Hum Ecol (2011) 39:465–478 DOI 10.1007/s10745-011-9393-z

Beyond Verticality: Fuelscape Politics and Practices in the Andes Keely Maxwell

Published online: 10 May 2011 # Springer Science+Business Media, LLC 2011

Abstract This paper analyzes the material, discursive, and biophysical dimensions of fuelscapes, or energy landscapes. Ethnographic and ecological fieldwork was conducted in the Machu Picchu Historic Sanctuary, Peru. Andean land use studies have focused on agricultural patterns such as vertical production zones. Fuelscapes are an important, energy-based means of producing and representing landscape. They show how uncultivated lands fit into livelihood strategies and reflect historic sedimentation of landscape. Fuelscapes are shaped by ecological characteristics, historic settlement patterns and property rights, gendered and intergenerational divisions in household labor, and state conservation policies. Conservation policies delimit fuelscapes to privilege live trees, but the resultant denudation of dead wood may carry implications for ecosystem health. This study elucidates how official policies intersect with household and communal resource use strategies to produce Andean fuelscapes. It provides insight into how uncultivated ecosystems fit into land use politics, practices, and representation. Keywords Firewood . Fuelscapes . Andes . Conservation . Energy . Peru

Introduction Andean landscapes are a wonder to behold. Snow-capped peaks loom above the horizon, forest greenery peeks out of steep ravines, and corn stalks reach to the sky from valley K. Maxwell (*) Franklin and Marshall College, Lancaster, PA 17604–3003, USA e-mail: [email protected]

bottoms. Evidence of human modification of the landscape is everywhere: irrigation canals snake down mountains and terraces extend up hillsides. If we zoom in our gaze, we find people scouring scrubby slopes, busy at the daily task of gathering firewood. Men, women, and children traverse the terrain, tugging at recalcitrant branches, chopping up logs, and loading wood onto donkeys. Plant biomass is a principal source of cooking energy throughout the Andes and is of critical importance to rural livelihoods. To date, however, we have little understanding of the political, socioeconomic, and biophysical factors that influence household energy strategies, nor how energy acquisition fits into human-environment relations at large. In this article, I analyze the landscape of energy production. I utilize ethnographic and ecological data to show how firewood serves as an important lens through which people perceive, represent, and interact with the landscape around them. I advance theoretical understanding of people’s relation with uncultivated ecosystems and fuel politics and practices. Spatial analyses of Andean land use historically have been informed by verticality, a concept utilized by John Murra to explain how pre-Columbian populations sought to “maximize vertical control over ecological floors” (Murra 2002 (1972):86). His study spurred ethnographic analysis of how contemporary populations divide land use into elevation-based production zones. Corn, potato, and livestock grazing take place in different zones. Each zone has a its own land use with distinct property rights over resources. Despite the proven variability of production zone spatiality on the ground (Zimmerer 1999), a vertical orientation continues to influence how Andean land use patterns are theorized and represented. Production zone research focuses on how agriculture and livestock grazing produce cultural landscapes. More recent works categorize

466

landscape as a matrix in which patchworks of land use with different degrees of anthropogenic influence are embedded (Young 2009). Such studies often break down land use on elevation lines and highlight how socioeconomic drivers, including population, cause land cover change (Jokisch and Lair 2003; Kintz et al. 2006; Postigo et al. 2008). They identify broad drivers of change but often fail to incorporate emic conceptions of landscape and local drivers of land use. Neither the production zone nor matrix approach addresses how firewood collection comprises part of land use patterns and property rights, or how fuel energy traverses ecosystems to support cultural enterprises. Nor do they pay significant attention to people’s relation with uncultivated ecosystems, a facet of Andean land use that remains understudied overall (Paulson 2003:247). Paulson utilized a gendered approach to conduct one of the few studies of uncultivated landscapes. She shows how it is important to livelihoods but utilized primarily by women and children, and thus neglected in land management projects (Paulson 2003:247). Other social forces that produce uncultivated lands remain unexplored. In this article, I follow material flows of firewood that link people with uncultivated lands. In so doing, I uncover how people produce and represent energy landscape, which is important to understand given the critical role energy plays in rural livelihoods. Globally, over 2.4 billion people depend on wood, agricultural residue, and dung to meet energy needs (International Energy Agency 2002:33). In Latin America, an estimated 23% of the population relies on such traditional biomass (International Energy Agency 2002:387). Notwithstanding the importance of biomass energy, it remained largely off the policy radar until Erik Eckholm’s book The Other Energy Crisis, Fuelwood (1975) called attention to a “woodfuel crisis” in developing countries. Eckholm highlighted the difficulties millions of people face in securing daily energy needs. A spate of support and critiques of the crisis hypothesis followed (Agarwal 1986; Dewees 1989; de Montelambert and Clement 1983). International institutions responded to the crisis by promoting plantations to increase supply and introducing improved stoves to reduce demand (Arnold et al. 2006). Even though biomass fuels are a significant component of people’s energy use and natural resource management, research on the topic has long been the ugly stepsibling to energy and forestry studies. There is increasing recognition that policy approaches that sketch a general picture of crisis or reduce the issue to a supply and demand problem are inadequate (Arnold et al. 2006). Despite such recognition, fuel studies continue to produce regional and global-scale estimates of wood demand instead of taking a wider human ecological perspective that captures the complexities of energy use (e.g., Broadhead et al. 2001). Studying the human ecology of firewood sheds

Hum Ecol (2011) 39:465–478

light on the conditions under which people make the resource use decisions they do and offers insights into human-environment relations more broadly. To investigate the Andean landscape of energy, I utilize the concept of “fuelscape,” a term that emerged during 1980s discussions on the woodfuel crisis. Crisis literature set the parameters of fuelwood discussion along the lines of physical wood availability and scarcity. Wisner et al. (1987:233) were the first to use the term fuelscape, characterizing it as an “unseen landscape” of wood scarcity and abundance. Wisner argues that material differences in fuel reflect socioeconomic conditions including class stratification, population growth, and wood commodification, not simply the presence or absence of trees (Wisner 1989:240). He utilizes an economic approach to show how competing demands on time, money, and land influence energy access in rural Kenya. Solutions to fuel scarcity, he argues, need to address underlying rural poverty instead of just trying to increase wood supply and decrease demand (Wisner 1989:245). Cline-Cole (1998) builds upon the fuelscape concept. He characterizes fuelscapes as hidden landscapes of scarcity and abundance produced through forestry institution reports and policies. Both Wisner and Cline-Cole illuminate how the social context of wood use has traditionally been “unseen” by crisis literature. Cline-Cole goes one step further to highlight how state reports and scientists do not merely report scarcity, they create it through estimates and presumptions. Broader social forces, he argues, influence how international and Nigerian foresters construct “knowledge claims” about fuelwood in Nigeria (Cline-Cole 1998:343). By keeping fuelwood within the realm of discourse, however, he fails to connect energy politics with how everyday Nigerians perceive and produce firewood. Here, I expand the notion of fuelscape yet again. I move beyond a focus on wood scarcity and abundance to incorporate how people shape and represent energy landscapes. I utilize ecological and ethnographic evidence collected in the village of Rayanpata, Department of Cusco, Peru, to show interconnections among the material, discursive, and biophysical dimensions of fuelscapes. Examining fuelscapes provides a way of understanding how a particular territory is shaped by a combination of ecological properties and regular activity patterns influenced by a broader social and political context. In Rayanpata, firewood is the principal source of cooking energy. It is a commons resource collected from cropland, scrub, forest, and puna (high grassland) ecosystems. Fuelscapes comprise well-trodden paths going from population centers to the monte (uncultivated ecosystems). Villagers do not conceptualize fuel collection as movement between vertical ecological floors or production zones, but as a journey from cultivated to uncultivated ecosystems.

Hum Ecol (2011) 39:465–478

Fuelscapes are coded with symbolic meaning, a means of producing and representing landscape. Their shape is not predicted by Andean land use or fuelwood literature. They do not completely track vertical ecological lines of more collection in lower, wood-rich ecosystems and less in higher ecosystems with less woody biomass. Nor do they follow a gradient of wood scarcity near human populations to abundance further away, as crisis literature predicts. Socioeconomic factors such as population and poverty are not the principal drivers of energy behavior in Rayanpata. I argue that the four principal factors that shape fuelscapes are: ecological characteristics, historic settlement patterns and property rights, gendered and intergenerational divisions in household labor, and state conservation policies that privilege live trees. I advance fuelscapes beyond a focus on scarcity and abundance to become a more powerful tool for analyzing human-environment relations. Examining people’s relations with the uncultivated ecosystems that comprise fuelscapes illuminates the historic and modern sedimentation of anthropogenic landscapes, including the spatial articulation of livelihood practices and state policies.

467

Forests grow in patches along irrigation canals and on steep slopes surrounding streams. The predominant overstory species is unca (Myrcianthes oreophyla, M. indifferens). Beneath the forest canopy, understory tree species such as maki maki (Oreopanax spp.) and pukañawi (Mauria ferroginia) grow. On the forest floor, herbs, grasses, and moss abound. There is a 10 cm deep layer of organic matter. Scrub hillsides turn into puna above approximately 3,700 m.a.s.l. Puna is characterized in ecological literature as grassland, but shrubs and stands of trees are also present. Bunch grasses and herbaceous mats form the ground cover. Rayanpateños graze cattle and mules here. The patchiness of these ecosystems reflects environmental and anthropogenic variability at macro- and micro-topographic scales. Factors influencing vegetation structure and composition include snow, wind, landslides, fire, agriculture and livestock. Vegetation in these ecosystems is comparable to genera and species that grow elsewhere in the Andes (Stern 1995: Young 1993; Young and Keating 2001), indicating that they are operating under similar biophysical and social processes.

Methodology Site Description The peasant community of Rayanpata extends across several steep river valleys ranging from 2,600 to 4,200 m. a.s.l. It is wholly enclosed in the Machu Picchu Historic Sanctuary, which is managed by the Natural Resources Institute, the National Cultural Institute, and the Machu Picchu Management Authority. Approximately 250 Quechua-speaking subsistence farmers who grow corn and potatoes live in Rayanpata. The cash economy in the village is directly tied to tourism to Machu Picchu and the Inca Trail. Men work as porters and cooks on trekking expeditions or as site wardens and archaeological reconstruction workers for Sanctuary institutions. Women sell refreshments to tourists and porters on the trail. Villagers collect firewood from cropland, scrub, forest, and puna (Fig. 1). Table 1 lists the predominant tree, shrub, and ground cover characteristics of these ecosystems. The population center at 3,000 m.a.s.l. is surrounded by cropland, which consists of a scattering of agricultural fields, grassy pastures, shrubs, and trees. Disturbancetolerant shrubs and planted trees are used for food, fodder, fences, and fuel. Beyond field boundaries, slopes steepen quickly and the vegetation turns to scrub. Scrub hillsides receive limited direct sunlight, but have few water sources and bake when the sun does hit. Little organic matter covers the ground. Vegetation grows as clumps of shrubs, herbs, and grasses, with occasional trees. Villagers graze livestock here, and animal paths are trodden into the soil.

The present study breaks new ground by utilizing mixed qualitative and quantitative methods to paint a comprehensive picture of the cultural and ecological dimensions of energy use. Much of my data was collected utilizing ethnographic methods. I began my study with semistructured interviews about fuelwood with randomly selected household heads, but found that opportunistic interviews on particular sub-topics elicited more substantial answers. I triangulated responses from a variety of informants. It took time to build trust on charged topics such as Sanctuary policies and communicate about abstract concepts such as landscape. I observed daily activity from different viewpoints in the village to see people’s patterns of movement. I also did participant observation, e.g. went on fuel collection trips, chatted with women and girls cooking over the fire, and patrolled with park rangers. To complement these qualitative data, I conducted a survey of 87 household heads (male and female) in Rayanpata and adjacent communities. Households were selected by stratified random sampling based on community residency. The survey utilized an open-ended questionnaire presented in Spanish or Quechua depending on the comfort of informants. Questions focused on socioeconomic aspects of fuel collection and wood preferences. I analyzed Sanctuary policies through interviews with officials and textual analysis of policy documents. I recorded time allocation in three households in Rayanpata using monthly visits over the course of a year

468

Hum Ecol (2011) 39:465–478

Fig. 1 Cropland, scrub, forest, and puna in Rayanpata. Firewood is collected from Llullucha, Huayruro, and Matara river valleys

(Gross 1984; Johnson 1975; Leatherman 1996). Households with similar age distributions but different income classes and offspring gender ratios were selected. I visited each house once a month from 5 a.m. to 8 p.m. and recorded fixed interval behavioral observations once an hour, for a total of 5,848 observations. Spot checks confirmed that the spectrum of activities and time allocated

to each were representative of other households. I coded behaviors and analyzed patterns according to age and gender. During these visits, I weighed all firewood collected, stored, and burned, separating wood by species. Four assistants and I collected vegetation data from forty-five 100 m2 plots, with three plots each in three scrub, seven forest, and four puna locations. Using a variation on

Livestock, firewood, medicinal plants, former potato, former houses

Livestock, firewood, medicinal plants, construction

Livestock, firewood, medicinal plants

Corn, potato, houses, domestic animals, firewood

469

Herbs, grasses, moss, leaf litter, organic matter

Stipa icchu, Bromus spp., Acaena spp

Forest (2600–3800 m.a.s.l.)

Puna (3,600–4,400 m.a.s.l.)

Eucalyptus globulus, Prunus serotina, n/a P. persica, Erythrina falcata, Alnus acuminata Escallonia resinosa, Hesperomeles Tree: 3,942 lanuginosea, Oreocallis grandiflora Shrub: 86,700 Dead: 3,8275 Understory trees: Mauria Overstory trees: Myrcianthes Tree: 9,344 ferruginea, Oreopanax ischnolobus, oreophyla, M. indifferens, Shrub: 41,956 Vallea stipularis, Nicotiana tomentosa Myrsine latifolia, Styloceras Dead: 2,407 laurifolium, Alnus acuminata Berberis spp., Barnadecia horrida., Vallea stipularis, Gynoxys spp. Tree: 2,017 Baccharis spp. Shrub: 60,124 Dead: 5,750 Cropland (2600–3200 m.a.s.l.) Pennisetum clandestinum, Acalypha aronoides, Baccharis spp., paths, agricultural fields, Berberis spp., Nicotiana tomentosa stone walls, live fences Scrub (2600–3600 m.a.s.l) Leaf litter, animal trails, Baccharis spp., Calceolaria spp., Bromus spp., Festuca spp. Minthostachys spp., Nicotiana tomentosa

Trees Shrubs Ground cover Ecosystem

Table 1 Cropland, scrub, forest, and puna characteristics. Top species & total stems/ha in each ecosystem

Stems/ ha

Land use

Hum Ecol (2011) 39:465–478

Whittaker and Long Thin Plot nested vegetation sampling methods (Schmida 1984; Stohlgren et al. 1995), we measured dbh (diameter at breast height) of all trees over 5 cm dbh in the 5×20 m plot, shrub basal diameters and tree regeneration in a 5 × 5 m subplot, and percent herbaceous ground cover and organic matter depth in five 1×1 m subplots. In the 5×5 m subplot, we measured the volume of standing dead wood and weighed fallen dead wood. We divided fallen dead wood by species, decomposition category (more and less decomposed, based upon visual and tactile characteristics), and diameter class (< 1 cm, 1–5 cm, 5–10 cm, and 10–20 cm). In cropland, we took 15 vegetation transects across fields and paths to collect shrub and ground cover data and measured all trees over 10 cm dbh.

Vertical Landscapes Murra’s (Murra 2002 (1972)) verticality theory provides the foundation for subsequent analyses of Andean land use patterns. In the 1950s, John Murra researched how five preColumbian societies worked to access the greatest possible number of ecological floors. He termed the resultant spatial pattern of territorial control “vertical archipelagos,” whereby a single ethnic group established colonies in multiple floors (Murra 2002 (1972):86). Verticality theory underscores how ecological gradients influence social organization. It has been modified to explain past and present land use patterns (Platt 1982; Webster 1971). This carryover from past to present fosters imaginaries of lo andino, the idea that a fundamentally Andean way of life and worldview has persisted across the centuries. Brush (1976:161–164) describes three spatial configurations of verticality in northern Peru: compressed, a single population accesses multiple crop zones; archipelago, exploited zones are widely separated requiring migration or colonization; and extended, zones are contiguous but distances are long, requiring market exchange. Although verticality has been debated and contested (van Buren 1996, Zimmerer 1996), a vertical orientation continues to imbue spatial analyses of Andean land use. Other studies of land use patterns have shifted terms from verticality to production zone, “a communally managed set of specific productive resources in which crops are grown in distinctive ways” (Mayer 2002:245). This definition emphasizes human agency over environmental adaptation, utilizes agriculture to distinguish zones, and highlights the communal context of land use decisions. Researchers have delineated three principal production zones in the Peruvian Andes, divided along elevation lines. In high altitude puna (páramo, jalka), farmers grow potatoes and other tubers. Above crop limits, pastoralists

470

graze livestock. In temperate quishuar (kishwar, kichwa), corn, wheat, and barley are grown. In lower altitude yungas (montaña, temple), tropical crops such as coca, coffee, citrus, and banana are grown (Brush 1976; Flores Ochoa 1985; Mayer 2002). Production zones are agricultural strategies, but also cultural representations of space with encoded meanings (Fioravanti-Molinie 1982; Orlove 1977; Zimmerer 1999). Andean cosmologies inform production zone symbolism through dualisms between upper and lower. Corn-growing quishuar residents depict puna uplands as containing a wild, savage environment and people, in contrast to domesticated valleys (FioravantiMolinie 1982:228; Isbell 1977:57). Rayanpata contains puna tuber and livestock grazing zones and quishuar corn zones. The population center lies at the upper boundary of the quishuar zone. Previously, most villagers resided in the puna and kept a second hut near corn fields. In recent years, tourism jobs and education opportunities have pushed the population center down to the quishuar zone. Rayanpateños grow a third- to halfhectare field of corn there. A few families continue to plant potato nearby, but no longer grow potato high in the puna since they don’t live close enough to protect fields from being overrun by livestock or tourist campsites. Property rights to land and resources vary with production zone. High pasture and tuber lands tend to be communally managed; lower corn and tropical crop fields tend to be individually managed (Guillet 1981). This trend does not hold true for all rural areas. Climate, topography, Inca and colonial resettlements, estate divisions, and agrarian reform combine to create diverse systems of land tenure. Moreover, distinctions between private and communal property are blurred at all elevations. Corn fields in the quishuar zone belong to individuals, but irrigation is managed communally (Gelles 1999). Tuber fields in the puna are planted by individuals, but follow communal sectorial fallowing schedules (Guillet 1981; Orlove and Godoy 1986). Household rights to communal pasturelands are fluid, not static (Postigo et al. 2008). Property rights to firewood in Rayanpata, as I will demonstrate, involve a similar blending of communal and individual rights. Zimmerer (1999) cautions against the presumption that production zones are environmentally determined and stacked one atop the other. Presumptions of verticality reduce complex agricultural strategies to environmental adaptations and turn land use into a static “spatial container for farming” (Zimmerer 1996:20). In Paucartambo, Peru, and Cochabamba, Bolivia, Zimmerer (1999) found that production zones are named after topographic features (e.g. hillside, valley), giving the impression that they are vertically organized. The spatial pattern of farming units on the ground, however, is one of overlapping patchworks rather than vertical floors (Zimmerer 1999). Patchiness

Hum Ecol (2011) 39:465–478

results from how zones are developed in response to socioeconomic as well as environmental factors. Production zones in Potosí, Bolivia are oriented towards a sociallydetermined center, not a vertical gradient, so a tuber zone may be at a lower elevation than a corn zone (Harris 1985). Despite the proven variability of production zone spatiality on the ground, a vertical orientation continues to influence how land use patterns are theorized and represented. Production zones are defined and perceived through the lens of agricultural activity. There has been little focus on how energy practices integrate into land use categories. Several analyses of production zones fail to mention fuels altogether (Dollfus 1982; Mitchell 1991; Webster 1971, Wiegers et al. 1999). Brush (1977:80) notes that firewood sources, while important to rural households, are not designated as a distinct zone in Uchucmarca, Peru. On the other side of the coin, ethnobotanical studies record whether individual species are used as fuel, but do not detail distribution and collection patterns (Franquemont 1988; Gade 1967; Bentley and Valencia 2003). Studies of Andean agriculture have revealed the many ways people shape the material and symbolic landscape of the Andes. Fuelwood practices, I posit, play similarly important roles in people’s production of Andean landscapes.

Firewood and Landscape Fuelscapes contain material and semiotic elements. Patterns of fuel availability reflect differences in ecosystem composition and productivity. Verticality is not absent from fuelscapes, but it is not the predominant geographic orientation, either. In high altitude puna grasslands, woody biomass is scarce and productivity is low. Inhabitants plant a few trees around their houses, but these are slow to mature and cannot sustain daily energy needs. People of the puna use three principal sources of fuel: dried livestock dung, collection trips to lower altitude ecosystems, and harvesting shrub stems and grasses on a near-daily basis (Allen 1988; Brush 1977; Johannessen and Hastorf 1990; Larme 1993; Weismantel 1988; West 1987; Winterhalder et al. 1974). In inter-Andean valleys, such as where Rayanpata lies, the predominant tree species is non-native eucalyptus (Eucalyptus globulus). Outside of eucalyptus groves, great quantities of woody biomass are not readily available. Cultivatable land is utilized for agricultural fields and patches of forests grow in inaccessible places. Rayanpata’s proximity to extensive uncultivated forest and scrub gives it an edge up over nearby peasant communities. Residents of valleys not far away have had to rely for decades on purchasing firewood in weekly markets, harvesting eucalyptus, and collecting brush from fallow agricultural fields (Gade 1967; Skar et al. 1982). Rayanpata

Hum Ecol (2011) 39:465–478

is also distinct from neighboring villages whose populations are declining due to migration to urban centers. Young couples stay in Rayanpata because of tourism work opportunities. Should the winds of tourism opportunity change, so will this trend. Even though villagers earn cash from tourism, they continue to utilize firewood instead of purchasing fuel. As such, overall wood use likely has increased in recent years. Villagers use the term leña (llant’a in Quechua) for firewood. Leña generally encompasses things that burn in the fire, but refers primarily to sizeable pieces of wood. Other categories of firewood are rajada (branches with leaves, such as those fallen from eucalyptus) and charamosca (a mix of small twigs and grasses). Rayanpateños utilize nearly 50 types of fuels at an average rate of 1.8 k per capita per day. Of these fuels, 35 are woody plant species; the rest include dried grasses and herbs, livestock dung, crop residues, and construction materials. Not all plants are equally valued as fuels. People prefer wood that “burns well:” it dries quickly, catches fire rapidly, burns for a long time, and produces little smoke. Villagers are embarrassed and annoyed when they lack adequate fuels: “it’s only charamosca” one cook lamented to me as she labored to fan flames of dried grass. Another woman who had to resort to burning an old fence post complained that her lazy husband had not brought her enough leña. Using undesirable fuels is cumbersome and suggests to visitors that household management is wanting. The majority of survey respondents prefer leña from native trees (59%) over smaller shrub stems (27%) and eucalyptus (31%).1 Those who prefer shrubs and eucalyptus live in communities with limited access to larger native trees; Rayanpateño residents listed only native trees (80%) and eucalyptus (33%) as preferred fuels. Table 2 lists the top 10 fuels burned in household fuel studies. Four preferred fuels in Table 2 come from the forest overstory: chachakomo, aliso, unca, and t’asta (Nos. 1, 3, 4, 10). Rayeta (No. 7) is a low, mounded plant from the puna. Non-preferred pukañawi, q’amasto, and p’ispita (Nos. 5, 6, 9) are smaller trees and shrubs that grow in scrub and forest understory. The extensive use of non-preferred fuels reflects their ready accessibility. They are ubiquitous in cropland and scrub close to people’s hearths, and their growth form lends itself to easy collection. P’ispita, for example, grows in a mass of stems, so biomass turnover provides a constant source of dead wood. It is easy for women walking by to snap off a few dead branches to cook lunch. The impact of fuel collection on the landscape takes place on the scale of individual trees rather than an overall reduction of forest cover. Villagers do fell entire trees for

1

Percentages are not exclusive in survey responses.

471

fuel, but the impact of their actions is mitigated since Andean trees tend to grow multiple trunks (Stern 1995; Valencia 1992; Young and Keating 2001). Also, people harvest wood from multiple ecosystems. Another strategy is to pollard trees in cropland. Pollarding involves regularly cutting back branches a few meters above the ground. It encourages regrowth while protecting new shoots from browsing livestock. A third strategy is to collect “dry wood,” or already dead wood. Seventy-four percent of survey respondents say they prefer to collect dead wood as it requires less time to dry before it can “burn well.” Firewood is an important means by which people perceive and represent landscape. Villagers use production zone terms quishuar and puna to describe general landscape divisions. When I surveyed household heads as to where they get their fuel, though, no one gave responses based on production zone. Forty percent of respondents named individual places where they went for fuel. Place names are key aspects of people’s understanding of landscape, as they both describe an area and signal appropriate resource access and use (Zimmerer 1999:153). In Rayanpata, each hilltop, ravine, and agricultural field has its own name that reflects landscape features, land use, plants, and cosmological dualisms of upper and lower. Along with place names, dead wood comprises part of people’s cognitive maps of their community (c.f. Franquemont 1988). Villagers make mental notes of good fuel sources. When I set out on fuel trips with people, they’d make a beeline for a particular place, saying “I saw some good firewood the last time I was by here.” When searching for fuel, people use dead wood as a filter through which they perceive the landscape: dried plant stems move to the foreground, and other resources are relegated to the background. Forty-nine percent of survey respondents used directionality to indicate fuel location. Informants would wave their hand towards the mountains and say they got fuel from “up there,” with a tone implying that it was a long and arduous journey. Traveling “up” from Rayanpata’s center to scrub, forest, and puna is an artifact of agrarian reform policies that gave the community access to multiple ecosystems. It also reflects intra-community migration down from the puna to reside in lower valley lands. Traveling “up” appears to imply verticality, but villagers do not categorize fuel sources by ecological floor or production zone. Instead, fuel collection is conceived as material flows from uncultivated to cultivated lands. Discussion with informants elicited a term that encompasses the places from which people get fuel: monte. Monte translates as scrubland, woodland, or mountain, a term that aptly describes the entangled mass of vegetation through which one must scramble to obtain firewood. Villagers said that they “went to the monte” when collecting fuels from scrub, forest, and puna, interchangeably.

472

Hum Ecol (2011) 39:465–478

Table 2 Top ten fuels used in Rayanpata. Ranked by kilos burned on stove in household fuel measurements. Preferences articulated by Rayanpata residents. Source ecosystems: C cropland; S scrub; FO forest overstory; FU forest understory; P puna Use Ranking

Common Name

Scientific Name

Family

Source Ecosystem

Preferred species?

1 2 3 4 5 6 7 8 9 10

Chachakomo Chikllur Aliso Unca Pukañawi Q’amasto Rayeta Eucalyptus P’ispita T’asta

Escallonia resinosa Vallea stipularis Alnus acuminata Myrcianthes oreophyla Mauria ferruginea Nicotiana tomentosa Azorella spp. Eucalyptus globalus Acalypha aronoides Escallonia myrtilloides

Grossulariaceae Elaeocarpaceae Betulaceae Myrtaceae Anacardiaceae Solanaceae Apiaceae Myrtaceae Euphorbiaceae Grossulariaceae

S, FO S, P, FU FO FO S, FU C, S, FU P C C, S FO, P

Yes Yes Yes Yes No No No Yes No Yes

Monte is vegetation that “grows by itself,” i.e. uncultivated trees and shrubs. Cultivated and uncultivated lands are distinguished in a manner distinct from vertical differentiations between wild, savage puna uplands and domesticated quishuar valleys. While the division between wild and domestic remains, monte has its own geographic configuration and cosmological symbolism. The monte is wild because it was not planted, but humans can and should use monte vegetation, and there is little spiritual or physical danger from doing so. Sherbondy (1986) similarly notes that Andean residents distinguish wild from planted trees, and have distinct symbolic and material relationships with each. Just because monte is uncultivated now doesn’t mean that it wasn’t previously used or impacted by people. Moreover, uncultivated areas are important sources of wild agrobiodiversity and other important crop inputs (Zimmerer 2010:351). Monte trees grow on former agricultural lands as well as in ravines never used for crops or grazing. Scrub, forest, and puna all reflect some degree of anthropogenic influence. Fuelscapes and production zones operate simultaneously as landscape organizing principles in Rayanpata. Examining people’s relations with monte provides a deeper understanding of the historic sedimentation of anthropogenic landscape than does attention to agricultural land use alone. The monte is an uncultivated landscape, but it is also a produced one.

Property Rights and Settlement Patterns Fuelscapes are contoured by vegetation availability and fuel preferences, then further delimited by settlement patterns that determine property rights to biomass fuels. Rights to trees vary with community residency, house and field ownership, and the degree of human labor in tree planting and harvest—a complex mix of ownership and usufruct

rights seen elsewhere in the world (e.g. Fortmann 1985). Planted trees belong to their owners, as is the case with eucalyptus groves. When native trees are planted they, too, become private property. Tree ownership does not necessarily entail exclusive use, though. People must purchase a eucalyptus tree to cut it down, but anyone may collect branches fallen on the forest floor to burn on the hearth. Harvesting wood confers exclusive usufruct rights to the harvester. Ownership of harvested wood is accrued by felling live trees or cutting up already dead wood. Rights remain whether the harvest is brought to the house or not. Two claimants to firewood once got into a physical fight when the person who had cut up wood and left it in the forest to dry claimed that another person had taken “his” wood. Monte is a commons resource since its trees “grow by themselves” and are not planted by anyone. Peasant community membership confers rights to harvest resources from the monte. People do travel a short distance across community borders to collect wood, creating a buffer zone of bi-communal plant resources. In village interiors, however, outsiders sneaking in to harvest wood would be dealt with harshly. Peasant community boundaries were created as large estates underwent inheritance subdivisions and agrarian reform, so fuelscapes reflect past territorial partitioning. Even though the monte is a commons open to all community members, informal spatial subdivisions influence where individuals go to collect fuel on a regular basis. These subdivisions are based on historic settlement patterns. Villagers collect fuel from one of Rayanpata’s three upland valleys, Llullucha, Huayruro, and Matara (Fig. 1). The valley from which a family collects fuel correlates to that family’s particular settlement history, namely in which upland valley they used to reside before moving down to the quishuar zone. Going “up” to collect fuel means traveling to the monte, but also traveling through a landscape of memory.

Hum Ecol (2011) 39:465–478

One must know the monte to successfully utilize its resources, and community membership alone does not confer this knowledge. Repeated travels to the same upland valley to graze livestock and harvest fuel promote intimate knowledge of the environment. On more than one occasion, villagers took me to part of the monte where they had just discovered an untapped source of firewood. Interestingly, these places were remarkably close to houses and trails. The location of new fuel sources close to population centers belies the crisis narrative of fuelwood scarcity in which firewood collection is said to create concentric circles of deforestation around population centers. Villagers who found new fuel sources asserted that fuel was not really scarce in Rayanpata: “there’s enough firewood if you know where to look; if you don’t know, you suffer, I’ve seen it; but it’s easy, if you know.” Those who don’t know where to look, these successful firewood finders imply, are responsible for creating their own fuelscapes of scarcity by lacking knowledge of the monte.

Household Labor and Fuelscapes The socioeconomic factor that most impacts daily energy decisions is the scarcity and abundance of household labor. Within a given household, family members of different ages and genders make distinct fuel collection trips (Fig. 2) (Maxwell 2011). Men and young adult males make the longest trips, which may be undertaken in conjunction with another activity such as checking on livestock. They hike 1 to 3 h “up” and haul 30 k or more of firewood on their backs, usually accompanied by a donkey with its own load. These trips produce larger logs of preferred species. They are particularly necessary to stock up on wood before the rainy season. Temporal delineation of fuelscapes involves a spike in wood collection before the rainy season begins and even “dry” wood gets wet.

473

Women and children may accompany men on long trips, but informants categorized these trips as belonging to the domain of male labor. When I surveyed household heads about who collected firewood, 42% said men went; 32% said everyone in the house went; 18% said male and female heads went; and 8% said women went. In one such exchange, the husband proclaimed: “we, the men, go to make firewood.” His assertion struck me as peculiar since previously, I’d only seen his wife carrying wood. Women and young adult females have less time for long collection trips. Most of their trips are made in conjunction with other chores, so they tend to collect fuel from nearby scrub, cropland, and forest. They gather smaller loads of dried branches and stems, often of easily accessible non-preferred species. Women’s role in fuel collection was visible to me as an outsider watching people’s movement across the landscape. Locals, however, tended to interpret collecting leña as bringing back large loads of preferred species from forest as men do, not an armful of non-preferred species from scrub as women do. The large quantity of nonpreferred species burned on the fire (Table 1) indicates that women do make a significant energy contribution to the hearth. Children and elderly villagers also do energy work (Fig. 2). Groups of children as young as four are sent to nearby scrub and forest to learn collection techniques and fuel knowledge. They also have a great deal of fun, and would proudly give me their small bundles of wood to weigh. Children collect dried branches of preferred and non-preferred species. Elderly villagers, too, bring energy for the hearth. They collect small loads of dead branches, twigs, and grass from cropland. While the quantity of fuel that children and seniors bring back adds relatively little energy to the fire, the labor they expend reinforces their role as an integral part of the household. Households at different stages of their life cycle produce and reproduce distinct fuelscapes. Adult offspring bring loads of

Fig. 2 Gendered and Intergenerational Fuelscapes. Children: ages 4–10, male and female. Young adult males and females: ages 11–23. Men and women: household heads. Seniors: over 75

474

firewood to elderly parents. Women with husbands who work on trekking trips are most likely to purchase eucalyptus. Families with young adult males have the most wood stored. Household composition and labor allocation strategies play a significant role in determining who goes to collect fuel, where they go, and what types of wood they bring back, producing gendered and intergenerational fuelscapes.

Conservation Policies & Firewood Collection Fuelscapes are a resource use strategy that has long been embedded in larger political economic systems. Inca and Spanish heads of state required subjects to plant trees and supply fuel as tribute (Ansión 1986; Chepstow-Lusty and Winfield 2000; Hastorf et al. 2005; Johannessen and Hastorf 1990; Sherbondy 1986). After Spanish conquest, mining center Potosí required huge quantities of fuel, leading to wood depletion for many kilometers around (Godoy 1984). Modern state policies also have altered access to fuel. Peasant cooperatives in Apurimac, Peru limited fuel access as part of territorial struggles during agrarian reform (Skar et al. 1982). The 1990 Fujishock economic policy to reduce hyperinflation by relaxing price controls caused the price of propane and kerosene to spike, leading to a temporary upsurge in firewood demand as people turned back to wood-fueled stoves. Today, one of the ways by which government actors seek to modify fuelscapes in Rayanpata is via conservation policies, making fuelscapes a co-production of rural land use and state politics. Conservation policy in the Machu Picchu Historic Sanctuary is guided by the 1998 Master Plan. It recommends that the Natural Resources Institute, the state agency responsible for managing natural resources in the Sanctuary, “prohibit, without exception, the extraction of vegetative species from the forest” (INRENA 1998:280). Sanctuary managers equate tree use with deforestation, defined in environmental reports as a reduction in forest cover. Sanctuary inhabitants are not allowed to fell live trees for construction or fuel; they may only collect already dead wood. Justification for this policy lies in how Sanctuary institutions utilize particular forms of Peruvian science to justify environmental “knowledge claims” (c.f. Cline-Cole) backed by scientific authority, claims that privilege live trees over ecological functioning. Two differences from Cline-Cole’s analysis emerge. He emphasizes the role of international forest consultants in Nigerian wood politics (Cline-Cole 1998:313). My analysis shows how institutional rhetoric emerged from a regional Peruvian science that privileges live trees. Second, I expand on his approach by delineating how state policies have fed back to impact village energy practices and politics.

Hum Ecol (2011) 39:465–478

Sanctuary institutions utilize a regional production of knowledge about Machu Picchu’s natural resources. Since the 1940s, the principal form of scientific study of Sanctuary ecosystems has been botanical inventories done by scientists from the University of Cusco. Even before global concern over biodiversity emerged, the way in which scientists and officials valued the Sanctuary’s natural resources was by counting live plants. Today, managers utilize environmental reports from university scientists as a basis for policy decisions. The other dimension of Sanctuary “knowledge claims” is the Master Plan’s dependence on vertical land use classifications that utilize climatic, geologic, and elevation criteria to determine what vegetation is supposed to be in a given area (Holdridge 1967; Tosi 1960). According to the Master Plan, the area around Rayanpata should be “very moist-montane subtropical forest.” Managers presume that deviation from this vegetation scheme is caused by unwarranted anthropogenic interference, so restricting human use of trees will allow the forest to return. Privileging live trees in firewood policies is problematic for two reasons. First, the presumption that firewood collection causes deforestation is unwarranted. My examination of fuel practices indicates that the principal scale of impact is at the level of individual trees, not a reduction in forest cover. Vegetation data show that 67% of the basal area of live trees and shrubs is made up of the top 10 fuel species; the other 33% consists of 24 other species. These data indicate that wood harvest has not depleted the living biomass of highly used fuel species. Cumulative felling may promote species that can withstand repeated cutting, but woody vegetation is not in decline despite a lengthy history of fuel harvest. Other vegetation research in the Andes corroborates that forest patches can be relatively stable over time (Byers 2000; Jameson and Ramsay 2007; Kintz et al. 2006). Second, by using live trees as a metric of management success, Sanctuary policies ignore ecosystem processes. Dead wood plays important ecological roles from nutrient cycling to wildlife habitat. The dead wood villagers collect today is a legacy from the past and a renewable resource only if plants that die now remain in the monte. Fallen dead wood of the top 10 fuel species comprises an average of 33% of all dead wood (Fig. 3), less than expected given that these species make up a substantial proportion of live plants. The dead wood that does remain on the ground consists of pieces that are too small or too decomposed to serve as fuel. Sanctuary managers run the risk of alienating villagers by restricting fuelscapes to dead wood. Furthermore, this wood policy could be detrimental to ecosystem health in the long term. The policy has impacted energy politics and performance more than actual wood use. People have not substantially altered the locations or quantities of firewood

Hum Ecol (2011) 39:465–478

gathered. The majority of survey respondents prefers to collect already dead wood and has always made a habit of harvesting it. Those who always cut live wood for fuel have continued that practice, as well. To hear most residents speak, though, state conservation policies have made acquiring energy much more arduous. Time and again informants asserted that only after the Natural Resources Institute came on the scene were they then forced to collect dead wood: “We hardly use natural firewood because it’s prohibited;” “they only let us bring dry wood, from far away;” “they watch, we have to look for charamosca (small twigs), they don’t let us use trees.” Other rural residents were more sanguine about fuel availability: “there’s enough wood, you just need to know how to find it.” These contradictory claims appear to be opposing stances about resource scarcity, but are really distinct rhetorical strategies to achieve the same policy goal, namely getting the Natural Resource Institute to relax its no-tree felling policy. The first set of speakers emphasizes the hardship current policy imposes on rural livelihoods; the second downplays the impact wood collection has on forest resources. Despite a strong desire to get forest policies changed, villagers take little overt political action. They are wary of engaging Sanctuary institutions over fuelwood in formal political arenas since this issue is bound up in larger debates over residency and resource use rights. When the issue of fuelwood came up in one community meeting with Sanctuary managers, officials threatened to impose more restrictions, not fewer. Tensions ran high for weeks and Fig. 3 Fallen dead wood of all species and top 10 fuel species in scrub, forest, puna

475

people turned temporarily to buying eucalyptus for fear they would be harassed on wood collection trips. The most common form of firewood politics is when villagers engage in performances of compliance with park rangers. Such political performance is found in state-local wood conflicts elsewhere (Haenn 2005; Mathews 2008). Villagers elide compliance by felling a tree and leaving it in the forest to dry so a few weeks later they can confidently walk past the ranger station with their “dead” wood. They also carry newly cut wood along irrigation canals instead of established trails to avoid detection. Going “up” to collect fuel is where the best wood is; it is also farther away from the watchful eye of park rangers. For their part, rangers tend not to stray far from assigned patrol routes into the depths of the monte. They also do not look too closely at the piles of “dead” wood people haul past their front door. Villager fuelscapes have become more covert as a result of their embroilment in broader resource use conflicts.

Conclusion The mundane practice of picking up sticks is central to Andean livelihood strategies. Furthermore, fuelscapes are an important way through which people represent and produce the landscape around them. This article advances understanding of Andean landscapes by illuminating the complex and layered dimensions of landscape production. Fuelscapes are informed by biophysical parameters, but are

476

not simply an adaptive response to vertically differentiated environmental conditions. Instead, they demonstrate people’s ingenuity in modifying and taking advantage of varied ecosystems that often contain limited woody resources. Fuelscapes reflect legacies from the past in the form of peasant community boundaries, historic settlement patterns, and woody biomass production. They are not, however, remnants of pre-Columbian lo andino ways of interacting with Andean environments. Rather, they are a co-production of livelihood strategies and modern political economic policies from agrarian reform to protected areas. Daily decisions about where to collect fuel go beyond simple reflections of wood scarcity and abundance. While broad socioeconomic drivers such as population and poverty do influence energy behavior, fuel collection patterns are primarily informed by gendered and intergenerational patterns of labor. Understanding these subtle drivers of land use and emic conceptions of landscape complements existing approaches to land use land cover change research. Fuelscapes serve alongside production zones as material and symbolic human interactions with Andean landscapes. They help us move beyond a focus on verticality and agriculturally-derived land use categories to provide new insights into the semiotic importance and anthropogenic production of uncultivated ecosystems. Going to the monte means traveling to where trees grow by themselves and traveling through landscapes of memory and meaning. The geographic form of monte fuelscapes depends on community boundaries that establish property rights to wood and, more informally, on individual family histories of land use. Energy connections to landscape are a new way of approaching fuelwood, which to date has been mired in policy research that frames it as a crisis of mismatched supply and demand. Wisner (1989) and Cline-Cole (1998) utilize the concept of fuelscape to show how wood scarcity and abundance are both material realities exacerbated by socioeconomic inequities and discursively manufactured by a myriad of institutions for their own ends. By only focusing on discourse, though, we are left without an idea of local fuel realities. This article advances how fuelscape can be used as a lens into fuelwood politics and practice by integrating its discursive, material, and biophysical dimensions. It also moves beyond preoccupations with wood scarcity and abundance. I show how collection patterns do not reflect ever-increasing circles of wood denudation around human settlements. Wood availability does differ among cropland, scrub, forest, and puna ecosystems, but household labor availability plays a greater role in determining where people go to acquire energy than ecological characteristics alone. Sanctuary conservation policies restrict any and all tree felling in the protected area, a strategy based on the privileging of live trees over

Hum Ecol (2011) 39:465–478

ecological function. The immediate impact of this policy on firewood practices is ambiguous, particularly since many villagers prefer collecting already dead wood. Villagers are leery of overt political confrontation with Sanctuary institutions. They rely on compliance—and the appearanceof-compliance—as a means of contesting official fuelscapes. The policy does have the potential to harm local livelihoods as well as ecological processes in the long run. This article points toward new directions that researchers and policy-makers need to undertake to better comprehend the complex nature of fuelwood. First, we need to cast off preoccupations with woodfuel crises, scarcity, and abundance. This narrow view fails to capture how people such as Andean residents have long existed in areas with limited wood availability and responded to shifting political economic contexts. Policy solutions that seek to increase wood supply and reduce demand perpetuate the notion that simple material changes can solve complex environmental problems. Human ecological studies of the diverse forces that shape energy decisions are necessary. Inattention to how official fuelscapes diverge from local realities can exacerbate rather than mitigate environmental and social problems. We also need to recognize that fuelwood is important for more than its energetic properties alone. Fuelscapes are a means by which over 2.4 billion people in the world interact with their surroundings. Tracing pathways from hearth to fuel source illuminates important dimensions of how Andean rural residents produce uncultivated ecosystems as places with distinctive symbolic significance, anthropogenic drivers of land use, and historic sedimentation of landscape. Acknowledgements Research for this article was funded by the National Science Foundation, Yale Center for International and Area Studies, Program in Agrarian Studies, and Tropical Resources Institute. I would like to thank the many institutions in Peru that supported my fieldwork, in particular the Machu Picchu Program, Natural Resources Institute, National Cultural Institute, and Machu Picchu Management Authority. I am particularly indebted to the Rayanpateños with whom I worked, especially the families who participated in household studies. I am extremely grateful for the helpful comments and suggestions given by Michael Billig, Enrique Mayer, and three anonymous reviewers on previous drafts of this article. Franklin and Marshall students Karina Bongaarts and Jessica Schwartz spent many hours organizing data. Mike Rahnis aided with figure-making and general support. I hope that I have accurately represented fuel practices in Rayanpata; any errors are mine.

References Agarwal, B. (1986). Cold Hearths and Barren Slopes: the Woodfuel Crisis in the Third World. Zed Books, London. Allen, C. J. (1988). The Hold Life Has: Coca and Cultural Identity in an Andean Community. Smithsonian Institutional Press, USA. Ansión, J. (1986). El Arbol y el Bosque en la Sociedad Andina. Ministerio de Agricultura, FAO, Lima.

Hum Ecol (2011) 39:465–478 Arnold, J. E., Kohlin, G., and Persson, R. (2006). Woodfuels, Livelihoods, and Policy Interventions: Changing Perspectives. World Development 34: 596–611. Bentley, J., and Valencia, J. (2003). Learning about trees in a Quechua-speaking Andean community in Bolivia. In Van Mele, P. (ed.), Way Out of the Woods: Learning How to Manage Trees and Forests. CPL Press, Newbury, pp. 69–134. Broadhead, J., Bahdon, J., and Whiteman, A. (2001). Past Trends and Future Prospects for the Utilization of Wood for Energy. FAO, Rome. Brush, S. B. (1976). Man's use of an Andean ecosystem. Human Ecology 4: 147–166. Brush, S. B. (1977). Mountain, Field, and Family: the Economy and Human Ecology of an Andean Valley. U. Pennsylvania Press, Philadelphia. Byers, A. C. (2000). Contemporary Landscape Change in the Huascaran National Park and Buffer Zone, Cordillera Blanca, Peru. Mountain Research and Development 20: 52–63. Chepstow-Lusty, A., and Winfield, M. (2000). Inca Agroforestry: Lessons from the Past. Ambio 29: 322–328. Cline-Cole, R. (1998). Knowledge Claims and Landscape: Alternative Views of the Fuelwood-Degradation Nexus in Northern Nigeria. Environment and Planning D-Society & Space 16: 311–346. de Montelambert, M. R., and Clement, J. (1983). Fuelwood Supplies in Developing Countries. FAO, Rome. Dewees, P. A. (1989). The Woodfuel Crisis Reconsidered - Observations on the Dynamics of Abundance and Scarcity. World Development 17: 1159–1172. Dollfus, O. (1982). Development of Land-Use Patterns in the Central Andes. Mountain Research and Development 2: 39–48. Eckholm, E. (1975). The Other Energy Crisis, Firewood. Worldwatch Institute, Washington, DC. Fioravanti-Molinie, A. (1982). Multi-levelled Andean society and market exchange: the case of Yucay (Peru). In Lehmann, D. (ed.), Ecology and Exchange in the Andes. Cambridge University Press, Cambridge, pp. 211–230. Flores Ochoa, J. L. (1985). Interaction and complementarity in three zones of Cuzco. In Masuda, S., Shimada, J., and Morris, C. (eds.), Andean Ecology and Civilization: An Interdisciplinary Perspective on Andean Ecological Complementarity. University of Tokyo Press, Tokyo, pp. 251–276. Fortmann, L. (1985). The Tree Tenure Factor in Agroforestry with Particular Reference to Africa. Agroforestry Systems 2: 229–251. Franquemont, C. R. (1988). Chinchero Plant Categories: an Andean Logic of Observation. Cornell University. Gade, D. (1967). Plant Use and Folk Agriculture in the Vilcanota Valley of Peru: a Cultural-Historical Geography of Plant Resources. University of Wisconsin. Gelles, P. (1999). Water and Power in Highland Peru. Rutgers University Press, New Brunswick. Godoy, R. A. (1984). Ecological Degradation and Agricultural Intensification in the Andean Highlands. Human Ecology 12: 359–383. Gross, D. R. (1984). Time allocation: a tool for the study of cultural behavior. Annual Review of Anthropology 13: 519–558. Guillet, D. W. (1981). Land Tenure, Agricultural Regime, and Ecological Zone in the Central Andes. American Ethnologist 8: 139–158. Haenn, N. (2005). Fields of Power, Forests of Discontent: Culture, Conservation, and the State in Mexico. University of Arizona Press, Tucson. Harris, O. (1985). Ecological duality and the role of the center: Northern Potosí. In Masuda, S., Shimada, I., and Morris, C. (eds.), Andean Ecology and Civilization: an Interdisciplinary Perspective on Andean Ecological Complementarity. University of Tokyo Press, Tokyo, pp. 311–336.

477 Hastorf, C. A., Whitehead, W. T., and Johannessen, S. (2005). Late Prehistoric Wood use in an Andean Valley. Economic Botany 59: 337–355. Holdridge, L. R. (1967). Life Zone Ecology. Tropical Science Center, San Jose. Costa Rica. INRENA. (1998). Plan Maestro del Manejo del Sanctuario Histórico de Machu Picchu. INRENA, INC. International Energy Agency (2002). World Energy Outlook 2002. OECD, Paris. Isbell, B. J. (1977). To Defend Ourselves: Ecology and Ritual in an Andean Village. University of Texas at Austin, Austin. Jameson, J. S., and Ramsay, P. M. (2007). Changes in High-Altitude Polylepis Forest Cover and Quality in the Cordillera de Vilcanota, Perú, 1956–2005. Biological Conservation 138: 38–46. Johannessen, S., and Hastorf, C. A. (1990). A History of Fuel Management (A.D. 500 to the Present) in the Mantaro Valley, Peru. Journal of Ethnobiology 10: 61–90. Johnson, A. (1975). Time allocation in a Machiguenga Community. Ethnology 14: 310–321. Jokisch, B. D., and Lair, B. M. (2003). One Last Stand? Forests and Change on Ecuador’s Eastern Cordillera. Geographical Review 92: 235–256. Kintz, D. B., Young, K. R., and Crews-Meyer, K. A. (2006). Implications of Land Use/Land Cover Change in the Buffer Zone of a National Park in the Tropical Andes. Environmental Management 38: 238–252. Larme, A. (1993). Work, Reproduction and Health in two Andean Communities (Department of Puno, Peru). Production, Storage, and Exchange (PSE) in a Terraced Environment on the Eastern Andean Escarpment. Retrieved from. Leatherman, T. L. (1996). A biocultural Perspective on Health and Household Economy in Southern Peru. Medical Anthropology Quarterly 10: 476–495. Mathews, A. S. (2008). State Making, Knowledge, and Ignorance: Translation and Concealment in Mexican Forestry Institutions. American Anthropologist 110: 484–494. Maxwell, K. (2011). Hearth and Household Economy in an Andean Rural Village. Human Organization 70: Forthcoming. Mayer, E. (2002). The Articulated Peasant: Household Economies in the Andes. Westview Press, Boulder. Mitchell, W. (1991). Peasants on the Edge: Crop, Culture, and Crisis in the Andes. University of Texas Press, Austin. Murra, J. V. (2002 (1972)). El ‘control vertical’ de un máximo de pisos ecológicos en la economía de las sociedades andinas. In El Mundo Andino: Población, Medio Ambiente y Economía. Instituto de Estudios Peruanos, Lima, pp. 85–126. Orlove, B. S. (1977). Integration Through Production: the Use of Zonation in Espinar. American Ethnologist 4: 84–101. Orlove, B. S., and Godoy, R. (1986). Sectoral Fallow Systems in the Central Andes. Journal of Ethnobiology 6: 169–204. Paulson, S. (2003). Gendered Practices and Landscapes in the Andes: the Shape of Asymmetrical Exchanges. Human Organization 62: 242–254. Platt, T. (1982). The role of the Andean ayllu in the reproduction of petty commodity regime in Northern Potosí (Bolivia). In Lehmann, D. (ed.), Ecology and Exchange in the Andes. Cambridge University Press, Cambridge, pp. 27–69. Postigo, J., Young, K. R., and Crews, K. (2008). Change and Continuity in a Pastoralist Community in the High Peruvian Andes. Human Ecology 36: 535–551. Schmida, A. (1984). Whittaker’s Plant Diversity Sampling Method. Israel Journal of Botany 33: 41–46. Sherbondy, J. (1986). Mallki: Ancestros y Cultivo de Arboles en los Andes. Documento de Trabajo #5, Proyecto FAO/ Holanda/INFOR. Skar, S. L., N. Samanex, and S. Cotarma. (1982). Fuel Availability, Nutrition and Women’s Work in Highland Peru. ILO. Retrieved

478 from. Stern, M. J. (1995). An inter-Andean Forest Relict - Vegetation Change on Pasochoa Volcano, Ecuador. Mountain Research and Development 15: 339–348. Stohlgren, T. J., Falkner, M. B., and Schell, L. D. (1995). A ModifiedWhittaker Nested Vegetation Sampling Method. Plant Ecology 117: 113–121. Tosi, J. A. (1960). Zonas de Vida Natural en el Perú. Instituto Interamericano de Ciencias Agricolas de la OEA, Peru. Valencia, N. (1992). Western slope cloud forests. In Young, K. R., and Valencia, N. (eds.), Biogeografía, Ecología y Conservación del Bosque Montano en el Perú. UNMSM, Lima, pp. 155–170. Van Buren, M. (1996). Rethinking the Vertical Archipelago. American Anthropology 98: 338–351. Webster, S. (1971). An Indigenous Quechua Community in Exploitation of Multiple Ecological Zones. Revista del Museo Nacional 37: 174–183. Weismantel, M. J. (1988). Food, Gender, and Poverty in the Ecuadorian Andes. Waveland Press, Prospect Heights. West, T. L. (1987). The burning bush: exploitation of native shrubs for fuel in Bolivia. In Browman, D. L. (ed.), Arid Land Use Strategies and Risk Management in the Andes. Westview Press, Boulder, pp. 151–167. Wiegers, E. S., Hijmans, R. J., Herve, D., and Fresco, L. O. (1999). Land use intensification and disintensification in the Upper Cañete Valley, Peru. Human Ecology 27: 319–339.

Hum Ecol (2011) 39:465–478 Winterhalder, B., Larsen, R., and Thomas, R. B. (1974). Dung as an Essential Resource in a Highland Community. Human Ecology 2: 89–104. Wisner, B. (1989). Power and Need in Africa. Africa World Press, Trenton. Wisner, B., Gilgen, H., Antille, N., Sulzer, P., and Steiner, D. (1987). A matrix-flow approach to rural domestic energy: a Kenyan case study. In Cocklin, C., Smit, B., and Johnston, T. (eds.), Demands on Rural Lands: Planning for Resource Use. Westview Press, Boulder, pp. 211–238. Young, K. R. (1993). Tropical Timberlines—Changes in Forest Structure and Regeneration Between two Peruvian Timberline Margins. Arctic and Alpine Research 25: 167–174. Young, K. R. (2009). Andean Land use and Biodiversity: Humanized Landscapes in a Time of Change. Annals of the Missouri Botanical Garden 96: 492–507. Young, K. R., and Keating, P. L. (2001). Remnant forests of Volcan Cotacachi, Northern Ecuador. Arctic Antarctic and Alpine Research 33: 165–172. Zimmerer, K. S. (1996). Changing Fortunes: Biodiversity and Peasant Livelihood in the Peruvian Andes. University of California Press, Berkeley. Zimmerer, K. S. (1999). Overlapping Patchworks of Mountain Agriculture in Peru and Bolivia: Toward a Regional-Global Landscape Model. Human Ecology 27: 135–165. Zimmerer, K. S. (2010). Woodlands and Agrobiodiversity in Irrigation Landscapes Amidst Global Change: Bolivia, 1990–2002. The Professional Geographer 62: 335–356.