Human Ecology Review, Vol. 18, No. 2, 2011 147
© Society for Human Ecology
Research in Human Ecology
Abstract
In the context of sustainable development, we investigate
four subsistence communities, one each from India, Bolivia,
Laos and Thailand, to understand the systemic interrelations
between the food production systems and related environmental
pressures. In doing so, we revisit Ester Boserup’s theory
of increasing land productivity at the expense of declining
labour productivity as a consequence of agricultural intensification.
Our data confirm Boserup’s assumptions within
the reach of traditional agriculture, but find them not to
apply to hunting & gathering communities and to agricultural
systems now increasingly dependent on fossil fuels and industrial
fertilizers. Instead we propose a theory of “sociometabolic
transitions” as being more appropriate to understanding
transitions in land and labour productivity
across a wider range of modes of subsistence.
Keywords: sociometabolic regimes, sociometabolic
transitions, farming systems, time use, labour and area productivity,
rural development
Introduction
Since the 1960s, the study of rural subsistence communities
has received considerable attention within anthropology
and human ecology. Interest in the allocation of land and
household labour in farming systems and the diversification
of rural livelihoods in response to ubiquitous development efforts
have been at the core of such enquiries (Boserup 1965,
Clark & Haswell 1967, Chayanov 1966, Geertz 1963, Lee &
DeVore 1968, Sahlins 1972, Wilkinson 1973, Ellis 1998,
Hunt 2000, Amanor & Pabi 2007). In the context of discussions
on global environmental change, these concerns have
evoked renewed interest among scholars concerned with social
and ecological sustainability (e.g. Gowdy 1998, Costanza
et al. 2007, Tainter 2006, Sieferle 1997b, Fischer-Kowalski
& Haberl 2007). As almost half of the world’s population
still lives in rural areas on subsistence agriculture, gathering,
hunting and fishing (UNDP 2007), the direction these communities
will take is extremely crucial in terms of future trajectories
of global resource and land use.
While the industrialized world seeks for pathways away
from and beyond fossil fuels, the dominant development
model for those rural communities is still the eventual indusSociometabolic
transitions in subsistence communities:
Boserup revisited in four comparative case studies
Marina Fischer-Kowalski1
Institute of Social Ecology, Vienna
Alpen Adria Universitaet, Austria
Simron J. Singh
Institute of Social Ecology, Vienna
Alpen Adria Universitaet, Austria
Christian Lauk
Institute of Social Ecology, Vienna
Alpen Adria Universitaet, Austria
Alexander Remesch
Institute of Social Ecology, Vienna
Alpen Adria Universitaet, Austria
Lisa Ringhofer
Hilfswerk Austria International
Vienna, Austria
Clemens M. GrĂĽnbĂĽhel
CSIRO Sustainable Ecosystems
Canberra, Australia
148 Human Ecology Review, Vol. 18, No. 2, 2011
Fischer-Kowalski, et al.
trialization of their agriculture — fuelled with fossil energy
— and the absorption of a large fraction of their population
into the industrial labour market of growing cities. This appears
to be the only chance for an escape from poverty, illhealth,
and illiteracy. In contrast, sustainable development requires
a broader search for pathways where short-and-long
term benefits for the people come at the lowest possible environmental
cost and avoids increasing the burden and stress
on the people in terms of working time (Haberl et al. 2004;
2011). To this end, there is an urgent need to look beyond
simple evolutionary sequences of responses to population and
market pressures and the adoption of modern technologies, as
well as beyond single variables of land and labour productivity
and return upon investment. Instead, an understanding of
the complex relationships within socio-ecological systems requires
looking more systematically at the broad range of dynamics
and pressures the social system exerts upon its environment,
and how these pressures change with development.
In this paper we compare four rural subsistence communities
in different stages of agricultural intensification to understand
the systemic interrelations between the food production
system and environmental pressures as a consequence.
In doing so, we revisit Ester Boserup’s (1965) theory
of agricultural change, in particular her hypothesis on rising
area productivity at the expense of declining labour productivity
in consequence of intensification in traditional
farming systems2. The relevance of Boserup’s contribution in
understanding agricultural change in traditional farming systems
is well acknowledged. However, our findings reveal
caveats in this theory when applying it to agricultural systems
now increasingly dependent on fossil fuels and industrial fertilizers.
Instead we propose a theory of sociometabolic transitions
as being more appropriate to understanding agricultural
development under contemporary conditions in the context
of discussions around ecological sustainability and global
land use change.
We begin by introducing our key theoretical assumptions
and concepts, followed by a brief description of cases and
methods used in data collection. We then present the main
findings, concluding with an evaluation of our hypothesis and
our theoretical framework.
Theoretical assumptions and concepts
Our point of departure is the theory of sociometabolic
regimes as developed by Sieferle (1997a, 2001) and further
elaborated by him and other authors since (Fischer-Kowalski
et al. 1997).The theory claims that, in world history, certain
modes of human production and subsistence can be broadly
distinguished that share, at whatever point in time and irrespective
of biogeographical conditions, certain fundamental
systemic characteristics derived from the way humans utilize
and thereby transform nature. Key to distinguishing sociometabolic
regimes, according to Sieferle (1997a) is the
source of energy used, and the main technologies of energy
conversion. Traditional subsistence systems such as hunters
& gatherers and the agrarian depend (almost) completely on
the solar energy flux and its fixation through plant photosynthesis.
The crucial difference between the energy regime of
hunting & gathering and agriculture is the conversion technology.
While hunters & gatherers are “passive” users of
solar energy, that is they utilize plant and animal biomass
wherever they find it, the agrarian regime relies mainly on an
“active” utilization by colonizing (see below) terrestrial
ecosystems. In other words, peasants try to channel solar energy
onto a few plant species they wish to use by changing
the land cover and seeking to monopolize selected quality
land for their food and feed purposes, albeit at the cost of
more human labour that further increases with agricultural intensification.
The industrial sociometabolic regime, on the
other hand, transcends the limitations inherent in relying on
the current flux of solar energy by utilizing fossil fuels, the
stock of historical solar fluxes accumulated over millions of
years.
The transitions between sociometabolic regimes are unlike
what is often understood as an incremental change in material
and energy use in societal development (White 1949).
Instead, the shift between energy regimes is associated with a
major transformation of society. The grand shifts in the past
(the Neolithic and the Industrial Revolution) have allowed for
an enormous increase in energy and material use, boosting
metabolic rates (Haberl et al. 2011). However, sociometabolic
regimes, according to this theory, are not something static.
Rather, they are constituted by a set of opportunities and constraints
within which certain dynamics take place. But if the
dynamics transcend or are pushed out of the boundary conditions
of the regime by exogenous forces, turbulence will
ensue with an unpredictable outcome anywhere between collapse
of the social system (Tainter 1988, Leemans & Costanza
2005) and a transition into another sociometabolic regime
(Fischer-Kowalski & Haberl 2007). Below we introduce two
interrelated concepts that allow us to describe the structure
and dynamics within and between sociometabolic regimes —
social metabolism and colonization of natural systems — and
how the varying forms of higher level interventions affect
these.
Social metabolism
Social metabolism draws on an organismic analogy by
claiming that any social system not only reproduces itself
culturally, by communication, but also biophysically (namely
its population, built infrastructure, man-made artefacts and
Human Ecology Review, Vol. 18, No. 2, 2011 149
Fischer-Kowalski, et al.
livestock) through a continuous energetic and material exchange
with the natural environment (and eventually with
other social systems). Social metabolism can be quantified in
terms of energetic and material flows per time period, usually
a year. The size of the flows required depends, on the one
hand, on the size of the biophysical structures (or stocks) of
the social system (i.e. the size of the human and livestock
population, and all human-made infrastructures), and on the
sociometabolic regime, on the other hand. Different sociometabolic
regimes have substantially different metabolic
profiles (i.e. quantity and quality of materials and energy
used). Metabolic profiles can be expressed as total quantities
for a complete social system (a society, a community, or, for
example, a household), and they can, for reasons of comparability,
be referred to the number of the human populations
the social system sustains, and are calculated as metabolic
rates (in terms of energy or materials required per person and
year). The higher the metabolic rate, the more resources per
inhabitant have to be extracted or imported and the more outflows
of wastes and emissions are produced, therefore the
higher is — other things being equal — the impact upon the
environment. Once adequate boundaries of the social system
are defined (and this has received a great deal of methodological
attention by a number of researchers; see for example
Fischer-Kowalski & HĂĽttler 1998; Matthews, et al. 2000;
Schandl, et al. 2002), biophysical structures (stocks), flows,
metabolic profiles and metabolic rates can and have been
measured or estimated in a comparable way for a number of
social systems (communities, societies) on various scales
across history (for an overview see Fischer-Kowalski &
Haberl 2007).
Colonization of natural systems
The second concept employed for characterizing the respective
society-nature interaction is colonization (FischerKowalski
& Haberl 1998). Social systems not only exchange
energy and materials with their natural environment, they
also deliberately intervene into natural systems with the intention
of transforming them in ways they consider more useful
for themselves. A classic colonizing intervention is
changing the land cover in favour of agriculture, but this term
can usefully be applied to processes as varied as animal
breeding, genetic modification or dam building. The important
commonality is that natural systems thus transformed by
human intervention are usually brought into a state far from
the relatively natural state, and need a continuous input of
human labour (and typically also energy and materials) to be
kept in that state. Thus a social system’s colonizing activities
are related not only to the effort that must be invested both in
terms of working time (quantity), but also to the use of certain
technologies (quality). The more a society modifies its
environment, the more metabolic returns it may expect, but
also the more efforts it has to expend to keep it in the desired
state — and this may create the need to invest even more
working time. Only by using the “subterranean forest” of fossil
fuels (Sieferle 2001), in connection with industrial tools
such as machines and synthetic fertilizers, can the relation
between intensification and increasing working time be bro-
ken.
The relationship between land use and labour has received
considerable attention since the publication of Ester
Boserup’s The Conditions of Agricultural Growth (1965)
where she argues that in traditional subsistence systems, in
response to population pressure, technological development
takes place leading to an intensified use of land sustaining
more people, but at the expense of a higher input of human
labour. While an advantage in terms of land productivity
(where land is a limiting variable), it is not so in terms of
labour input per unit of harvest. In this sense, intensive agriculture
needs to maintain a larger pool of labour to maintain
and reproduce the colonized system. With less of a positive
connotation, Geertz (1963, p.80) had termed this same
process of increasing land productivity through increasing
labour input for feeding more people just at the same level as
before “agricultural involution” and considered it an ultimately
self defeating process.
Notwithstanding, it is imperative that there is a positive
net energetic return upon investment (EROI) between labour
invested and harvest gained on a system level (measured in
energy units). If a society invests more energy than what is
gained in terms of crop harvest, this system cannot be sustained
for long, unless it compensates this loss from another
energy source that produces a surplus. Thus, traditional
agrarian systems, with biomass as the main source of energy
will strive to retain a positive EROI at an aggregate system
level even as labour productivity declines but as long as land
productivity increases. With the introduction of fossil fuels
and related technology into agriculture, the decoupling of
land, labour and energy occurs.
In other words, land is no longer a limiting variable
when it comes to harnessing energy. Market prices and subsidies
for fossil fuels play an important role in maintaining
low, if not negative overall EROI.3 Such a system is unsustainable
from the point of view of both future energy availability
as well as the environmental impact these technologies
have.
Interactions between system scales
In the age of fossil fuels, industrial production and improvements
in transport and communication, we cannot expect
our case studies dating at the beginning of the 21st century
to comply with the typical pre-industrial mode of pro-
150 Human Ecology Review, Vol. 18, No. 2, 2011
Fischer-Kowalski, et al.
duction in full, however isolated they may be. Therefore, interventions
from higher scale systems already affected by the
industrial transformation must be considered. They most likely
have an impact on the metabolic profile and on the social
dynamics of the local subsistence system under consideration.
Four types of interventions were found to be most common
which we categorize as follows: (a) provision of services
such as health and education, (b) regulatory mechanisms
from the state such as introduction of legal instruments and
market conditions, (c) supply of (often subsidized) fossil-fuelbased
technologies either through agricultural extension programmes
(machines, mineral fertilizers) and/or by introducing
electricity and communication network and transport infrastructure
in the region, thus enhancing opportunities for
marketing produce, buying commodities from outside, and
labour migration thereby modifying the local production and
consumption patterns, and (d) supply of specific aid and subsidies
either as part of general welfare policies, or as part of
famine or disaster relief. Such interventions buffer communities
from responding to extreme events using their own traditional
and self-organization capacities as may have been in
the past.
Description of the four cases and context
The first case is Trinket Island in the Nicobar archipelago
(India) with 399 inhabitants in 2001. The community exhibits
an economic portfolio that combines hunting, gathering,
fishing, pig and chicken rearing, growing coconuts and
bartering copra (dried coconut meat used as raw material in
the extraction of oil) in lieu of rice, sugar, cloth, kerosene,
and other necessities on markets located on neighbouring islands.
A few families maintain food gardens where they grow
an assortment of crops such as bananas, pineapples, yam,
sugarcane, oranges, lemons, papaya and jackfruit. Simple
metal tools such as sickles, axes and spades are used for
clearing, planting, or harvesting. Since the 1980s, the Indian
government has gradually introduced a variety of welfare
programmes for the Nicobarese such as education, health services,
transport infrastructure and a variety of subsidies including
the sale of cheap diesel and kerosene (Singh &
Schandl 2003, Singh et al. 2001; Singh & GrĂĽnbĂĽhel 2003).4
The second case is Campo Bello (Bolivia) with a population
of 231 inhabitants in 2004. Campo Bello is characterized
by swidden agriculture, fishing, hunting, gathering and
raising poultry. Rice is the most important crop, but the villagers
also grow plantains, maize, manioc and others of lesser
importance, such as peanuts, sugar cane, citrus and varieties
of sweet potatoes. The technology employed in agriculture
is simple, using only machetes, hoes and rice seeders for
the sowing of rice. A certain amount of rice and plantains are
sold in the market for cash or barter. Wage labour is also sold
by younger men. The village has witnessed a number of development
projects introduced by the local administration
and non-governmental agencies. Development efforts include
the construction of a school building made of concrete in
1993, the installation of electricity for the school building, a
solar panel for the operation of a communal telephone, and
the installation of several individual latrines and concrete
wells. In 2006 and still ongoing, a new project involved various
families in the cultivation of beans and the raising of
poultry.
The third case studied is the multi-ethnic community of
Nalang (Laos) with a population of 702 people in 2001.
Nalang is again a subsistence economy dominated by rainfed
rice farming, primarily through permanent paddy farming
and also shifting cultivation. In very productive years, a harvest
surplus may either be sold to other villages or to outside
traders. In the late 1990s, the production of cucumber was introduced
as an important cash crop during the dry season.
Gathering, fishing & hunting activities are also common. To
catch fish, people apply a variety of techniques ranging from
line fishing to casting nets. Hunting is carried out either by
using traditional hunting devices like traps or bows or more
elaborate home-made guns. Some buffaloes are still reared
for use in agricultural work and general transport. The arrival
of the motor-plough in the mid 1990s, however, has diminished
the need for buffaloes, and for meat production, buffaloes
have largely been replaced by cattle, since their maturing
times are more rapid. The construction of a road in 1980
allowed some to engage in wage labor.
The village of Sang Saeng (Thailand) with 171 inhabitants
was investigated in 1998. Sang Saeng’s economy revolves
around permanent rice farming, raising livestock, foraging
and labour migration. While the glutinous rice is grown
only for subsistence, non-glutinous varieties are specially
produced for the market. Villagers own vegetable gardens and
keep chicken and ducks for their own consumption. Buffaloes
are raised as working animals, but cattle are destined for the
local market. Hunting and gathering activities also feature
prominently: foraging helps to diversify the diet, especially
during the dry season, when gardening is limited or made virtually
impossible. During the dry season, temporary migration
for wage labour to urban centres or to coffee and rubber
plantations in southern Thailand is now rather common, with
the labourers returning in time for the peak agricultural period
for preparation of rice nurseries, ploughing, transplanting,
and harvesting. Threshing machines, electrical water pumps,
non-potable water supply, electricity, and a semi-permanent
road are a consequence of government interventions.
Human Ecology Review, Vol. 18, No. 2, 2011 151
Fischer-Kowalski, et al.
Methods
The four cases were studied by members of our team between
1998 and 2006. The field research in each of the communities
extended over several months. To all of the communities,
there were also follow-up visits to cross-check on
data.5 Here we focus only on describing the methodology for
outlining the social metabolic profiles of these communities
as well as aspects of their land-use and labour input.
To define the characteristic metabolic profile for each of
the social systems, we adapted the standard national Material
and Energy Flow Accounting (MEFA) toolbox as prescribed
by the Statistical Office of the European Union (Eurostat
2001; Eurostat 2007) to the local context. Local material
and energy flow accounts were created following a number
of steps: (a) defining the systems boundary of the socioecological
system under investigation according to systemic
and pragmatic considerations, in particular concerning population
and the territory the social system is entitled to exploit
(legally or by tradition), (b) identifying the biophysical stocks
that the community maintains and reproduces year after year,
such as human population, domesticated livestock and manmade
artefacts (buildings, wells, pathways, boats, and machines),
(c) quantifying flows of materials and energy that the
society organizes to maintain and reproduce its biophysical
stocks. These flows may originate from the domestic environment
(domestic extraction) or they may be imported from
other social units. Equally, on the output side we differentiate
between wastes and emissions that are deposited onto the domestic
environment and exports to other social units (for details
on local MEFA methods, see Singh et al. 2010).
Local-level data is not as readily available as for national
accounts. They have to be generated using a combination
of innovative, often labour and time-consuming techniques of
on-site quantification. The stock account of the social system
is based on an inventory of the local population, livestock and
the most important human-made structures. Livestock and
artefacts are expressed in metric tons: this was done by actual
weighing or by drawing on factors from externally available
sources. Flow data were generated by sample weighing
and estimations. All biomass flows were calculated both in
terms of weight of dry matter and fresh weight when harvested6
or traded. For energy flows and energy conversion
processes, the same system boundaries as in material flow accounting
were applied (Haberl 2001, 2002). Material flow
data for biomass and fossil fuels were converted to energy
units by using calorific values. The indicators generated are
standard indicators developed in the framework of material
and energy flow analysis on the level of national economies
(Eurostat 2007; Haberl et al. 2004), and they express the
amounts actually used by a social system during the course of
a year. Derived indicators express these flows per capita population
as metabolic rates.
Land use was studied according to the same system
boundaries. First, we accounted for the �total area’ controlled
and used by the social system, corresponding to the territory
as defined above (Singh et al. 2010). We mapped the total
land cover of this territory according to use and according to
ecosystem types, such as primary and secondary forests,
grasslands, mangroves, horticultural gardens and agricultural
fields. Data for land-cover and ecosystem types were either
taken from official statistics (especially in case of forests,
grasslands, mangroves and beaches) or measured by the researcher
(as in the case of settlements, agricultural fields and
coconut plantations). Various indicators are generated by referring
material, energy and labour time flows to respective
areas.
While elaborate time allocation studies were undertaken
for all four cases for a sustainability analysis (Fischer-Kowalski,
et al. 2010), in this paper we only focus on labour time in
staple food production. Such activities were repeatedly observed
(sample size 3-5 observations for each activity) and
records were made how long they lasted and who (in terms of
gender and age) participated in them. These activities were
then weighted according to their annual frequency and thus
the average daily hours could be calculated. In order to arrive
at system level data, the frequency of these processes across
the year was estimated and used for weighing. Labour patterns
were taken into account to accommodate for seasonal
fluctuations of time use allocation, and data eventually needed
to be adjusted.
Findings
We organize the presentation of our findings in the following
manner: first, we present the basic demographic and
sociometabolic data that will allow us to classify our cases by
— to use Boserup’s term — degree of agricultural intensification.
In a next step, we go more deeply into the issues of
land and labour productivity and seek to test Boserup’s formulation
of the dynamics of agricultural intensification.
Table 1 gives an overview of basic demographic and sociometabolic
features of the four communities. If we consider
population density as a certain indication of population
pressure on land, we find Trinket to have by far the lowest
density (11 persons / km2), along with a low rate of population
growth. Campo Bello and Nalang have an intermediate
population density (around 40 persons / km2), and fairly high
population growth rates (3-4% annually). Sang Saeng has a
high population density (93 persons / km2), but its population
growth rate is similar to Trinket.7
Concerning the sociometabolic parameters (table 1), all
152 Human Ecology Review, Vol. 18, No. 2, 2011
four communities are characterized by very low metabolic
rates, which are typical for traditional subsistence communities.
Their energy metabolic rates of 20-40 GJ are by an order
of magnitude smaller than the metabolic rates of roughly 200
GJ in the European Union or even 400 GJ per contemporary
US citizen (Haberl et al. 2006)8. Similarly, the material metabolic
rates range between 1.6 and 3.7 tons per capita —this
compares to 13 tons per contemporary inhabitant of the European
Union (Weisz, et al. 2006)9. It is apparent that the
communities are largely self-sustaining: of all materials that
are being consumed annually, only a very small share are
(imported) industrial products (between less than 1 up to 14%
in terms of weight), and the energy metabolism is largely
based on biomass; fossil fuels amount to less than 10%. Thus,
we are dealing with communities far away from the industrial
sociometabolic level.
Beyond those shared characteristics, some numbers
stand out. Trinket’s energy metabolic rate is higher than in
both Campo Bello and Nalang, and its share of fossil fuels in
total energy consumption at 6.4% also exceeds the respective
share in Campo Bello and Nalang (table 1). Fossil fuels are
only used in transportation: there is government supply of
diesel for running motored boats that the Nicobarese use for
transporting copra to trade for rice and other goods. And as
they get this diesel delivered as imports, it by definition
counts as part of their domestic consumption. In the case of
Campo Bello, in contrast, it is the traders who visit the village
for trading rice with pasta or alcohol, so their transport
fuel does not count as part of Campo Bello’s metabolism. Another
noteworthy difference concerns Sang Saeng: here the
per capita energy use is twice as high as in Campo Bello, the
fossil fuel share approaches 10%, and industrial commodities
are used much more frequently than anywhere else.
Information on the food system allows an insight in the
relative position of each of the communities. In all four communities,
there is about the same amount of food available for
daily consumption (i.e. between 11,000 and 13,000 kJ, or
2,700-3,200 kcal per day and inhabitant). The origin of this
food is different, though. All communities gain a certain fraction
of their food from hunting/fishing/gathering; while this
amounts to 16% or less in Campo Bello, Nalang and Sang
Saeng, it is 69% of all food intake on Trinket (Figure 1). Thus
in terms of the food consumption system, Trinket stands out
as a community predominantly based on a hunting and gathering
mode of subsistence.
Another important feature of these communities is their
degree of self-sufficiency versus market integration. Clearly,
Nalang is the most self-sufficient community: it only imports
Fischer-Kowalski, et al.
Table 1. Demographic features, basic sociometabolic characteristics and the food system
Trinket Campo Bello Nalang Sang Saeng
hunting + shifting cultivation intensive rice intensive rice
gathering (short fallow periods) cultivation cultivation
demographic features
Population (cap) 399 231 702 171
Size of territory (ha) 3626 615 1630 184
Pop density (cap/km2) 11.0 37.6 43.1 92.9
Pop growth (%/yr) 1.5 3.8 3.0 1.4
Share of population below age 15 (%) 39 61 45 n.a.
basic sociometabolic parameters
Material metabolic rate (t/cap/yr) 3.7 1.6 2.6 3.6
Share of industrial products in the materials used (%) 0.3 1.3 0.4 13.9
Energy metabolic rate (GJ/cap/yr) 29.5 20.6 26.3 40.5
Share of fossil fuel in energy used (%) 6.4 1.0 1.5 8.3
food production and consumption
food consumption (GJ/yr) 1752 940 3320 666
imported food (% of consumption) 29 20 1 49
food production (GJ/yr) 2820 1840 3752 2213
exported food (% of production)1 48 38 18 85
staple food production (GJ/yr) 2979 1590 3101 2098
area for sfp, incl. fallow (ha) 29 200 139 146
labour hours for sfp (1000h/yr) 25 114 213 83
fossil energy inputs in sfp (GJ/yr) 0 0 171 446
Notes: Material metabolic rate = annual domestic extraction of biomass + minerals for construction plus all imported materials, minus exported materials, in tons,
divided by population numbers. Energy metabolic rate is the analogue, expressed in Gigajoule per capita. Staple food is rice, cereals and tubers; in the case of
Trinket it is copra.
Human Ecology Review, Vol. 18, No. 2, 2011 153
1% of the food it consumes, and it exports only 18% of its
production (Table 1). At the same time, Sang Saeng stands
out in market integration, with imports amounting to 49% of
its food consumption, and exports as high as 85% of food
production. All four communities
are net exporters of food (if consumption
is set in relation to production,
see Table 1), and thus they
are suppliers of food to the society
at large, but to varying degrees.
For the remaining analysis, we
shall focus on staple food production10
and standard agroecological
indicators to secure comparability
among our cases and with other
similar data (see Clark & Haswell
1967). While the basic information
can be found in Table 1, we visualize
our main findings in Figures 2-4.
We will from now on pursue the key
Boserupian (1965, 1981) hypothesis
of an endogenous process of agricultural
intensification (in the
course of mounting population pressure
and development) leading to
increased yields per unit area, at the
expense of increased labour input
and possibly declining labour productivity. In our tables, we
have tentatively ranked the communities studied along a
“Boserupian axis” from Trinket to Sang Saeng; across the
fairly similar cases of Campo Bello and Nalang. The data
presented so far have to a certain degree
confirmed such a ranking. But
what we can see in Figure 2, displaying
land and labour productivity
across the four cases definitely
warrants a different interpretation.
The results presented in Figure
2 demonstrate for Trinket substantially
more favourable conditions
than for any other of the four communities.
Both land productivity (in
the sense of how much land is required
to produce a certain amount
of nutritional energy) and labour
productivity (signifying how much
work is required to realize this energy
harvest) are far higher than in
any of the other communities.11
Looking at this from another
angle, it appears that no incremental
evolutionary pathway of “agricultural
intensification” would lead
from a — however untypical — sociometabolic
system of hunting and
Fischer-Kowalski, et al.
Figure 1. Annual food consumption by its origin from agriculture or foraging
Notes: this includes all the food consumed by the members of the community, irrespective of it having been extracted
domestically or imported from outside
Figure 2. Labour productivity and land productivity in staple food production
Notes: Labour productivity is measured as energy content of the annual harvest of the staple food per hour worked for
staple food production. Land productivity is measured as annual harvest of the staple food per hectare agricultural land
(including fallow land). Labour hours are annual hours worked for staple food production per unit area of staple food
production.
154 Human Ecology Review, Vol. 18, No. 2, 2011
gathering like Trinket to anything like the other communities.
It would appear completely absurd to invest additional labour
time into additional land for so relatively
little return. Thus, a sociometabolic
system like in Trinket
will keep on or else collapse - but
cannot gradually be transformed
into an agrarian system like we see
in the other cases.12 In effect, rather
our hypothesis of distinct sociometabolic
regimes is confirmed:
communities like Trinket adhere to
a sociometabolic regime of hunter
& gatherers, however untypical, and
there is no continuous, non-disruptive
pathway leading from this
regime to an agrarian regime. It
takes a major transformation, a
“transition” (Fischer-Kowalski &
Rotmans 2009) for a community to
transcend this mode of subsistence.
We can see that Campo Bello is
a “traditional” (Boserup 1981) production
system in the sense of not
using fossil fuel based inputs or
even animal traction. In Nalang there is already some fossil
fuel input (1.2 GJ/ha), while Sang Saeng uses 3.1 GJ of fossil
fuel per hectare. In industrial cereal
production systems, the corresponding
use of fossil fuels amounts
to a value of between 3 and 7 GJ/ha
(e.g. Bonny 1993; Golley et al.
1990; Swanton et al. 1996; Tsatsarelis
1993)13 Thus Sang Saeng is
already an industrialized agricultural
production system. And of
course, a tractor is an enormously
labour saving device in agriculture!
Taking into account fossil fuel use,
the relation between Campo Bello,
Nalang and Sang Saeng would
probably comply to the Boserupian
hypothesis: without fossil fuels,
Nalang and Sang Saeng would
probably have a much lower labour
productivity than apparent in Figure
3. Beyond that, there is a lower productivity
of land in Sang Saeng than
in all other cases. This may have geographic
reasons (Sang Saeng lies
in a relatively arid region), but it
may also be causally related to long
term over-exploitation of the land.
Fischer-Kowalski, et al.
Figure 3. Labour productivity, land productivity and fossil energy input in staple food production
Notes: Labour productivity is defined as the annual staple food harvest (in energy units) per labour hour invested for its
production. Land productivity is defined as the annual staple food harvest (in energy units) per unit land used for staple
food production (incl. fallow land). Fossil energy input is confined to direct input into staple food production (i.e.
diesel for agricultural machines).
Figure 4. Energy input, energy return on investment (EROI), and land and labour productivity in staple food produc-
tion
Notes: Energy input is the sum total of labour input (in energy units) plus fossil fuel input. EROI is defined as annual
staple food harvest (in energy units) divided by the energy input into its production (labour hours in energy units, plus
fossil fuels). Based on the according numbers given by Smil (1991, pp.86-89), we derive the energy input per labour
hour by multiplying the basic metabolic rate (BMR) of 288 kJ/hour for males and 234 kJ/hour for females by a coefficient
for moderate activities of 5.45 for reference males and 4.85 for reference females. The average value for all agricultural
work of 1.46 MJ/hour is based on the assumption that three fourths of all agricultural labour is done by men.
1 In case of Trinket, referring to staple food only!
Human Ecology Review, Vol. 18, No. 2, 2011 155
How can we capture this combined effect of (latently)
declining labour productivity masked or compensated for by
the use of fossil fuel driven machinery? In Figure 4 we make
an effort in this direction: we add two indicators to the already
presented indicators of land and labour productivity.
One indicator is total energy input into the food production
system: it is the sum of human labour (calculated in energy
units) and fossil fuel input, both per unit area. This indicator,
as was to be expected, shows a clear rise in the direction of
“agricultural intensification”: while in Campo Bello energy
investment per hectare amounts to 0.8 GJ/ha/year, it is about
four times as much in Nalang (3.4 GJ/ha/year) and Sang
Saeng (3.0 GJ/ha/year).
If on top of this we wish to capture the effect of lower
area productivity (as apparent for Sang Saeng), we need to
turn the attention to the “energy return on investment”
(EROI). This classic parameter relates energy output (in the
form of harvest) to energy input (labour and fossil fuels).
What we then see is a clear decline along a pathway of intensification
and industrialisation. While in Campo Bello, the
energy harvested is almost ten times as much as the energy
input, it is only 6.5 times as much in Nalang and not even five
times as much in Sang Saeng.
Conclusion
Does the theory of sociometabolic regimes allow ordering
our four case studies? Does it make sense to distinguish
between “endogenous” and “exogenous” dynamics, and can
we explain regularities and irregularities reasonably that
way? Our findings tend to comply much better with the theory
of sociometabolic transitions than with the classical
Boserupian theory. There seem to be substantial qualitative
differences between the hunting & gathering regime that only
thrives if there is a high land productivity that requires little
work (here exemplified by Trinket in a number of ways), the
regime of non-fossil-fuel-based (subsistence) agriculture as
exemplified by Campo Bello and Nalang, and a fossil-fuelbased
industrial regime which Sang Saeng is stepping into.
With increasing inputs of fossil fuels into agriculture, the
Boserupian link between decreasing labour productivity and
increasing population density is overridden by the industrial
link between increasing use of fossil fuels and industrial
technology increasing labour productivity.
Our results point in the direction of our theory of regime
transitions, and the crucial role of working time. While of
course there are numerous small steps that lead from the sociometabolic
regime of hunting & gathering to agriculture
(and in most agrarian communities there are still elements of
hunting & gathering preserved), in terms of labour this seems
to imply a major transition, requiring maybe shocks or strong
pressure beyond gradualism. Similarly, as far as our few
cases warrant even a tentative conclusion, the “Boserupian”
endogenous dynamics looks plausible among agrarian communities,
but as soon as exogenous variables substantially
come into play, such as fossil fuels and other development interventions
in the form of infrastructure and markets, a dynamics
towards some kind of “transition” leading out of the
trap of labour intensification occurs, but into an increasing
energy requirement for agriculture, and a declining energy return
on investment (EROI).
Despite the limited number of case studies, the Boserupian
theory of agricultural development as synthesized in
Sieferle’s theory of sociometabolic regime transitions give adequate
guidance to interpret our findings. In contrast to
Boserup’s continuous developmental process from a foraging
mode to intensive agriculture, our findings comply better with
the assumption of qualitative transitions between foraging, the
— pre-industrial — agrarian mode and finally fossil- fuelled
intensive agriculture. These transitions reflect themselves in a
number of sociometabolic indicators, and particularly in
labour time. Labour time is low with foragers, rises substantially
with agricultural colonization and then with each step of
intensification, but can in a next transition be lowered through
the use of fossil fuels. However, the pressure on the environment
(at least in terms of anthropogenic mass and energy
flows per unit area) rises from one transition to the next.
What can be harvested from these findings for a potential
next transition, and potential policies in support of it? The
key insight is the systemic character of society-nature interactions.
Even seemingly trivial and well-meant interventions
may trigger a whole chain of consequences in the sociometabolic
system that can be detrimental either to the social or to
the ecological balance, or both. Traditional development and
aid policies tend to overlook the intricate ties between demography,
labour time, land degradation and subsistence / income.
Which pathways are viable, and which will be beneficial
in the long term, needs a thorough consideration leading
to maybe very different conclusions for different socioecological
systems. Classical sectoral planning to boost the efficiency
of one variable alone (such as increasing land productivity
or enhancing dairy output) will in effect lead to an overall
system change, some of which may not be desirable or
sustainable in the long run. Development policies must therefore
be sensitive to these systemic interrelations and the
trade-offs involved therein.
Acknowledgements
The authors are grateful to several colleagues who provided stimulus
to our ideas and engaged in lengthy discussions during the course of our research
and in the preparation of this paper. In particular, we would like to
Fischer-Kowalski, et al.
156 Human Ecology Review, Vol. 18, No. 2, 2011
thank John Gowdy, Helmut Haberl, Fridolin Krausmann and Willi Haas for
their valuable comments on the paper drafts and to Irene Pallua for her
technical support. We are also grateful to the various donors who made
these field studies possible over the last decade within the framework of
various projects. They are the Austrian Science Fund (Project RECOVER L
275-G05), the Austrian National Bank, and the European Commission.
In addition, we acknowledge individual grants and fellowship from the
Government of India, the Austrian Ministry of Science, Education and Culture,
the START visiting fellowship nominated by the IHDP-IT programme,
and the RAI fellowship in Urgent Anthropology granted to Simron
Jit Singh.
Endnote
1 Marina.Fischer-Kowalski@aav.at
2 For a recent review of the lasting impact of Boserup’s work, see
(Turner & Fischer-Kowalski 2010)
3 Several studies have documented a massive decline in EROI when
moving from traditional to industrial agriculture (e.g. Pimentel, et al.
1990, Rambo 1984). In the case of Austrian agriculture, the EROI declined
from 6:1 in 1830 to 1:1 in the year 2000, mainly due to the
heavy inputs of fossil fuels and nitrogenous fertilizers (Krausmann,
et al. 2004).
4 Trinket Island was completely destroyed by the 2004 tsunami, and
the surviving inhabitants moved to another island. The data presented
here is from previous years, though some information on household
working time was corroborated after the tsunami.
5 Each of the field surveys resulted in a doctoral thesis covering many
more aspects of the communities than selected for this comparative
analysis (Mayrhofer-GrĂĽnbĂĽhel 2004, Ringhofer 2007, Singh 2003).
6 Livestock grazing was estimated and considered as part of the har-
vest.
7 We have only limited knowledge on how population dynamics is regulated.
In Trinket, historically, there were traditional birth control
measures in place, while nowadays, the Nicobarese women willingly
participate in India’s sterilization programme. In the past, excess
population tended to settle on other islands not yet inhabited. In
Campo Bello, we know from interviews that they complain about
having to shorten fallow times because of population pressure (Ringhofer
2010); we assume the strong growth is mainly due to a recent
reduction in mortality rates because of improved medical services.
For Nalang and Sang Saeng we have no further information.
8 They are even low in relation to the 70 GJ/cap calculated for historical
Austrian villages in 1830 (Krausmann 2004)
9 The material metabolic rates amount to more than 5 tons for the historical
Austrian villages mentioned (Krausmann 2004).
10 With Trinket, there is a certain definition problem concerning the
“key staple food”. They do not agriculturally produce any staple
food. What the Nicobarese do is collect coconuts, produce copra, and
exchange that copra for rice, which is then the staple food they consume.
Nevertheless, we treat copra as their staple food. For Campo
Bello and Nalang, where there is shifting cultivation, the fallow areas
are included in the area of staple food production, as they functionally
are indispensable for this mode of production. Labour time for all
three agrarian communities in Table 1 encompasses only the time for
the production of the staple food, which is rice in the case of Sang
Saeng and Nalang, and plantains, rice, manioc and maize in the case
of Campo Bello (all products of the shifting cultivation system). Only
fossil fuels used in the production process of the staple food are accounted
for.
11 Although the inhabitants of Trinket are sedentary, they do not intensively
colonize the island’s terrestrial ecosystems. Coconut palms
grow on the sandy beaches and bear coconut around the age of 10,
and remain productive for almost 100 years without much caretaking
by humans. The only occasional work is to dig a coconut into the
sand at the right season and a convenient place and protect it from
pigs for the first few years. The coconuts are “harvested” by letting
them fall down and gathering them, or occasionally by climbing the
trees and lopping them down. For a long time, the Nicobarese have
exchanged these nuts with ships passing by and bartered them for
rice. In the absence of rice, they used pandanus and other tubers as a
staple food (Singh 2003). Today, copra (dried coconut flesh) has replaced
the nuts as exchange. This requires breaking the nuts open,
scooping out the flesh, and collecting firewood. This is untypical
agricultural work. The only activity that comes close to agriculture is
the feeding of the pigs with coconut flesh. However, this feedstuff
comprises only 30% of the pigs’ diet — the remaining 70% is scavenged
in the forest. Furthermore, pork forms an insignificant part of
the diet of the Trinket inhabitants (4gms/cap/day), and pigs are being
slaughtered almost exclusively for ceremonies and festivals.
12 Paradoxically, a collapse of this system actually occurred: By the end
of 2004 a tsunami disrupted the whole region, killing one third of the
islanders, breaking Trinket apart and eliminating all coconut palms.
The islanders asked us in for help, but our efforts to motivate them to
horticultural activities were completely in vain. Instead, the island
population lived on Indian State and international aid until now, completely
changing its lifestyle towards consuming food from aid programmes
and industrial products they buy from their compensation
payments (motor cycles, mobile phones, fancy clothing, etc.). It will
take a number of years for new coconut palms to bear fruit, and we
consider it an unintended field experiment to see how these communities
will survive from now on since the aid is terminated.
13 This includes only the direct input of fossil fuels (e.g. for the machines
used in the fields). Industrial systems additionally depend on
the indirect use of fossil fuels used for the production process of fertilizers
and agrochemicals.
References
Amanor, K.S. & Pabi, O. (2007). Space, Time, Rhetoric and Agricultural
Change in the Transition Zone of Ghana. Human Ecology, 35, 51-67.
Bonny, S. (1993). Is Agriculture Using More and More Energy? A French
Case Study. Agricultural Systems, 43, 51-66.
Boserup, E. (1965). The conditions of agricultural growth. The economics
of agrarian change under population pressure. Chicago:
Aldine/Earthscan.
Boserup, E. (1981). Population and Technology. Oxford: Basil Blackwell.
Chayanov, A.V. (1966). The Theory of Peasant Economy. Edited by
Thorner, Daniel; Smith, R.E.F.; Kerblay, B. Homewood: American
Economic Association.
Fischer-Kowalski, et al.
Human Ecology Review, Vol. 18, No. 2, 2011 157
Clark, C. & Haswell, M. (1967). The Economics of Subsistence Agriculture.
London: McMillan.
Costanza, R., Graumlich, L.J. & Steffen, W. (2007). Sustainability or Collapse?
An Integrated History and Future of People on Earth. Cambridge,
MA, London, UK: The MIT Press.
Ellis, F. (1998). Household strategies and rural livelihood diversification.
Journal of Development Studies, 35, 1-38.
Eurostat. (2001). Material use indicators for the European Union, 1980-
1997. Economy-wide material flow accounts and balances and derived
indicators of resource use. Working Paper No. 2/2001/B/2.
Luxemburg: European Commission/Eurostat, prepared by Bringezu,
S. and SchĂĽtz, H. -109.
Eurostat. (2007). Economy-wide Material Flow Accounting. A Compilation
Guide. Luxembourg: European Statistical Office.
Fischer-Kowalski, M. & Haberl, H. (1998). Sustainable Development:
Socio-Economic Metabolism and Colonization of Nature. International
Social Science Journal, 158, 573-587.
Fischer-Kowalski, M. & Haberl, H. (2007). Socioecological transitions
and global change: Trajectories of Social Metabolism and Land Use.
Cheltenham, UK; Northhampton, USA: Edward Elgar.
Fischer-Kowalski, M., Haberl, H., HĂĽttler, W., Payer, H., Schandl, H.,
Winiwarter,V. & Zangerl-Weisz, H. (1997). Gesellschaftlicher Stoffwechsel
und Kolonisierung von Natur. Ein Versuch in Sozialer
Ă–kologie. Amsterdam: Gordon & Breach Fakultas.
Fischer-Kowalski, M. & Hüttler, W. (1998). Society’s Metabolism. The Intellectual
History of Material Flow Analysis, Part II: 1970-1998.
Journal of Industrial Ecology, 2, 107-137.
Fischer-Kowalski, M. & Rotmans, J. (2009). Conceptualizing, Observing
and Influencing Social-Ecological Transitions. Ecology and Society,
14, 1-3. [online] URL: http://www.ecologyandsociety.org/vol4/
iss2/art3.
Fischer-Kowalski, M., Singh, S.J., Ringhofer, L., GrĂĽnbĂĽhel, C.M., Lauk,
C. & Remesch, A. (2010). Sociometabolic regimes in indigenous
communities and the crucial role of working time: A comparision of
case studies. Social Ecology Working Paper, 121. Vienna: IFF Social
Ecology, 1-41. 1726-3808.
Geertz, C. (1963). Agricultural Involution. Berkeley: University of California
Press.
Golley, F.B., Ruiz, A.C. & Bellot, J. (1990). Analysis of resource allocation
to irrigated maize and wheat in northern Spain. Agriculture, Ecosystems
& Environment, 31, 313-323.
Gowdy, J. (1998). Limited Wants, Unlimited Means: A Reader on HunterGatherer
Economics and the Environment. Washington D.C:
Haberl, H. (2001). The Energetic Metabolism of Societies, Part I: Accounting
Concepts. Journal of Industrial Ecology, 5, 11-33.
Haberl, H. (2002). Economy-Wide Energy Flow Accounting. In H. Schandl,
C.M. GrĂĽnbĂĽhel, H. Haberl & H. Weisz (Eds.), Handbook of
Physical Accounting. Measuring Bio-Physical Dimensions of SocioEconomic
Activities. MFA - EFA - HANPP. Vienna: Federal Ministry
of Agriculture and Forestry, Environment and Water Management,
30-48.
Haberl, H., Fischer-Kowalski, M., Krausmann, F., Martinez-Alier, J. &
Winiwarter, V. (2011). A socio-metabolic transition towards sustainability?
Challenges for another Great Transformation. Sustainable
Development, 19, 1-14.
Haberl, H., Fischer-Kowalski, M., Krausmann, F., Weisz, H. & Winiwarter,
V. (2004). Progress Towards Sustainability? What the conceptual
framework of material and energy flow accounting (MEFA) can offer.
Land Use Policy, 21, 199-213.
Haberl, H., Weisz, H., Amann, C., Bondeau, A., Eisenmenger, N., Erb, K.H.,
Fischer-Kowalski, M. & Krausmann, F. (2006). The energetic metabolism
of the EU-15 and the USA. Decadal energy input time-series
with an emphasis on biomass. Journal of Industrial Ecology, 10,
151-171.
Hunt, R.C. (2000). Labor Productivity and Agricultural Development:
Boserup Revisited. Human Ecology, 28, 251-277.
Krausmann, F. (2004). Milk, Manure and Muscular Power. Livestock and
the Industrialization of Agriculture. Human Ecology, 32, 735-773.
Krausmann, F., Haberl, H., Erb, K.-H. & Wackernagel, M. (2004). Resource
flows and land use in Austria 1950-2000: Using the MEFA
framework to monitor society-nature interaction for sustainability.
Land Use Policy, 21, 215-230.
Lee, R.B. & DeVore, I. (1968). Man the Hunter. Chicago: Aldine.
Leemans, R. & Costanza, R. (2005). Integrated History and future Of People
on Earth (IHOPE). IHDP Newsletter, 2/2005, 4-5.
Matthews, E., Amann, C., Fischer-Kowalski, M., Bringezu, S., HĂĽttler, W.,
Kleijn, R., Moriguchi, Y., Ottke, C., Rodenburg, E., Rogich, D.,
Schandl, H., SchĂĽtz, H., van der Voet, E. & Weisz, H. (2000). The
Weight of Nations: Material Outflows from Industrial Economies.
Washington, D.C.: World Resources Institute.
Mayrhofer-GrĂĽnbĂĽhel, C. (2004). Resource Use Systems of Rural Smallholders.
An Analysis of Two Lao Communities. Universität Wien,
Ethnologie.
Pimentel, D., Dazhong, W. & Giampietro, M. (1990). Technological
Changes in Energy Use in U.S. Agricultural Production. In S.R.
Gliessman (Ed.), Agroecology, Researching the Ecological Basis for
Sustainable Agriculture. New York: Springer, 305-321.
Rambo, A.T. (1984). No free lunch: A reexamination of the energetic efficiency
of swidden agriculture. An introduction to human ecology research
on agricultural systems in Southeast Asia, 154-163.
Ringhofer, E. (2007). The Tsimané in their environment: a socio-ecological
analysis of the environmental relations of an indigenous community
in the Bolivian Amazon. IFF Soziale Ă–kologie.
Ringhofer, L. (2010). Fishing, Foraging and Farming in the Bolivian Amazon.
On a Local Society in Transition. New York: Springer.
Sahlins, M. (1972). Stone Age Economics. New York: Aldine de Gruyter.
Schandl, H., GrĂĽnbĂĽhel, C.M., Haberl, H. & Weisz, H. (2002). Handbook
of Physical Accounting. Measuring bio-physical dimensions of
socio-ecomonic activities. MFA - EFA - HANPP. 9. Vienna: Federal
Ministry of Agriculture and Forestry, Environment and Water Management,
1-75.
Sieferle, R.P. (1997a). Kulturelle Evolution des Gesellschaft-Natur-Verhältnisses.
In M. Fischer-Kowalski, H. Haberl, W. HĂĽttler, H. Payer,
H. Schandl, V. Winiwarter & H. Zangerl-Weisz (Eds.),
Gesellschaftlicher Stoffwechsel und Kolonisierung von Natur. Ein
Versuch in Sozialer Ă–kologie. Amsterdam: Gordon & Breach Fakultas,
37-53.
Sieferle, R.P. (1997b). RĂĽckblick auf die Natur: Eine Geschichte des Menschen
und seiner Umwelt. MĂĽnchen: Luchterhand.
Sieferle, R.P. (2001). The Subterranean Forest. Energy Systems and the InFischer-Kowalski,
et al.
158 Human Ecology Review, Vol. 18, No. 2, 2011
dustrial Revolution. Cambridge: The White Horse Press.
Singh, S.J. (2003). In the Sea of Influence: A World System Perspective of
the Nicobar Islands. Lund: Lund University Press.
Singh, S.J. & GrĂĽnbĂĽhel, C.M. (2003). Environmental Relations and Biophysical
Transitions: The Case of Trinket Islands. Geografiska Annaler,
Series B, Human Geography, 85 B, 187-204.
Singh, S.J., GrĂĽnbĂĽhel, C.M., Schandl, H. & Schulz, N. (2001). Social Metabolism
and Labour in a Local Context: Changing Environmental
Relations on Trinket Island. Population and Environment, 23, 71-
104.
Singh, S.J., Ringhofer, L., Haas, W., Krausmann, F., Lauk, C. & FischerKowalski,
M. (2010). Local Studies Manual: A researcher’s guide for
investigating the social metabolism of rural systems. Social Ecology
Working Paper, 120. Vienna: IFF Social Ecology, 1-69. 1726-3808.
http://www.uni-klu.ac.at/socec/downloads/WD120_Web.pdf
Singh, S.J. & Schandl, H. (2003). Socio-Economic Metabolism in the
Nicobar Islands. Empirical Research in Society-Nature Interactions.
In B. Benzing & B. Herrmann (Eds.), Exploitation and Overexploitation
in Societies Past and Present. IUAES-Intercongress 2001 Goettingen.
MĂĽnster: LIT Publishing House, 169-184.
Swanton, C.J., Murphy, S.D., Hume, D.J. & Clements, D.R. (1996). Recent
improvements in the energy efficiency of agriculture: case studies
from Ontario, Canada. Agricultural Systems, 52, 399-418.
Tainter, J.A. (1988). The collapse of complex societies. Cambridge: Cambridge
University Press.
Tainter, J.A. (2006). Social complexity and sustainability. Ecological Complexity,
3, 91-103.
Tsatsarelis, C.A. (1993). Energy inputs and outputs for soft winter wheat
production in Greece. Agriculture, Ecosystems & Environment, 43,
109-118.
Turner, B.L. & Fischer-Kowalski, M. (2010). Ester Boserup: An interdisciplinary
visionary relevant for sustainability. Proceedings of the National
Academy of Sciences of the United States of America, 107,
21963-21965.
UNDP. Human Development Report 2007/2008. Internet . 2007. 15-1-
2008. (GENERIC) Ref Type: Electronic Citation
Weisz, H., Krausmann, F., Amann, C., Eisenmenger, N., Erb, K.-H.,
Hubacek, K. & Fischer-Kowalski, M. (2006). The physical economy
of the European Union: Cross-country comparison and determinants
of material consumption. Ecological Economics, 58, 676-698.
White, L.A. (1949). Energy and the Evolution of Culture. In L.A.White
(Ed.), The Science of Culture: A Study of Man and Civilization. New
York: Grove Press, 363-393.
Wilkinson, R.G. (1973). Poverty and Progress: An Ecological Model of
Economic Development. London: Methuen.
Fischer-Kowalski, et al.