The Connected Past @ SAA 2013
Session 57: Evening Thursday April 4 at SAA 2013 in Honolulu, Hawaii
Download the full SAA programme here.
Chaired by Tom Brughmans (University of Southampton) and Barbara Mills (University of Arizona)
Discussant: Ian Hodder (Stanford University)
Presenters:
Mark Golitko and Gary Feinman
Herb Maschner, Jennifer Dunne and Spencer Wood
Ethan Cochrane
Shawn Graham
Barbara Mills, Matthew Peeples, Wm. R. Haas, Jr., Lewis Borck, and Jeffery Clark
Tom Brughmans, Simon Keay and Graeme Earl
Tim Kohler, Stefani Crabtree and Michael Berry
Angus Mol, Corinne Hofman and Menno Hoogland
Fiona Coward
Koji Mizoguchi
Session Abstract
Over the last decade the number of published archaeological applications of network methods and theories has increased significantly. This session will build on this increasing interest in networks among archaeologists by highlighting a number of research themes that deserve further exploration. Firstly, it aims to illustrate how particular archaeological research contexts can drive the selection and adaptation of formal network methods from the wide range of existing approaches, where possible through interdisciplinary collaboration. Secondly, papers in this session will address the role archaeological data can play in network methods, the decisions we are faced with when defining nodes and ties, and how our theoretical approaches can be expressed through formal methods incorporating empirical data. Thirdly, the session will address the compatibility of network theories and methods. Lastly, the potential of incorporating materiality within existing network approaches and the study of long-term network evolution will be discussed.
This session will address these themes through methodological or theoretical papers, and will further illustrate the potential of a networks perspective for archaeology in a number of innovative case-studies. It hopes to illustrate that approaches with an interdisciplinary scope but dominated by archaeological research contexts yield the most critical and useful archaeological network studies.
Presentations
Mark Golitko and Gary Feinman
Network analysis of Classic and Postclassic obsidian distribution in Mesoamerica
Changes in the structure of Mesoamerican trade networks between the Classic (~AD 250-800) and Postclassic periods (~AD 800-1500) have been raised as a key factor in the rises, falls, and transitions in Mesoamerican political geography, for instance the collapse of major inland centers in the Maya area around AD 850, termed the Classic Maya “collapse.” We evaluate these postulated shifts in the prehispanic economy using network analysis of obsidian distribution patterns on a Mesoamerica-wide scale, building on earlier work focused specifically on the Maya area of southeastern Mesoamerica. Methodologically, we address the utility of network analysis for analyzing prehistoric exchange systems and explore limitations of network reconstruction using archaeological materials.
Herb Maschner, Jennifer Dunne and Spencer Wood
Food-webs as network tools for investigating historic and prehistoric roles of humans as consumers in marine ecosystems
Humans lived on Sanak Island, Alaska, for over 6000 years. This fact motivated us to assemble a food-web describing the trophic interactions among species in the marine ecosystems of the Sanak Archipelago, integrated over thousands of years, based on a combination of field observation, experimentation, zooarchaeology, and ethnographic data. The food web is constructed of 513 taxa, 6774 feeding links, and an average of 13 links per taxon. We show that the humans are super-generalists, feeding directly on 122 taxa in the marine web. People are also super-generalists, extremely connected to other species, and highly omnivorous. They are #2 in path length and #5 in omnivory, and have short path lengths from all other species (1.76 links on average). 491 of 513 (96%) of species are within 2 links of humans. By feeding on many taxa across all trophic levels, humans have the potential to influence the persistence and stability of marine ecosystems. We present the largest food web ever created with humans as a key component of the total food web and will discuss these results and ways that food-web analyses can inform research on the ecology of humans in marine ecosystems.
Ethan Cochrane
Artifact Classification and Networks: A Case Study from the Southwest Pacific
The vast and rapid geographic spread of Lapita pottery across more than 50 islands, 200 sites, and 4,000 km of ocean in the southwest Pacific has stimulated research on Lapita networks for 50 years. Much of this research concerns the cultural relatedness of Lapita populations, ranges in its involvement with networks from dalliance to the explicit tools of graph theory, and has at times produced contradictory results. I argue here that to evaluate contradictory results and fruitfully use networks at any level of involvement we must pay special attention to archaeological classification (including node definition) and its links to explanatory theory. My argument is supported by comparisons of my own and others research on Lapita networks.
Shawn Graham
Reanimating Networks with Agent Modeling
In this paper, I present thoughts on how to reanimate networks extracted from Roman materials, using agent based models. In particular, I look at networks extracted from the epigraphy of the Roman construction trade in the Tiber Valley. I then use these as the substrate for the pattern of interactions in an agent based model to explore the likely consequences for information flow in such a(n artificial) society. Conversely, I explore then how agent models can be used to simulate past patterns of interaction, generating networks which we would could then use to map against archaeological materials. Many different processes could create the patterns we see archaeologically; agent based models used in this generative capacity allow us to narrow the field.
Barbara Mills, Matthew Peeples, Wm. R. Haas, Jr., Lewis Borck, and Jeffery Clark
Multiscalar Perspectives on Social Networks in the Late Prehispanic Southwest
The application of social network analysis (SNA) to archaeology is closely tied to historical trajectories and interactions occurring across widely varying social and spatial scales. Rather than seeing this as an impediment to the application of social network analysis in archaeology, we show how changing the regional scale of inquiry can lead to different yet complementary interpretations about the relationships among settlements. Using decorated ceramic frequency data from the Southwest Social Networks Project we present the analysis of three different spatial scales over time to show how the same social processes of migration in the 13th century followed by widespread religious movements in the 14th and 15th centuries were expressed in terms of their network characteristics. In the southern Southwest these processes resulted in a highly connected network with many long-distance connections, while in the northern Southwest networks were more discrete with more short-distance connections. We look at several network measures to better understand how these different network “textures” emerged and compare them to independently documented differences in ceramic production, population density, and migration histories.
Tom Brughmans, Simon Keay and Graeme Earl
Just points and lines? Exploratory network analysis from a Roman archaeology perspective
Many archaeological applications of formal network techniques consist of an exploration of empirically attested archaeological entities linked by relationships (of whatever nature the researcher considers meaningful). Among the most common issues with these exploratory approaches are how different data types can be used to create networks or validate hypothetical relational processes and how long-term change in connectivity can be explored. This paper will discuss these issues from a Roman archaeology perspective through a case study on urban connectivity in Roman Southern Spain.
Traditional approaches to the archaeology of Roman southern Spain have neglected the study of inter-urban connections. Iron Age (ca. 5th c.BC to 3rd c.BC) and Roman (ca. 3rd c.BC to 5th c.AD) sites as well as different archaeological data types are often studied independently, which is necessary for a critical understanding of these different sources. However, all these sources were also once part of a single long-term cultural process. A multi-scalar exploratory network method is introduced that aims to explore aspects of the changing interactions between 190 sites dated to a range of ten centuries as evidenced through ten archaeological data types. In doing this the potential and limitations of such an approach will be critically evaluated.
Tim Kohler, Stefani Crabtree and Michael Berry
Secrets of the Southwest Solved by Walkative Tree Rings
With appropriate cautions, tree-ring date distributions can provide proxies for human population size and rates of increase. Using the available corpus of tree-ring dates for the Southwest divided into a number of regions and time slices of varying length, we ask “what flows among which regions would best account for the distributions of tree-ring dates in each region?” Network analyses allow us to identify periods of relatively high, and low, flows, to characterize which regions are most connected at various times, and whether there is any sense in which some regions are more central than others. Our results will be compared with estimates of regional birth rates derived from an independent bioarchaeological proxy, Bocquet-Appel’s 15P5 measure.
Angus Mol, Corinne Hofman and Menno Hoogland
Remotely Local: A network model of the 14th century settlement of Kelbey’s Ridge, Saba
The settlement of Kelbey’s Ridge is located on Saba in the heart of the Northeastern Caribbean archipelago. During the past 25 years Saba has been the focus of intensive and extensive archaeological fieldwork undertaken by the Caribbean Research Group, Leiden University. Building on the archaeological relational datasets that have been collected during this period, this paper will investigate the role of Saba, specifically the site of Kelbey’s Ridge, in the Late Ceramic Age network of the Northeastern Antilles. Several unique feature of the island and the site testify to the fact that, although Saba itself is small (5 sq mi/13km2), its inhabitants were taking part in patterns of mobility and interaction that took place at the local, regional and interregional level. Through an ego-network approach the island of Saba will be shown to be represent a microcosm of overarching, 14th century network processes and dynamics. By doing so this paper will contribute to the evolving view of Caribbean Late Ceramic Age patterns of interactions, approaches that seek to integrate varied archaeological relational datasets, and discussions on the status of “islands as units of analysis” in archaeological network studies and beyond.
Fiona Coward
Getting to grips with the very earliest social networks: the challenges of using network methodologies to tackle Palaeolithic datasets
A series of recent analyses have demonstrated an expansion in the scale and complexity of social networks as larger and more permanent social aggregations grew up with the adoption of a more sedentary, village-based and ultimately agricultural lifestyle during the late Palaeolithic and early Neolithic. However, these developments may represent only a relatively late stage in a much more long-term and gradual process of scaling-up of social networks which took place over the course of hominin and human evolution, from the small, intimate and highly local communities of other primates, to the globally-connected, city-dwelling Homo urbanus of the contemporary world. In this paper I will argue that increasing engagement with material culture was central to this process, providing a mechanism by which humans were able to expand the scale of social relations well beyond those of their closest relatives. Network methodologies provide the best opportunity to investigate this; however, the nature of Palaeolithic datasets makes such analyses challenging; in this paper I will discuss the problems and also some potential ways forward, through which we may be able to gain vital insights into some of the very earliest social networks and the ways in which they subsequently evolved.
Koji Mizoguchi
Prestige goods and social hierarchization revisited: A formal network approach to the hierarchization of intercommunal relations in the Middle Yayoi period in northern Kyushu, Japan
This paper shows that the monopolization of contacts with the Other, signified by the monopolistic acquirement and distribution of prestige goods, can indeed be the “prime mover” in social hierarchization. This was demonstrated by examining discrepancies between the intercommunal hierarchy simulated by centrality analysis methods of formal network analysis and that indicated by the differences in the contents of grave goods from the richest burials in the individual polities comprising the northern Kyushu Yayoi cultural horizon. The former shows that the highest centrality scores are achieved by those polities that occupy the geographical core of the northern Kyushu region whereas the latter shows that one of the highest-ranked burials of the region is actually located on the northwestern periphery of the network of interactions reconstructed by the distribution of prestige goods, such as bronze mirrors, imported from the Han Chinese outpost of Lelang. The polity where the burial existed neither had the largest population concentration nor enjoyed any particular advantage in food/material production, strongly suggesting that its position at the top of the hiearchized network was achieved by its geographical location, advantageous for contacts with Lelang.