The Potosí principle: religious prosociality fosters selforganization of larger commnities under extreme natural and economic conditions

The Potosı ´ principle: religious prosociality fosters selforganization of larger commnities under extreme natural and economic conditions Juan Luis Sua ´rez and Shiddarta Va ´squez The CulturePlex Lab, Faculty of Arts and Humanities, University of Western Ontario, 1151 Richmond Street, London, ON, Canada N6A 3K7 Fernando Sancho-Caparrini Department of Computer Science and Artificial Intelligence, E.T.S. Ingenierı ´a Informa ´tica, Universidad de Sevilla, Av. Reina Mercedes, s/n, 41012, Sevilla, Spain Abstract We show how in colonial Potosı ´ (present-day Bolivia) social and political stabil-ity was achieved through the self-organization of society through the repetition of religious rituals. Our analysis shows that the population of Potosı ´ develops over the time a series of cycles of rituals and miracles as a response to social upheaval and natural disasters and that these cycles of religious performance become crucial mechanisms of cooperation among different ethnic and religious groups. Our methodology starts with a close reading and annotation of the Historia de Potosı ´ by Bartolome ´ Arzans. Then, we model the religious cycles of miracles and rituals and store all social and cultural information about the cycles in a multirelational graph database. Finally, we perform graph analysis through traversals queries in order to establish facts concerning social networks, historical evolution of behaviors, types of participation of miraculous characters according to dates, parts of the city, ethnic groups, etc. It is also important to note that the religious activity at the group level gave native communities a way to participate in the social life. It also guaranteed that the city performed its role as producer of silver in the global economic structure of the Spanish empire. This case proves the importance of religion as a mechanism of stability and self-organization in periods of social or political turbulence. The multidisciplinary methodology combining traditional humanistic techniques with graph analysis shows a great potential for other sociological, historical, and literary problems. became a crucial place for Spain’s economic interests in America (Bakewell, 1975; Flynn and Gira ´ldez, 2002; Hamilton, 1934) and a testing laboratory for Wright’s hypothesis. First, the growth of the population was uncontrolled from the outset, resulting in chaos with regard to urban planning and health (Arza ´ns de Orsu ´a y Vela, 1965), with its cyclic corrective elements in natural disasters. Second, Potosı ´ became a key node in the economic structure not only of the Spanish Crown but also of the first global trade network (Bakewell, 1975; Elliott, 1963; Flynn and Gira ´ldez, 2002). Third, in social terms, the number of different groups multiplied very quickly, resulting in a Potosı ´that never was locally self-sufficient and that turned increasingly complex and interdependent (McNeill and McNeill, 2003), with constant conflicts and high levels of violence between groups, including among Spanish groups from different peninsular origin (Diamond, 2005), and in particular over the native populations working in the mines (Bakewell, 1984; Cole, 1985; Helmer, 1960). All of these factors had an immense impact on the religious behavior of the people of Potosı ´, who constantly turned to the supernatural to alleviate natural disasters, the abuses of the mita system of mining on the indigenous population, and the human disorders resulting from what were perceived to be ‘moral errors’ of the inhabitants. Religious prosociality is the idea that religions facilitate certain acts that benefit others even when there is a personal cost to the actor (Tandeter, 1993). The ‘Potosı ´principle’ refers to the radical case that colonial Potosı ´presents in terms of the survival of a community constantly under threat from natural causes and human violence, but whose position in the imperial and global economic network does not allow it to disappear. The ‘Potosı ´ principle’ proves the importance of a sustained activity of religious rituals for the spread of successful behavioral strategies at the group level in a context of conquest and assimilation (Norenzayan and Shariff, 2008). A sustained ritualistic activity in Potosı ´ guaranteed communication with the supernatural and miraculous participation of the divinity in the survival of the community, while at the same time 1 Background It is normally assumed that there is a contradiction between religiously oriented communities and the ability of these societies to self-organize beyond hierarchical structures imposed and maintained by religious and political elites. On the contrary, certain forms of religiosity contribute to the stability of larger groups made up of socially and ethnically diverse communities. 1 It is a mistake to think that it is possible to acquire this stability only through top-down mechanisms imposed from the highest level of the social, religious, and political institutions in place. The conclusions of this line of thought usually imply that a predominantly religious society is always synonymous with a fundamentalist one in which clerics and officers control all forms of power, and that religion does not allow for the emergence of self-organization from the bottom-up. 2 The case we study here proves the opposite. In colonial Potosı ´ (present-day Bolivia), an overwhelmingly religious society is able to self-organize under adverse natural conditions and maximum economic stress in order to foster a much-needed intergroup stability and guarantee the survival of the larger community. Henrich et al. (2010) have shown that the existence of norms to maintain fairness among strangers is linked with the dissemination of institutions such as market integration and participation in world religions. Our case confirms the hypothesis that modern world religions may have contributed to the sustainability of large societies and of large-scale interactions and that they still have to be taken into account when explaining present-day political and economic dynamics. Wright (2009) has demonstrated that there is a connection between the forming of multinational empires such as the Hispanic Monarchy (Elliott, 1963; Elliott, 2006; Kamen, 2003) that have to accommodate within them different ethnic and religious groups, the survival of which depends on the perception of their relationships in terms of a nonzero sum, and the ability of the Abrahamic god to grow beyond the initial political community. Immediately after the discovery of silver by the Spanish in 1545, Potosı ´ (present-day Bolivia) allowed for the coming together of the different social and ethnic groups (Boyd and Richerson, 2002). In colonial Potosı ´, a growth of the Abrahamic god (Wright, 2009) coincided with the reinforcement of religious prosociality (McNeill and McNeill, 2003); the role played by the Catholic religion and its moralizing god (Heinrich et al., 2006) in maintaining a political community of planetary dimensions; as well as its role as a facilitator of social cohesion and in-group solidarity (Roes and Raymond, 2003). This is an extreme case of religious prosociality because rituals and miracles act as a constant cultural response to natural disasters, physical violence, and economic instability over the colonial history of the city on the part of the people of Potosı ´. 2Methodology As the main source of our data, we have used the History of Potosı ´(HP) (Arza ´ns de Orsu ´a y Vela, 1965), a text written by Bartolome ´de Arza ´ns (1676–1736), who compiles in 1343 pages the history of the famous city from 1545 to 1736. We have carried out a manual annotation of the text in order to select all the parts related to the cycle of miracles and rituals that we described below. Our model is defined over a multirelational graph database (Chartrand and Lesniak, 1986; Rodriguez and Shinavier, 2009), following the schema graph represented by Fig. 1 and allowing for the detailed inclusion of the objects in question as well as their semantic relations. These objects relate to a repeated cultural phenomenon in the history of Potosı ´by which a chaotic situation provoked by natural disasters, moral deviations, or social unrest is followed by a ritual to engage the divinity, then the performance of a miracle that restore order, and a final thanksgiving ritual. This has enabled us to define a homogeneous structure to represent all of the contextual sequences and relations expressed in the text by making use of the following elements: Miracle of the Chronicle, Chaotic Situation,Ritual of Request,Miraculous Character,Miracle,Ritual of Thanks,Characters, Place,Date, and Location in the Chronicle. From each of the 107 miracles—manually annotated—in the chronicle, we obtain a graph that represents the objects and specific relations occurring in the description formed in the HP of each of the miracles. The combination of all of these small graphs, making use of the same semantic context, provides us with a large network of interconnections to show the projection of the socio-religious network that Arza ´ns describes in his history (Fig. 2). Since, our model is not defined over a standard relational schema, but over a multirelational graph database (a graph with a classification on the edges and nodes, following the scheme represented by Fig. 1) (Chartrand and Lesniak, 1986; Mendelzon and Wood, 1989), we apply a special query system. As in the case of the query system based on relational algebra in Relational Databases (Rodriguez and Neubauer, 2010a) (resulting on the SQL language) (Rodriguez and Shinavier, 2009), in the case of graph databases an algebra is being developed in order to make queries and graph transformations. Traversals are the most common and useful operations on these query systems for graph databases (Mendelzon and Wood, 1989; Rodriguez and Geldart, 2009; Rodriguez and Neubauer, 2010a, b). They are just as important for graph databases as joins in the SQL queries are for relational ones. A traversal in a graph database is uniquely described by two data items: a start node and a specification for the schema of the path that will be used for the traversal. And the result always leads to a set of nodes. Depending on the structure of the graph being traversed, and the traversal you are interested in, that set may contain many, one or zero nodes or may even contain repeated nodes, indicating that more than one path, which meets the definition of the traversal, connecting the two nodes have been found. For example, if in the previous scheme, we traverse from a node representing a miraculous character to the nodes representing the required rituals performed with them, we will find that for some specific Miraculous Characters there are no required rituals while others obtain a set of repeated ones. For example, the following traversal (we do not explicit the types of the edges, since in our case they are determined by the types of the pairs of nodes they connect): <‘‘Miraculous Character Clas.’’, e 1 , ‘‘Miraculous Character’’, e 2 , ‘‘Chronicle Miracle’’, e 3 , ‘‘Miracle’’, e 4 , "Miracle Clas."> When applied to the node ‘Virgin’, this produces the following multiset (a set with repeated elements) of Miracle Classification nodes: {Apparition: 16, Healing:16, Salvation: 34, Nature: 7, Resurrection: 7} By applying the previous traversal to the complete graph (from the complete set of nodes of type ‘Miraculous Character Clas.’, to the set of nodes of type ‘Miracle Clas.’), we obtain the weighted graph in Fig. 3. It shows the different types of miracles that the types of miraculous characters have performed. In Fig. 4a–d, we show some more simple queries, we can make following the same methodology (a frequency filter has been applied to show only the more relevant items in the resulting graphs, where size of the node represents the total frequency of the node in the database, and thickness of edges show the weight of it). From the information obtained from the query, we can also obtain standard chart representations of the same data. Figures 5a and b show the relations between rituals and the main groups of characters participating in them. Other samples of data analysis with this methodology are represented in Figs 6–8d; Fig 9a and b. 3Discussion According to the HP, 107 miracles occurred in Potosı ´between 1545 and 1730. Each of these is associated with at least one ritual of request that individuals or groups use to request the intervention of the divinity and afterwards, it is associated with at least one ritual of thanksgiving. The inhabitants of Fig. 1 Schema Graph supporting the data extracted from the History of Potosı ´. Potosı ´relate to the Catholic divinity through a diversity of forms such as God, Jesus, Saints, and various denominations of Virgin Mary. Different individuals and social groups turn to different dedications and forms of saints and virgins in order to communicate with the divinity. This is consistent with Atran and Norenzayan’s (Atran and Norenzayan, 2004; Shariff and Norenzayan, 2007) assertions about the role of religion as a cultural innovation that aided by ritual makes possible the existence of large and cooperative moral communities of genetically unrelated individuals. By scanning the network of miracles and miraculous characters intervening on each occasion, together with the parallel network of social groups Fig. 2 Network showing the socio-religious network that Arza ´ns describes in his history. Fig. 3 Result of applying the simple path query from ‘Miraculous Char. Class’ to ‘Miracle Class’. participating in rituals (Norenzayan, 2010) and benefiting from the miracles, we are provided with an overall view of the religious network of colonial Potosı ´as well as its chronological evolution. The analysis of the resulting graph shows that in all cases, the trigger is an episode of chaos in which nature or human violence leads people to immediately turn to the divinity and to perform collective ritualistic acts in order to restore social and natural balance in the city. An example of a natural chaotic situation is the overflowing of the Kari Kari lagoon that in 1626 resulted in the flooding of Potosı ´and a huge loss of life and property. According to the chronicle of Arza ´ns, eleven miracles occurred in the city throughout that year. However, accidents relating to working in the mine placed many inhabitants of Potosı ´in life-threatening situations are numerous throughout the history of the city. The miraculous characters who intervened the most in the life of the city are the Candelaria Virgin (under several of her names) and God. Fig. 4 The traversals used for these queries are simple paths connecting the nodes: (a) Chaotic Situation–Character; (b) Miraculous Character–Character; (c) Miracle Class–Miraculous Character; (d) Rituals–Character. With regard to the characters in the social network, the groups of neighbors, 3 the people of Potosı ´, women, children, and the indigenous people are those with the greatest preponderance in the socio-religious life of Potosı ´. In Fig. 10, we can see how the miraculous characters are connected with the different social groups that participate in the rituals relating to the miracles resulting from them. In this figure, we can appreciate that the group of women has a special bond with the Candelaria Virgin of San Martı ´n, one of the parish churches of Potosı ´belonging to an area of indigenous people. When there are various ethnic and social groups participating in the ritual—what we have called the ‘people of Potosı ´’—then God the Father is the character who intervenes most in a miraculous event. Another important character with a variety of connections and her strong link with the neighbors is the Candelaria Virgin of another one of the parish churches of the Indians, that of San Pedro. If we turn our attention to the use of different types of rituals, we shall see that the ritual of thanksgiving is very often used by the neighbors, women, indigenous people, and the people of Potosı ´, whereas the Spanish and Creoles never appear as separate Fig. 5 (aand b) Relations between rituals and main groups of characters. groups, although they do appear on certain occasions as part of the ‘people of Potosı ´’ group. The case of Potosı ´highlights the decisive role played by religious group behaviors and rituals in maintaining inter-group cohesion and confirms how religious rituals, while costly for the members of the group, serve to reinforce the cooperative norms, ascertain the intentions of other individuals and subgroups in relation to the group’s activities and to facilitate the existence and stability of larger groups (Norenzayan and Shariff, 2008). First, a religious prosociality from the bottom-up emerges from Potosı ´’s history as different ethnic and social groups start the rituals in each particular case reinforcing inter and intra-groups solidarity and cooperation (Sossis and Alcorta, 2003; Sossis and Bressler, 2003) under difficult natural and social conditions. In other words, every group that initiates the religious relationship with the intention of solving a specific problem turns to one or more particular miraculous characters, but they are all part of the many Catholic Christianities of the Atlantic (Ferna ´ndez-Armesto, 2007; Mills, 2007). Fig. 6 Temporal distribution of miracles grouped by ‘Miracle Class’. Fig. 7 Chart comparing the chronological distribution of miracles and the silver production in Potosı ´(silver production according to Bakewell, 1975). Second, religious behavior allows for a better understanding of the mechanisms that explain self-organization and stability under ethnic and cultural variety in Potosı ´(Klein, 1992). Third,thecaseoftheVirginoftheCandelariashows a figure who has undergone a process of hybridization and presents both European and pre-Hispanic features. This helped indigenous peoples within the Spanish empire to domesticate the invaders’ beliefs and maintain a certain degree of continuity with their own past (Ferna ´ndez-Armesto, 2007). History teaches that there are different ways in which social and political action, and economic and religion dynamics may relate to one another. One of those cases is that of present-day political movements in favor of democracy in North African and Middle-East countries in which collective rituals organized around religion help to galvanize Fig. 8 (a) Participation of Characters in rituals related to the Candelaria Virgin, considering all the different denominations; (b) Main chaotic situations related to miracles by the Candelaria Virgin; (c) Participation of Characters in rituals related to the Candelaria Virgin, grouped by denominations; (d) Main chaotic situations and related miracles by the Candelaria Virgin grouped by denominations.