(and ulluco)? - Springer Link

Agriculture and Human Values 20: 107–123, 2003. © 2003 Kluwer Academic Publishers. Printed in the Netherlands.

Just small potatoes (and ulluco)? The use of seed-size variation in “native commercialized” agriculture and agrobiodiversity conservation among Peruvian farmers Karl S. Zimmerer Department of Geography, University of Wisconsin – Madison, Madison, Wisconsin, USA

Accepted in revised form November 25, 2002

Abstract. Farmers of the Peruvian Andes make use of seed-size variation as a source of flexibility in the production of “native commercial” farmer varieties of Andean potatoes and ulluco. In a case study of eastern Cuzco, the use of varied sizes of seed tubers is found to underpin versatile farm strategies suited to partial commercialization (combined with on-farm consumption and the next season’s seed). Use of seed-size variation also provides adaptation to diverse soil-moisture environments. The importance and widespread use of seed-size variation among farmers is demonstrated in the emphasis and consistency of linguistic expressions about this trait. Small and small-medium seed is typically sown in the community’s “Hill” unit of sub-humid, upper-elevation agriculture. Seed tubers of medium-size and larger are needed for drought-stressed locales in lower-elevation landscape units. Farm-level preferences for the seed-size of tubers also suggest potential relations to resource endowments of farm households and gender-related management, although these tendencies were not statistically significant in the study. An intra-varietal, landscape-environmental perspective on seed-size management, which includes an emphasis on within-field versatility, helps to strengthen the research support of local seed production in policies and programs aiming for in situ agrobiodiversity conservation, marketing capacity, and food security. Key words: Andean potatoes, Drought adaptation, Farmer selection, Food security, in situ agrobiodiversity conservation, Market-based conservation, Participatory Crop Improvement, Peruvian agriculture, Ulluco, Seed traits Karl Zimmerer is Professor in the Department of Geography, the Institute for Environmental Studies, the Program in Development Studies, and the Land Tenure Center at the University of Wisconsin–Madison. He is Co-Director of the Environment-and-Development Advanced Research Circle (EDARC). He received his MS and PhD in geography at the University of California at Berkeley. His current research includes projects on the relations of agrobiodiversity to irrigation and development change among Quechua and non-Quechua farmers in Bolivia and Peru and seed networks of maize and potato growers in Mexico and the Andean countries. He is also undertaking a study of the idea of eco-regional planning and the role of agriculture. Introduction A large number of farmers in the Andean highlands of Peru utilize a style of farming that relies on the partly commercial production of diverse farmer varieties (FVs) of the mainstay food plants of potatoes and ulluco.1 The persistence and renewal of this form of low-input agriculture is attributed, most directly, to the worsening markets for producers of the staple foodstuffs of the Peruvian Andes (Mayer, 2002; Mayer and Glave, 1999). The real price of potatoes, for example, has declined by more than 200% since the late 1980s. Market terms for the growers of ulluco and other minor Andean tuber crops were worsened also, although less steeply (Bianco and Sachs, 1998). One downward pressure on the price levels of these staple foodstuffs is

due to rebounding supply in the wake of nearly disappeared threats of Shining Path and civil war violence (Bentley et al., 2001: 325). The substantial increase of input costs also contributes to general pressures for the persistence and renewal of low-input agriculture in Peru as well as in other developing countries (Almekinders and Louwaars, 1999: 24; Nellithanam et al., 1998). Unfavorable markets of both their farm products and inputs are reinforcing a field-level production strategy among many potato and ulluco growers in Peru’s highlands who cultivate the diverse FVs that are referred to as “native commercialized” (nativa comercializada) (Brush, 1992; Fano, 1999; Fano and Benavides, 1992; Mayer and Glave, 1999). Cultivation of the “native commercialized” category is dependent

108

K ARL S. Z IMMERER

on a moderate number of varietal types (10–15) within an agricultural region (CIP and PNUMA, 2000). Since “native commercialized” types are preferred locally for their yield and reliability, and since they are biologically diverse, this cultivation is an important form of agriculture for Andean farmers and for the maintenance of Andean food plants. Assessing the production of “native commercialized” varieties is relevant to both in situ agrobiodiversity conservation and food security, two main goals of sustainability policies in agricultural development (Brookfield, 2001; Cleveland et al., 1994; Mayer, 1994; Rhoades, 2001; Thrupp, 1998; Tripp, 1996). My research is focused on the management of seedsize in FVs of potatoes and ulluco under the “native commercialized” category of agricultural production in eastern Cuzco, Peru. This article examines four facets of seed-size management. First, it investigates the farmers’ preference for size characteristics of the seed tubers of potato and ulluco types in the “native commercialized” category. Second, how are the preferences for seed-size verbally expressed in everyday language that communicates local value and use. The third facet is the incorporation of seedsize preference into overall household-level agricultural strategies, gender differences among household heads, and the variation of growing environments in the local farm landscape. Fourth is an assessment of the possible relation between seed-size management and the resource-level of farm households. Recent advances in knowledge of each of the above facets of farmers’ selection and management of seed (with special reference to seed-size traits) are used to guide the design and analysis of the case study (see Literature Review in the next section). The general importance of “native commercialized” potato and ulluco varieties stems from commonness, local value, and biological diversity. “Native commercialized” production accounts for 15–40% of agricultural plantings in the central and southern portions of the Andean highlands of Peru, and it thus offers special promise to market-based approaches to in situ agrobiodiversity conservation (Brush, 1992, 1995; Fano, 1999). Also, potatoes and ulluco that belong to the “native commercialized” category are well adapted to a variety of growing sites, since they are grown in fields that cover a moderate-tolarge range of agroenvironments, partly as a result of seed networks (Zimmerer, 1998, 2002a, 2003). These diverse varieties are recognized as staple foodstuffs that are widely used in local dishes. “Native commercialized” types, combined with other potato and ulluco varieties, are a staple of approximately 8–10 million small farmers in the Andean countries, many of them peasant and indigenous farmers (Horton,

1987; Rhoades and Nazarea, 1999). Biodiversity of the “native commercialized” types is inclusive of a high level of intravarietal genetic variation (Zimmerer and Douches, 1991). They contribute to the total numbers of FVs that surpass 2,000 potato types (CIP, 1997) and 100 ulluco variants in the Andean countries (Rousi et al., 1986; Rousi et al., 1988). Viability of the field cropping of diverse Andean potato and ulluco varieties is central to current in situ agrobiodiversity conservation (CIP and PNUMA, 2000; Iriarte et al., 1998; NRC, 1989). This article’s research on seed-size management is also intended to aid the support of potato and ulluco agriculture under “native commercialized” production through the further development of local seed production and participatory crop improvement/plant breeding (PCI/PPB). This pair of approaches prioritizes the involvement of farmers in the goals of agrobiodiversity conservation, food security, and marketbased development in a growing number of projects, programs, and policies (Eyzaguirre and Iwanaga, 1996; Friis-Hansen and Sthapit, 2000; McGuire et al., 1999; Witcombe et al., 1998). One joint focus for these two approaches is addressing the range of farmers’ knowledge and preferences in their selection and management of diverse FVs, such as those of the Andean crops (Ashby and Sperling, 1995; Bentley, 1994; Cleveland and Soleri, 2002; Rhoades, 2001; Rhoades and Nazarea, 1999; Thiele et al., 2001). In addition, understanding farmers’ intra-varietal selection of these diverse crop types is crucially important to the involvement of social and agroecological science in local seed production and participatory plant breeding (Soleri et al., 2000; Zimmerer, 2002a). My research is designed to provide an integrated emphasis on intra-varietal seed management, farm practice at the within-field level, and a landscape-environmental perspective. Broadly its goal is to contribute a dimension of farmer knowledge and agroecology that can strengthen the analytical frameworks for local seed production and participatory plant breeding.

Rethinking seed size, agroecology, and Andean farm management The analysis and support of small-farm seed management is vital to the cultivation of FVs of both potato and ulluco. Failure of seed supply is increasingly regarded as a chief limit on viability of diverse food plants in the southern Peruvian Andes and other Andean regions (Bianco and Sachs, 1998; Dueñas et al., 1992; Fano, 1999; Fano and Benavides, 1992; Zimmerer, 2003). Recent findings, which are discussed in this section, demonstrate the importance

J UST SMALL POTATOES ( AND ULLUCO )?

of seed size traits as a principal focus of participatory approaches to crop improvement (PPB/PCI) and small-farm management. Use of seed size must be considered in relation to agroecological, cultural, and economic conditions within Andean farm communities. Size-related preference is an important feature of seed management in the potato and ulluco crops. Agronomic research has determined that best-use seed tubers of the potato crop are intermediate in size and at a balance point between large and small extremes (Caligari, 1992; Harris, 1992; Horton, 1987; on general seed-size specifications, see Almekinders and Louwaars, 1999: 88). Larger seeds store more photosynthetic reserves that are available to sprouting plants. Diminishing benefits are reached at a certain size where the seed’s increased storage confers nominal added growth potential to new plants (Harris, 1992). Studies of Andean agriculture suggest that existing preferences for tuber-size may, however, be significantly more varied than conventional agronomic recommendations. “Large tuber size” is identified as a chief criterion of the selection of potato seeds in the Bolivian Andes (Watson et al., 1992, in Thiele et al., 1997). By contrast, small-size seed is used for Andean potatoes in at least certain parts of the Peruvian Andes (Ortega, 1997) and those of other Bolivian locales as well (Bentley and Vasques, 1998). These references to importance of seed-size do not, however, address the use of seed-size variation. Nor do the existing studies address the role of the use of seed-size variation that is related to farm management styles (such as “native commercialized” production) or in terms of landscape-environmental and broadly social factors (including cultural and economic dimensions). Nor do the current works on seed-size treat the management of ulluco, which as a regionally important tuber crop merits comparison to the primary staple of potatoes. Linguistic expressions are revealing indicators of the cultural and social dimension of agricultural resource management (Nazarea, 1998; Richards, 1996). Given the large lexicon of Andean farming (Beyersdorff, 1984; Cusihuamán, 2001; Mannheim, 1991), everyday speech is hypothesized to treat seed-size management in detail. The application of commonplace language and terminology is typically useful for understanding the rationales that guide farmers in routine activities (such as seed-size use) (Alcorn, 1989; Richards, 1996). Key phrases and terms are components of the “agricultural scripts” that add to the unspoken elements of farm and resource practice (Alcorn, 1989: 65). In the case of seed-size management, the detailed description of FV management is expected to draw on distinguishing features at

109

the intra- or sub-varietal level (Berlin, 1992). (Seedsize is unlikely to serve as the sole referent, for Andean farmers may also typically make intra-varietal distinctions based on other tuber qualities such as health and disease status, shape, color; Thiele et al., 1997, Zimmerer, 1996.) The farmers’ verbal expressions offers a view of their seed management that can be combined with the analysis of seed traits and agroecological relations in order to show a dual perspective of both science (agronomic and environmental) and social constructivism (Cleveland, 2001; Cleveland and Soleri, 2002; Soleri et al., 2000). Finally, the content and frequency of everyday linguistic expressions can reveal the characteristics of cultural significance, social commonness, and general economic importance (Croll and Parkin, 1992). Seed-size use is situated within the overall economic strategy of the agricultural household and may include those factors associated with combined market and subsistence objectives, gender relations, and access to varied growing environments (Butler Flora et al., 2001). In the case of economic rationales, the farm household must manage its seed supplies as part of overall agricultural strategy (Brookfield and Padoch, 1994; Mayer, 2002; Netting, 1993). Seed management may be combined with market and consumption objectives in the same field (Brush, 1992), or these goals may be segregated spatially in the landscape, as illustrated in the prevalence of fieldor chacra-based management within Andean agriculture (Gonzáles, 2000). Gender relations are central to seed management in the Peruvian Andes, since men and women undertake divisions of labor that differ with respect to farming, food processing, marketing, and seed selection (Fernandez et al., 2000; Hamilton, 1998; Iriarte et al., 2000; Paulson, 1998; Tapia and de la Torre, 2000; Weismantel, 1988). It is an important empirical question whether these gender-related differences of farming act as factors that exert an influence on seed-size management. The farm household’s access to particular growing environments is also a potential influence, since agroecological conditions could influence the choice of certain seed-size traits. Evaluation of seed-size adaptations from an evolutionary-ecological perspective has shown that large seed-size is widespread as an adaptation to environments that are subject to the regular stress of moisture deficits (Blumler, 1998; Damania et al., 1996; Soleri et al., 2000; Westoby, 1979/1980). Irregular water supply, by contrast, leads to a large number of smaller-size seeds as the most common adaptation. The adaptive dynamics of seed-size agroecology are not considered, thus far, in the case of tuber-bearing food plants (to-date, the focus is on cereals, such as wheat, as well as uncultivated seed-

110

K ARL S. Z IMMERER

Figure 1. Study areas of eastern Cuzco (with seed-size preferences; inset [map A] of Mollomarca community shows landscape-environmental units).

bearing species). Peruvian farmers’ management of seed-size in the Andean potato and ulluco crops might include similarities to one, both, or neither of the above evolutionary-ecological strategies. Soil moisture is a logical focus, since it frequently functions as a key agroecological factor for small-scale farmers, and it is commonly used among Andean farmers as a gauge of farmland suitability (Altieri, 1995; Brookfield and Stocking, 1999; Carney, 1991; Gliessman, 2001; Mayer, 2002; Winterhalder, 1994). The possible relation between farmers’ management of seed size and household resource-level is supported through a number of studies on Andean agriculture. For example, farmers that are better-off by local standards are more likely to cultivate diverse FVs of mainstay foodstuffs (Brush et al., 1992; Bianco

and Sachs, 1998; Zimmerer, 1996). Hypothetically, resource-level differences may relate to seed-size preferences (better-off farmers can afford larger seed that is more potentially productive).

Study methods and materials The design of this research is a case study among farmers in the region of eastern Cuzco that belongs to Paucartambo Province (locations in Figure 1). The region’s farmers are bilingual Quechua–Spanish speakers who identify themselves as both Quechua Indians and campesinos (“peasants”). The case study involves research in three communities, Mollomarca (District Paucartambo), Colquepata (District Colque-

111

J UST SMALL POTATOES ( AND ULLUCO )? Table 1. Potato and ulluco farmer varieties and tuber samples in interview. Local name

Common name

Scientific name

Total no. tubers

Range of weights (g)

Range of shapes

Range of colors

Qompis

Andean potato

Solanum tuberosum subsp. andigena

15

15–130

Nearly spherical (varied flattening at ends)

Light yellow-red (varied pattern)

Mariva

Andean potato

Solanum tuberosum subsp. andigenum

15

13–103

Oblong, flattened slightly, varied tapering at end

Dark red-black

Qompis lisas

Ulluco

Ullucus tuberosus

15

8–82

Nearly spherical– oblong

Solid orange-yellow, some with red speckles

pata), and Majopata (District Challabamba) (Figure 1). Each community counts upwards of 100 households as members. Together the communities were chosen to represent a range of environmental and social differences that might influence seed-size management. The community of Colquepata contains a larger area of semi-arid environments and includes a higher degree of integration into potato and ulluco markets (Fano and Benavides, 1992). Majopata and the Challabamba communities encompass a greater area of sub-humid climates and specialization in production for offseason potato and ulluco markets (Brush, 1992), while Mollomarca shows intermediate characteristics. The field study was conducted in 1993 and 1998 using a structured interview of seed-size preferences and farm management, a semi-structured portion of the interview that inquired about the rationales for seedsize preference, ethnographic participant-observation, and the environmental analysis of soil moisture in a range of growing environments. The seed-size interviews were conducted with 33 small farmers who are active growers of potatoes and ulluco. In the structured portion, farmers were asked to inspect a representative sample of tubers of three common FVs in the “native commercialized” category, two of potatoes (qompis, mariva) and one of ulluco (qompis lisas) (Table 1).2 Farmers were asked to identify the seed(s) they would typically choose for sowing.3 To address their weighing of size criteria relative to other factors, farmers were asked to explain their seed tuber preferences in overall terms (including such factors as tuber shape, color, and tuber health or disease status). Tubers in the sample were not considered in the analysis of seed-size preferences if they were commonly chosen (either for or against) on the basis of non-size factors (according to one half or more of persons interviewed). In the semi-structured portion of the field inter-

view, we asked an open-ended question about the farmer’s general rationales for his or her preferences. A local field assistant aided with this stage of the interview, which was conducted in the QuechuaSpanish mix that is typically spoken in eastern Cuzco. The farmers were encouraged to respond in everyday language or the style of speech used when conversing with “people they know” (conocidos) about agriculture. Their responses were transcribed, to the extent possible, at the time of the interview. Ethnographic participant-observation was focused on seed activities in a series of three micro-locales where farmers undertake their main decisions about the management of seed size in Andean potatoes and ulluco. These sites were (i) the field at harvest; (ii) the storeroom when seed was deposited; and (iii) the storeroom at the time of removing seed for planting. Two social attributes were accounted for in the design of the interview sample: gender and household resource-level. The sample of farmers participating in the interviews included 16 men and 17 women.4 Each person was asked who in their household typically selects seed, whether they work alone in choosing seed, and whether size-related decisions differ among spouses. Household resource-level was estimated through a set of three techniques. First quantitative information was elicited in the interview on the number, type, and area of agricultural fields, the number and type of livestock herds, and the number and ages of household laborers. (These resource-level indicators are used widely in studies of socioeconomic assets and agroenvironmental management among Andean farmers; see Bianco and Sachs, 1998; Deere, 1990; Ellis-Jones and Moore, 1999; Mayer and Glave, 1999; Zimmerer, 1996.) Second, the person interviewed was asked to describe his or her own assets as “poor,” “medium,” or “rich,” local terms

112

K ARL S. Z IMMERER

that are widely used as designations of relative wealth. Finally, a key informant, typically a community official, was asked to provide a relative wealth-ranking of the persons interviewed in each study area.5 Agroecological methods involved the use of landscape-environmental mapping and soil moisture measurement. The former involved visits to field sites and transect walks in order to gain familiarity with the coverage of four main landscape-environmental units that farmers designate as Hill, Ox Area, Early Planting, and Valley. Farmers’ designation of these within-community units, which are also readily recognized among researchers and development agents, are utilized as a means of spatially organizing their farm resource management (Brush, 1992; Mayer, 2002; Zimmerer, 2002a, b, 2003); these farm-management and agroecological distinctions would be expected to influence seed-size use. To focus this analysis, the community of Mollomarca was chosen for detailed sampling of soil moisture. Sampling of soil moisture measurements was designed to estimate the main spatial and temporal variations of potential importance to “native commercialized” agriculture. A ceramic-tip soil moisture probe (tensiometer) was used to record soil suction at 15 cm depth, a standard technique for the measurement of soil moisture potential (Hubbell and Sisson, 1998). Measurements of soil moisture, in centibars, were taken at the nearest field site at each 100-meter interval (3100–4000 m.a.s. l). Readings were recorded at 3 meters from the corners of each field, and values were averaged. Soil moisture samples, taken at four times during the year (January 15, April 15, July 15, October 15), were then used to calculate annual mean values and variability. Fields tested for soil moisture also were identified on the basis of landscape-environmental unit (Hill, Ox Area, Early Planting, Valley).

Findings: Management of tuber size in potato and ulluco seed Most farmers in the communities of eastern Cuzco produce the “native commercialized” category of field containing the Andean potatoes. This field type, sown with a single FV such as mariva or qompis, is combined in the household repertoire with other field styles that utilize either modern varieties (MVs) or varietal polycultures.6 “Native commercialized” fields are located primarily in the farm unit known locally as the Early Planting (irrigated) and Ox Area (rainfed, mid-elevation) (Figure 1). Also “native commercialized” fields occur in smaller numbers in the rainfed Hill area at upper-elevations (Figure 1). Ulluco is produced on approximately two-thirds of the farms

of eastern Cuzco. Ulluco field production is located primarily in the landscape units known as Hill and Ox Area. Ulluco growing is based entirely on FVs such as qompis lisas. Approximately one-half of ulluco producers rely on field production that is at least partly marketed, which thus closely resembles the “native commercialized” category of Andean potatoes. Results of the structured interview, given in Table 2, show the preferences for those tubers that were regularly chosen for or against on the basis of seed-size (see also Figure 1 for inter-community comparisons and the geographical distribution of seedsize preferences).7 For the qompis variety, farmers’ most preferred choice weighs 28 grams (25.0% of interviews), their second most common preference was a tuber weighing 52 grams (22.2% of interviews), while the third weighed 15 grams (13.9% of interviews). The preferred choice (28 grams) was typically sown in a group of two similar-size tubers per planting hole, a practice associated with the name iskay muhu (“two-seed”) given to this tubersize. The tubers that ranked second and third most preferred were described, respectively, as “only-seed” (sapan muhu) and “three-seed” (kimsa muhu). Several tubers with weights that are intermediate in this series (e.g., 44 grams) were chosen less frequently, farmers explained, since they fall in between the main seedsize categories.8 For the mariva variety, a tuber weighing 39 grams was the size most commonly chosen for this criterion (35.4% of interviews) (Table 2). The tuber that was preferred second most frequently weighed 76 grams (25.0% of interviews); preferred third was the tuber of 60 grams (22.9% of interviews); and preferred fourth was a tuber of 25 grams (16.7% of interviews). Seedsize categories of these four tubers were described as: “two-seed” (iskay muhu; 39 grams); “only-seed” (sapan muhu; 60 grams, 76 grams); and “three-seed” (kimsa muhu; 25 grams). Farmers’ primary reason for their preference of a larger size of seed tuber in mariva, when compared to qompis, is that the former has fewer eyes per unit surface area. To gain more eyes per tuber in mariva they select a larger seed-size.9 Preference for a larger seed tuber in the mariva variety was also explained as due to agroecological reasons, namely its planting in environments within the farm landscape that are more subject to seasonal drought stress (as discussed further below). For the qompis lisas variety of ulluco, the interviewed farmers tended to prefer a tuber weighing 23 grams (45.2% of interviews) (Table 2). Second most commonly chosen was a 19-gram tuber (21.4% of interviews), while a significant number of other persons preferred the smaller tubers of 8 grams (11.9% of interviews) and 10 grams (19.0% of interviews).

113

J UST SMALL POTATOES ( AND ULLUCO )? Table 2. Seed tubers and size preferences of Andean potatoes and ulluco (n = 33 farmers).a Qompis (potato farmer variety) (total = 36 first choices)

Mariva (potato farmer variety) (total = 48 first choices)

Qompis Lisas (ulluco farmer variety) (total = 45 first choices)

Weight (g)

Frequency Chosen first

Weight (g)

Weight (g)

Frequency Chosen first

15 22 28 33 34 41 52 63 75

13.9% 8.3% 25.0% 11.1% 11.1% 5.6% 22.2% 2.8%

13 20 25 39 60 76 97 100

8 10 19 23 47 50 82

11.9% 19.0% 21.4% 45.2%

Frequency Chosen first

16.7% 35.4% 22.9% 25.0%

2.4%

a Note that several farmers voiced a preference for two tubers as their first choice based on the criterion of size. This tendency led the total number of first choices expressed for each variety (40 in qompis; 48 in mariva; 45 in qompis lisas) to be greater than the size of the interview sample (n = 33). Also note that the analysis in this table includes only those tubers in which the size parameter was referred to in the interviews as the dominant factor (nine tubers of qompis, eight tubers of mariva, and seven tubers of qompis lisas; see text for further discussion of this method).

The latter tubers were described as “three-seed” types (kimsa muhu) and the 23-gram and 19-gram samples as “two-seed” (iskay muhu). Preference for smallsize ulluco seed was described as befitting the lower value of the crop, which typically is managed as a minor planting, usually in the second year of rotation (known locally as kallpa). Compared to their potato crop, ulluco growing is allocated less field preparation, cultivation work, and input endowments (manure, irrigation). In summary, the chief similarity of potato and ulluco seed is that farmers hold a variety of seed-size preferences (“only-seed,” “twoseed,” “three-seed”), with the decrease in seed-size corresponding to the increased number of seed tubers per hole.10 In responses to the open-ended question, farmers regularly made reference to tuber-size specifications and tuber health-status as the most important criteria of seed choice (Table 3). Their responses pointed to the importance of seed-size as a central part of field- and farm-level management strategy in order to maximize the planting area relative to the amount of material that was sown. This “covering efficiency” was referred to, without elicitation, in each of the interviews. The farmer mostly phrased his or her preference for seed samples of the medium- or small-medium size (the seed-size types that were most commonly selected) by describing how these choices make for an efficient use of harvest. Five phrases were commonly used to voice this reasoning: “it walks far”; “it walks high”; “it walks

well”; “it advances well”; “it covers the mountain”; or “enters a large area” (Table 3). The detailed responses of farmers offer insights also into their view of the trade-off between yield and costs that is related to seed-size. Many persons said they would ideally choose a larger-size seed tuber since it would help ensure a number of favorable plant-growth and tuber-yield characteristics (Table 3). The advantage of large-size seed was, however, more than offset by its lower “covering efficiency,” which interviewees often referred to in negative terms (for example, “it walks a short ways”). For example, one farmer said “only-seed’ (a large-size tuber) produces well, though it [the seed supply] covers little of the fields” (sapan muhu allinta rurun, chakra pisiman purin). At the opposite end of the size spectrum, the seed of small tubers was perceived as beset with manageable drawbacks (Table 3). The “threeseed” tubers, for example, were said to amplify the risk of poor yield although this material is widely used when slightly larger seed is unavailable. One farmer expressed this trade-off as “the sowing of small seed produces only small plants” (khullu muhuman tarpusqa khulluta rurun). Farmers’ practice of choosing small-medium and medium-size tubers for seed was confirmed through ethnographic participant-observation. The general separation of tubers (both in potatoes and ulluco) takes place primarily in a storage area although sometimes begins in the field after harvest. Healthy large-size

114

K ARL S. Z IMMERER

Table 3. Examples of common expressions (and frequency mentioned) for seed-size preferences (frequency mentioned = n/33 interviews). “Covering” efficiency: frequency mentioned: 100% Walks far (karuman purin), walks high (aswan hatunman purin or hatunman purin); walks well (allinta purin), advances well (avance bien); it plants a lot (hatunman tarpusqa); covers the mountain (cubre el orqo); enters a large area (hatunman haycun or hatunman tarpusqa); walks a short ways (uchuyman purin, pisilluman purin, or manam karuman purinchu) Size-categories and trade-offs: frequency mentioned: 100% One-seed or “only seed” (sapan muhu); two-seed (iskay muhu); three-seed (kimsa muhu); too small (khullu; unfavorable), for example, “three-seed goes far as seed, although it yields little” (kimsa muhu awsan hatunman purin, ichaqa khullunta rurun) Size-related growth and yield characteristics: frequency mentioned: 73% Has a good plant (allin yura); grows well (allinta wiñan); yields well (allinta rurun hatunchata); produces many tillers (aswanta wachan); produces beautifully (munayta rurun); branches abundantly (rap’a-rap’achamanta); plentiful stems (llusaq tulluyoq); good tuberization (allin ch’ampata wiñan); stands up well, “thick stem” (rakhu tullu) Environmental adaptation: frequency mentioned: 85% Grows well in varied environments, versatile (allinta qachan wiñan); stands up well to sun (resistata allinta); grows well in off-season (mawaypi kalllpachawan wiñan)

Table 4. Access and allocations to varied end-uses of “native commercial” fields per farm household (n = 33 households). Mean field numbers refer to total fields per household. Potato: Mean number fields (total)

Mean number single farmer variety fields

Ulluco: Mean number fields

Allocations of farmer variety potatoes Market-bound Food (kg) (kg)

Seed (kg)

4.2

1.8

0.70

905

583

tubers, locally referred to as the “first class” (primera clase) or “only seed” (sapan muhu) category, are typically set aside for consumption and sale. Tubers for household consumption are called “cost potatoes” (gasto papayku) – “cost” in this case refers to home consumption and its usage in this common term reflects the ethos of Andean farmers who tend to idealize the self-provisioning of subsistence (Gudeman and Rivera, 1990; Mayer, 2002). (As a result, gasto papayku can be translated and interpreted to mean “our potatoes to eat” or “the potatoes we eat at home.”) Tubers slated for market sales are referred to as “money potato” or qolqe papa. Healthy tubers that still remain unselected as this point typically belong to “second class” (segunda clase) and “third class” (tercera clase). They comprise the stored stock from which seed is chosen. (Once becoming seed these designations correspond, respectively, to tubers of the “two-seed” and “three-seed” categories.) Fields of the “native commercialized” category are multi-purpose insofar as this production is typically allocated among the varied end-uses of marketing, food, and seed (Table 4). Marketing is commonly supplied with one part of the harvest of these fields

1118

(approximately one-third on average). Only a few farmers in the interviews designated their “native commercialized” fields for the sole purpose of selling. Instead, the three-way allocation to each of the diverse end-uses is common to the large majority of “native commercialized” fields (mean values of 905 kg to market sales, 1,118 kg to food consumption, and 583 kg to seed; Table 4). To small farmers this custom of cropping for multiple end-uses offers a series of benefits. The most important advantage is the flexibility conferred by combining modest amounts of income (by providing a marketable commodity) with the bolstering of food security (by providing household food) and the reduction of input costs and uncertainty (by self-provisioning seed). Size preferences for seed tubers take place within the context of this three-way allocation, which in effect pits the allocation of planting material against the other high-priority purposes of marketing and food. Preference for medium and small-medium seed tubers helps to maximize the allocation potential of harvest among the three main end-uses. This preference also helps ensure the household’s supply of its own seed, which for the diverse FVs is the predominant source

J UST SMALL POTATOES ( AND ULLUCO )?

of planting material, followed by procurement through farmer-farmer exchange and market purchases. This self-supply of seed tubers of potato and ulluco is held as necessary for an input that farmers believe could otherwise be too expensive or poor quality (Fano and Benavides, 1992; Mayer and Glave, 1999). Costs of potato and ulluco seed are calculated to account for 10–30%, on average, of the direct costs of production in eastern Cuzco. Moreover, the existing markets for seed of the FVs are both unreliable and insufficient. Poor availability of the preferred “two-seed” and “three-seed” categories is typical of markets at nearby villages (e.g., Paucartambo) and the city of Cuzco. Landscape-environmental variation, evident in the parameter of soil moisture, is influential in farmers’ management of seed-size. Farmers’ responses frequently described their preferences as related to the soil-moisture conditions of the intended growing environment. The most common preference was for the use of larger-size seed in planting sites that are prone to regular moisture-stress at the beginning of the growing season (October–November). Farmers treat a range of planting sites as too dry for sowing the small size of seed tubers that belong to the “threeseed” category. These drier growing environments include the inner canyons and the mid-slopes of the main valleys. According to their own landscapeenvironmental perspective, the farmers of eastern Cuzco refer to many of these seasonally drought-prone sites as belonging to the “Ox Area” (yuñlla tarpuy), which is a growing area where ox teams (rather than hand tools) are utilized for tillage and planting (see inset of Figure 1). A number of the farmers interviewed described how “three-seed” tubers would not be suited to the growing environments of the local “Ox Area.” At the same time, these seed tubers of a smaller size are deemed practical for planting at nearby upper elevations (the “Hill” area in farmer’s landscapeenvironmental perspective) and in sites with the most reliable irrigation.11 The correspondence of seed-size preferences to soil moisture is of agroecological significance for a pair of reasons. First, the relation of seed-size to soil moisture is an indication of the importance of intra-varietal tuber management for this commonplace parameter of potato and ulluco agroenvironments. The growing environments of inner canyons, river bottoms, and mid-slope sections of Andean valleys typically receive less rainfall and are subject to higher temperatures and drought stress (Gelles, 1999; Guillet, 1992; Knapp, 1988; Tapia, 1996; Winterhalder, 1994). Extending, intensifying, or relocating the cultivation of potato and ulluco agrobiodiversity to these drought-stressed locales of Andean valleys (referred to as an “internal” frontier of Andean agriculture),

115

Figure 2. Mean annual soil suction by farm-landscape unit.

as might be driven by development change or global warming, would reinforce the need for the adaptive capacity of seed-size variation. Models of global climate change, for example, predict that increased drought and water shortages as well as greater climate variability will beset tropical mountains (including the Peruvian Andes) (Arnell, 1999). Second, the measurements of soil moisture suggest the broad adaptive strategy that is associated with farmers’ management of seed-size variation. Soil moisture measurements of the study area in eastern Cuzco, as shown in Figure 2, indicated that fields of the Hill area were endowed with sufficient soil moisture (annual mean = 41.1 centibars of soil suction; 30–40 centibars is recommended for potato growing). Also, sampled sites in the Hill area showed relatively minor within-area variation (among seasons and within-Hill sites; standard deviation = 3.7 centibars) (Figure 2). Hill fields, which are located in rainfed, upper-elevation locales, provide sites where farmers typically make use of the seed-sizes associated with the “two seed” and “three seed” categories. The farmers’ discussions indicated that these smaller sizes are considered suited to the moderate uncertainty of the soil moisture regime of Hill fields. By contrast, the Ox Area and the Early Planting demonstrate significantly larger standard deviations (Figure 2). More pronounced inter-site and seasonal variation of drought stress corresponds to farmers’ preference for large-size seed tubers in these field areas. Farmers’ management of seed-size in Andean potatoes and ulluco thus bears a general relation to the evolutionary ecology of seedsize adaptations; the smaller categories of seed-size are suited to moderate risk (as in rainfed Hill fields), while larger categories of seed size are relied upon where the risk of drought stress is greater (as in Ox Area fields

116

K ARL S. Z IMMERER Table 5. Field holdings and resource-level in eastern Cuzco (n = 33 farmers; 11 farmers per community). SES indicator

Total fields Cultivated fields Irrigated fields Less-valuable fields “Native commercial” potato fields

Mean value

Standard deviation (in-community)

Standard deviation (inter-community)

Families above Mean (%)

Range

14.79 8.00 1.09 5.85 1.81

5.94 3.92 1.13 3.15 1.68

9.73 4.84 1.31 4.58 2.12

42.4% 39.4% 24.2% 42.4% 45.4%

1–45 1–23 0–5 0–17 0–9

Table 6. Gender, resource-level, and the frequency of seed-size preferences. Socioeconomic category

Qompis potatoes Size Other Small-middle First First First

Middle-large First

Mariva potatoes Size Other Small-middle First First First

Middle-large First

“Poor” Men (n = 8) Women (n = 12)

0.75 0.83

0.25 0.17

0.75 1.00

0.25 0.00

0.75 0.67

0.25 0.23

0.67 1.00

0.33 0.00

“Middle” Men (n = 5) Women (n = 5)

0.60 0.80

0.40 0.20

0.40 0.60

0.60 0.40

0.60 0.60

0.40 0.40

0.40 0.60

0.60 0.40

“Rich” Men (n = 1) Women (n = 2)

0.00 0.50

1.00 0.50

0.00 0.00

1.00 1.00

0.00 0.50

1.00 0.50

0.00 0.00

1.00 1.00

and the cloud-forest-type Early Planting; see Majopata community and the surrounding area in Figure 1). The variation of resource-levels was considerable with respect to cultivation indicators as well as other socioeconomic markers (Table 5).12 On average, farmers sow 8 fields each year (mean extension of a household’s total fields measures less than 2.0 hectares). Ownership of “native commercialized” potato fields demonstrates a pattern of variation, based on standard deviation and range, that is similar to the other cultivation indicators. Despite the commonness of “native commercialized” production in eastern Cuzco, therefore, significant difference in access to this field type is present at the farm household-level. Such variation of cropping resources (combined with livestock and labor resources) is incorporated into the local designations of “poor peasants,” “middle peasants,” and “rich peasants” within the group of small farmers.13 While all three main socioeconomic groups are active in the growing of “native commercialized” potatoes and ulluco, their preferences for seed-size are not identical. Results show a slight tendency for house-

hold resource-level to correspond to preference for tuber seed-size (Table 6), although this relationship is a weak correlation that is not statistically significant. As shown, “poor” and “middle” growers were most likely to rank the factor of size first in the evaluation of their selection criteria. By contrast, so-called “rich” farmers, the relatively better-off by local standards, also put major weight on the parameter of size, although they prefer to select medium or larger seed. Minor gender differences also were found to cooccur, to a degree, in conjunction with the variation of preferences for seed-size. Smaller seed-size was only slightly preferred in interviews with women in each of the categories (“poor,” “middle,” “rich”) (Table 6). While interpretation is preliminary due to small sample size, the suggestion of a minor-level of gender-based differences is noteworthy.14 Although women farmers of eastern Cuzco typically undertake the within-storeroom management of tubers, the members of a household often work together in the actual seeding of fields and also in the separation of harvest into size-based categories including seed. The similarity of seed-size preferences among men and

J UST SMALL POTATOES ( AND ULLUCO )?

women may also be influenced through the highly varied array of women’s livelihood activities in eastern Cuzco (on- and off-farm activities include commercial crop production, livestock-raising, migrant labor, and extra-community social networks). As a result, seed-size management is subject to shared information and strategy-making within and among farm households. Moderate or relatively minor gender-related differences may thus be associated with economic diversification and the flexible definition of gender roles that are increasingly common (Collins, 1988; Deere and León, 1998; Gisbert et al., 1994; Hamilton, 1998; Paulson, 1998; Weismantel, 1988).

Just small potatoes (and ulluco)?: Reflections on seed management, the “native commercialized” category, and participatory agricultural development and agrobiodiversity conservation A high degree of management versatility is gained through farmers’ use of seed-size variation in the “native commercialized” category of the Andean potato and ulluco crops that is farmed in eastern Cuzco. Their management of seed-size variation (based on the distinction of three main size-based categories) contributes to versatility through the capacity for varied within-field allocations to future seed, home consumption, and market sales. This versatility is important in the highly diversified farm economies of the Peruvian highlands where, similar to neighboring Ecuador and Bolivia, a substantial growth of diversification has occurred during the past decade (Bebbington, 2001; Escobal, 2001; Reardon et al., 2001). Seed-size management, as a form of intravarietal selection, also reduces the risk of crop loss through adjustments to soil-moisture conditions. Use of size-based categories of seed in response to landscape-environmental variation enables farmers to undertake agriculture in sub-optimal environments, including drought-stressed sites, which are locally utilized as important places of food production and which are valued globally as areas of concentrated, in situ agrobiodiversity conservation (Rhoades and Nazarea, 1999). These contributions of seed-size management to farming versatility are notable for a pair of reasons. First, they entail within-field crop management, whereas previous examinations of field-crop versatility (in agrobiodiversity-farming) were based on the interpretation of strategies that coordinate the use of multiple fields. In the Andean countries, for example, this multi-field emphasis is prominent in the interpretation of household’s combined sowing of FVs and MVs in different field sites (Brush, 1992, 1995;

117

Richards, 1995; Tripp, 1996). While multi-field cropping strategies remain a major component, the contribution of seed-size management to within-field versatility is another main cornerstone of current Andean farm strategies. Secondly, seed-size management is notable, since it takes place at the intravarietal level. By contrast, the variety-level of management practices, illustrated in growers’ varietal preferences, is emphasized in previous research and policy interpretations of agrobiodiversity management in such settings as Andean agriculture (Rhoades and Nazarea, 1999; Wood and Lenné, 1997; Zimmerer, 1996). In sum, this article’s research indicates that field-level and intravarietal management are important elements of the widespread success of the “native commercialized” category of Andean potatoes and ulluco production. These important components should become more fully incorporated into considerations of future seed and agrobiodiversity management in Andean farming. Size-based seed management at the within-field and intravarietal levels is also a source of sharpening our understanding of the economic and sociocultural dimension of “native commercialized” agriculture in the Peruvian Andes. From the view of farm household economics, size-based seed management is a demonstration that the current management styles for “native commercialized” potatoes and ulluco are based on marketing levels that are partial (since the harvested product usually is allocated to seed and home consumption as well) and contingent (since the share to market sales is typically determined at the time of end-use allocation). As a result, the proper interpretation of “native commercialized” agriculture as a type of in situ agrobiodiversity conservation is as neither fully market-based (advocated or assumed by certain conservation proponents) nor as a full-fledged scenario of market-retreat (suggested in an alternative depiction of the so-called “return of the native seeds”; Nellithanam et al., 1998). In general, the intermediate interpretation that corresponds to the strategic makeup of “native commercialized” agriculture resembles the more complex and less polarized understanding of MVs and FVs that has been advanced (Tripp, 1996; Cleveland, 2001). The sociocultural significance of my research includes the high level of similarity in the verbal “scripts” that farmers within the region of eastern Cuzco use in order to explain and discuss their seed-size preferences. The repeated commonness of certain expressions is not the consequence of a depauperate vocabulary, since Andean farmers are known to possess ample vocabularies of agricultural terms and metaphors (Beyersdorff, 1984). Instead, the regular use of key words and phrases is taken to reflect the notable level of sharing conversation

118

K ARL S. Z IMMERER

about farming that takes place in and among Andean farm communities, notwithstanding differences due to socioeconomic contrasts and gender relations (Mayer, 2002). The sharing of seed-size scripts is significant enough to outweigh tendencies toward different seed-size preferences among genders or resource-level groups. Social commonality and its contribution to viability of the “native commercialized” category of Andean potatoes and ulluco suggest that seed-size management is of immediate relevance to agricultural development initiatives. Local seed production is currently the target of a number of projects, programs, and policies undertaken through participatory research and development with farmers (Almekinders et al., 1994; Huamán, 2000; Iriarte et al., 1998). The growth of new institutional arrangements for seed production in the Andean countries is based on private seed certification committees (CIALS), farmer organizations, special seed projects, and NGOs (Bentley et al., 2001; Thiele et al., 2001). My research seeks to contribute to these institutional strategies, which fill a void due to failure of previous formal seed programs that in Peru, Bolivia, and Ecuador have generally failed to account for more than 5% of potato seed production (Bentley et al., 2001; Crissman and Uquillas and PROSEMPA in Thiele, 1999). Programs to-date for local production of seed tubers of the Andean potato and ulluco crops have prioritized providing farmers with access to seed of suitable quality in terms of varietal characteristics and the disease-free or reduced disease-load condition of seed tubers (CIP and PNUMA, 2000; Iriarte et al., 1998; Thiele et al., 1997). Andean farmers’ management of seed-size characteristics should be more fully incorporated into the programs of local seed production. As demonstrated, farmers’ production of their own potato and ulluco seed is influenced through their multi-faceted use of seed-size variation, which includes the extensive use of small-medium and small-size tubers. Certain Andean agricultural experts have pointed to such preferences in the high-agrobiodiversity farming of peasant and indigenous farmers of the Peruvian and Bolivian Andes (Bauman, 1987; Bentley and Vasques, 1998; Ortega, 1997). To date, however, this awareness has not appeared in the social science and participatory research literatures or in field studies of Andean farmers’ production of potatoes and ulluco in relation to agrobiodiversity conservation. Design of the institutional distribution of seed tubers to farmers, which is central to several seed-support programs, should consider the creation of size-related guidelines. In the case of many peasant and indigenous small farmers of Peru (such as those of eastern Cuzco), size-related guidelines would identify the specific

importance of small-medium and medium-size seed tubers. In addition to fitting with farmers’ existing practice, this adaptation and its evaluation in seedsupport programs could lead to the sizeable reduction of project costs (since smaller seed-size represents lower expense). Also, the programs for seed production should consider the general need for research on local size-based preferences (including the trade-offs they entail) as part of the basic array of project design and implementation activities. Preference for smaller seed tubers is also of significance to participatory plant breeding (PPB), which has the goal of providing FVs that are locally adapted to farmers’ production conditions, sociocultural circumstances, and market demand (Almekinders and Louwaars, 1999; Cleveland and Soleri, 2002; Eyzaguirre and Iwanaga, 1996; McGuire et al., 1999; Soleri et al., 2000; Tripp, 1996). My findings highlight the potential implications of modifications to tubersize that could occur through the new technical innovations of PPB. A set of contextual points prefaces this comment.15 First, the application of PPB to Andean potatoes and ulluco is still in an early stage, so its full impact on tuber-size trends is not yet possible to determine. Second, conventional planting breeding for tuber-size in potatoes (and presumably ulluco) is complex, since the trait of tuber-size is determined through a combination of genetic and phenotypic changes.16 While the yield increases of the modern potato crop are primarily the result of breeding for response to high-input field environments, this application of scientific plant breeding also targets the modification of genetic factors that exert control over tuber traits, including that of tuber-size (Caligari, 1992; D.S. Douches, Michigan State, University, personal communication [February 20, 2001]; Douches et al., 1996; H. DeJong, Agriculture and Agri-Food Canada, personal communication [October 23, 2000]). Third, the individual potato plant characteristically produces a variety of tuber-sizes, so the analysis of tuber-size trends must focus on directional changes incorporating tuber-size variation. Given these facets of potato breeding, this article’s findings suggest that participatory approaches should become more aware of the possible conflict of interest and preference with respect to the trait of tubersize. This suggestion stems from the observation that participatory plant breeding makes use of many of the same techniques and assumptions as conventional plant breeding (Almekinders and Louwaars, 1999; Eyzaguirre and Iwanaga, 1996; McGuire et al., 1999), and that the main difference is that farmers “participate.”17 The suggestion that tuber-size trends are of potential concern is grounded in the observation that conventional breeding has been prone to produce

J UST SMALL POTATOES ( AND ULLUCO )?

potato types that yield large-size tubers, a trait that is preferred among urban consumers and agroindustry such as French fry manufacture (Bentley et al., 2001; Douches et al., 1996; Horton, 1987: 46; Thiele et al., 1997: 284). The trend to larger tuber-size is linked to a tendency for the progressive reduction of the range of tuber-size variation (per yield of individual plant) in modern potato breeding and variety improvement (Howard, 1970; Louwes and Neele, 1987). While certain potato farming sectors would benefit through the breeding of larger tuber-size or reduced tuber-size variation (for example, the large growers of French-fry potatoes), these trends would poorly serve, and could disadvantage, the growers of Andean potatoes and ulluco FVs, such as those of eastern Cuzco. Yet a trend to large-size tubers may already operate in conventional potato breeding. It may become incorporated, perhaps implicitly, as a part of the mainstream of participatory breeding. As a result, new attention should be focused on potential for the divergence or a possible conflict of existing seed-size preferences, namely those of the farmers of the Andean potato and ulluco crops and the projected outcomes of plant breeding trends.

Acknowledgments Fieldwork was conducted with the generous cooperation of Peruvian farmers and a regional farmer organization (Liga Agraria Miguel Quispe), faculty and students of Peruvian universities (especially Ramiro Ortega of CICA in Cuzco), local communities, NGOs (particularly CIDA and CERAS), and the International Potato Center (CIP). Preliminary findings were reported at the International Congress of Ethnobiology in Athens, Georgia (October 24, 2000), the Congreso Nacional de Geografía in Lima, Peru (November 29, 2000), and in a panel on “Ethnography and the Theory of Science” at the annual meeting of the Association of American Geographers (March 1, 2001). A draft version of the paper was presented in the “Andean Livelihoods and Sustainability” session at the meeting of the Latin American Studies Association (LASA) in Washington, DC (September 9, 2001) and the symposium of the Madison Ecology Group (MEG) (September 21, 2001). I am grateful for the comments on presentations and the suggestions on early versions of this paper that were offered by David Preston, Janis Alcorn, Diana Liverman, Tad Mutersbaugh, David Cleveland, Daniela Soleri, Brent Berlin, Fausto Sarmiento, and the editor and anonymous reviewers of Agriculture and Human Values. Thanks also for helpful advice from David Douches, Henry de Jong, David Spooner, Mark Blumler, and Joshua Posner.

119

Notes 1. Potatoes in the “native commercialized” category belong primarily to the subspecies Solanum tuberosum subsp. andigenum. Ulluco types belong to the species Ullucus tuberosus, known as lisas and papalisas in Peru. It is a tuber-bearing food plant that belongs to the nasturtium family, and is not a close relative of the potato. 2. Each variety studied is grown widely as a “native commercialized” type in Andean regions of Peru. Qompis is a target variety in participatory plant breeding (near Lake Titicaca, Peru), and mariva and qompis lisas could likewise be chosen for such programs given their good yields, marketability, and biological diversity (Valdivia et al., 1996) Mariva was selected by the National Potato Program of Peru, and released through the formal seed system as a modern variety in central and southern Peru in the 1960s or early 1970s (Whyte, 1977). Mariva was already being grown in eastern Cuzco by the mid-1970s. Since then it has been widely incorporated into farmers’ own seed production systems, and thus has been converted into a farmer variety. 3. The specification of preference for tuber size per se was not designed as a control-type experiment, since sample tubers displayed differences in more than one parameter. 4. The following numbers of persons were interviewed in a group of three communities: Mollomarca (11); Majopata (11); and Colquepata (11). Officials in each community were asked to review the membership roster and estimate the socioeconomic status of each member household. This information was then used to select a sample of households for the main interview so that the full spectrum of socioeconomic ranks was included. 5. Results of the three wealth-estimating techniques were compiled and combined while in the field. The individual cases with conflicting data were re-checked in order to make a final determination. 6. Significantly less common than the “native commercial” fields of varietal monoculture is the production of farmer-variety potatoes in highly diverse mixtures (varietalpolyculture) of more than 20–30 varieties per field. Varietal polycultures, while biologically diverse, have become less common due to on-farm labor-time shortages and disintensification of this planting style (Zimmerer, 1996). 7. In the interviews, farmers chose for or against the tubers shown in Table 2 on the basis of seed-size, which involved 10 tubers of qompis, 9 tubers of mariva, and 8 tubers of qompis lisas. The remaining tubers of the initial sample, as shown in Table 1, were chosen for or against primarily on the basis of non-size factors. 8. The most preferred of the qompis tubers (28 grams) was also mentioned favorably for having a sufficient number of eyes for numerous tuber sprouts and, relatedly, the right shape of tuber, which is referred as spherical or rump’u in the case of the qompis variety. Farmers are aware that the concentration of eyes per unit of tuber surface area is greater in smaller seed than larger seed. The effectiveness of this seed-size advantage depends on the healthy conditions of eyes. Favorable properties of tuber eyes are described as many eyes (ñawi cape or ñawi ñawi); an

120

9.

10.

11.

12.

13.

14.

K ARL S. Z IMMERER

apical eye (uman ñawi); a well-formed eye (allin ruhusa ñawiyog); a round eye (muyu ñawi); healthy or diseaseless eyes (ch’uya ñawi, lluska ñawi) generally good eyes (allin ñawi, chawfra ñawi) or flowerlike eyes (t’ika ñawi). Unfavorable conditions of the eyes are termed as protruding (pata ñawi, nunupasaq ñawi); over-recessed (t’oqo ñawi), liplike (wak’a ñawi); elongate (wasqa ñawi) eyes-about-tosprout (ch’iqchi ñawi); malformed or worn (qolla), and one eye (ñawisapa). Mariva seed tubers were also chosen for the proper shape, which in mariva is a tapered, avocado-like shape known locally as palta. The larger mariva tubers that are preferred (e.g., 39 grams, 60 grams) approach the size of seed that is common in the potato farming systems of North America and Europe, where seed tubers (or cut sections) typically weigh 70 grams or more (Allen et al., 1992). The latter is a form of planting density (Allen and Wurr, 1992). The other type of planting density, the spacing of plants, which is a function of the distance between planting holes, was not referred to in either the preference interview or the open-ended responses and ethnographic participantobservation. The most reliable irrigation is provided by ditch canals and subsurface wetland flows, while less reliable sites depend on variations in fog drip and cloud cover. For the latter farmers prefer the large-size of seed tubers, either “two-seed” or “only-seed” types. Inter-household differences within the individual farm communities account for much of the variation, as estimated through standard deviations (Table 5, second column). Farm-level variation of resources is added to, to varying minor degrees, when household holdings are compared at the inter-community (Table 2, third column from the left). These self-designations were supplied in the interview responses of farmers. They were checked with the quantitative data obtained which, in most cases, confirmed their self-designations. “Poor peasants” were those with access rights to less than 8 fields and small herd sizes (fewer than 15 sheep or camelids and fewer than 8 horses, cows, or pigs). “Rich peasants” hold access rights to 15 fields or more and large herds sizes (greater than 50 sheep or camelids and greater than 15 horses, cows, or pigs). The wealth-status designations were also checked with the relative wealth-ranking of the key informant in each area. In a total of 13 cases where disparate categorization occurred in the data, the person’s household was considered at the lower level (for example, the “medium category” if the ranks of “medium” and “rich” were both present in the classification). This finding may differ in other crops and farming regions. In the case of common beans in Colombia, for example, Ashby et al. (1989: 116) found that farmers’ “most important criterion for grain acceptability was grain size; women by contrast preferred traditionally small-grain varieties . . . more flavourful and higher yielding. Women viewed a small grain type . . . as desireable from the point of view of sustaistence and consumption ohbjectives of the small farm.”

15. The ulluco crop has not yet been modified through the widespread use of conventional breeding. Broad resemblances to the potato crop suggest that the future modification of ulluco might adopt a similar breeding strategy that would result in large-size tubers. 16. The conventional breeding of modern crop varieties has been based, in general, on selection for the increased size and number of yield components such as seed and tubers. In conventional potato breeding, this relation is extremely complex since yield increase is due principally to resistance breeding (to pests and disease), response to nitrogen fertilizers and irrigation, and earliness (Bradshaw and Mackay, 1984; CIP, 1984; Douches et al., 1996; Harris, 1992; Howard, 1970; Jellis and Richardson, 1987; Salaman, 1985). In turn, the increased yields of the modern potato varieties are tightly correlated with tuber weight and size as well as number per plant (Howard, 1970). 17. “Participation” is a much-used term in agricultural development that can be ambiguous and is much debated (Ashby and Sperling, 1995; Bentley, 1994; Tripp, 1996). On recent discussions of participation in development-andenvironment trends see for example Agrawal and Gibson (1999) and Zerner (2000).

References Agrawal, A. and C. C. Gibson (1999). “Enchantment and disenchantment: The role of community in natural resource conservation.” World Development 27(4): 629–649. Alcorn, J. B. (1989). “Process as resource: The traditional agricultural ideology of Bora and Huastec resource management and its implications for research.” Advances in Economic Botany 7: 63–77. Allen, E. J. and D. C. E. Wurr (1992). “Plant density.” In P. M. Harris (ed.), The Potato Crop: The Scientific Basis for Improvement, 2nd edition (pp. 292–333). New York: Chapman & Hall. Allen, E. J., P. J. O’Brien, and D. Firman (1992). “Seed tuber production and management.” In P. M. Harris (ed.), The Potato Crop: The Scientific Basis for Improvement, 2nd edition (pp. 247–291). New York: Chapman & Hall. Almekinders, C. and N. Louwaars (1999). Farmers’ Seed Production: New Approaches and Practices. London: Intermediate Technology Publications. Almekinders, C., N. P. Louwaars, and G. H. de Bruijn (1994). “Local seed systems and their importance for an improved seed supply in developing countries.” Euphytica 78: 207–216. Altieri, M. A. (1995). Agroecology: The Science of Sustainable Agriculture, 2nd edition. Boulder, Colorado: Westview Press. Arnell, N. W. (1999). “Climate change and global water resources.” Global Environmental Change 9: 31–49. Ashby, J. A. and L. Sperling (1995). “Institutionalizing participatory, client-driven research and technology development in agriculture.” Development and Change 16(4): 753–770. Ashby, J. A., C. A. Quiros, and T. M. Rivers (1989). “Farmer participation in technology development: work with crop varieties.” In R. Chambers, A. Pacey, and L. A. Thrupp (eds.),

J UST SMALL POTATOES ( AND ULLUCO )? Farmer First: Farmer Innovation and Agricultural Research (pp. 115–132). London: IT Publications. Bauman, D. (1987). “El programa de semilla del Nor Yauyos.” In E. Malpartida and H. Poupon (eds.), Sistemas agrarios en el Perú (pp. 147–153). Lima, Peru: UNALM and ORSTOM. Bebbington, A. (2001). “Globalized Andes? Livelihoods, landscapes, and development.” Ecumene 8(4): 414–436. Bentley, J. W. (1994). “Facts, fantasies, and failures of farmer participatory research.” Agriculture and Human Values 11: 140–150. Bentley, J. W. and D. Vasques (1998). The Seed Potato System in Bolivia: Organisational Growth and Missing Links. London: ODI Agricultural Research and Extension. Bentley, J. W., R. Tripp, and R. Delgado de la Flor (2001). “Liberalization of Peru’s formal seed sector.” Agriculture and Human Values 18: 319–331. Berlin, B. (1992). Ethnobiological Classification: Principles of Categorization of Plants and Animals in Traditional Societies. Princeton, New Jersey: Princeton University Press. Beyersdorff, M. (1984). Léxico agropecuario quechua. Cuzco, Peru: CBC. Bianco, M. and C. Sachs (1998). “Growing oca, ulluco, and mashua in the Andes: Socioeconomic differences in cropping practices.” Agriculture and Human Values 15: 267–280. Blumler, M. A. (1998). “Evolution of caryopsis gigantism and agricultural origins.” In Research in Contemporary and Applied Geography: A Discussion Series 22 (1–2) (pp. 1–46). New York: Binghamton University. Bradshaw, J. E. and G. R. Mackay (eds.) (1994). Potato Genetics. Wallingford, UK: CAB International. Brookfield, H. (2001). Exploring Agrodiversity. New York: Columbia University Press. Brookfield, H. and C. Padoch (1994). “Appreciating agrodiversity: A look at the dynamism and diversity of indigenous farming practices.” Environment 36: 7–11, 37–45. Brookfield, H. and M. Stocking (1999). “Agrodiversity: Definition, description, and design.” Global Environmental Change 9: 77–80. Brush, S. B. (1991). “A farmer-based approach to conserving crop germplasm.” Economic Botany 39: 310–325. Brush, S. B. (1992). “Ethnoecology, biodiversity, and modernization in Andean potato agriculture.” Journal of Ethnobiology 12(2): 161–185. Brush, S. B. (1995). “In situ conservation of landraces in centers of crop diversity.” Crop Science 35: 346–354. Brush, S. B., J. E. Taylor, and M. Bellon (1992). “Technology adoption and biological diversity in Andean potato agriculture.” Journal of Development Economics 39: 365–387. Butler Flora, C., F. Larrea, M. Ordoñez, S. Chancay, S. Baez, and F. Guerrero (2001). “Production strategy typology for sustainable agriculture and natural resource management.” In R. E. Rhoades (ed.), Bridging Human and Ecological Landscapes: Participatory Research and Sustainable Development in an Andean Agricultural Frontier (pp. 193–214). Dubuque, Iowa: Kendall/Hunt Publishing. Caligari, P. D. S. (1992). “Breeding new varieties.” In P. M. Harris (ed.), The Potato Crop: The Scientific Basis for Improvement, 2nd edition (pp. 334–372). New York: Chapman & Hall. Carney, J. (1991). “Indigenous soil and water management in

121

Senegambian rice farming systems.” Agriculture and Human Values 8: 37–48. CIP (Centro Internacional de la Papa) (1984). Potatoes for the Developing World: A Collaborative Experience. Lima, Peru: CIP. CIP (Centro Internacional de la Papa) (1997). Diversidad de papas nativas en los Andes. Lima, Peru: CIP. CIP (Centro Internacional de la Papa)/PNUMA (Programa de las Naciones Unidas para el Medio Ambiente) (2000). Efectividad de las estrategias de conservación in situ y el conocimiento campesino en el manejo y uso de la biodiversidad. Lima, Peru: CIP. Cleveland, D. A. (2001). “Is plant breeding science objective truth or social construction? The case of yield stability.” Agriculture and Human Values 18: 251–270. Cleveland, D. A. and D. Soleri (2002). Collaborative Plant Breeding Integrating Farmer and Scientist Knowledge and Practice. Wallingford, UK: CAB International. Cleveland, D. A., D. Soleri, and S. E. Smith (1994). “Do folk crop varieties have a role in sustainable agriculture?” BioScience 44: 740–751. Collins, J. (1988). Unseasonal Migrations: The Effects of Rural Labor Scarcity in Peru. Princeton, New Jersey: Princeton University Press. Croll, E. and D. Parkin (1992). Brush Base: Forest Farm: Culture, Environment, and Development. London: Routledge. Cusihuamán, Antonio (2001). Diccionario quechua, CuzcoCollao, 2nd edition. Cuzco, Peru: CBC. Damania, A. B., L. Pecetti, C. O. Qualset, and B. O. Humeid (1996). “Diversity and geographic distribution of adaptive traits in Triticum turgidum L. (durum) wheat landraces from Turkey.” Genetic Resources and Crop Evolution 43(5): 409– 422. Deere, C. D. (1990). Households and class relations: Peasants and landlords in Northern Peru. Berkeley / Los Angeles: University of California Press. Deere, C. D. and M. León (1998). “Gender, land, and water: From reform to counter-reform in Latin America.” Agriculture and Human Values 15: 375–386. Douches, D. S., D. Maas, K. Jastrzebsi, and R. W. Chase (1996). “Assessment of potato breeding progress in the USA over the last century.” Crop Science 36: 1544–1552. Dueñas, A., R. Mendivil, G. Lobaton, and A. Loaiza (1992). “Campesinos y papas: A propósito de la variabilidad y erosión genética en comunidades campesinas del Cusco.” In C. I. De Gregori, J. Escobar, and B. Martitorena (eds.), Perú: el problema agrario en debate, SEPIA IV (pp. 287–309). Lima, Peru: Universidad Nacional de la Amazonia Agraria. Ellis-Jones, J. and T. Moore (1999). “Livelihood strategies and assets of small farmers in the evaluation of soil and water management practices in the temperate inter-Andean valleys of Bolivia.” Mountain Research and Development 19(3): 221–234. Escobal, J. (2001). “The determinants of nonfarm income diversification in rural Peru.” World Development 29(3): 497– 508. Eyzaguirre, P. and M. Iwanaga (1996). Participatory Plant Breeding. Rome: IPGRI. Fano, H. (1999). La papa en el Perú. Lima, Peru: Ministerio de Agricultura.

122

K ARL S. Z IMMERER

Fano, H. and M. Benavides (1992). Los cultivos andinos en perspectiva: producción y utilización en el Cusco. Lima, Peru: CIP and Cusco: CERES. Fernandez, M., P. Shrestha, and P. Eyzaguirre (2000). “Integrating gender analysis for participatory genetic resource management: Technical relevance, equity, and impact.” In E. Friis-Hansen and B. Sthapit (eds.), Participatory Approaches to the Conservation and Use of Plant Genetic Resources (pp. 44–53). Rome: IPGRI. Friis-Hansen, E. and B. Sthapit (2000). Participatory Approaches to the Conservation and Use of Plant Genetic Resources. Rome: IPGRI. Gelles, P. H. (1999). Water and Power in Highland Peru: The Cultural Politics of Irrigation and Development. New Brunswick: Rutgers University Press. Gisbert, M. E., M. Painter, and M. Quitón (1994). “Gender issues associated with labor migration and dependence on offfarm income in rural Bolivia.” Human Organization 53(2): 110–123. Gliessman, S. R. (2001). Agroecosystem Sustainability: Developing Practical Strategies. Boca Raton, Florida: CRC Press. Gonzáles, T. A. (2000). “The cultures of the seed in the Peruvian Andes.” In S. B. Brush (ed.), Genes in the Field: Onfarm conservation of crop diversity (pp. 193–216). Rome: IPGRI. Gudeman, S. and A. Rivera. 1990. Conversations in Colombia: The Domestic Economy in Life and Text. Cambridge: Cambridge University Press. Guillet, D. W. (1992) Covering Ground: Communal Water Management and the State in the Peruvian Highlands. Ann Arbor: University of Michigan Press. Hamilton, S. (1998). The Two-Headed Household: Gender and Rural Development in the Ecuadorian Andes. Pittsburgh: University of Pittsburgh Press. Harris, P. M. (ed.) (1992). The Potato Crop: The Scientific Basis for Improvement. London: Chapman & Hall. Horton, D. (1987). Potatoes: Production, Marketing, and Programs for Developing Countries. Boulder, Colorado: Westview Press. Howard, H. W. (1970). Genetics of the Potato, Solanum tuberosum. New York: Springer-Verlag. Huamán, Z. (2000). Semilleros comunales de papas nativas del Perú. Revista AgroNoticias (Lima) 251: 28–31. Hubbell, J. M. and J. B. Sisson (1998). “Advanced tensiometer for shallow or deep soil water potential measurements.” Soil Science 163(4): 271–277. Iriarte, V., F. Terrazas, and G. Aguirre (1998). Memoria: Primer encuentro taller sobre el mantenimiento de la diversidad de tubérculos andinos en sus zonas de origen. Cochabamba, Bolivia: Poligraf. Iriarte, V., L. Lazarte, J. Franco, D. Fernandez, and P. Eyzaguirre (2000). “The role of gender in the conservation, location, and management of genetic diversity in potatoes, tarwi, and maize in Pocoata, Bolivia.” In E. Friis-Hansen and B. Sthapit (eds.), Participatory Approaches to the Conservation and Use of Plant Genetic Resources (pp. 131–139). Rome: IPGRI. Jellis, G. J. and D. E. Richardson (eds.) (1987). The Production of New Potato Varieties: Technological Advances. Cambridge: Cambridge University Press.

Knapp, G. (1988). “The effects of variation in mean temperature and frost risk.” In M. L. Parry, T. R. Carter, and N. T. Konjin (eds.), The Impact of Climatic Variation on Agriculture. Volume 2. Dordrecht, The Netherlands: Kluwer Academic Publishers. Louwes, K. M. and A. E. F. Neele (1987). “Influence of weight of seed tubers on selection of first year clones: Preliminary results.” In G. J. Jellis and D. E. Richardson (eds.), The Production of New Potato Varieties: Technological Advances (pp. 78–80). Cambridge: Cambridge University Press. Mannheim, B. (1991). The Language of the Inka since the European Invasion. Austin: University of Texas Press. Mayer, E. (1994). “Recursos naturales, medio ambiente, tecnología, y desarrollo.” In O. Dancourt, E. Mayer, and C. Monge (eds.), Perú: el problema agrario en debate, SEPIA V. Lima, Peru: SEPIA. Mayer, E. (2002). The Articulated Peasant: Household Economies in the Andes. Boulder, Colorado: Westview Press. Mayer, E. and M. Glave (1999). “Alguito para ganar (a little something to earn): Profits and losses in peasant economies.” American Ethnologist 26(2): 344–369. McGuire, S., G. Manicad, and L. Sperling (1999). Technical and Institutional Issues in Participatory Planting BreedingDone from a Perspective of Farmer Plant Breeding: A Global Analysis of Issues and of Current Experience. Cali, Colombia: CIAT. Nazarea, V. D. (1998). Cultural Memory and Biodiversity. Tucson: University of Arizona Press. Nellithanam, R., J. Nellithanam, and S. S. Samiti (1998). “Return of the native seeds.” The Ecologist 28(1): 29–33. Netting, R. (1993). Smallholders, Householders: Farm Families and the Ecology of Intensive, Sustainable Agriculture. Palo Alto, California: Stanford University Press. Ortega, R. (1997). “Peruvian in situ conservation of Andean crops.” In N. Maxted, B. V. Ford-Lloyd, and J. G. Hawkes (eds.), Plant Genetic Conservation: The in Situ Approach (pp. 303–314). London: Chapman & Hall. NRC (National Research Council) (1989). Lost Crops of the Incas: Little-Known Plants of the Andes with Promise for Worldwide Cultivation. Washington, DC: National Academy Press. Paulson, S. (1998). Desigualdad social y degradación ambiental en América Latina. Quito, Ecuador: Abya-Yala. Reardon, T., J. Berdegué, and G. Escobar (2001). “Rural nonfarm employment and incomes in Latin America.” World Development 29(3): 395–409. Rhoades, R. E. (2001). Bridging Human and Ecological Landscapes: Participatory Research and Sustainable Development in an Andean Agricultural Frontier. Dubuque, Iowa: Kendall/Hunt Publishing. Rhoades, R. E. and V. D. Nazarea (1999). “Local management of biodiversity in traditional agroecosystems.” In W. W. Collins and C. O. Qualset (eds.), Biodiversity in Agroecosystems (pp. 215–236). Boca Raton, Florida: CRC Press. Richards, P. (1995). “The versatility of the poor: Indigenous wetland management systems in Sierra Leone.” Geo-Journal 35(2): 197–203. Richards, P. (1996). “Culture and community values in the selection and maintenance of African rice.” In S. B. Brush and D. Sabinskey (eds.), Valuing Local Knowledge: Indigenous

J UST SMALL POTATOES ( AND ULLUCO )? People and Intellectual Property Rights (pp. 209–229). Washington, DC: Island Press. Rousi, A., J. Salo, R. Kalliola, P. Jokela, L. Pietilla, and M. Yli-Rekola (1986). “Variation pattern in ulluco (Ullucus tuberosus-Basillaceae): A supposedly asexual Andean tuber crop.” Acta Horticulturae 182: 145–152. Rousi, A., P. Jokela, R. Kalliola, L. Pietilla, J. Salo, and M. YliRekola (1988). “Morphological variation among clones of ulluco (Ullucus tuberosus, Basillaceae) collected in southern Peru.” Economic Botany 43(1): 58–72. Salaman, R. N. (1985). The History and Social Influence of the Potato. Cambridge: Cambridge University Press. Soleri, D., S. E. Smith, and D. A. Cleveland (2000). “Evaluating the potential for farmer and plant breeder collaboration: A case study of farmer maize selection in Oaxaca, Mexico.” Euphytica 116(1): 41–57. Tapia, M. E. (1996). Ecodesarrollo en los Andes altos. Lima, Peru: Fundación Friedrich Ebert. Tapia, M. E. and A. de la Torre (2000). La mujer campesina y las semillas andinas. Lima, Peru: FAO and IPGRI. Thiele, G. (1999). “Informal potato seed systems in the Andes: Why are they important and what should we do with them?” World Development 27(1): 83–99. Thiele, G., G. Gardner, R. Torrez, and J. Gabriel (1997). “Farmer involvement in selecting new varieties: Potatoes in Bolivia.” Experimental Agriculture 33: 275–290. Thiele, G., E. van de Fliert, and D. Campilan (2001). “What happened to participatory research at the International Potato Center?” Agriculture and Human Values 18: 429–446. Thrupp, L. A. (1998). Cultivating Diversity: Agrodiversity and Food Security. Washington, DC: Worldwatch Institute. Tripp, R. (1996). “Biodiversity and modern crop varieties: Sharpening the debate.” Agriculture and Human Values 13(4): 48–63. Valdivia, R., E. Huallpa, V. Choquehuanca, and M. Holle (1996). “Monitoring potato and oxalis varieties in mixtures grown on farm family fields in the Titicaca Lake basin, Peru, 1990–1995.” In P. Eyzaguirre and M. Iwanaga (eds.), Participatory Plant Breeding (pp. 144–150). Rome: IPGRI. Weismantel, M. (1988). Food, Poverty, and Gender in the Ecuadorian Andes. Philadelphia: University of Pennsylvannia Press. Westoby, M. 1979/1980. “Elements of a theory of vegetation dynamics in arid rangelands.” Israel Journal of Botany 28: 169–194. Whyte, W. F. (1997). “Potatoes, peasants, and professors: A development strategy for Peru.” Sociological Practice 2(1): 7–23.

123

Winterhalder, B. (1994). “The ecological basis of water management in the Central Andes: Rainfall and temperature in southern Peru.” In W. P. Mitchell and D. Guillet (eds.), Irrigation at High Altitudes: The Social Organization of Water Control Systems in the Andes. Washington, DC: American Anthropological Association. Witcombe, J., D. Virk, and J. Farrington (eds.) (1998). Seeds of Choice: Making the Most of New Barieties for Small Farmers. London: Intermediate Technology Publications. Wood, D. and J. M. Lenné (1997). “The conservation of agrobiodiversity on-farm: Questioning the emerging paradigm.” Biodiversity and Conservation 6: 109–129. Zerner, C. (2000). People, Plants, and Justice: The Politics of Nature Conservation. New York: Columbia University Press. Zimmerer, K. S. (1996). Changing Fortunes: Biodiversity and Peasant Livelihood in the Peruvian Andes. Berkeley / Los Angeles: University of California. Zimmerer, K. S. (1998). “The ecogeography of Andean potatoes: versatility in farm regions and fields can aid sustainable development.” BioScience 48(6): 445–454. Zimmerer, K. S. (2002a). “Social-agroenvironmental variability of seed production and the potential collaborative breeding of potatoes in the Andean countries.” In D. Cleveland and D. Soleri (eds.), Farmers, Scientists, and Plant Breeding: Integrating Knowledge and Practices (pp. 83–107). Wallingford, UK: CAB International. Zimmerer, K. S. (2002b). “Common field agriculture and development change in Latin America.” Journal of Cultural Geography 19(2): 37–63. Zimmerer, K. S. (2003). “Geographies of seed networks and approaches to agrobiodiversity conservation.” Society & Natural Resources 16 (forthcoming). Zimmerer, K. S. and D. S. Douches (1991). “Geographical approaches to crop conservation: The partitioning of genetic diversity in Andean potatoes.” Economic Botany 45(2): 176– 189. Address for correspondence: Karl S. Zimmerer, Department of Geography, Institute for Environmental Studies, Program in Development Studies, University of Wisconsin – Madison, 384 Science Hall, 550 North Park Street, Madison, WI 53706-1491, USA Phone: +1-608-262-2138; Fax: +1-608-265-3991; E-mail: [email protected]