Zum Inhalt springen

Modeling and Simulating Past Minds and Networks: Dynamics of Religious Beliefs and Practices in the Graeco-Roman World

25-334| Tuesday, 3:30 p.m. | 128
Panel Chair: Esther Eidinow

Although recent scholarship in the social and cognitive sciences provide theoretical perspectives concerning the transmission of religious beliefs and practices across a population, these hypotheses have yet to be integrated. For instance, while epidemiology of representations is undoubtedly correct to stress the importance for a successful representation to be fitting to the human cognitive architecture, our understanding is incomplete without considering the nature of social links. Such links can be analyzed using novel approaches in network theory. Both epidemiological and network based approaches have been adopted within the study of ancient Graeco-Roman religions, it represents an ideal environment to integrate their respective implementations. In order to proceed with such an integration, historians must enrich their methodological arsenal. This panel introduces case studies offering ways to utilize and combine epidemiological and social network approaches to historical materials, while stressing the limits of particular tools, with particular reference to the scarce evidence available.

Justin Lane

Multi-Agent AI as a tool for understanding historical data: Modeling the formation of Early Christianity

This presentation offers an example of how theory, history, and multi-agent artificial intelligence (MAAI) can create an interdisciplinary approach to the study of historical religions, using Early Christianity as a case study. It presents a novel MAAI model, which utilizes the theory of divergent modes of religiosity or DMR (Whitehouse, 1995, 2000, 2004), that relies on empirical evidence to revise earlier attempts at modeling the theory (McCorkle & Lane, 2012; Whitehouse, Kahn, Hochberg, & Bryson, 2012). The model tests the theory’s generalizability and validity using historical data, namely the case of Early Christianity. By testing the model in conjunction with biblical, historical, and archaeological sources, we can begin to create a clearer picture of the possible dynamics within Early Christian religious groups. Specifically it utilizes social network approximation techniques drawing from cognitive and social approaches to the rise of Christianity (Stark, 1996). This incorporates both estimations of population sizes (Grove, Pearce, & Dunbar, 2012; Hill & Dunbar, 2003), limits on social network clusters (Dunbar, Duncan, & Nettle, 1995; Gonçalves, Perra, & Vespignani, 2011; Mcpherson, Smith-lovin, & Cook, 2001; Roberts, Wilson, Fedurek, & Dunbar, 2008), and the complexity of early Christian beliefs (Lane, 2013). This presentation concludes by addressing the methodological and theoretical issues inherent in the use of computer modeling of historical data as well as its ability to apply novel theories to historical data.

Vojtěch Kaše

Modelling Ritual Dynamics against the Data: Early Christian Meal Practices as a Test Case

Any invention, modification and selection of a ritual practice in a group and its stabilization over time or transmission of it from one group to another depends on a lot of factors. Method of agent-based modelling enables to consider relative influence of them in an explicit way and to “re-grow” particular processes of interest in an artificial computational environment. Despite of its artificialness, a comparison of simulation results with real world data can help to demonstrate probability of some suggested historical trajectories and, in that respect, to test particular hypotheses. Drawing on recent experimental evidence concerning intuitive evaluation of ritual efficacy, this paper takes into consideration relative influence of the so-called cognitive attraction of a ritual behavior. One agent-based simulation is discussed and compared to the data derived from literary sources concerning early Christian meal practices in the first four centuries CE.

Dalibor Papoušek

Pro-Jewish and Counter-Jewish Trends in the Spread of Early Christianity: Construction of Network Models

This paper reconsiders the influence of the Jewish heritage in the spread of early Christianity throughout the Greco-Roman Mediterranean. Following the new dating of Lukan writings up to the first half of the 2nd century (Pervo 2006), it assumes Marcionite Christianity to be a contemporaneous Christian trend using other networks for its spreading than that of Luke. Despite weak evidence, it is obvious that these two trends maintained different attitudes to the Jewish background. This paper tries to design models using Jewish diaspora networks (Collar 2013) for Lukan Christianity, which remained open to the Jewish tradition, and trade and maritime networks for Marcionite Christianity, which might use the infrastructure provided by its founder´s shipping company. Critical evaluation of these models can help solve the question to what extent the spread of Christianity was influenced by the Jewish diaspora networks and how other networks may have been employed in this process

István Czachesz

Cognitive Science and Network Theory in the Study of Early Christian Origins

This presentation offers a case study of the application of network theory to the analysis of historical texts. I will speak about the generation and analysis of word co-occurrence networks in the Greek text of the New Testament, using examples from past and ongoing studies, and pay particular attention to the historical and psychological validity of such models. Whereas word co-occurrence networks can be thought of as statistical models of (large) textual corpora, at this place I will outline an approach working with smaller textual units, based on reading-theory and memory studies. As I have shown in previous publications, node and link centrality measures in word co-occurrence networks of biblical passages yield interesting observations about central thematic and compositional features of such passages (that go beyond the insights gained from traditional methods based on word frequency and concordances) as well as open up ways to reconsider how familiarity with certain texts influences the reading of other texts (offering new, quantitative perspectives on intertextuality). In terms of the psychological validity of the models, the cultural context of the first-century Mediterranean has to be taken into account. This implies that the underlying psychological model of textual reception needs to be informed by insights from orality studies. Previous cognitive psychological work on memory in oral transmission is especially helpful in building culturally informed models that do justice to the ways contemporary audiences typically processed the text of the Gospels. Further, research on working memory, including empirical results on working memory span in sentence processing, can be integrated into the model in order to increase its cognitive psychological validity. Finally, I will consider how the syntactic features of Koine Greek (the dialect of the New Testament) can be taken into consideration to fine-tune the creation of networks. The textual examples are selected from well-known passages of the synoptic Gospels, such as the Sermon of the Mount (Matthew 5-7) and its Lukan parallels (Luke 6:17-49). Based on these examples, the relationship between network models, on the one hand, and traditional philological approaches, on the other hand, will be addressed.