation
Flower bed development
Recent, well-established
In situ observation
Vegetable garden areaa
o10%, none
In situ observation
Tree areaa
>50%, 1050%, o10%
In situ observation
Spatial distribution of trees
Isolated, in rows, grouped, alongside woodlot, woodlot
In situ observation
Tree development
Recent planting, mature planting
In situ observation
Property condition
Excellent, good, poor
In situ observation
Ornamental elements
Old farming implements,

decorative stone walls, pond
In situ observation
Present, absent
In situ observation
Area under forest cover (1997)
o1%, 149%, 5080%, >80%
Aerial photographs
Area of abandoned land (1997)b
o1%, 110%, >10%
Aerial photographs
Area under cultivation (1997)
o1%, 125%, >25%
Aerial photographs
Area under pasture (1997)b
o1%, 110%, >10%
Aerial photographs
Garden shed
b
a
b
Relative to the overall area of residential use.

Relative to the overall lot area.
To characterise the context of these practices, we take

advantage of the socio-demographic and residential
history information and the visual analysis data already
collected at the lot scale (Table 1). Then, we used cluster
analysis to discern signicant relationships existing
between these residential practices (active indicators),
their visual landscape contexts and resident proles
(illustrative indicators).
the converging actions of the local and the migrant

populations (Non-farming residential lot and Farming lot trajectories), the others are signicantly
associated with clear migration phenomena and landscape context characteristics (i.e. Landscape aesthetic
lot and Declining lot trajectories).
2.3.2. Domestic landscape trajectories

Seven residential practice clusters are identied
(Fig. 6). From these clusters dominant tendencies were
revealed through the lots use (ex. residential, agricultural) or its evolution (ex. revival, decline). These
tendencies enabled us to identify four domestic landscape trajectories (Fig. 6). While the rst two result from
Non-farming lot trajectory. Given their number and

their visual proximity, non-farming residential lots
have greatly modied rural landscapes by introducing
suburban dwelling models into rural areas. This
development is an example of the ongoing loss of the
local landscapes uniqueness. This trajectory tends to
be implemented by locals and, to a lesser extent, by
migrant residents.
ARTICLE IN PRESS
436
Fig. 6. Overview of residential practice clusters, their associated socio-demographic proles and landscape contexts as well as the resulting domestic
landscape trajectories.
*
Farming lot trajectory. The imprints of domestic

farming are still perceptible in these lots, but are no
longer exclusively associated with the traditional
local farmer. Both migrant and local residents are
associated with this trajectory. It is the extent of
agricultural activities, rather than the origins of its
residents, that seems a determining factor in the
residential practices observed. The lower the importance of farming activities, the greater the emphasis
on domestic beautication (owers, shrubs, etc.) and
enhancement of traditional farming landmarks (stone
walls, isolated mature trees).
Landscape aesthetic lot trajectory. The combination
of particular domestic practices (home maintenance;
yard landscaping; home remoteness) with specic
landscape contexts (upland lots, panoramic views)
and their residents proles (urban background,
professionals) characterised the landscape aesthetic
trajectory. Criteria such as views, proximity to
nature, and privacy herald the transformation of
rural landscapes into places to live rather than
exclusively to farm. Paradoxically, some residential

practices (e.g. abandoned farm buildings, obstruction
of views through reforestation of lot fronts) tend to
accelerate the loss of local landscape character.
Declining lot trajectory. In contrast to the activities
that create the previous trajectories, poor maintenance, fewer home and outbuilding renovations, the
absence of ornamental vegetation, and the lack of
traditional rural elements (isolated trees or hedgerows) are indicative of a lots decline. While less
associated with accessibility to views, some residential
settings are intimately associated with migrant
populations from urban backgrounds that settled in
the township over the past few years (after 1990). In
contrast to the previous lot trajectory, informal
interviews suggest that these less afuent migrants
are mobile refugees (Brunet, 1980; Fitchen, 1994)
seeking lower cost rural housing.
The last two trajectories, which take place within

distinct landscape contexts, stand for the most visible
ARTICLE IN PRESS
domestic landscape changes associated with urban-torural migration. These opposing trajectories emerge
from the conjunction of both social and physical
landscape dynamics. Generally, our results suggest that
the migrants relocation into rural areas, while often
associated with landscape aesthetic motivations and
lifestyle choices (Coppack, 1988; Halliday and
Coombes, 1995; Swafeld and Fairweather, 1998;
Walmsley et al., 1998), do not produce unique consequences, such as a systematic rural gentrication
movement. The results also provide some insight into
the various underlying motives of migration as well as
their distinct impacts on domestic-scale landscape
dynamics.
3. Regional, local and domestic dynamics relationships

Beyond identifying the diverse phenomena expressed
at regional, local and domestic scales, it is necessary to
examine the way these processes are revealed at distinct
spatial levels. This rst section looks at local and
domestic-scale interactions. Then, we proceeded to show
how these small-scale insights consequently shed light
on larger scale territorial dynamics.
3.1. Local and domestic cross-scale examination
Many authors have already recognised the diverse
manifestations of rural in-migration (Kayser, 1990;
Halfacree and Boyle, 1998). When considered from a
landscape dynamics perspective, some rural migration
patterns require small-scale examination, particularly
those emerging within well-dened landscape contexts
or involved with agricultural development.
Rural migrations have been associated with landscape
aesthetic motives (Halliday and Coombes, 1995; Swafeld and Fairweather, 1998), usually through large-scale
surveys. Our approach allows smaller-scale processes
and changes to become visible. The approach reveals
distinct migrants residential settlement patterns and
how they are related to specic landscape contexts at the
lot scale (Fig. 5). More importantly, via residential
practices, these migrant relocations introduced specic
domestic landscape trajectories (Fig. 6). For example,
the evolution of upland lots settled by migrants
exemplies the shift from farming to residential functions when contemplative attitudes towards the landscape are promoted (Fig. 6, Landscape aesthetic lot
trajectory). These domestic practices are expressed
through home beautication (including a high level of
home maintenance and more elaborate landscaping of
yards), the enhancement of privacy, and proximity to
nature (remoteness of the home, and its visual inaccessibility from the road) as well as aesthetic consumption
values, such as a view-oriented house (Fig. 7a and b).
437
Specic residential location patterns as well as distinct

uses and visible imprints at the domestic scale are
indicative of the new social demands placed on rural
areas. They exemplify the transition of these areas as
places that emphasise basic material needs and primary
production into places valued for non-material aesthetic
values and consumption that characterise urban and
non-farming interests in the countryside (Hervieu and
Viard, 1996; Boyle and Halfacree, 1998; ORourke,
1999). However, these rural in-migration manifestations
do not represent for one all-encompassing process; other
forms of social recomposition coexist with these new
values. Migrant settlement within agricultural lowlands,
and their role in the evolution of agricultural landscape
exemplies this phenomenon.
Our eld observations show that some migrants
cultivate their land. Agricultural activities for a similar
proportion of migrants persist in collaboration
with local farmers through rental agreements. For
the lots where the information could be obtained (20),
this informal cooperation between migrants and
local farmers was concentrated more on uplands (9/12)
than in lowlands (2/8). These distinct spatial patterns
could be indicative of different motivations in upland
versus lowland migrant landowners. More fundamentally, given that many indications of agricultural
persistence are observed in parallel with in-migration
settlement, do they suggest a relative dissociation
between residential dynamics and land use development
(Fig. 7c)?
The following two examples illustrate the interactions
between migrant establishment and land evolution by
way of an examination of local landscape trajectories
and their links with residential practices. For lots where
the advanced agricultural abandonment trajectory
was identied at the local scale only a quarter were
considered to be a neglected agro-forested lot at the
domestic scale. Half (6/12) of these lots were identied
as well-kept agro-forested lots or remote vieworiented lots. Thus, the signs of abandoned-land
use dynamics observed at the local level (Fig. 7d)
are not always associated with a corresponding
decline at the residential level (Fig. 7e). Of the remote
view-oriented lots identied at the domestic scale,
more than 64% of them (9/14) maintain agricultural
activities at the local scale. All cultivation on these
lots is through lease agreements with local farmers
(Fig. 7c). This indicates that evidence of the landscape aesthetic domestic trajectory can coexist with
continued agricultural exploitation. Domestic changes
and local land use development clearly do not exist
in a one-way relationship. Faster and more visible
transformations in residential practices could evolve in
conjunction with more gradual local agricultural
changes. In this context, the time lag between sociospatial processes and their potential inuence on
ARTICLE IN PRESS
438
Fig. 7. Some small-scale landscape illustrations of: remote view-oriented residential settlements with (a) no and (b) ltered visual links to the road;
(c) a well-kept migrants residence with abandoned farm buildings, the lots are cultivated through rental agreements with local farmers; (d) a case of
advanced agricultural abandonment; (e) preservation of a stone wall illustrating traditional farming landmarks and (f) a conventional farming lot
with new farm buildings.
agricultural activities needs to be investigated in greater

detail.
3.2. Regional, local and domestic cross-scales trends
Regional scale dynamics (Fig. 3) are characterised by
an increasing dissociation between agricultural and
socio-demographic trajectories. As discussed in the
previous section, a similar dissociation occurs at the
local scale. From this viewpoint, the landowners sociodemographic proles do not seem to be a determining
factor with regard to local landscape trajectories. As
previously recognised at local (Domon et al., 1993;
Paquette and Domon, 1997; Pan et al., 1999) and
regional scales (Meeus et al., 1990; Paquette and
Domon, 1999), biophysical attributes (e.g., geomorphological deposits, topography) are more signicantly
associated with land use changes. It could be argued
ARTICLE IN PRESS
that as long as farming is viable, it will be practiced,

whether by local or migrant landowners (Fig. 7f).
Therefore, considering the extent of occupational
changes involved with rural recomposition, an everincreasing number of lots will be cultivated through
rental agreements (Primdahl, 1999, p. 145). One must,
however, question the longevity of such agricultural
dynamics. At what proportion of farming to migrant
population might this agricultural dynamic be disturbed? The combined effect of in-migration and decline
in farming population would ultimately create an
irreversible imbalance between the proportions of
farming and non-farming residents.
Migrants also induce smaller-scale changes that reect
their identity and unveil their meanings of rurality. This
study reveals two opposing domestic-scale trajectories
associated with migrant relocation (Fig. 6). It has
previously been shown how some exurban dwellers
search for a rural idyll (Swafeld and Fairweather,
1998). This attitude seems to have found expression in
specic residential practices and locations (e.g., upper
hillside settlements) within the study area. For other
migrants, rural areas are associated with very different
aspirations. For the less afuent and mobile urban
migrants (Brunet, 1980; Fitchen, 1994), the attractiveness of low-cost rural housing seems to represent the
most likely reason for migration. Despite the revitalisation caused by some types of migration, a greater
number (Hugo and Bell, 1998) is associated with the
declining lot trajectories. Behind the rational
decision-making that appears to motivate such migrant
moves, could other forms of the rural idyll exist? As
Ilbery (1998, p. 4) suggests, people living in rural areas
behave in different ways, aspire to different goals and
have unequal access to what they desire. Clearly, rural
recomposition no longer seems to be synonymous with a
single gentrication process or a rural renaissance.
These previous examples are the most noticeable
expressions of a more complex process. While largescale statistical portraits help to discern the dominant
tendencies, small-scale social recomposition dynamics
and their conjunction with particular landscape contexts
need to be documented via a micro-level based
investigation.
The Havelock case study exemplies the transitional
rural areas shown in the previous regional typology
(Fig. 3). Its specic situation has a number of elements
in common with the agricultural, the periurban and
the amenity archetypal rural trajectories. In fact,
Havelock offers no special leisure or second-home
development appeal (e.g., ski resorts, open water, etc.),
and is removed from other well-known areas with these
amenities (e.g., parts of the Eastern townships and
Laurentians in southern Que! bec; the Georgian Bay area
near Metropolitan Toronto in Ontario; or New York
States Hudson River Valley). The extent, rhythm, and
439
intensity of the recomposition process in Havelock is

clearly of a lesser magnitude than in the aforementioned
areas. However, relative to traditional agricultural
areas, Havelock has experienced a denite farm-toresidential transition within its upland sectors. Therefore, it could be argued that the varied and spatially
delimited landscape qualities (e.g. upper hillsides,
agricultural lowlands) found in Havelock Township
are signicant factors underlying its evolution.
Given the diverse situations revealed through this
multi-scale study, it is useful to identify the dominant
tendencies suspected of shaping future changes. We will
also explore how these trends confront rural citizens
with conicting images of the countryside, as well as
how they challenge conventional planning approaches.
4. Towards ongoing rural changes

Rural areas have experienced continuous transformations with social recomposition contributing to these
changes in many ways. This section discusses three
related trends, demographic evolution, landscape dynamics, and rural community changes that appear to
determine the future of rural areas. It also discusses
planning issues that the rural population must consider
over the next decades.
(i) Demographic evolution trends: Since the 1960s,
many rural areas have experienced demographic
growth (Fuguitt, 1985; Kayser, 1990; Robinson,
1990). Even if turnaround, reversal and rebound
episodes have been identied (Joseph et al., 1988;
Kayser, 1992; Fuguitt et al., 1998; Johnson, 1998),
the increase in rural populations no longer seems a
temporary situation. The best-documented migration ows are those of the second-home owners and
the elderly; commuters migration ows are slightly
less documented (Clout, 1986; Jean, 1989; Dahms
and McComb, 1999). Commuting may become
more important given current labour restructuring
tendencies (job diversication, part-time work,
etc.). Many factors suggest an increase in retirement
migration in the future. As stated by Kontuly
(1998, p. 66), the ageing of the population [means]
a growing pool of potential migrants. The shift of
many Baby Boomers from an economically
active to inactive status could well lead to an
imminent intensication of migration due to retirement. Currently no single theory explains the
redistribution of the rural population (Champion,
1998; Kontuly, 1998; Dahms and McComb, 1999).
Rather, a combination of factors including the
improvement of transportation (road networks)
and communication infrastructure (telecommuting), and their ability to inuence residential
ARTICLE IN PRESS
440
preferences are recognised as supporting elements

in this demographic growth. In addition, changes in
collectively shared environmental attitudes and
better economic conditions for an increasing
amount of future retirees foreshadow a probable
increase in rural recomposition trends.
(ii) Landscape dynamic trends: In southern Que! bec
(Paquette and Domon, 1997, 1999; Pan et al.,
1999), as elsewhere in many western countries
(Meeus et al., 1990; Simpson et al., 1994), the
decline in agricultural land use is concentrated on
uplands characterised by uneven topography and
morainic deposits. In conjunction with this decline,
these areas offer some of the most prized residential
settings for in-migration. Like forestry and agriculture during the past two centuries, the landscape
may now be a resource that contributes to the
redistribution of population and rural redevelopment (Domon, 1997). However, the landscape
cannot be reduced to a single purpose resource,
nor can its planning be similar to traditional
resource management. Nevertheless, one can assume that the rural landscape is close to what
Hirsch calls a positional good, whose enjoyment
depends upon its scarcity (1975; quoted by Newby,
1990, p. 633; see also Murdoch and Day, 1998, p.
190).Agricultural regression provides an opportunity for re-appropriation by in-migrants, and
allows for reforestation, which in turn produces a
landscape homogenisation. For example, forest
cover areas increased by nearly 30% in Havelock
Township between 1968 and 1992. More than 58%
of this growth is a result of pasture-to-forest land
use change. Therefore, the inevitable reforestation
of vast stretches of land induced by agricultural
restructuring limits the diversity of the landscape
resource since it reduces the opportunities of
valued experiences. However, such an evolution
may induce different responses depending upon
whether the observer is a local or a migrant
(Hunziker, 1995, p. 401). While the former appears
to associate land in fallow with a loss of feeling at
home, the latter interprets it more positively, as a
gain of naturalness. The actions of some
migrants may also enhance the loss of distinct
rural landscapes through their own domestic
practices (Fig. 7). These observations call for the
need to better understand landscape dynamics and
their related planning issues. Within the context of
rural repopulation, such issues inevitably entail new
collective aspirations and debates.
(iii) Community changes and rural planning challenges:
As Halfacree and Boyle note (1998, p. 10), some
rural in-migration streams tend to be closely linked
to the hegemonic success of idyll-type social
representations of the rural [...]. Moreover, these
representations tend to introduce new local power

balances within rural community. These emerging
struggles are well expressed by these authors
(Halfacree and Boyle, 1998, p. 11):
[...] the migrants experienced reality of the rural
environment is often perceived to be at odds with
their representational blueprint. Such dissonance is
reected in an increasing number of disputes between
farmers and incomers unhappy with the sights,
sounds and smells of the countryside.
For some, the relative importance of migration inux
within a rural community may be indicative of specic
socio-political trajectories. Thus, the predominance of
migrants over long-standing residents may well reect
the prevalence of preservationist attitudes over developmental ones (Marsden, 1995). From this sociopolitical perspective, many rural changes are typically
characterised by a shift from an agricultural veto to an
in-migrant dominated preservationist veto (Halfacree
and Boyle, 1998, p. 11). However, a more complex social
polarisation may hide behind this overly simplied
preservationist/developmentalist dichotomy. From an
examination of some landscape appropriation practices
in Que! bec, Tremblay and Poullaouec-Gonidec (2002)
suggest that an aesthetic community evolves over time
according to the way in which this community attaches
importance to a given space. In this context, the
migrant-local opposition may not exactly reect the
polarisation that leads to the construction of these
communities. For instance, migrants associated with
conventional farming practices and those associated
with landscape aesthetic consumption practices
(Fig. 6) may project views of the rural landscape that
are at odds with one another. As shown in Havelock
Township (Paquette and Domon, 2001a, b) and elsewhere (Allan and Mooney, 1998), this is all the more
plausible since these social groups are heterogeneous. As
such, how would rural recomposition shape new local
socio-politic conditions? Under which circumstances
and in which manner (e.g. through political actions,
local associative groups, etc.) could a given minority
social group, like urban migrants, inuence the way the
countryside is perceived? How do these perceptions
reect values that may ultimately conict with existing
rural planning strategies and policies?
Undoubtedly, rural repopulation gives rise to new
opportunities as well as new threats for housing,
demographic and economic conditions (Newby, 1990;
Stockdale et al., 2000). Even more importantly, this
repopulation gives rise to new demands for future rural
planning (Paquette and Domon, 2000). The increasing
diversity of aspirations resulting from social recomposition, challenges traditional land (and landscape) planning methods (e.g. farming and forestry development
Vos and Meekes, 1999; Mather, 2001). In southern
ARTICLE IN PRESS
Que! bec, new rural development initiatives that integrate

landscape dimensions have conrmed the necessity of
multiple-use rural planning (Paquette and Domon,
2000). Such collective action has promoted the agreement and the support of individuals (landscape producers as well as landscape consumers) formerly
advocating divergent views. In order to manage landscape changes as well as the various experiences and
functions they support, future rural development must
move beyond the role of traditional regulation models
promoted by sector-specic institutions and policies.
Within this development model, rural residents are also
invited to clarify what kind of landscapes (and communities) they want to create, or even redene (Domon and
Paquette, 1999), for the benet of both migrant and
local residents (Stockdale et al., 2000).
5. Conclusion
At the regional, local and domestic scale, this
exploratory study revealed specic trends in social
recomposition and landscape dynamic relationships.
Regional typological outlines show the important
diversity of contemporary rural areas. Moreover, they
illustrate the increasing dissociation between agricultural and socio-demographic trajectories. The paradoxical
demographic growth of some rural municipalities that
nevertheless show a regression in agricultural practices
exemplies this tendency.
The changes in lot occupation documented within the
Havelock Township case study revealed new residential
settlement patterns induced by the inux of migrants.
Some of these in-migration ows appear to be signicantly associated with specic landscape contexts.
Although it cannot be explained by a single process, the
appeal of some landscape characteristics seem to be a
determining force shaping the social recomposition of
rural communities.
Rural in-migration shows dissimilar inuences on
ongoing landscape dynamics. With the exception of
conned agricultural abandonment trajectories, migrant
relocation does not seem to imprint singular landscape
changes at the local scale. In contrast, our results
revealed that migrants are signicantly associated with
specic domestic practices once settled. These results
suggest that social in-migration dynamics and local land
use development do not evolve in a one-way relationship. More fundamentally, the phenomena observed are
indicative of the new residents identities in rural places
and generate specic values for the landscapes qualitative dimensions. From a planning perspective, these new
social interests challenge resource-based development
policies and actors to transcend their productionoriented goals (Marsden, 1998; Mather, 2001; Erickson
et al., 2002). Future planning recommendations must be
441
more concerned with social interests as well as their

explicit and implicit inuences on landscape transformations and other valued rural experiences.
This exploratory study paves the way for future
research. Beyond recognising the complex relationships
between social and landscape dynamics, it is crucial to
better understand how these dynamics emerge and
which combination of factors sustains them.
Since landscape attractiveness appears to be signicantly associated with specic in-migration ows, a
more in-depth investigation of the underlying motives
for migration is required (Halliday and Coombes, 1995).
Many researchers stress the need for an analysis of
discourse and emphasise its heuristic nature in unveiling
rural migratory processes (Boyle and Halfacree, 1998).
In this sense, individual narratives are helpful in
revealing how specic social representations shape
distinct images of the rural landscape for migrants and
locals. Exploring individual motives will also help
understand how landscape consumption-based images
compete with, and/or inuence, more everyday and
deep-rooted images (Lowenthal, 1997). For landscape
planning, such an analysis is helpful in anticipating
occurrences of social polarisation and creating innovative strategies able to reconcile the divergent views
expressed by various social networks.
Ultimately, it is the relative inuence of the landscape
(in both physical and symbolic forms) as a driving force
in social recomposition that must be understood. Such
an evaluation must be conducted from both spatial and
temporal perspectives. As shown in our case study,
landscape attractiveness is a contributing factor in
shaping spatially selective recomposition processes.
More investigation is needed to document the spatial
recurrence of this process. At the same time, we have to
evaluate how such manifestations may show themselves
within distinct time lags from one region to the other. As
the landscape refers to an everchanging set of culturally
constructed meanings (Poullaouec-Gonidec, 1994; Poullaouec-Gonidec and Domon, 1999), it is also necessary
to emphasise the way collective representations and
emblems of the rural landscape, both old (pastoral,
picturesque, etc.) and new (such as the environmental
aestheticTremblay and Poullaouec-Gonidec, 2002),
persist or change.
A detailed examination of smaller-scale practices
reects conscious and unconscious motives (Lewis,
1979). Given that some domestic practices result from
non-verbalised motives or unconscious decision-making,
a comparative analysis of residents practices and
motivations allows for a better assessment of the gap
that potentially exists between discernible practices and
their verbalised expression. Such analysis may help to
anticipate future landscape transformations by identifying latent changes, namely, potential practices expressed
through residents discourse and aspirations. For
ARTICLE IN PRESS
442
instance, it could help to clarify the role of migrants,

local landowners, and tenants with respect to agricultural activities by distinguishing between their respective
intentions and real participation in these activities. This
empirical material could allow us to show how informal
collaborations between migrant landowners and local
tenant farmers contribute to a spontaneous social
convergence for the benet of the maintenance of
landscape attributes. Conversely, it could also better
dene when and where farming practices are vulnerable
to agricultural abandonment. This knowledge is essential to formulate landscape planning recommendations.
Acknowledgements
We gratefully acknowledge Tracey Hesse and Kevin
Marks for editing, Martin Joly and Karine Faure for
their help on cartography. We would like to thank two
anonymous reviewers for their helpful comments on an
earlier draft of this manuscript. Fellowships from the
Social Sciences and Humanities Research Council of
Canada (SSHRC), the Universite! de Montre! al and the
Canadian Forest Service helped Sylvain Paquette to
complete this research paper. Ge! rald Domon received
research grants from the SSHRC and the Fonds pour la
Formation de Chercheurs et lAide a" la Recherche
!
(FCAR) of the ministe" re de lEducation
du Que! bec.
References
Abu-Ghazzeh, T.M., 2000. Environmental messages in multiple-family
housing: territory and personalization. Landscape Research 25,
97115.
Allan, J., Mooney, E., 1998. Migration into rural communities:
questioning the language of counterurbanisation. In: Boyle, P.,
Halfacree, K. (Eds.), Migration into Rural Areas: Theories and
Issues. Wiley, Chichester, pp. 280302.
Archambault, L., 1995. Les fermes centenaires du canton de Brome
comme exemple de mutation des paysages agro-forestiers du
champ urbain de Montr!eal. Unpublished M.Sc.A. Thesis, Facult!e
de lam!enagement, Universit!e de Montr!eal.
Baudry, J., 1993. Landscape dynamics and farming systems:
problems of relating patterns and predicting ecological changes.
In: Bunce, R.G.H., Ryszkowski, L., Paoletti, M.G. (Eds.), Landscape Ecology and Agroecosystems. Lewis Publishers, Boca Raton,
pp. 2140.
Beesley, K.B. (Ed.), 1991. Rural and Urban Fringe Studies in Canada.
Geographical Monographs, Vol. 21. Atkinson College, York
University, North York.
Berque, A., 1990. M!ediances, de milieux en paysages. Reclus,
Montpellier.
Boyle, P., Halfacree, K., 1998. Migration into rural areas: a collective
behaviour framework. In: Boyle, P., Halfacree, K. (Eds.), Migration into Rural Areas: Theories and Issues. Wiley, Chichester, pp.
303316.
Brunet, Y., 1980. Lexode urbain, essai de classication de la
population exurbaine des Cantons de lEst. Canadian Geographer
24, 385405.
Brunger, A.G., Bowles, R.T., Wurtele, S., 1991. Patterns of fringe

community: post world war 2 settlement near Peterborough,
Ontario. In: Beesley, K.B. (Ed.), Rural and Urban Fringe Studies
in Canada, Geographical monographs No. 21. Atkinson College,
York University, North York, Canada, pp. 183207.
Bryant, C.R., Russwurm, L.H., McLennan, A.G., 1982. The Citys
Countryside: Land and its Management in the Rural-Urban
Fringe. Longman, London.
Champion, T., 1998. Studying counterurbanisation and the rural
population turnaround. In: Boyle, P., Halfacree, K. (Eds.),
Migration into Rural Areas: Theories and Issues. Wiley, Chichester, pp. 2140.
Clark Labs for Cartographic Technology and Geographic Analysis,
1997. Idrisi Geographic Analysis System. Version 2.0, Worcester,
USA.
Clout, H., 1986. Population changes in rural Britain: a review. Espace,
Populations et Soci!et!es 3, 1932.
Coppack, P.M., 1988. The role of amenity. In: Coppack, P.M.,
Russwurm, L.H., Bryant, C.R. (Eds.), Essays on Canadian Urban
Process and Form III: The Urban Field, Department of Geography
Publication Series No. 30. University of Waterloo, Waterloo, pp.
4155.
Dahms, F., Hallman, B., 1991. Population change, economic activity
and amenity landscapes at the outer edge of the urban fringe. In:
Beesley, K.B. (Ed.), Rural and Urban Fringe Studies in Canada,
Geographical monographs No. 21. Atkinson College, York
University, North York, Canada, pp. 6790.
Dahms, F., McComb, J., 1999. Counterurbanization; interaction and
functional change in a rural amenity areaa Canadian example.
Journal of Rural Studies 15, 129146.
Despr!es, C., 1991. The meaning of home: literature review and
directions for future research and theoretical development. Journal
of Architecture and Planning Research 8, 96115.
Domon, G., 1997. Mise en valeur des paysages et petites collectivit!es:
enjeux et perspectives. In: Actes du Symposium sur la ruralit!e et le
d!eveloppement des petites collectivit!es. UQAT, Chaire en d!eveloppement des petites collectivit!es, pp. 110116.
Domon, G., Bouchard, A., Gari!epy, M., 1993. The dynamics of the
forest landscape of Haut-Saint-Laurent (Qu!ebec, Canada): interactions between biophysical factors, perceptions and policy.
Landscape and Urban Planning 25, 5374.
Domon, G., Paquette, S., 1999. Les zones rurales du sud du
Qu!ebec: territoires de mouvance, territoires a" reconstruire. In:
Poullaouec-Gonidec, P., Gari!epy, M., Lassus, B. (Eds.), Le
paysage: territoire dintentions. Harmattan, Montr!eal and Paris,
pp. 5578.
Dubot, F., Lizet, B., 1995. Pour une ethnologie du paysage. In:
Paysage au pluriel. Pour une approche ethnologique des paysages.
Minist"ere de la Culture, Mission du Patrimoine ethnologique.
!
Editions
de la Maison des sciences de lhomme, Paris, pp. 225240.
Egan, A.F., Luloff, A.E., 2000. The exurbanization of Americas
forests: research in rural sociology. Journal of Forestry 98, 2630.
Erickson, D.L., Ryan, R.L., De Young, R., 2002. Woodlots in the
rural landscape: landowner motivations and management attitudes
in a Michigan (USA) case study. Landscape and Urban Planning
58, 101112.
Fitchen, J.M., 1994. Residential mobility among the rural poor. Rural
Sociology 59, 416436.
Fuguitt, G.V., 1985. The nonmetropolitan turnaround. Annals of the
Review of Sociology 11, 259280.
Fuguitt, G.V., Beale, C.L., Fulton, J.A., Gibson, R.M., 1998. Recent
trends in nonmetropolitan cities and villages: from the turnaround,
through reversal, to the rebound. In: Schwarzweller, H.K., Mullan,
B.P. (Eds.), Research in Rural Sociology and Development: Focus
on Migration, Vol. 7. Jai Press Inc., Stamford (Ct) and London
(England), pp. 121.
ARTICLE IN PRESS
Gerardin, V., Lachance, Y., 1997. Vers une gestion int!egr!ee des bassins
versants. Atlas du cadre e! cologique de r!ef!erence du bassin versant
de la rivi"ere Saint-Charles, Qu!ebec, Canada. Minist"ere de
lEnvironnement et de la Faune du Qu!ebec, Minist"ere de
lEnvironnement du Canada, Qu!ebec.
Gorton, M., White, J., Chaston, I., 1998. Counterurbanisation,
fragmentation and the paradox of the rural idyll. In: Boyle, P.,
Halfacree, K. (Eds.), Migration into Rural Areas: Theories and
Issues. Wiley, Chichester, pp. 215235.
Halfacree, K., Boyle, P., 1998. Migration, rurality and the postproductivist countryside. In: Boyle, P., Halfacree, K. (Eds.),
Halliday, J., Coombes, M., 1995. In search of counterurbanisation:
some evidence from Devon on the relationship between patterns of
migration and motivation. Journal of Rural Studies 11, 433446.
Halseth, G., Rosenberg, M., 1990. Conversion of recreational
residences: a case study of its measurement and management.
Canadian Journal of Regional Science 13, 99115.
Hart, J.F., 1998. The Rural Landscape. The Johns Hopkins University
Press, Baltimore, London.
Hervieu, B., Viard, J., 1996. Au bonheur des campagnes (et des
!
provinces). Editions
de lAube, La Tour dAigues, France.
Hugo, G., Bell, M., 1998. The hypothesis of welfare-led migration to
rural areas: the Australian case. In: Boyle, P., Halfacree, K. (Eds.),
Hummon, D.M., 1989. House, home, and identity in contemporary
American culture. In: Low, S.M., Chambers, E. (Eds.), Housing,
Culture, and Design: A Comparative Perspective. University of
Pennsylvania Press, Philadelphia, pp. 207228.
Hunziker, M., 1995. The spontaneous reafforestation in abandoned
agricultural lands: perception and aesthetic assessment by locals
and tourists. Landscape and Urban Planning 31, 399410.
Ilbery, B., 1998. The Geography of Rural Change. Prentice-Hall,
England.
Ilbery, B., Bowler, I., 1998. From agricultural productivism to postproductivism. In: Ilbery, B. (Ed.), The Geography of Rural
Change. Prentice-Hall, England, pp. 5784.
Jacobs, P., Gaudreau, R., Lalonde, G., 1986. M!ethode danalyse
visuelle pour lint!egration des infrastructures de transport. Minist"ere des Transports du Qu!ebec, Service de lEnvironnement.
Jean, B., 1989. La question rurale: la ruralit!e et sa sociologie.
Recherches sociologiques 20, 287309.
Jean, B., 1997. Territoires davenir: Pour une sociologie de la ruralit!e.
Presses de lUniversit!e du Qu!ebec, Qu!ebec.
Johnson, K.M., 1998. Renewed population growth in rural
America. In: Schwarzweller, H.K., Mullan, B.P. (Eds.),
Research in Rural Sociology and Development: Focus on
Migration, Vol. 7. Stamford (Ct), and London (England), Jai
Press Inc, pp. 2345.
Jollivet, M., 1988. Du paysan a" lagriculteur: le changement social dans
le monde rural. In: Mendras, H., Verret, M. (Eds.), Les champs de
la sociologie fran@aise. Armand Colin, Paris, pp. 4961.
Jollivet, M. (Ed.), 1997. Vers un rural postindustriel: rural et
environnement dans huit pays europ!eens. LHarmattan, Paris,
Montr!eal.
Joseph, A.E., Keddie, P.D., Smit, B., 1988. Unravelling the population
turnaround in rural Canada. Canadian Geographer 32, 1730.
Kayser, B., 1990. La Renaissance Rurale: Sociologie des campagnes du
monde occidental. Armand Collin, Paris.
Kayser, B., 1992. La renaissance rurale a" l!epreuve du recensement de
1990. In: H!erin, R. (Ed.), G!eographie Sociale: Quelles campagnes
pour demain? Centre de Publications de lUniversit!e de Caen,
Angers, pp. 397404.
443
Kontuly, T., 1998. Contrasting the counterurbanisation experience

in European nations. In: Boyle, P., Halfacree, K. (Eds.),
Langlois, A., Phipps, M., Leroux, D., 1993. Lanalyse des paysages
ethnoculturels de lOntario de lEst et du Pontiac (Qu!ebec) par
lanalyse des correspondances. Cahiers de G!eographie du Qu!ebec
37, 539551.
Lewis, P.F., 1979. Axioms for reading the landscape: some guides to
the American scene. In: Meinig, D.W. (Ed.), The Interpretation of
Ordinary Landscapes: Geographical Essays. Oxford University
Press, New York, pp. 1132.
Lowenthal, D., 1997. European landscape transformations: the
rural residue. In: Groth, P., Bressi, T.W. (Eds.), Understanding
Ordinary Landscapes. Yale University Press, New Haven, London,
pp. 180188.
Lucy, W.H., Phillips, D.L., 1997. The post-suburban era comes to
Richmond: city decline, suburban transition, and exurban growth.
Landscape and Urban Planning 36, 259275.
Mari!e, M., Viard, J., 1988. La campagne invent!ee. Actes Sud, Arles,
France.
Marsden, T., 1995. Beyond agriculture? Regulating the new rural
spaces. Journal of Rural Studies 11, 285296.
Marsden, T., 1998. New rural territories: regulating the differentiated
rural spaces. Journal of Rural Studies 14, 107117.
Mather, A.S., 2001. Forests of consumption: postproductivism,
postmaterialism, and the postindustrial forest. Environment and
Planning C: Government and Policy 19, 249268.
Meeus, J.H.A., Wijermans, M.P., Vroom, M.P., 1990. Agricultural
landscape in Europe and their transformation. Landscape and
Urban Planning 18, 289352.
Mougenot, C., 1986. Les populations rurales: e! volutions et perceptions
en Wallonie. Espace, Populations et Soci!et!es 3, 1118.
Murdoch, J., Day, G., 1998. Middle class mobility, rural communities
and the politics of exclusion. In: Boyle, P., Halfacree, K. (Eds.),
Murdoch, J., Pratt, C.A., 1993. Rural studies: modernism, postmodernism and the post-rural. Journal of Rural Studies 9, 411427.
Newby, H., 1990. Revitalizing the countryside: the opportunities and
pitfalls of counter-urban trends. Journal of Royal Society of Arts
138, 630636.
ORourke, E., 1999. Changing identities, changing landscapes:
human-land relations in transition in the Aspre, Rousillon.
Ecumene 6, 2950.
Pan, D., Domon, G., de Blois, S., Bouchard, A., 1999. Temporal
(19581963) and spatial patterns of land use changes in Haut-SaintLaurent (Qu!ebec, Canada) and their relation to landscape physical
attributes. Landscape Ecology 14, 3552.
Paquette, S., Domon, G., 1997. The transformation of the agroforestry
landscape in nineteenth century: a case study in southern Quebec
(Canada). Landscape and Urban Planning 37, 197209.
Paquette, S., Domon, G., 1999. Agricultural trajectories (19611991),
resulting agricultural proles and current sociodemographic
proles of rural communities in southern Qu!ebec (Canada): a
typological outline. Journal of Rural Studies 15, 279295.
Paquette, S., Domon, G., 2000. Le paysage comme agent de
recomposition des communaut!es rurales du sud du Qu!ebec:
# e,
nouvelles possibilit!es, nouvelles exigences. In: Carrier, M., Cot!
! ements dun d!ebat
S. (Eds.), Gouvernance et territoires ruraux: El!
sur la responsabilit!e du d!eveloppement. Presses de lUniversit!e du
Qu!ebec, Qu!ebec, pp. 189222.
Paquette, S., Domon, G., 2001a. Trends in rural landscape development and sociodemographic recomposition in southern Quebec
(Canada). Landscape and Urban Planning 55, 215238.
ARTICLE IN PRESS
444
Paquette, S., Domon, G., 2001b. Rural domestic landscape changes: a

survey of the residential practices of local and migrant populations.
Landscape Research 26, 367395.
Phipps, M., Langlois, A., Jiang, W., 1994. Les marqueurs de lidentit!e
ethno-linguistique dans les paysages ruraux: LOntario de lest et le
Pontiac (Qu!ebec). Canadian Geographer 38, 6175.
Poudevigne, I., Alard, D., 1997. Landscape and agricultural patterns
in rural areas: a case study in the Brionne Basin, Normandy,
France. Journal of Environmental Management 50, 335349.
Poullaouec-Gonidec, P., 1994. Esth!etique des paysages de la modernite! . Trames 9, 2934.
Poullaouec-Gonidec, P., Domon, G., 1999. Nature et contribution du
suivi de l!evolution des paysages: le cas du SMVP (Syst"eme de
monitoring visuel des paysages). In: Gervais, D. (Ed.), Pr!evoir
lavenir: les e! valuations dimpacts pour un nouveau si"ecle. Les actes
du 4e"me Colloque international des sp!ecialistes francophones en
e! valuation dimpacts. Secr!etariat francophone de lAssociation
internationale pour l!evaluation dimpacts, University of Strathclyde, Glasgow, pp. 213234.
Primdahl, J., 1999. Agricultural landscapes as places of production and
for living in owners versus producers decision making and
implications for planning. Landscape and Urban Planning 46,
143150.
Riebsame, W.E., Gosnell, H., Theobald, D.M., 1996. Land use and
landscape change in the Colorado Mountains. 1. theory, scale and
pattern. Mountain Research Development 16, 395405.
Robinson, G.M., 1990. Conict and Change in the Countryside: Rural
Society, Economy and Planning in the Developed World. Belhaven
Press, London, New York.
Ryan, R.L., 1998. Local perceptions and values for a Midwestern
River Corridor. Landscape and Urban Planning 42, 225237.
Schein, R.H., 1997. The place of landscape: a conceptual framework

for interpreting an American scene. Annals of the Association of
American Geographers 87, 660680.
Simpson, J.W., Boerner, R.E.J., Demers, M.N., Berns, L.A.,
Artigas, F.J., Silva, A., 1994. Forty-eight years of landscape
change on two contiguous Ohio landscapes. Landscape Ecology 9,
261270.
Stockdale, A., Findlay, A., Short, D., 2000. The repopulation of rural
Scotland: opportunity and threat. Journal of Rural Studies 16,
243257.
Swafeld, S., Fairweather, J., 1998. In search of Arcadia: the
persistence of the rural Idyll in New Zealand rural subdivisions.
Journal of Environmental Planning and Management 41, 111127.
Theobald, D.M., Gosnell, H., Riebsame, W.E., 1996. Land use and
landscape change in the Colorado Mountains. 2. A case study of
the East River Valley. Mountain Research Development 16,
407418.
Thomson, L.M., Mitchell, C.J.A., 1998. Residents of the urban eld: a
study of Wilmot Township, Ontario, Canada. Journal of Rural
Studies 14, 185201.
Tremblay, F., Poullaouec-Gonidec, P., 2002. Contre le tout paysage:
pour des e! mergences et ... des oublis. Cahier de g!eographie du
Qu!ebec 46, 345355.
Vos, H., Meekes, H., 1999. Trends in European cultural landscape
development: perspectives for a sustainable future. Landscape and
Urban Planning 46, 314.
Walmsley, D.J., Epps, W.R., Duncan, C.J., 1998. Migration to the
New South Wales North Coast 19861991: lifestyle motivated
counterurbanisation. Geoforum 29, 105118.
Willits, F.K., Luloff, A.E., 1995. Urban residents views of rurality and
contacts with rural places. Rural Sociology 60, 454466.

Changing Ruralities, Changing Landscapes:exploring Social Recomposition Using A Multi-Scale Approach

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Changing Ruralities, Changing Landscapes:exploring Social Recomposition Using A Multi-Scale Approach

Hochgeladen von

Copyright:

Verfügbare Formate

ARTICLE IN PRESS

Journal of Rural Studies 19 (2003) 425444

Changing ruralities, changing landscapes: exploring social

aesthetic, etc.). These new demands are primarily

S. Paquette, G. Domon / Journal of Rural Studies 19 (2003) 425444

of what individuals value in rural places and territories

2.1. The evolution of rural municipalities in southern

Fig. 1. Multiple-scale research design.

an investigation of residential practices. This domicile

2. Rural recomposition and landscape dynamics: a multiscale perspective

2.1.1. Study area and database development: a regionalscale study framework

S. Paquette, G. Domon / Journal of Rural Studies 19 (2003) 425444

identication of agricultural trajectories3 (196191),

activity. Given their dissimilar and crucial inuence on

S. Paquette, G. Domon / Journal of Rural Studies 19 (2003) 425444

land use, agricultural vocations are no longer

Our examination of agricultural trajectories and their

1982; Coppack, 1988) as important factors of attraction

visual analysis: width of the visual eld, depth of the

using 1:20 000 cadastral (19665) and topographical

Open, ltered, closed

Depth of the visual eld

o100 m, 100 m2 km, >2 km

Elevated, horizon, closed

Summit, upper/lower hillside, hillock,

Very small, small, intermediary, large,

Digital terrain model

High, intermediary, low, none

o1%, 1%50%, >50%

Areas remaining under cultivation

o1%, 1%25%, >25%

Areas remaining under pasturea

o1%, 1%15%, >15%

o1%, 1%20%, >20%

o1%, 1%20%, >20%

Constructed before 1965, after 1965

Original farm building dynamics

No building before 1965, all original

New building dynamics (1965)

No new building, one new building,

Full/part-time farmer, farm worker, small

Areas remaining under forest covera

business, salaried worker, professional,

Havelock, adjacent municipalities,

Havelock, adjacent municipalities,

Previous place of residence

Havelock, adjacent municipalities,

Residential occupation type

Permanent, weekly, seasonal,

1824, 2544, 4564, 65 years and over

Year of personal acquisition

Before 1960, 196069, 197079, 198089,

Relative to the overall lot area.

Fig. 5. Resulting landscape contextsHavelock township (N=181).

unstable economic conditions) and the reasons

on residential settlement. Cluster analysis also allowed

S. Paquette, G. Domon / Journal of Rural Studies 19 (2003) 425444

act distinctively on the overall recomposition process.

selective residential settlements previously observed on

Home characteristics. Observations have allowed the

Excellent, good, poor

Brick, synthetic, wood, stone, other

Facing road, facing view