Advertisement

Sustainability Science

, Volume 13, Issue 1, pp 59–70 | Cite as

Cultural multi-level selection and biological market theory explain the coupled dynamics of labor exchange cooperation and social support

  • Shane J. MacfarlanEmail author
  • Mark Remiker
Special Feature: Original Article Applying Cultural Evolution to Sustainability Challenges
Part of the following topical collections:
  1. Special Feature: Applying Cultural Evolution to Sustainability Challenges

Abstract

Smallholders rely on labor exchange and social support networks; however, little is known about the cooperative dynamics of these interlinked systems. Whereas cultural multi-level selection (CLMS) predicts group membership and changes in the dominant level of selection modulates cooperation, biological market theory (BMT) posits market size fluctuations affect cooperation. We assess these predictions by examining two dimensions of labor exchange, competitive helping and labor reciprocity, and their downstream impacts on social support in a Dominican community between 2007 and 2010. First, we analyze within-community labor organization. Next, we analyze how international regulatory change and the 2008–2009 World Trade Collapse affected local labor organization and its impacts on competitive helping and labor reciprocity. Finally, we show how labor dynamics affected social support. Analyses reveal that (1) village labor initially involved two levels (labor contracting and labor exchange) and the presence of a structured group who maintained higher rates of reciprocal labor relative to non-group members; (2) changes in the international commodities market reduced labor contracting, the size of the labor exchange market, and the dominant level of selection, resulting in less competitive helping, lower rates of reciprocity for group members, and more cliquish social support; and (3) as the global market for bay oil ameliorated, labor organization shifted back to a pre-recession structure, resulting in a larger labor market with more competitive helping and higher rates of reciprocity amongst group members. We highlight the utility of an integrated CMLS and BMT framework for analyzing cooperative dynamics and socio-economic systems sustainability.

Keywords

Reciprocity Labor exchange Social support Smallholders Cultural multi-level selection theory Biological market theory 

Introduction

The world’s approximately one billion rural, poor smallholders (people who make a living from less than ten hectares of land) are economically, socially, and politically marginalized, lacking access to assets, infrastructure, information, cash and credit, and technology (IFAD 2013; Murphy 2010). Because formal institutions are often weak and markets imperfect in these settings, smallholders rely on locally based, informal institutions for making a living and achieving individual- and group-level goals (Cervantes-Godoy et al. 2013; Fafchamps 2006). Developmental economists and anthropologists have identified labor exchange and social support networks as two informal institutions that are essential to the economic and social well-being of smallholders (Downey 2010; Erasmus 1956; Fafchamps and Gubert 2007; Macfarlan et al. 2012). Whereas labor exchange allows smallholders to generate labor pools when cash and credit are rare or unavailable (Erasmus 1956; Moore 1975; Macfarlan et al. 2012), social support networks allow individuals and communities to buffer or mitigate risks including illness (Sugiyama 2003; Sugiyama and Sugiyama 2003), food shortages (Jaeggi et al. 2016), climate change driven natural disasters (Harvey et al. 2014; Morton 2007), and income shortfalls (Fafchamps and Gubert 2007; Cervantes-Godoy et al. 2013). A number of scholars have postulated that labor exchange and social support networks represent coupled systems in smallholder populations (Dirks 1972; Erasmus 1956; Fafchamps and Gubert 2007; Horowitz 1967; Ponte 2000; Provinse 1937; Swindell 1985), and recent quantitative analyses confirm this proposition—people who assist one another in labor exchange are more likely to assist each other in times of need and vice versa (e.g., Jaeggi et al. 2016; Lyle and Smith 2014; Macfarlan 2010; Macfarlan and Lyle 2015). However, questions remain about the dynamics of these systems and their resiliency. Uncovering the dynamics and resiliency of these socio-economic systems is vital for understanding how globalization, economic price collapse, and natural disasters impact the sustainability of smallholder populations and helps practitioners integrate smallholder populations more appropriately into national and international political-economic contexts where formal mechanisms dominate.

It has been argued that labor exchange networks are highly resilient, group-level adaptive responses to environmental dilemmas (Downey 2010). Between-group differences in rates of labor assistance and reciprocity represent differential responses to natural resource management problems. Communities with lower levels of labor assistance and reciprocity indicate groups that are willing to punish individuals who violate rules regarding commons, while those with higher rates indicate communities composed of individuals who follow social rules regarding the management of commons (Downey 2010). While it is likely that labor exchange networks are adaptive and resilient, empirical and theoretical issues remain with such explanations. First, such analyses have been unable to demonstrate mechanistically how village-level rates of reciprocity articulate with individual-level villager punishment and how such group-level adaptations might emerge given the conflicts of interest present at the level of the individual (e.g., Waring et al. 2015). Second, most analyses are cross-sectional in nature, making assessments about longitudinal adaptations and resiliency difficult. Furthermore, recent theoretical advances across the biological and social sciences related to the evolution of cooperation [e.g., cultural multi-level selection (Waring et al. 2015, 2017) and biological market theory (Barclay 2011; Macfarlan 2016; Macfarlan et al. 2012; Noe and Hammerstein 1994, 1995)] provide alternative mechanisms by which cooperative labor dynamics might evolve and persist. Finally, despite a large literature demonstrating a relationship between labor exchange and social support (Dirks 1972; Erasmus 1956; Horowitz 1967; Jaeggi et al. 2016; Lyle and Smith 2014; Macfarlan 2010; Macfarlan and Lyle 2015; Ponte 2000; Provinse 1937; Swindell 1985), the above-mentioned research has yet to identify how the dynamics of labor exchange relate to social support.

To clarify the preceding issues, our research seeks to explain the dynamics of two dimensions of labor exchange, competitive helping and labor reciprocity, as well as their impacts on social support. We do so by assessing predictions derived from cultural multi-level selection and biological market theory using data from a Dominican village over a 3-year period. Results suggest that an integrated perspective best explains how institutional changes impacted competitive helping and labor reciprocity and their downstream impacts on social support networks.

Biological market theory and cultural multi-level selection theory

Biological market theory (BMT) considers the effects of variability and partner choice on cooperative outcomes (Noe and Hammerstein 1994, 1995). When individuals differ in cooperative ability and have choice over partners, then selection favors cooperation over defection as individuals can discriminate high-quality from low-quality partners, resulting competition for access to high-quality partners and high-quality partners receiving more than low-quality ones (Barclay 2004; Barclay and Willer 2006; Fu et al. 2008; Macfarlan et al. 2012, 2014; McNamara et al. 2008; Nesse 2007; Roberts 1998). Formal models demonstrate that the level of competitive helping in biological and social market contexts is dependent on the size of the market place (Barclay 2011). Competitive helping increases in environments where a greater number of individuals exist and/or individuals have greater ability to move between partners (Barclay 2011). However, as social markets decrease in size, either due to decreases in available partners or opportunities to interact, competitive helping decreases (Barclay 2011). Reciprocity is postulated to evolve under particular market contexts (Barclay 2011; Macfarlan et al. 2012). When individuals have limited choice over new partners (i.e., smaller market sizes), retention of existing partners through the control of rewards and punishments could lead to contingent reciprocity. On the other hand, as market sizes increase, individuals should respond by expanding partner choice relative to partner control, thereby decreasing rates of reciprocity. In reference to labor exchange, as labor market size decreases, individuals should respond by assisting fewer people leading to higher rates of contingent reciprocity; however, as the size of the labor exchange market increases, individuals should assist more people leading to lower rates of reciprocity. Because labor exchange and social support are linked systems, it is predicted that as competitive helping in labor exchange decreases, so too should social support.

Cultural multi-level selection theory (CMLS) examines the evolutionary dynamics of individuals in nested group settings (Waring et al. 2015). CMLS envisions a universe where individuals assort, cooperation evolves, and institutions emerge, leading to the formation of stable social groups (Waring et al. 2017). Once stable social groups form, they can then operate as levels of selection in the evolution of institutions and cooperation at multiple levels of hierarchy (Boyd and Richerson 2009; Waring et al. 2015). Changes to ecology, population structure, institutional diversity, or cultural transmission modulate the extent to which cooperation emerges or dissipates at multiple levels of social organization (Waring et al. 2015). In particular, CMLS theory predicts that for any social dilemma, individualistic strategies will be favored over group-functional strategies when the dominant level of selection is below that of the dilemma (Waring et al. 2015). However, when the dominant level of selection occurs above that of the dilemma, selection on groups favors cooperative, group-functional traits. Because labor exchange represents a social dilemma (Macfarlan et al. 2012, 2013; Macfarlan and Lyle 2015), the CMLS framework predicts that labor reciprocity should decrease when changes to the organizational structure of labor cause the selective environment to shift from higher to lower levels of social organization and should increase when the selective environment shifts from lower to higher levels of labor organization. Furthermore, because social support represents a group-functional trait (Duhaime et al. 2004) that is linked to labor exchange, we predict that cooperative dynamics in social support networks should track changes to cooperative dynamics in labor exchange.

The CMLS framework (Waring et al. 2015) provides a set of conceptual tools for analyzing how cultural traits should respond to disturbance in socio-ecological systems, involving the identification of (1) a focal trait—competitive helping and labor reciprocity; (2) the organizational context of the trait; (3) historical forces operating on trait; and (4) the levels of selection operating on the trait. Here, we apply this framework to analyze cooperation in labor exchange and social support a Dominica village.

Study site in the CMLS framework

Dominica is an independent Caribbean nation located in the Lesser Antilles between the French Departments of Guadeloupe and Martinique (Honychurch 1995). It is relatively underdeveloped (Caribbean Development Bank 2003), has some of the highest poverty rates in the Caribbean (IMF 2004), and is highly vulnerable to exogenous shocks related to climate change and international economic activity (Caribbean Development Bank 2010). Bwa Mawego (pseudonym) is one of the nation’s poorest villages, located on the eastern coast in the St. David Parish (Quinlan 2004). In 2007, the village contained approximately 400 residents living in 180 households in 12 hamlets with another 100–200 individuals residing part time (Macfarlan and Quinlan 2008; Macfarlan 2011). Villagers are descended from a mix of West African, European, and native Carib populations (Quinlan and Flinn 2005). Typical of many Afro-Caribbean communities, village social life has a matrifocal orientation (Macfarlan 2011; Quinlan 2006). Village economy is a mix of slash-and-burn horticulture, fishing, and small-scale commercial agriculture. The primary cash crop grown is bay trees. Agricultural land in the community is owned in common by patrilineages and lineage members have usufruct land rights. While this ownership strategy makes accessing credit and loans difficult due to lack of collateral (Caribbean Development Bank 2003), it anchors the community and tethers social relationships.

Focal traitcompetitive helping and labor reciprocity Labor exchange is as a system of rural labor whereby small groups of community members (e.g., 2–15 people) self-organize to assist one another in agricultural work that requires collective action (Macfarlan et al. 2012; Moore 1975). Labor groups have been characterized as operating like a rotating and saving credit association (Gilligan 2004)—a single individual acts as a labor leader (referred to as a chief-for-a-day or CFAD) and group members assist this CFAD in his/her agricultural work. When the work is completed, another group member acts as the CFAD. The process is repeated until all members have had the opportunity to act as the CFAD. CFADs do not share the goods produced a labor exchange event with those who provide assistance. However, CFADs may share the fruits of their labor with other community members if labor contracting exists (Macfarlan and Lyle 2015). As such, labor exchange represents an iterated, sequential, N-player, mutual aid dilemma (Macfarlan et al. 2012, 2013; Macfarlan and Lyle 2015). All individuals are better off receiving labor when they act as the CFAD; however, such a system requires individuals to provide costly support to others who act as the CFAD. As such, assortment mechanisms are necessary to support reciprocity in labor exchange. Although many qualitative ethnographic reports suggest that labor exchange systems are highly reciprocal (Erasmus 1956; Moore 1975), quantitative analyses suggest that within- and between-community rates of competitive helping and labor reciprocity vary considerably (Downey 2010; Macfarlan et al. 2012). We seek to understand the dynamics of competitive helping and reciprocity in labor exchange.

Organizational context Labor—Within Bwa Mawego, labor exchange occurs in the context of bay oil distillation. Bay is a native Caribbean tree (Pimenta racemosa [Miller] J.W. Moore), the leaves of which are harvested every 10 months and steam distilled to produce essential oil of bay (Ames et al. 1971; Macfarlan et al. 2012, 2013; Macfarlan and Lyle 2015; McHale et al. 1977; Tucker et al. 1991). Individuals farm bay trees and distill bay oil, then sell raw bay oil to the Dominican Essential Oils Cooperative. The minimum land holding to produce bay oil is a quarter acre, which nets approximately 7.5 L of oil and is sold to the cooperative for $176 US (Macfarlan 2010). Land holdings vary considerably within the village—the average adult has a quarter acre of land, a few have upwards of 12 acres, and others no land. Because of land differentials, individuals with large land holdings will contract others to work the land for them. Approximately 70% of all bay oil production involves labor contracting (Macfarlan and Lyle 2015). Bay oil production is organized in the following manner: landowners give labor contracts to others who work the land (the CFAD), who must in turn generate a labor pool to complete work. Landowners and CFADs compete for access to one another, resulting in mutually beneficial partnerships between a landowner and a CFAD, who split the cash earnings of bay oil distillation events at the rate of 2/3–1/3, respectively. Males are more likely to be chosen as a CFAD and five times as many males are engaged in bay oil distillation relative to women (Macfarlan et al. 2012; Macfarlan and Lyle 2015). Males with a reputation for labor competency are selected more often for labor contracts (Macfarlan and Lyle 2015). Although a landowner may provide labor contracts to multiple individuals and individuals may obtain labor contracts from multiple landowners simultaneously, relationships between a landowner and a CFAD can remain stable for multiple years (Macfarlan and Lyle 2015). Labor contracting thus represents a principal–agent dilemma.

Once a labor contract is formed, and due to the difficulty of the task, CFADs must generate a labor pool to assist in production (Macfarlan et al. 2012, 2013). This labor pool forms the basis of the labor exchange environment. Individuals helped by the CFAD in the past are supposed to assist. Because bay oil distillation is a highly conspicuous task, villagers realize when they should help. Individuals who have not been helped by the CFAD are also allowed to assist if they seek to forge a new labor exchange partnership. Helpers receive nothing from the distillation event except for the possibility of receiving labor reciprocally and social support in the future (Macfarlan 2010). Previous quantitative research suggests that males compete with one another for access to labor exchange relationships (Macfarlan et al. 2012). Males who provide assistance to a greater number of CFADs receive a reputation for prosociality (Macfarlan et al. 2013). Individuals with good reputations achieve larger labor pools relative to those with bad reputations (Macfarlan et al. 2012). Larger labor pools allow high-quality farmers to be selective about with whom to form reciprocal labor exchange partnerships and induce competition between helpers to curry the favor of the CFAD. Individuals who give greater days of labor to the CFAD are more likely to be selected as a reciprocal exchange partner (Macfarlan et al. 2012).

In Bwa Mawego, the competitive market for reciprocal labor exchange partners resulted in the formation of single, un-named group of highly prosocial individuals who assisted one another in bay oil production and other tasks requiring collective action (e.g., house repair and planting). The members established a set of norms for admittance, maintenance, and expulsion. Admittance into the group was contingent on an individual (1) demonstrating high prosociality, giving multiple days of labor to existing group members, (2) being willing to pay an entry fee of $25 Eastern Caribbean into a common pool, and (3) understanding the group’s “vibe” [i.e., a set of norms related to work requirements: “cigarettes, rum, and talking shit” (SJM field notes: June 15, 2006)]. Once admitted to the group, members could put forth requests for assistance in tasks requiring collective action. Members would work every Thursday on a single task that was deemed highest priority by the collective. The individual selected to receive assistance had to supply a pack of cigarettes and a bottle of rum for a half-day of work or cigarettes, rum, and food for a full day of labor. Group members who failed to supply assistance were penalized $10 EC on their first infraction and were expelled by the group after the second infraction; however, they were not penalized for failing to assist if they had a legitimate excuse (e.g., illness and injury). Group members reported that no permanent leader existed and group membership evolved over the first 3 years of existence (2003–2006) between 8 and 13 community members. By 2007, the group contained 11 members.

Social support—consistent with the larger Afro-Caribbean culture-area (Dirks 1972; Horowitz 1967; Wilson 1969, 1971, 1973), within-community social support networks are sexually differentiated and appear to be adaptive responses to poverty and economic uncertainty (Dirks 1972; Horowitz 1967; Macfarlan 2010; Macfarlan and Quinlan 2008; Quinlan 2005). Whereas female social relationships are affectively close, long lasting, and predicated upon kinship, the domestic unit, or church groups, males form highly flexible, short-term individual alliances with genetic kin and other men who are close in age, live in close proximity, have similar reputations, and who exchange labor (Dirks 1972; Macfarlan 2010; Wilson 1969, 1971, 1973). Male friendships tend to be ego-centered which provides flexibility and rapid decision-making in response to economic opportunities (Dirks 1972). When work occurs, males tap existing networks and escalate social relationships or forge new relationships that may improve economic opportunities (Macfarlan 2010). Not surprisingly, men with greater access to work opportunities are reported to build larger social networks as most other men vie for the opportunity to create a partnership with them (Macfarlan 2010). Thus, poverty and economic uncertainty create a selective environment for male relationships that are highly individualistic, despite the group-level benefits that accrue to the community when social support networks are highly cohesive.

Historical and political-economic forces The production of bay oil, and thus the degree to which competitive helping and labor reciprocity exists, is situated within a larger historical and political-economic context. Bay oil is distilled by villagers and sold to the nation’s lone essential oils cooperative, who in turn refines the oil and sells it on the international commodities market for $30 US per pound as an ingredient in the cosmetics and food flavoring industries. Dominica is the world’s largest exporter of bay oil (Ames et al. 1971; McHale et al. 1977; Tucker et al. 1991) and essential oils are one of Dominica’s major export commodities (United Nations Statistical Division 2017). International regulations and global trade determine the amount of bay oil exported from Dominica, and therefore, the amount the national cooperative purchases from smallholders, which in turn affects within-community labor dynamics. Competitive helping and labor reciprocity are situated within these dynamics.

Within-community structured labor groups—while systematic data collection on the history of structured labor groups in Bwa Mawego does not exist, anecdotal evidence suggests that at least since the 1970s, a small number of community members (e.g., three to ten individuals) would coordinate activity periodically to form well-structured groups that assisted one another in bay oil production and other tasks. Some groups were composed strictly of males, others of females. These groups would exist for several months to several years and provide a stable supply of reciprocal labor. Group dissolution normally occurred in the context of witchcraft accusations. Villagers currently report that at least one such group of males exist in the community and that group membership is partially fluid.

European Union (EU) regulatory change—in 1992, the United Nations Conference on Environment and Development issued an international mandate for the development of a single, globally harmonized system for addressing the classification of chemicals, labels, and safety datasheets in an effort to facilitate global trade as well as consumer identification of hazardous chemicals (European Chemicals Agency 2009; United Nations 2011). In 2002, the globally harmonized system (GHS) for the classification and labeling of chemicals was completed and endorsed by EU member states at the World Summit for Sustainable Development. The GHS was promulgated in 2009 (European Chemicals Agency 2009) through multiple pieces of legislation (Regulation (EC) No 1272/2008 2008; Regulation (EC) No 1334/2008 2008; Regulation (EC) No 1223/2009 2009). The net effect of these institutional changes was that essential oil of bay required labeling to indicate that it has the potential to harm human and environmental health and it cannot be added as an ingredient in food preparations, because it contains hazardous chemicals (i.e., d-limonene, linalool, eugenol, and methyl eugenol) (Demyttenaere 2012, 2016).

International Trade Collapse—between 2008 and 2009, the world economy experienced its most severe financial shock since the Great Depression of the 1930s. Concomitant with the Great Recession of 2008–2009 was a Great Trade Collapse (Bems et al. 2013), whereby global trade declined rapidly between the third quarter of 2008 and the second quarter of 2009 due to a reduction in global demand for products (Baldwin 2009). This decline in global trade was the largest experienced over the previous 40 years (Shelburne 2010) and affected the demand for a number of commodity classes (Bems et al. 2013), including essential oils. Global exports of essential oils decreased by 14.7% between 2008 and 2009 on the international commodities market (United Nations Statistical Division 2017). Not surprisingly, Dominican essential oil export volume dropped precipitously between 2008 and 2009 (United Nations Statistical Division 2017). By the third quarter of 2009 through 2010, global trade ameliorated and essential oil exports returned to pre-recession volumes both globally and for Dominica specifically.

Together, EU legislation and the Great Trade Collapse affected the international trade in essential oils and Dominican essential oil exports. According to Garner Eloi (Director of the Dominican Essential Oils and Spice Cooperative Society), fears over the potential toxicity of bay oil, in conjunction with the contraction of international trade, reduced international demand for bay oil, which forced the cooperative to reduce purchasing bay oil from smallholders (personal communication June 14, 2010). As the market began to ameliorate following the third quarter of 2009, demand for bay oil increased on the international commodities market and the cooperative increased purchases of raw bay oil from smallholders.

Levels of selection Prior to the promulgation of EU regulations and the World Trade Collapse, within-community bay oil production regimes were organized at two levels: between landowners and CFADs for labor contracts and between CFADs and helpers for reciprocal labor. Quantitative analyses demonstrate that selection was operating at both levels (Macfarlan et al. 2012; Macfarlan and Lyle 2015); however, labor contracting precedes labor reciprocity and thus has primacy for understanding how cooperation at one level of social organization impacts another. Here, we examine how the above-mentioned historical and political-economic forces modulated the organizational structure of labor within the community and, therefore, the selective environment for competitive helping and labor reciprocity in bay oil production and its downstream consequences on social support.

Methods

Bay oil production data were collected over three 10-month time periods: July 2007 through April 2008 (hereafter “T1”), July 2008 through April 2009 (hereafter “T2”), and May 2009–February 2010 (hereafter “T3”). One author (SJM) and a village resident performed daily instantaneous scan sampling (Bernard 2002) of the village’s eight distilleries. During distillery scans, we recorded the owner of the bay leaves, their sex and age, as well as the focal farmer (CFAD), all helpers, their sex, age, and residence. Across the 30 months of data collection, we recorded 243 individuals engaged in 701 distillation events. Descriptive statistics associated with bay oil production in each 10-month period can be found in Tables 1 and 2. A subset of 48 males were selected to have their social support networks identified following T1 and T2. Social support networks were assessed by two village members (one male and one female) who rated pairs of farmers as having positive, neutral, or negative relationships using the following prompt: “If person X was in a time of need, would person “Y” provide assistance?” Positive ratings indicated that a person would always provide assistance, neutral ratings indicated a person might provide assistance, but only under certain conditions, and negative ratings indicated that the person would never help the focal individual. Agreement between raters was analyzed using a QAP correlation with Ucinet version 6.523 (Borgatti et al. 2002). Raters had moderate agreement between networks following T1 (r = 0.41; p < 0.001; n = 2256) and T2 (r = 0.31; p < 0.001; n = 2256), and social support network data were averaged resulting in five ordinal categories (positive, moderately positive, neutral, moderately negative, and negative) (Table 3).
Table 1

Descriptive statistics associated with bay oil production in each 10-month period

 

Events

Total people involved

Total land owners

# landowners contracting labor

# of labor contracts given

Total CFADs

Total helpers

Total days labor

T1

241

150

70

55

161

63

101

728

T2

191

139

60

44

124

60

97

607

T3

269

168

72

57

198

59

118

889

Table 2

Descriptive statistics associated with cooperation in bay oil distillation

 

n

Mean (SD)

Median

Min/max

Group size T1

235

3 (1.3)

3

1/10

Group size T2

191

3 (0.9)

3

2/6

Group size T3

269

3.3 (1)

3

2/6

# Helpers per CFAD

63

5 (3.6)

4

0/19

# Helpers per CFAD

60

4 (2.3)

3.5

1/11

# Helpers per CFAD

59

5.6 (4)

4

1/18

# of CFADs assisted T1

101

3 (3)

2

1/17

# of CFADs assisted T2

97

2.5 (2.5)

2

1/15

# of CFADs assisted T3

118

2.8 (3)

2

1/16

Table 3

Frequencies of male dyadic social support relationships following T1 and T2

 

Positive

Moderately positive

Neutral

Moderately negative

Negative

T1

196

322

1143

591

4

T2

234

165

692

484

681

Results

Labor organization and the initial environment for competitive helping and labor reciprocity

In T1, prior to EU regulatory change and the Global Trade Collapse, within-community bay oil production involved cooperation at two stages, labor contracting and labor exchange. In T1, 150 people were involved in bay oil across 241 events. Of these, 28 events lacked information regarding the landowner and six events lacked information on group size resulting in 213 events with complete information. Of these events, 75% involved labor contracting—55 landowners allocated 161 labor contracts to CFADs. With respect to labor exchange, 63 CFADs organized distillation events involving 101 assistants who together labored for 728 person days. Mean group size was three individuals and people provided assistance to five CFADs on average. Twenty-nine percent of all dyads who could have engaged in labor reciprocity actually did so (84 of 288 dyads). A generalized estimating equation demonstrates that the odds of reciprocal labor exchange were five times higher for members of the structured group of 11 men in T1 relative to non-group members [Wald X 2 = 12.3; p = 0.0004; n-observations = 288; n-groups = 144; odds ratio ± (RSE) = 5.1 ± (2.4); z = 3.5].

International market dynamics, labor organization, competitive helping, and labor reciprocity

In T2, after EU regulatory change and the Global Trade Collapse, a significant shift in within-community labor relationships occurred. First, as the national cooperative reduced bay oil purchases, villagers responded by producing less. A 20% reduction in bay oil distillation events occurred in T2 (Table 1). Ten fewer landowners (15%) contracted labor in T2 and logistic regression shows a significant decrease in the odds that a landowner awarded a labor contract to another between the two periods [Wald X 2 = 4.8; pseudo R 2 = 0.01; p = 0.02; n = 404; odds ratio ± (RSE) = 0.6 ± (0.1); z = −2.3]. Some landowners removed themselves from the production of bay oil altogether, while others opted to work the land themselves. A zero-inflated Poisson regression model shows that the group of 11 men obtained significantly more labor contracts relative to non-group members during the economic recession [Wald X 2 = 8.9; p = 0.003; n = 178; IRR ± (RSE) = 2.3 ± (0.6); z = 2.9] (Fig. 1). Despite a reduction in labor contracting, roughly, the same number of CFADs and helpers participated in T2. Thus, the contraction of the labor market resulted in fewer opportunities to interact, not a reduction in the number of people with whom to interact.
Fig. 1

Box plot relationship between time period, group membership, and labor contracting

BMT posits that a decrease in market size should reduce competitive helping. In fact, helpers responded to the shift in the organizational structure of bay oil production by interacting with fewer CFADs. A generalized estimating equation examining CFAD assistance between T1 and T2 demonstrates a significant decrease in competitive helping [Wald X 2 = 7.1; p = 0.008; n-observations = 264; n-groups = 132; IRR = 0.82 ± (0.06); z = −2.7] (Fig. 2).
Fig. 2

Scatter plot relationship between Dominican essential oils exports and labor exchange cooperation

BMT predicts that when markets contract in this manner, an increase in contingent reciprocity should occur as helpers have less partner choice. In contrast, CMLS predicts that changes to the organizational structure of labor that weakens selection at higher levels of social organization (i.e., between landowners and CFADs) should produce decreases in cooperative behavior at lower levels of social organization (i.e., rates of reciprocity between CFADs and helpers). We find that rates of contingent labor reciprocity decreased slightly between T1 and T2 at the population level [T1: 84 of 288 dyads reciprocated (29%); T2: 74 of 278 dyads reciprocated (27%)]. Furthermore, the odds of establishing a reciprocal labor exchange relationship were not statistically different between members of the structured group and non-group members in T2 [Wald X 2 = 0.7; p = 0.37; n-observations = 278; n-groups = 139; odds ratio ± (RSE) = 1.6 ± (0.9); z = 0.9]. This was due to the fact that members of the structured group proportionally decreased rates of labor reciprocity in T2 relative to T1, while non-group members increased it. Together, these findings suggest that (1) non-group members shifted strategies from partner choice to partner control as the labor market contracted thereby making labor more cliquish and (2) rates of labor reciprocity decreased amongst group members because of a shift in labor organization and the dominant level of selection.

Amelioration of the international bay oil market, labor organization, competitive helping, and labor reciprocity

In T3, following the amelioration of trade on the international commodities market, Dominica essential oils export volumes increased and the national cooperative increased purchases of raw oil from smallholders. This resulted in a concomitant increase in bay oil production locally. Seventy-eight (~30%) more distillation events occurred in T3 relative to T2 as landowners put more land into production through labor contracting. Between T2 and T3, 12 more landowners contracted that labor and logistic regression shows an increase in the odds a landowner contracted labor to another [Wald X 2 = 5.2; pseudo R 2 = 0.01; p = 0.03; n = 457; odds ratio ± (RSE) = 1.6 ± (0.3); z = 2.2] (Fig. 2).

BMT posits that an increase in market size should increase competitive helping. In fact, the shift in the organizational structure of labor towards greater labor contracting resulted in greater amounts of competitive helping by helpers towards CFADs. A generalized estimating equation examining CFAD assistance behavior between T2 and T3 shows a significant increase in competitive helping [Wald X 2 = 11.3; p = 0.0008; n-observations = 324; n = groups = 162; IRR ± (RSE) = 1.3 ± (0.1); z = 3.4; p = 0.001] (Fig. 2). As the labor market expanded and provided more opportunities to interact, individuals responded by increasing the number of CFADs with whom they interacted.

BMT predicts that the expansion of the labor exchange market should have decreased labor reciprocity, while CMLS predicts that the shift in labor organization towards greater labor contracting should have increased rates of labor reciprocity between CFADs and helpers through an alteration in the dominant level of selection. Between T2 and T3, we find that rates of labor reciprocity remained roughly equal at the population level [T2: 74 of 278 dyads (27%); T3: 96 of 344 dyads (28%)]. However, consistent with CMLS (and in contrast to BMT), the structured group of 11 achieved higher rates of labor reciprocity relative to non-group members in T3 [Wald X 2 = 3.8; p = 0.052; n-observations = 344; n-groups = 172; odds ratio ± (RSE) = 2.3 ± (0.9); z = 1.95] by proportionally increasing rates of labor reciprocity. This suggests that a group-functional trait—labor reciprocity—increased when the dominant level of selection moved above that of the social dilemma.

The dynamics of competitive helping and labor reciprocity on social support

Scholars have long posited a relationship between labor exchange and social support (Dirks 1972; Erasmus 1956; Horowitz 1967; Ponte 2000; Swindell 1985), and recent quantitative analyses demonstrate these two systems are interlinked even after controlling for a number of confounding variables, such as genetic and social kinship, spatial proximity, and age and reputation matching, and not-in-kind labor (Jaeggi et al. 2016; Lyle and Smith 2014; Macfarlan 2010; Macfarlan and Lyle 2015). To our knowledge, no research has examined the coupled dynamics of labor exchange and social support systems over time. As linked socio-economic systems, cooperative dynamics in one domain should precipitate into the other. In fact, we find a significant reduction in social support between T1 and T2 concomitant with the decrease in competitive labor assistance (Table 4) (Fig. 3). A comparative analysis of social support network density using Ucinet software version 6.523 (Borgatti et al. 2002) shows a significant decrease in social support over time (T1 = 0.03; T2 = −0.23; t = 6.1; p = 0.0002; n = 2256). While social support declined generally in T2, the group of 11 men experienced higher rates of social support relative to the population at large (pseudo R 2 = 0.11; Wald X 2 = 15.4; n = 47; IRR ± (RSE) = 2.2 ± 0.4; z = 3.9) (Fig. 3). The general decrease in social support appears directly tied to changes in labor exchange. A generalized estimating equation shows that one’s in-degree social support was positively related to the number of CFADs which they assisted in the previous 10-month time period [Wald X 2 = 27.9; p < 0.0001, n-observations = 96, n-groups = 48] (Table 5). Because individuals assisted less people in labor in T2 and non-group members shifted strategies from competitive helping to contingent reciprocity, social support networks decreased in size, as well. An increase in cliquish behavior in labor translated to more cliquish social support. The buffering of social support amongst the group of 11 appears to be related to the fact that they received more labor contracts in T2 and thus increased their likelihood of interacting with others.
Table 4

Descriptive statistics associated with the 48 men who were involved in bay oil distillation and had social support measured

 

N

Mean (SD)

Median

Min/max

In-degree social support T1

48

10.8 (6)

10.5

0/21

In-degree social support T2

48

8.3 (7)

7.5

0/26

Age

48

44 (12)

43

18/68

CFADs helped T1

48

4 (4)

2

0/17

CFADs helped T2

48

3 (3.5)

2

1/15

Fig. 3

Box plot relationship between time period, group membership, and social support

Table 5

Generalized estimating equation regression coefficients associated with in-degree social support outcomes

 

IRR (RSE)

z

p

Time

0.8 (0.08)

−2.2

0.027

Age

0.98 (0.006)

−3.0

0.003

CFADs assisted

1.04 (0.02)

2.6

0.009

Constant

35.7

14.2

<0.001

Discussion

This research was motivated by an empirical gap with broad public policy implications—little is known about the dynamics of labor exchange and its impact on social support networks. Because state- and market-sponsored institutions are weak in rural areas, smallholders rely on informal institutions such as labor exchange and social support networks to generate commodities and buffer risk. The current research was also motivated by an opportunity to assess predictions derived from CMLS and BMT concerning the evolutionary dynamics of competitive helping and reciprocity. While the CMLS framework suggests that cooperative dynamics in social dilemmas are modulated by group membership and changes to the dominant level of selection, BMT posits that changes in market size affects cooperative dynamics. Based on the preceding analyses it appears that both frameworks are necessary for explaining how cooperative dynamics emerge and persist. Institutional changes modulated the organization of labor and, therefore, the dominant level at which selection operated, causing social markets to expand and contract, affecting the dynamics of competitive helping, labor reciprocity, and social support.

In the case of Bwa Mawego, international regulatory change affecting the classification and labeling of chemicals, in conjunction with the global collapse in trade between 2008 and 2009, attenuated international demand for bay oil, which caused the national Dominican essential oils cooperative to reduce purchasing from smallholders. Landowners from Bwa Mawego responded to this contraction by reducing the number of labor contracts they awarded to CFADs. With fewer opportunities to interact, the labor exchange market contracted and helpers responded by assisting fewer CFADs generally. The contraction of bay oil production differentially affected group and non-group members. Non-group members responded by proportionally increasing rates of reciprocity—rather than focusing on strategies that prioritized partner choice, non-group members concentrated labor activity to retain the existing partners. As for the structured group of 11 men, rates of reciprocity decreased because of a shift in the organizational structure of labor that caused selection to weaken at a level above that of the social dilemma. As the social market for labor relationships became more cliquish, individual-level social relationships suffered. People assisted less CFADs in the second time period, and as a result, the number of people which one could rely in a time need also decreased. This shift translated at the community level to a village with fractured social cohesion, putting them at greater risk of exogenous shocks, because the primary mechanism for mitigating risk became weakened. This is particularly problematic for communities like Bwa Mawego that have seen a dramatic increase in exogenous shocks over the last several decades (Anderson et al. 2011). The ability of this socio-economic system to persist in the face of these exogenous shocks became compromised by institutional changes occurring at higher levels of human social organization.

The structured group of 11 men experienced a number of benefits that non-group members did not. First, they achieved higher rates of labor reciprocity in T1 and T3 relative to non-group members due to norms that promoted within-group cooperation. Consistent with the CMLS framework (and contrary to that of BMT), rates of reciprocity were higher for group members when the dominant level of selection (between landowners and CFADs) existed above that of the social dilemma (between CFADs and helpers). Thus, it appears that market size dynamics alone cannot account for the evolution of reciprocity when cultural group selection is at play. Second, group members received a greater number of labor contracts in T2 as the international market for bay oil collapsed which allowed them to work more than the population at large. These extra labor opportunities translated into larger social support networks relative to non-group members in T2. In this respect, selection for the evolution of groups existed, albeit in this case for the evolution of a single group. We believe that this finding is important as it sets the stage for the evolution of multiple groups, which is the grist mill upon which cultural selection across groups operates. Why other groups did not emerge at this time is a perplexing issue and requires further analysis.

As international trade ameliorated following the third quarter of 2009 and into 2010, bay oil production increased locally as landowners put more land into production and allotted more labor contracts. As labor organization shifted to greater labor contracting, opportunities to interact increased causing helpers to assist a greater number of CFADs. Consistent with Downey’s (2010) perspective, we find that Bwa Mawegan labor exchange networks appear resilient to exogenous shocks, especially those where norms and punishment lead to the formation of structured groups. However, we posit a different set of mechanisms that produce such resiliency. It is possible that the different mechanisms at play in the two contexts represent unique cultural adaptations to separate economic influences—in Bwa Mawego, Dominica commodities are made for the global market, while in Mayan Q’eqchi villages, commodities are produced for within-village consumption. Although we lack data on social support outcomes following the third time period, we predict that the shift away from contingent reciprocity to competitive altruism should have produced greater amounts of social support and, therefore, a more socially cohesive community. Future research will be necessary to clarify this relationship.

While our analyses are consistent with CMLS and BMT, we acknowledge that additional mechanisms play a part in the dynamics of intracommunity social support and labor relationships. For example, increases in technology (i.e., T.V. and Internet) and extra-community wage labor opportunities have affected how Bwa Mawegans interact with one another (Caribbean Development Bank 2010). We view these as complementary processes. Hopefully, future research will identify how modernization, in addition to global market forces, affects cooperative outcomes through the combined lens of CMLS and BMT.

Finally, political economists have long identified that global market forces affect within-community social relationships (Rosenberry 1988). However, a precise accounting of how these dynamics unfold has been lacking (e.g., Borgerhoff-Mulder and Coppolillo 2005). We believe that a synthetic CMLS and BMT perspective provides a mechanistic accounting of how political-economic forces translate into within-community social relationships. We hope that our research spurs complementary analyses.

Conclusion

Cultural multi-level selection and biological market theory explain the dynamics of competitive helping and reciprocity in labor exchange and their downstream impacts on social support. International institutional changes affected the local structure of labor organization which altered the selective environments for competitive helping. The relationships between market size dynamics shift in the dominant level of selection, and rates of reciprocity depended on whether individuals existed in a structured group. BMT better explained reciprocity dynamics for those who did not exist in a structured group, while CMLS better explained reciprocity dynamics for those who did. Because labor exchange and social support systems are interlinked, social support outcomes ebbed and flowed with the dynamics of competitive helping and labor reciprocity. The evolution of a structured group supported by norms and punishment insulated group members from the negative impacts of international regulatory change. A synthetic CMLS and BMT framework was necessary to understand the coupling of these systems and their dynamics. Although many anthropologists and political economists rightfully worry about the negative impacts of globalization on informal institutions in smallholder society, our analyses suggest that negative outcomes are not the only possible outcome. A combined CMLS-BMT framework suggests that under the right institutional conditions, greater cooperation and social cohesion can be achieved in a globalized world.

Notes

Acknowledgements

The authors thank Jeremy Brooks, Adrian Bell, and two anonymous reviewers for valuable comments that substantially improved the manuscript, the people of Bwa Mawego, Dominica for permitting this research, Juranie Durand for his expertise in bay oil distillation, and Robert Quinlan and Mark Flinn for introducing us to the site. SJM also thanks the Wasatch Experience, University of Utah, for discussions on socio-ecological systems sustainability.

Supplementary material

11625_2017_481_MOESM1_ESM.xlsx (193 kb)
Supplementary material 1 (XLSX 193 kb)

References

  1. Ames GR, Barrow M, Borton C, Casey TE, Matthews WS, Nabney J (1971) Bay oil distillation in Dominica. Trop Sci 13:13–25Google Scholar
  2. Anderson MG, Holcombe E, Blake JR, Ghesquire F, Holm-Nielsen N, Fisseha T (2011) Reducing landslide risk in communities: evidence from the Easter Caribbean. Appl Geogr 21(2):590–599CrossRefGoogle Scholar
  3. Baldwin RE (2009) The great trade collapse: causes, consequences, and prospects. Center for Economic Policy Research and VOXEU.org, GenevaGoogle Scholar
  4. Barclay P (2004) Trustworthiness and competitive altruism can also solve the “tragedy of the commons”. Evol Hum Behav 25:209–220CrossRefGoogle Scholar
  5. Barclay P (2011) Competitve helping increases with the size of biological markets and invades defection. J Theor Biol 281:47–55CrossRefGoogle Scholar
  6. Barclay P, Willer R (2006) Partner choice creates competitive altruism in humans. Proc R Soc B Biol Sci 274:749–753CrossRefGoogle Scholar
  7. Bems R, Johnson RC, Yi KM (2013) The great trade collapse. Annu Rev Econom 5:375–400CrossRefGoogle Scholar
  8. Bernard HR (2002) Research methods in anthropology: qualitative and quantitative approaches, 3rd edn. Altimira Press, Walnut Creek, CAGoogle Scholar
  9. Borgatti SP, Everett MG, Freeman LC (2002) Ucinet for windows: software for social network analysis. Analytic Technologies, HarvardGoogle Scholar
  10. Borgerhoff-Mulder M, Coppolillo P (2005) Conservation: linking ecology, economics, and culture. Princeton University Press, PrincetonGoogle Scholar
  11. Boyd R, Richerson PJ (2009) Culture and the evolution of human cooperation. Philos Trans R Soc B 364(1533):3281–3288CrossRefGoogle Scholar
  12. Caribbean Development Bank (2003) Country poverty assessment—final report, vol 1. Halcrow Group Limited, LondonGoogle Scholar
  13. Caribbean Development Bank (2010) Country poverty assessment—Dominica, vol 1. Kairi Consultants Limited, TunapunaGoogle Scholar
  14. Cervantes-Godoy D, Kimura S, Antón J (2013) Smallholder risk management in developing countries. OECD food, agriculture and fisheries papers, No. 61. OECD Publishing, Paris. http://dx.doi.org/10.1787/5k452k28wljl-en
  15. Demyttenaere JCR (2012) The new European Union Flavouring Regulation and its impact on essential oils: production of natural flavouring ingredients and maximum levels of restricted substances. Flavour Frag J 27:3–12CrossRefGoogle Scholar
  16. Demyttenaere JCR (2016) Recent EU legislation on flavors and fragrances and its impact on essential oils. In: Hünsü K, Baser C, Buchbauer G (eds) Handbook of essential oils: science, technology, and applications, 2nd edn. CRC Press, Boca Raton, pp 1071–1088Google Scholar
  17. Dirks R (1972) Networks, groups, and adaptation in an Afro-Caribbean community. Man 4:565–585CrossRefGoogle Scholar
  18. Downey SS (2010) Can properties of labor-exchange networks explain the resilience of swidden agriculture? Ecol Soc 15(4):15CrossRefGoogle Scholar
  19. Duhaime G, Searles E, Usher PJ, Myers H, Fréchette P (2004) Social cohesion and living conditions in the Canadian arctic: from theory to measurement. Soc Indic Res 66:295–317CrossRefGoogle Scholar
  20. Erasmus CJ (1956) Culture, structure, and process: the occurrence and disappearance of reciprocal farm labor. Southwest J Anthropol 12(4):444–469CrossRefGoogle Scholar
  21. European Chemicals Agency (2009) Questions & Answers on Regulation (EC) No 1272/2008 on classification, labelling and packaging of substances and mixtures. ECHA, HelsinkiGoogle Scholar
  22. Fafchamps M (2006) Development and social capital. J Dev Stud 42(7):1180–1198CrossRefGoogle Scholar
  23. Fafchamps M, Gubert F (2007) The formation of risk sharing networks. J Dev Econ 83:326–350CrossRefGoogle Scholar
  24. Fu F, Hauert C, Nowak MA, Wang L (2008) Reputation-based partner choice promotes cooperation in social networks. Phys Rev E 78:1–8Google Scholar
  25. Gilligan DO (2004) The economics of agricultural labor exchange with evidence from Indonesia. Unpublished dissertation. University Park, University of MarylandGoogle Scholar
  26. Harvey CA, Rakotobe ZL, Rao NS, Dave R, Razafimahatratra H, Rabarijohn RH, Rajaofara H, MacKinnon JL (2014) Extreme vulnerability of smallholder farmers to agricultural risks and climate change in Madagascar. Philos Trans R Soc B 369(1639):20130089CrossRefGoogle Scholar
  27. Honychurch L (1995) The Dominica story: a history of the Island. Macmillian Education, OxfordGoogle Scholar
  28. Horowitz MM (1967) Morne-Paysan: peasant village in Martinique. Holt, Reinhart, and Winston, New YorkGoogle Scholar
  29. IFAD (2013) Smallholders, food security, and the environment. International Fund for Agricultural Development, RomeGoogle Scholar
  30. IMF (2004) Dominica: interim poverty reduction strategy paper. International Monetary Fund, Washington, DCGoogle Scholar
  31. Jaeggi AV, Hooper PL, Beheim BA, Kaplan H, Gurven M (2016) Reciprocal exchange patterned by market forces helps explain cooperation in a small-scale society. Curr Biol 26:2180–2187CrossRefGoogle Scholar
  32. Lyle HF III, Smith EA (2014) The reputational and social network benefits of prosociality in an Andean community. Proc Natl Acad Sci USA 111(13):4820–4825CrossRefGoogle Scholar
  33. Macfarlan SJ (2010) The logic of labor exchange in a dominican village: competitive altruism, biologic markets, and the nexus of male social relationships. Unpublished doctoral dissertation, Washington State University, Pullman, WAGoogle Scholar
  34. Macfarlan SJ (2011) The dual-role method and ultimatum game performance. Field Method 23(1):102–114CrossRefGoogle Scholar
  35. Macfarlan SJ (2016) Social evolution: the force of the market. Curr Biol 26(16):R756–R758CrossRefGoogle Scholar
  36. Macfarlan SJ, Lyle H (2015) Multiple reputation domains and cooperative behavior in two Latin American communities. Philos Trnas R Soc B 370(1683):20150009CrossRefGoogle Scholar
  37. Macfarlan SJ, Quinlan RJ (2008) Kinship, family, and gender effects in the ultimatum game. Hum Nat 19:294–309CrossRefGoogle Scholar
  38. Macfarlan SJ, Remiker M, Quinlan RJ (2012) Competitive altruism explains labor exchange variation in a Dominican village. Curr Anthropol 35(1):118–124CrossRefGoogle Scholar
  39. Macfarlan SJ, Quinlan RJ, Remiker M (2013) Cooperative behaviour and prosocial reputation dynamics in a Dominican village. Proc R Soc B Biol Sci 280(1761):20130557. doi: 10.1098/rspb.2013.0557 CrossRefGoogle Scholar
  40. Macfarlan SJ, Walker RS, Flinn MV, Chagnon NA (2014) Lethal coalitionary aggression and long-term alliance formation among Yanomamö men. Proc Natl Acad Sci USA 111(47):1662–16669CrossRefGoogle Scholar
  41. McHale D, Laurie WA, Woof MA (1977) Composition of west Indian bay oil. Food Chem 2(1):19–25CrossRefGoogle Scholar
  42. McNamara JM, Bara Z, Fromhage L, Houston AI (2008) The coevolution of choosiness and cooperation. Nature 451:189–192CrossRefGoogle Scholar
  43. Moore MP (1975) Co-operative labour in peasant society. J Peasant Stud 2(3):270–291CrossRefGoogle Scholar
  44. Morton JF (2007) The impact of climate change on smallholder and subsistence agriculture. Proc Natl Acad Sci USA 104(50):19680–19685CrossRefGoogle Scholar
  45. Murphy S (2010) Changing perspectives: small-scale farmers, markets and globalization. International Institute for Environment and Development (IIED), LondonGoogle Scholar
  46. Nesse RM (2007) Runaway social selection for displays of partner value and altruism. Biol Theory 2(2):143–155CrossRefGoogle Scholar
  47. Noe R, Hammerstein P (1994) Biological markets: supply and demand determine the effect of partner choice in cooperation, mutualism, and mating. Behav Ecol Sociobiol 35:1–11CrossRefGoogle Scholar
  48. Noe R, Hammerstein P (1995) Biological markets. TrEE 10(8):336–339Google Scholar
  49. Ponte S (2000) From social negotiation to contract: Shifting strategies of farm labor recruitment in Tanzania under market liberalization. World Dev 28(6):1017–1030CrossRefGoogle Scholar
  50. Provinse JH (1937) Cooperative ricefield cultivation among the Siang Dyaks of Central Borneo. Am Anthropol 39(1):77–102CrossRefGoogle Scholar
  51. Quinlan MB (2004) From the bush: the frontline of health-care in a Caribbean village. Wadsworth, BelmontGoogle Scholar
  52. Quinlan RJ (2006) Gender and risk in a matrifocal Caribbean community. Am Anthropol 106(3):464–479CrossRefGoogle Scholar
  53. Quinlan RJ, Flinn MV (2005) Kinship, sex, and fitness in a Caribbean community. Hum Nat 16(1):32–57CrossRefGoogle Scholar
  54. Regulation (EC) No 1223/2009 of the European Parliament and of the Council of 30 November 2009 on cosmetic products (2009) Off J Eur Union L342:59–209Google Scholar
  55. Regulation (EC) No 1272/2008 of the European Parliament and of the Council on classification, labeling, and packaging of substances and mixtures, amending and repealing Directives 67/548/EEC and 1994/45/EC, and amending Regulation (EC) No 1907/2006 (2008) Off J Eur Union L353:1Google Scholar
  56. Regulation (EC) No 1334/2008 of the European Parliament and of the Council of 16 December 2008 on flavourings and certain food ingredients with flavouring properties for use in and on foods and amending Council Regulation (EEC No 1601/91, Regulations *EC) No 2232/92 and (EC) No 110/2008 and Directive 2000/13/EC (2008) Off J Eur Union L354:34–50Google Scholar
  57. Roberts G (1998) Competitive altruism: from reciprocity to the handicap principle. Proc R Soc Biol Sci B 265:427–431CrossRefGoogle Scholar
  58. Rosenberry W (1988) Political economy. Ann Rev Anthropol 17:161–185CrossRefGoogle Scholar
  59. Shelburne RC (2010) The global financial crisis and its impact on trade: The world and the European emerging economies. Discussion papers series 2010.2. United Nations Economic Commission for Europe, GenevaGoogle Scholar
  60. Sugiyama L (2003) Illness, injury, and disability among Shiwiar forager-horticulturalists: implications of health-risk buffering for the evolution of human life history. Am J Phys Anthropol 123(4):371–389CrossRefGoogle Scholar
  61. Sugiyama LS, Sugiyama MS (2003) Social roles, prestige, and health risk: social niche specialization as a risk buffering strategy. Hum Nat 14(2):165–190CrossRefGoogle Scholar
  62. Swindell K (1985) Farm labour. Cambridge University Press, CambridgeGoogle Scholar
  63. Tucker AO, Maciarello MJ, Adams RP, Landrum LR, Zanoni TA (1991) Volatile leaf oils of Caribbean Myrtaceae. I. Three varieties of Pimenta racemosa (Miller) J. Moore of the Dominican Republic and the commercial bay oil. J Essent Oil Res 3(5):323–329CrossRefGoogle Scholar
  64. United Nations (2011) Globally harmonized system of classification and labeling of chemicals (GHS), 4th edn. United Nations, New YorkGoogle Scholar
  65. United Nations Statistical Division (2017) Data retrieved 1 Jan 2017, http://comtrade.un.org/
  66. Waring TM, Kline MA, Brooks JS, Goff SH, Gowdy J, Janssen MA, Smaldino PE, Jacquet J (2015) A multilevel evolutionary framework for sustainability analysis. Ecol Soc 20(2):34CrossRefGoogle Scholar
  67. Waring TM, Goff SH, Smaldino PE (2017) The coevolution of economic institutions and sustainable consumption via cultural group selection. Ecol Econ 131:524–532CrossRefGoogle Scholar
  68. Wilson PJ (1969) Reputation and respectability: a suggestion for Caribbean ethnology. Man 4(1):70–84CrossRefGoogle Scholar
  69. Wilson PJ (1971) Caribbean crews: peer groups and male society. Caribb Stud 10(4):18–34Google Scholar
  70. Wilson PJ (1973) Crab antics: the social anthropology of English-speaking Negro societies of the Caribbean. Yale University Press, New HavenGoogle Scholar

Copyright information

© Springer Japan KK 2017

Authors and Affiliations

  1. 1.Department of AnthropologyUniversity of UtahSalt Lake CityUSA
  2. 2.Oregon Rural Practice-based Research NetworkOregon Health and Science UniversityPortlandUSA

Personalised recommendations