Reconstructing Archaeological Networks with Structural Holes.

Area of Science:

• Archaeological Network Analysis
• Computational Social Science
• Social Network Analysis

Background:

• Traditional methods for inferring archaeological networks rely on dyadic assumptions (e.g., distance, size), limiting the incorporation of complex tie dependencies.
• Existing models often fail to capture nuanced relationships, leading to less plausible reconstructions of historical networks.

Purpose of the Study:

• To investigate the application of network models that explicitly incorporate tie dependence for reconstructing unobserved archaeological networks.
• To demonstrate the utility of exponential random graph models (ERGMs) in capturing complex network structures relevant to social theories.

Main Methods:

• Utilized exponential random graph models (ERGMs) to analyze network structures with explicit tie dependencies.
• Applied ERGMs to reconstruct networks consistent with Burt's theories on social capital, specifically closure and structural holes.
• Illustrated the methodology using empirical data from the Middle Bronze Age Aegean.

Main Results:

• Demonstrated that ERGMs can effectively model tie dependencies, leading to more plausible network reconstructions.
• Showcased the ability of these models to capture network configurations aligned with theoretical concepts like closure and structural holes.
• Provided a data-driven example of network reconstruction for the Middle Bronze Age Aegean.

Conclusions:

• Network models incorporating tie dependence offer a more sophisticated approach to reconstructing unobserved archaeological networks.
• ERGMs provide a powerful framework for generating historically coherent network structures based on social theory.
• This methodology enhances our understanding of past societal organization and interaction patterns.