| Definitions |
|---|
| Agent-based model (ABM): A computational method that simulates the interactions of autonomous agents. |
| Geographic Information Systems (GIS): A system of software and hardware used to visualize, analyze, and interpret spatial data. |
| Geometric Morphometrics (GMM): A set of methods used to quantify and analyze the shape of objects. |
| Machine Learning (ML): A set of methods used to classify and predict patterns in data. |
| Network Science: A set of methods used to analyze relationships between entities. |
Evaluating the Relationship Between Material Culture and Social Networks in Archaeology
A Dissertation Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy
Social network analysis (SNA) is used to study the structure and dynamics of interactions among social actors at a variety of scales. It provides a robust set of methods and theoretical expectations linking network structures and positions to different kinds of outcomes for actors within a network (Peeples 2019, p. 6), and it is growing rapidly in archaeological research (Brughmans and Brandes 2017; Collar et al. 2015).
Network approaches offer a means of understanding how social affiliations develop and change through time. Importantly, these methods can accommodate analyses across multiple scales without reifying the traditional, spatially bounded categories associated with the culture-historical paradigm (Feinman and Neitzel 2020; Holland-Lulewicz 2021). Archaeological network studies typically use characteristics of material culture—such as source, style, or technology—as proxies for social relations. However, important questions remain regarding the relationship between specific types of material evidence and the social relations that network studies aim to capture.
For example, inferences based on structural positions such as brokerage (e.g., Peeples and Haas 2013) require strong supporting evidence for the social connections they assume. Because direct relationships are difficult to document in the archaeological record, relations identified through patterns in material culture should be viewed probabilistically (Peeples 2019, p. 29). This dissertation explicitly explores the complex relationships between social networks and the material proxies archaeologists use to model them.
While network science is the primary focus, this dissertation also employs a variety of computational methods—including agent-based modeling (ABM), geographic information systems (GIS), geometric morphometrics (GMM), and machine learning (ML). Definitions for these terms are provided in Table 1.
The term “network” (Figure 1) is used in two ways: as a reference to actual social relations among actors, and as a formal analytical construct (nodes and edges) used to abstract and analyze those relationships. In archaeology, networks built from material cultural patterns are not social networks in a strict sense. Rather, they represent proxies that map imperfectly onto social relationships. Clarifying the connection between these two uses of network is one of the primary goals of this dissertation.
To maintain clarity throughout the chapters, I refer to the patterns of social interaction as “interaction networks,” and to those built from material culture as “material culture networks” or “archaeological similarity networks” (Collar et al. 2015, p. 16). Defining and analyzing any kind of network in archaeology requires abstraction: a past phenomenon of interest must first be conceptualized in network terms and then translated into data—nodes and edges—that represent that concept (Brandes et al. 2013; Collar et al. 2015).
Figure 2 illustrates this process using a stylized model of technological diffusion, where network ties represent exchange pathways. Although network approaches have proven useful for understanding the structure of social relations in the past, it remains unclear how material culture-derived networks relate to the social processes they are intended to reflect (Peeples et al. 2016). Without stronger guiding assumptions, there is a risk of overinterpreting “networks” as social phenomena when they may reflect other processes.
The methods archaeologists use to construct material culture networks vary widely, including similarities in object counts, stylistic attributes, geochemical sourcing, and production techniques (e.g., Birch and Hart 2018; Bischoff 2018; Borck et al. 2015; Golitko et al. 2012; Mills, Clark, et al. 2013; Mills, Roberts, et al. 2013; Peeples 2018). There is also growing recognition that different kinds of interaction may leave different signatures in material culture (e.g., Gosselain 2000; Hegmon 2000; Hodder 1982; Wiessner 1983). Highly visible features may reflect emulation or stylistic convergence, while low-visibility traits often signal more direct forms of interaction, such as shared learning environments.
Any network analysis must ask: what kind of relationship is being modeled? Is it exchange, migration, shared production techniques, aesthetics, kinship, or a combination? This dissertation evaluates the relationship between material culture and social networks using both a virtual experiment (an agent-based model) and an empirical analysis focused on the Western Pueblo region of the U.S. Southwest (see Figure 3).
0.1 Style, Interaction, and Material Culture
The relationship between social interaction and material culture is a broad and complex issue, but one that must be addressed before any network analysis can be undertaken. This study was built on a theoretical model described by Peeples (2018) for exploring the trajectories of groups sharing collective social identities at large scales (Calhoun 1997; see also Calhoun 1995; Somers 1994; Stokke and Tjomsland 1996; Tilly 1978, 2001, 2002, 2004, 2005; White 1992, 2008). According to this perspective, the overall process of social identification can be understood as comprising two modes, relational and categorical identification (Peeples 2018, p. 8).
Relational identification refers to the process through which individuals identify with others through direct interactions such as kinship, exchange, and other kinds of frequent, direct interactions. Categorical identification refers to the process through which individuals identify with groups based on perceived similarities with others in those groups, such as ethnicity, religion, and nationality. Importantly, categorical identification does not necessarily involve direct interaction and thus may link many people who are unknown to each other. These modes of identification and their interplay provide a useful framework for interpreting how identities and group membership are expressed in different types and features of material culture, in part through the concept of style.
Style has been defined in several ways (e.g., Hegmon 1992, pp. 517–518; Macdonald 1990; Wiessner 1983, 1984, 1985). Carr (1995a, pp. 164–167) defines what he calls “material style” using several criteria, but primarily it is a material pattern with a restricted range of forms and a restricted spatio-temporal and contextual distribution. Other characteristics include the simultaneous expression of multiple styles and the existence of functionally equivalent alternatives—called isochrestic variation by Sackett (1982). Carr (1995a, p. 166) adds a few qualifications to Sackett’s original concept: functionally equivalent does not necessarily mean that the styles have identical performance characteristics, conscious choice is not necessary for different styles, and enculturation is only one of many factors involved in stylistic differences. From these definitions, it follows that patterns in the material culture in this study that are not restricted in space or time can be disregarded. Patterns that show restricted ranges can be used to infer social connections. The expression of multiple styles also indicates that evidence for both categorical and relational identities may be observed in a single artifact. These attributes should be distinguished in order to infer the types of social interaction that underlie patterns of similarity.
Technology is closely related to style, and it too has many definitions (Lechtman 1977; Lemonnier 1986, 1989, 1992; Pfaffenberger 1988, p. 41). Carr (1995a, p. 157) defines technology as “raw materials and production procedures.” The definition of technology used here includes several aspects: the processes and techniques of making or using a thing, the social and idiosyncratic context of the behavior, the material and ecological constraints, and the chaîne opératoire (Torres 2002) of the thing’s use-life. Understanding the modifications made throughout the life history of an object is essential for material culture studies, as the original form of the object can be obscured by subsequent use, modification, and discard (resharpening is particularly important for projectile points). All aspects of material culture technology should be considered before social inferences can be made.
Many studies investigated relationships between material culture and social identification (e.g., Carr 1995a, 1995b; Clark 2001; Dietler and Herbich 1998; Gosselain 1998, 2000, 2016; Hodder 1982; Huntley 2008; Lemonnier 1986; Lyons 2003; Neuzil 2008; Sassman and Rudolphi 2001; Stark et al. 1998; Wiessner 1983, 1997). Many of these studies found that variation in technological practices involved in processes such as building homes and making pottery, weapons, and other items are sometimes meaningful expressions of group social identity and sometimes are not. Meaningful variation must be identified empirically and case by case. There are some regularities, however, including that visibility is often helpful for understanding material style and social behaviors (Carr 1995b, p. 173; Wobst 1977). For example, the spread of low visibility technologies such as architectural construction techniques are interpreted as evidence of migration, while the spread of highly visible styles is more likely to represent emulation or diffusion (e.g., Clark 2001; Clark et al. 2008). Many of these same arguments can also be linked to the modes of identification described above. Categorical identification is not based on direct interactions among members of such groups but is often symbolized with highly visible material such as public architecture or objects used in public contexts (Peeples 2018, pp. 18–39). Relational identities, on the other hand, are formed through direct interactions and are more likely to be evidenced low-visibility attributes such as forming techniques or features found in domestic spaces see also (Carr 1995b; Clark 2001, pp. 6–22; Peeples 2018, pp. 18–39).
The visibility of an attribute also affects the level of transmission. Low-visibility attributes are more difficult to observe and copy than high-visibility attributes, although a sufficient degree of skill may be required. Figure 4 describes how this would affect material culture. Whereas high-visibility attributes may be easily copied, low-visibility attributes usually require careful observation of the crafting process. Communities of practice generate commonalities in low-visibility attributes representative of the strong relational connections within them. Constellations of practice represent greater diversity in practice but still share relational connections through historical, kinship, geographical, or other associations (Van Oyen 2016; Wenger 1998, pp. 126–133). As discussed, sharing high-visibility attributes indicates a categorical connection but does not require frequent interaction. Thus, levels of visibility can be used to infer categorical and relational connections among communities (see Upton 2019).
Understanding how material culture reflects social interaction requires both theoretical clarity and methodological precision. As discussed above, the concepts of style, technology, and visibility provide important frameworks for interpreting the social meaning of material variation. However, translating those frameworks into network models poses significant challenges. Not all material similarities indicate interaction, and the type of attribute—whether highly visible or embedded in production techniques—can lead to different interpretations. This ambiguity highlights the need for a systematic evaluation of how different forms of material culture relate to the social processes we seek to understand.
To address this challenge, the following chapters combine computational modeling with empirical case studies to explore how interaction networks can be inferred from material culture. The research design brings together network science, geometric morphometrics, agent-based modeling, and machine learning in an effort to build, test, and interpret archaeological networks in ways that account for both theory and complexity. The chapter overview below outlines how each stage of the dissertation contributes to this broader goal.
0.2 Research Design and Chapter Overview
This dissertation follows a hybrid format in which three standalone articles (Chapters 3–5) are integrated into a cohesive framework through this introduction and the concluding chapter. Each article contributes to the overarching question of how material culture networks can be constructed, validated, and interpreted in archaeological research, although chapter 3, as will be explained, is more targeted towards generating an appropriate dataset.
To evaluate the relationship between material culture networks and interaction networks, Chapter 2 introduces the ArchMatNet agent-based model (Bischoff et al. 2024; Bischoff and Padilla-Iglesias 2023; Padilla-Iglesias and Bischoff 2024). The model simulates social interactions and material production using two different material culture proxies: projectile points and ceramics. These proxies are parameterized to reflect different patterns of use and transmission. The simulation allows controlled testing of how well material culture networks reflect the underlying social networks—something not possible in empirical data alone.
Chapter 3 serves as an intermediate step between the agent-based model and its application. It presents a regional analysis of over 3,000 projectile points using geometric morphometrics and machine learning, primarily to develop a dataset that can be used for empirical network construction. This chapter also evaluates existing projectile point typologies and introduces an alternative classification based on measurable attributes and shape data.
Chapter 4 integrates projectile points, ceramics, and architectural data to build multilayer material culture networks for the Tonto Basin. It shows that different types of material culture generate different networks, likely reflecting differences in how those materials—and the knowledge to produce them—circulated in past communities. Given that pottery and projectile points are often associated with specific genders in the archaeological literature, these differences may even reflect gendered social networks.
Chapter 5 concludes the dissertation by evaluating the alignment between the empirical results and the patterns simulated by ArchMatNet. It also discusses broader implications for archaeological network analysis, particularly in light of growing interest in computational and reproducible research.
In many ways, this dissertation is an exercise in computational archaeology. As the field evolves, new methods are being adopted rapidly—but best practices for using them are still being developed. This project is an effort to combine several emerging approaches to address a core archaeological question: how do patterns in material culture relate to patterns of social interaction? At the same time, it critically evaluates the assumptions and outcomes of those methods, with the goal of contributing to the growing body of work on networks in archaeology.
Birch, J., & Hart, J. P. (2018). Social Networks and Northern Iroquoian Confederacy Dynamics. American Antiquity, 83, 13–33. https://doi.org/10.1017/aaq.2017.59
Bischoff, R. J. (2018). A Temporal and Spatial Analysis of San Juan Red Ware. Master’s thesis, Department of Anthropology, Brigham Young University, Provo, Utah. https://scholarsarchive.byu.edu/etd/7003/
Bischoff, R. J., & Padilla-Iglesias, C. (2023). A description and sensitivity analysis of the ArchMatNet agent-based model. PeerJ Computer Science, 9, e1419. https://doi.org/10.7717/peerj-cs.1419
Bischoff, R. J., Padilla-Iglesias, C., & Gravel-Miguel, C. (2024). Evaluating the Effects of Randomness on Missing Data in Archaeological Networks. Accepted to CAA 2021 Conference: Limassol, Cyprus. https://doi.org/10.31235/osf.io/vdy9b
Borck, L., Mills, B. J., Peeples, M. A., & Clark, J. J. (2015). Are Social Networks Survival Networks? An Example from the Late Pre-Hispanic US Southwest. Journal of Archaeological Method and Theory, 22, 33–57. https://doi.org/10.1007/s10816-014-9236-5
Brandes, U., Robins, G., McCranie, A., & Wasserman, S. (2013). What is Network Science? Network Science, 1, 1–15. https://doi.org/10.1017/nws.2013.2
Brughmans, T., & Brandes, U. (2017). Visibility Network Patterns and Methods for Studying Visual Relational Phenomena in Archeology. Frontiers in Digital Humanities, 4.
Calhoun, C. J. (1995). Critical Social Theory: Culture, History, and the Challenge of Difference. Blackwell. https://play.google.com/store/books/details?id=_WgDnwEACAAJ
Calhoun, C. J. (1997). Nationalism. Minneapolis, MN: University of Minnesota Press.
Carr, C. (1995b). A Unified Middle-Range Theory of Artifact Design. In C. C. & J. E. Neitzel (Eds.), Style, Society, and Person (pp. 171–258). Boston: Springer. https://doi.org/10.1007/978-1-4899-1097-4_6
Carr, C. (1995a). Building a Unified Middle-Range Theory of Artifact Design. In C. Carr & J. E. Neitzel (Eds.), Style, Society, and Person (pp. 151–170). Boston: Springer. https://doi.org/10.1007/978-1-4899-1097-4_6
Clark, J. J. (2001). Tracking Prehistoric Migrations: Pueblo Settlers Among the Tonto Basin Hohokam (Vol. 65). Tucson: University of Arizona Press. https://play.google.com/store/books/details?id=27Tk3xNT-6UC
Clark, J. J., Lyons, P. D., Hill, J. B., Neuzil, A. A., & Doelle, W. H. (2008). Immigrants and Population Collapse in the Southern Southwest. Archaeology Southwest, 22.
Collar, A., Coward, F., Brughmans, T., & Mills, B. J. (2015). Networks in Archaeology: Phenomena, Abstraction, Representation. Journal of Archaeological Method and Theory, 22, 1–32. https://doi.org/10.1007/s10816-014-9235-6
Dietler, M., & Herbich, I. (1998). Habitus, Techniques, Style: An Integrated Approach to the Social Understanding of Material Culture and Boundaries. In M. T. Stark (Ed.), The Archaeology of Social Boundaries (pp. 232–263). Washington, DC: Smithsonian Institution.
Feinman, G. M., & Neitzel, J. E. (2020). Excising Culture History from Contemporary Archaeology. Journal of Anthropological Archaeology, 60, 101230. https://doi.org/10.1016/j.jaa.2020.101230
Golitko, M., Meierhoff, J., Feinman, G. M., & Williams, P. R. (2012). Complexities of collapse: the evidence of Maya obsidian as revealed by social network graphical analysis. Antiquity, 86, 507–523. https://doi.org/10.1017/s0003598x00062906
Gosselain, O. P. (1998). Social and Technical Identity in a Clay Cristal Ball. In M. Stark (Ed.), The Archaeology of Social Boundaries (pp. 78–106). Washington, D.C.: Smithsonian Institution Press.
Gosselain, O. P. (2000). Materializing Identities: An African Perspective. Journal of Archaeological Method and Theory, 7, 187–217. https://doi.org/10.1023/A:1026558503986
Gosselain, O. P. (2016). The World is Like a Beanstalk: Historicizing Potting Practice and Social Relations in the Niger River Area. In A. P. Roddick & A. B. Stahl (Eds.), Knowledge in Motion: Constellations of Learning Across Time and Place (pp. 36–66). Tucson: University of Arizona Press. https://muse.jhu.edu/book/44808. Accessed 11 March 2019
Hegmon, M. (1992). Archaeological Research on Style. Annual Review of Anthropology, 21, 517–536. https://doi.org/10.1146/annurev.anthro.21.1.517
Hegmon, M. (Ed.). (2000). The Archaeology of Regional Interaction: Religion, Warfare, and Exchange across the American Southwest and Beyond. University Press of Colorado.
Hodder, I. (1982). Symbols in Action: Ethnoarchaeological Studies of Material Culture. Cambridge: Cambridge University Press. https://market.android.com/details?id=book-TSE9AAAAIAAJ
Holland-Lulewicz, J. (2021). From Categories to Connections in the Archaeology of Eastern North America. Journal of Archaeological Research, 29, 537–579. https://doi.org/10.1007/s10814-020-09154-w
Huntley, D. L. (2008). Ancestral Zuni Glaze-Decorated Pottery: Viewing Pueblo IV Regional Organization through Ceramic Production and Exchange (Vol. No. 72). Tucson: University of Arizona Press.
Lechtman, H. (1977). Style in Technology, Some Early Thoughts. In H. Lechtman & R. S. Merrill (Eds.), Material Culture: Style, Organization, and Dynamics of Technology (pp. 3–20). New York: West Publishing.
Lemonnier, P. (1986). The Study of Material Culture Today: Toward an Anthropology of Technical Systems. Journal of Anthropological Archaeology, 5, 147–186. https://doi.org/10.1016/0278-4165(86)90012-7
Lemonnier, P. (1989). Bark Capes, Arrowheads and Concorde: On Social Representations of Technology. In I. Hodder (Ed.), The Meaning of Things: Material Culture and Symbolic Expressions (pp. 156–171). London: Hyman, Unwin.
Lemonnier, P. (1992). Elements for an Anthropology of Technology. Ann Arbor: Museum of Anthropology, University of Michigan. https://philpapers.org/rec/LEMEFA
Lyons, P. D. (2003). Ancestral Hopi Migrations (First Edition edition.). Tucson: University of Arizona Press.
Macdonald, W. K. (1990). Investigating Style: An Exploratory Analysis of some Plains Burials. In Uses of Style in Archaeology (pp. 52–60). Cambridge: Cambridge Univ. Press.
Mills, B. J., Clark, J. J., Peeples, M. A., Haas, W. R., Roberts, J. M., Hill, J. B., et al. (2013). Transformation of social networks in the late pre-Hispanic US Southwest. Proceedings of the National Academy of Sciences of the United States of America, 110, 5785–5790. https://doi.org/10.1073/pnas.1219966110
Mills, B. J., Roberts, J. M., Jr, Clark, J. J., Haas, W. R., Jr, Huntley, D. L., Peeples, M. A., et al. (2013). The Dynamics of Social Networks in the Late Prehispanic U.S. Southwest. In C. Knappett (Ed.), Network Analysis in Archaeology: New Approaches to Regional Interaction (pp. 181–202). Oxford: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199697090.003.0008
Neuzil, A. A. (2008). In the Aftermath of Migration: Renegotiating Ancient Identity in Southeastern Arizona. Tucson: University of Arizona Press.
Padilla-Iglesias, C., & Bischoff, R. J. (2024). Hunter-gatherer mobility patterns influence the reconstruction of social networks from archaeological assemblages. Journal of Archaeological Science: Reports, 59, 104798. https://doi.org/10.1016/j.jasrep.2024.104798
Peeples, M. A. (2018). Connected Communities: Networks, Identity, and Social Change in the Ancient Cibola World. Tucson: University of Arizona Press. https://market.android.com/details?id=book-fyw-DwAAQBAJ
Peeples, M. A. (2019). Finding a Place for Networks in Archaeology. Journal of Archaeological Research, 1–49. https://doi.org/10.1007/s10814-019-09127-8
Peeples, M. A., & Haas, W. R. (2013). Brokerage and Social Capital in the Prehispanic U.S. Southwest. American anthropologist, 115, 232–247. https://doi.org/10.1111/aman.12006
Peeples, M. A., Mills, B. J., Haas, W. R. J., Clark, J. J., & Roberts, J. M. J. (2016). Analytical Challenges for the Application of Social Network Analysis in Archaeology. In T. Brughmans, A. Collar, & F. Coward (Eds.), The Connected Past: Challenges to Network Studies in Archaeology and History (pp. 59–84). Oxford: Oxford University Press.
Pfaffenberger, B. (1988). Fetishised Objects and Humanised Nature: Towards an Anthropology of Technology. Man, 23, 236–252. https://doi.org/10.2307/2802804
Sackett, J. R. (1982). Approaches to Style in Lithic Archaeology. Journal of Anthropological Archaeology, 1, 59–112.
Sassman, K. E., & Rudolphi, W. (2001). Communities of Practice in the Early Ceramic Traditions of the American Southeast. Journal of Anthropological Research, 57, 407–425.
Somers, M. R. (1994). The narrative constitution of identity: A relational and network approach. Theory and society, 23, 605–649. https://doi.org/10.1007/bf00992905
Stark, M. T., Elson, M. D., & Clark, J. J. (1998). Social Boundaries and Technical Choices in Tonto Basin Prehistory. In M. T. Stark (Ed.), The Archaeology of Social Boundaries (pp. 208–231). Washington, D.C: Smithsonian Institution Press.
Stokke, H., & Tjomsland, M. (1996). Collective Identities and Social Movements. Bergen, Norway: Chr. Michelsen Institute.
Tilly, C. (1978). From Mobilization to Revolution. Reading, Massachusetts: Addison-Wesley.
Tilly, C. (2001). Mechanisms in Political Processes. Annual Review of Political Science, 4, 21–41. https://doi.org/10.1146/annurev.polisci.4.1.21
Tilly, C. (2002). Stories, Identities, and Political Change. Lanham, Maryland: Rowman & Littlefield. https://play.google.com/store/books/details?id=x465AAAAIAAJ
Tilly, C. (2004). Social Boundary Mechanisms. Philosophy of the Social Sciences, 34, 211–236. https://doi.org/10.1177/0048393103262551
Tilly, C. (2005). Identities, Boundaries, and Social Ties. Boulder, Colorado: Paradigm Publishers.
Torres, M. M. (2002). Chaîne Opératoire: The concept and its applications within the study of technology. Gallaecia, 21, 29–44. https://dialnet.unirioja.es/descarga/articulo/265493.pdf
Upton, A. J. (2019). Multilayer Network Relationships and Culture Contact in Mississippian West-central Illinois, A.D. 1200 - 1450 (PhD thesis). PhD dissertation, Department of Anthropology, Michigan State University, East Lansing.
Van Oyen, A. (2016). Historicising Material Agency: from Relations to Relational Constellations. Journal of Archaeological Method and Theory, 23, 354–378. https://doi.org/10.1007/s10816-015-9244-0
Wenger, E. (1998). Communities of Practice: Learning, Meaning, and Identity (Vol. 9). Cambridge: Cambridge University Press. https://valenciacollege.edu/faculty/development/teaching-learning-academy/documents/CommunityofPractice.pdf
White, H. (1992). Identity and Control: A Structural Theory of Social Action. Princeton, New Jersey: Princeton University Press.
White, H. (2008). Identity and Control: How Social Formations Emerge. Princeton, New Jersey: Princeton University Press.
Wiessner, P. (1983). Style and Social Information in Kalahari San Projectile Points. American Antiquity, 48, 253–276. https://doi.org/10.2307/280450
Wiessner, P. (1984). Reconsidering the Behavioral Basis for Style: a Case Study among the Kalahari San. Journal of Anthropological Archaeology, 3, 190–234. https://doi.org/10.1016/0278-4165(84)90002-3
Wiessner, P. (1985). Style or Isochrestic Variation? A Reply to Sackett. American Antiquity, 50, 160–166. https://doi.org/10.2307/280643
Wiessner, P. (1997). Seeking Guidelines through an Evolutionary Approach: Style Revisited among the !Kung San of the 1990s. In C. M. Barton & G. A. Clark (Eds.), Rediscovering Darwin (Vol. 7, pp. 157–176). Arlington: American Anthropological Association.
Wobst, H. M. (1977). Stylistic Behavior and Information Exchange. In C. E. Cleland (Ed.), For the Director: Research Essays in Honour of the Late James B.Griffin (Vol. 61, pp. 317–342). Ann Arbor: University of Michigan.