In a recent article, Fumihiro Sakahira and Hiro'omi Tsumura (2023) used social network analysis methods to analyze change in obsidian trade networks in Japan throughout the 13,000-year-long Jomon period. In the paper recommended here (Sakahira and Tsumura 2024), Social Network Analysis of Ancient Japanese Obsidian Artifacts Reflecting Sampling Bias Reduction they revisit that data and describe additional analyses that confirm the robustness of their social network analysis. The data, analysis methods, and substantive conclusions of the two papers overlap; what this new paper adds is a detailed examination of the data and methods, including use of bootstrap analysis to demonstrate the reasonableness of the methods they used to group sites into clusters.

Both papers begin with a large dataset of approximately 21,000 artifacts from more than 250 sites dating to various times throughout the Jomon period. The number of sites and artifacts, varying sample sizes from the sites, as well as the length of the Jomon period, make interpretation of the data challenging. To help make the data easier to interpret and reduce problems with small sample sizes from some sites, the authors assign each site to one of five sub-periods, then define spatial clusters of sites within each period using the DBSCAN algorithm. Sites with at least three other sites within 10 km are joined into clusters, while sites that lack enough close neighbors are left as isolates. Clusters or isolated sites with sample sizes smaller than 30 were dropped, and the remaining sites and clusters became the nodes in the networks formed for each period, using cosine similarities of obsidian assemblages to define the strength of ties between clusters and sites.

The main substantive result of Sakahira and Tsumura’s analysis is the demonstration that, during the Middle Jomon period (5500-4500 cal BP), clusters and isolated sites were much more connected than before or after that period. This is largely due to extensive distribution of obsidian from the Kozu-shima source, located on a small island off the Japanese mainland. Before the Middle Jomon period, Kozu-shima obsidian was mostly found at sites near the coast, but during the Middle Jomon, a trade network developed that took Kozu-shima obsidian far inland. This ended after the Middle Jomon period, and obsidian networks were less densely connected in the late and last Jomon periods.

The methods and conclusions are all previously published (Sakahira and Tsumura 2023). What Sakahira and Tsumura add in Social Network Analysis of Ancient Japanese Obsidian Artifacts Reflecting Sampling Bias Reduction are:

·       an examination of the distribution of cosine similarities between their clusters for each period

·       a similar evaluation of the cosine similarities within each cluster (and among the unclustered sites) for each period

·       bootstrap analyses of the mean cosine similarities and network densities for each time period

These additional analyses demonstrate that the methods used to cluster sites are reasonable, and that the use of spatially defined clusters as nodes (rather than the individual sites within the clusters) works well as a way of reducing bias from small, unrepresentative samples. An alternative way to reduce that bias would be to simply drop small assemblages, but that would mean ignoring data that could usefully contribute to the analysis.

The cosine similarities between clusters show patterns that make sense given the results of the network analysis. The Middle Jomon period has, on average, the highest cosine similarities between clusters, and most cluster pairs have high cosine similarities, consistent with the densely connected, spatially expansive network from that time period. A few cluster pairs in the Middle Jomon have low similarities, apparently representing comparisons including one of the few nodes on the margins on the network that had little or no obsidian from the Kozu-shima source. The other four time periods all show lower average inter-cluster similarities and many cluster pairs have either high or low similarities. This probably reflects the tendency for nearby clusters to have very similar obsidian assemblages to each other and for geographically distant clusters to have dissimilar obsidian assemblages. The pattern is consistent with the less densely connected networks and regionalization shown in the network graphs. Thinking about this pattern makes me want to see a plot of the geographic distances between the clusters against the cosine similarities. There must be a very strong correlation, but it would be interesting to know whether there are any cluster pairs with similarities that deviate markedly from what would be predicted by their geographic separation.

The similarities within clusters are also interesting. For each time period, almost every cluster has a higher average (mean and median) within-cluster similarity than the similarity for unclustered sites, with only two exceptions. This is partial validation of the method used for creating the spatial clusters; sites within the clusters are at least more similar to each other than unclustered sites are, suggesting that grouping them this way was reasonable.

Although Sakahira and Tsumura say little about it, most clusters show quite a wide range of similarities between the site pairs they contain; average within-cluster similarities are relatively high, but many pairs of sites in most clusters appear to have low similarities (the individual values are not reported, but the pattern is clear in boxplots for the first four periods). There may be value in further exploring the occurrence of low site-to-site similarities within clusters. How often are they caused by small sample sizes? Clusters are retained in the analysis if they have a total of at least 30 artifacts, but clusters may contain sites with even smaller sample sizes, and small samples likely account for many of the low similarity values between sites in the same cluster. But is distance between sites in a cluster also a factor? If the most distant sites within a spatially extensive cluster are dissimilar, subdividing the cluster would likely improve the results. Further exploration of these within-cluster site-to-site similarity values might be worth doing, perhaps by plotting the similarities against the size of the smallest sample included in the comparison, as well as by plotting the cosine similarity against the distance between sites. Any low similarity values not attributable to small sample sizes or geographic distance would surely be worth investigating further.

Sakahira and Tsumura also use a bootstrap analysis to simulate, for each time period, mean cosine similarities between clusters and between site pairs without clustering. They also simulate the network density for each time period before and after clustering. These analyses show that, almost always, mean simulated cosine similarities and mean simulated network density are higher after clustering than before. The simulated mean values also match the actual mean values better after clustering than before. This improved match to actual values when the sites are clustered for the bootstrap reinforces the argument that clustering the sites for the network analysis was a reasonable result.

The strength of this paper is that Sakahira and Tsumura return to reevaluate their previously published work, which demonstrated strong patterns through time in the nature and extent of Jomon obsidian trade networks. In the current paper they present further analyses demonstrating that several of their methodological decisions were reasonable and their results are robust. The specific clusters formed with the DBSCAN algorithm may or may not be optimal (which would be unreasonable to expect), but the authors present analyses showing that using spatial clusters does improve their network analysis. Clustering reduces problems with small sample sizes from individual sites and simplifies the network graphs by reducing the number of nodes, which makes the results easier to interpret.

Reference

Sakahira, F. and Tsumura, H. (2023). Tipping Points of Ancient Japanese Jomon Trade Networks from Social Network Analyses of Obsidian Artifacts. Frontiers in Physics 10:1015870. https://doi.org/10.3389/fphy.2022.1015870

Sakahira, F. and Tsumura, H. (2024). Social Network Analysis of Ancient Japanese Obsidian Artifacts Reflecting Sampling Bias Reduction, Zenodo, 10057602, ver. 7 peer-reviewed and recommended by Peer Community in Archaeology. https://doi.org/10.5281/zenodo.7969330

The archaeological record comes in many forms. Some, such as buried sites from volcanic eruptions or other abrupt sedimentary phenomena are perhaps the only ones that leave relatively clean snapshots of moments in the past. And even in those cases time is compressed. Much, if not all other archaeological record is a messy affair. Things, whatever those things may be, artifacts or construction works (i.e., features), moved, modified, destroyed, warped and in a myriad of ways modified from their behavioral contexts. Do we at some point say the record is worthless? Not worth the effort or continuing investigation. Perhaps sometimes this may be justified, but as Daffara and colleagues show, heavily impacted archaeological remains can give us clues and important information about the past. Thoughtful and careful prehistorians can make significant contributions from what appear to be poor archaeological records. 

            In the case of Daffara and colleagues, a number of important theoretical cross-sections can be recognized. For a long time surface archaeology was thought of simply as a way of getting a preliminary peak at the subsurface. From some of the earliest professional archaeologists (e.g., Kidder 1924, 1931; Nelson 1916) to the New Archaeologists of the 1960s, the link between the surface and subsurface was only improved in precision and systematization (Binford et al. 1970). However, at Hatchery West Binford and colleagues not only showed that surface material can be used more reliably to get at the subsurface, but that substantive behavioral inferences can be made with the archaeological record visible on the surface.

            Much more important are the behavioral implications drawn from surface material. I am not sure we can cite the first attempts at interpreting prehistory from the surface manifestations of the archaeological record, but a flurry of such approaches proliferated in the 1970s and beyond (Dunnell and Dancey 1983; Ebert 1992; Foley 1981).  Off-site archaeology, non-site archaeology, later morphing into landscape archaeology all deal strictly with surface archaeological record to aid in understanding the past. With the current paper, Daffara and colleagues (2024) are clearly in this camp. Although still not widely accepted, it is clear that some behaviors (parts of systems) can only be approached from surface archaeological record. It is very unlikely that a future archaeologist will be able to excavate an entire human social/cultural system; people moving from season to season, creating multiple long and short term camps, travelling, procuring resources, etc. To excavate an entire system one would need to excavate 20,000 km2 or some similarly impossible task. Even if it was physically possible to excavate such an enormous area, it is very likely that some of contextual elements of any such system will be surface manifestations. 

Without belaboring the point, surface archaeological record yields data like any other archaeological record. We must contextual the archaeological artifacts or features weather they come from surface or below. Daffara and colleagues show us that we can learn about deep prehistory of northern Italy, with collections that were unsystematically collected, biased by agricultural as well as other land deformations agents. They carefully describe the regional prehistory as we know it, in particular specific well documented sites and assemblages as a means of applying such knowledge to less well controlled or uncontrolled collections.

References

Binford, L., Binford, R. S. R., Whallon, R. and Hardin, M. A. (1970). Archaeology of Hatchery West. Memoirs of the Society for American Archaeology, No. 24, Washington D.C.

Daffara, S., Giraudi, C., Berruti, G. L. F., Caracausi, S. and Garanzini, F. (2024). First evidence of a Palaeolithic frequentation of the Po plain in Piedmont: the case of Trino (north-western Italy), OSF Preprints, pz4uf, ver. 6 peer-reviewed and recommended by Peer Community in Archaeology. https://doi.org/10.31219/osf.io/pz4uf

Dunnell, R. C. and Dancey, W. S. (1983). The siteless survey: a regional scale data collection strategy. In Advances in Archaeological Method and Theory, vol. 6, edited by Michael B. Schiffer, pp. 267-287. Academic Press, New York.

Ebert, J. I. (1992). Distributional Archaeology. University of New Mexico Press, Albuquerque.

Foley, R. A. (1981). Off site archaeology and human adaptation in eastern Africa: An analysis of regional artefact density in the Amboseli, Southern Kenya. British Archaeological Reports International Series 97. Cambridge Monographs in African Archaeology 3. Oxford England.

Kidder, A. V. (1924). An Introduction to the Study of Southwestern Archaeology, With a Preliminary Account of the Excavations at Pecos. Papers of the Southwestern Expedition, Phillips Academy, no. 1. New Haven, Connecticut.

Kidder, A. V. (1931). The Pottery of Pecos, vol. 1. Papers of the Southwestern Expedition, Phillips Academy. New Haven, Connecticut.

Nelson, N. (1916). Chronology of the Tano Ruins, New Mexico. American Anthropologist 18(2):159-180.

This paper (Sosic et al. 2024) explores the Neolithic landscape of the Sopot culture in Đakovština, Eastern Slavonija, revealing a network of settlements through a multi-faceted approach that combines aerial archaeology, magnetometry, excavation, and field survey. This strategy facilitates scalable research tailored to the particularities of each site and allows for improved representations of buried archaeology with minimal intrusion. 

Using the site of Tomašanci-Dubrava as an example of the overall approach, the study further explores the use of drone imagery for 3D surface modeling, revealing a consistent correlation between crop surface elevation during full plant growth and ground terrain after ploughing, attributed to subsurface archaeological features. Results are correlated with magnetic survey and test-pitting data to validate the micro-topography and clarify the relationship between different subsurface structures.

The results obtained are presented in a comprehensive way, including their source data, and are contextualized in relation to conventional cropmark detection approaches and expectations. I found this aspect very interesting, since the crop surface and terrain models contradict typical or textbook examples of cropmark detection, where the vegetation is projected to appear higher in ditches and lower in areas with buried archaeology (Renfrew & Bahn 2016, 82). Regardless, the findings suggest the potential for broader applications of crop surface or canopy height modelling in landscape wide surveys, utilizing ALS data or aerial photographs.  

It seems then that the authors make a valid argument for a layered approach in landscape-based site detection, where aerial imagery can be used to accurately map the topography of areas of interest, which can then be further examined at site scale using more demanding methods, such as geophysical survey and excavation. This scalability enhances the research's relevance in broader archaeological and geographical contexts and renders it a useful example in site detection and landscape-scale mapping.

References

Renfrew, C. and Bahn, P. (2016). Archaeology: theories, methods and practice. Thames and Hudson. 

Sosic Klindzic, R., Vuković, M., Kalafatić, H. and Šiljeg, B. (2024). Digital surface models of crops used in archaeological feature detection – a case study of Late Neolithic site Tomašanci-Dubrava in Eastern Croatia, Zenodo, 7970703, ver. 4 peer-reviewed and recommended by Peer Community in Archaeology. https://doi.org/10.5281/zenodo.7970703

Studies of the economy of the Roman Empire have become increasingly interdisciplinary and nuanced in recent years, allowing the discipline to make great strides in data collection and importantly in the methods through which this increasing volume of data can be effectively and meaningfully analysed [see for example 1 and 2]. One of the key aspects of modelling the ancient economy is understanding movement and transport costs, and how these facilitated trade, communication and economic development. With archaeologists adopting more computational techniques and utilising GIS analysis beyond simply creating maps for simple visualisation, understanding and modelling the costs of traversing archaeological landscapes has become a much more fruitful avenue of research. Classical archaeologists are often slower to adopt these new computational techniques than others in the discipline. This is despite (or perhaps due to) the huge wealth of data available and the long period of time over which the Roman economy developed, thrived and evolved. This all means that the Roman Empire is a particularly useful proving ground for testing and perfecting new methodological developments, as well as being a particularly informative period of study for understanding ancient human behaviour more broadly. This paper by Page [3] then, is well placed and part of a much needed and growing trend of Roman archaeologists adopting these computational approaches in their research. 

Page’s methodology builds upon De Soto’s earlier modelling of transport costs [4] and applies it in a new setting. This reflects an important practice which should be more widely adopted in archaeology. That of using existing, well documented methodologies in new contexts to offer wider comparisons. This allows existing methodologies to be perfected and tested more robustly without reinventing the wheel. Page does all this well, and not only builds upon De Soto’s work, but does so using a case study that is particularly interesting with convincing and significant results. 

As Page highlights, Northern Italy is often thought of as relatively isolated in terms of economic exchange and transport, largely due to the distance from the sea and the barriers posed by the Alps and Apennines. However, in analysing this region, and not taking such presumptions for granted, Page quite convincingly shows that the waterways of the region played an important role in bringing down the cost of transport and allowed the region to be far more interconnected with the wider Roman world than previous studies have assumed.  

This article is clearly a valuable and important contribution to our understanding of computational methods in archaeology as well as the economy and transport network of the Roman Empire. The article utilises innovative techniques to model transport in an area of the Roman Empire that is often overlooked, with the economic isolation of the area taken for granted. Having high quality research such as this specifically analysing the region using the most current methodologies is of great importance. Furthermore, developing and improving methodologies like this allow for different regions and case studies to be analysed and directly compared, in a way that more traditional analyses simply cannot do. As such, Page has demonstrated the importance of reanalysing traditional assumptions using the new data and analyses now available to archaeologists. 

References

[1] Brughmans, T. and Wilson, A. (eds.) (2022). Simulating Roman Economies: Theories, Methods, and Computational Models. Oxford. 

[2] Dodd, E.K. and Van Limbergen, D. (eds.) (2024). Methods in Ancient Wine Archaeology: Scientific Approaches in Roman Contexts. London ; New York. 

[3] Page, J. (2024). Rivers vs. Roads? A route network model of transport infrastructure in Northern Italy during the Roman period, Zenodo, 7971399, ver. 3 peer-reviewed and recommended by Peer Community in Archaeology. https://doi.org/10.5281/zenodo.7971399

[4] De Soto P (2019). Network Analysis to Model and Analyse Roman Transport and Mobility. In: Finding the Limits of the Limes. Modelling Demography, Economy and Transport on the Edge of the Roman Empire. Ed. by Verhagen P, Joyce J, and Groenhuijzen M. Springer Open Access, pp. 271–90. https://doi.org/10.1007/978-3-030-04576-0_13

The paper entitled 'Social Network Analysis, Community Detection Algorithms, and Neighbourhood Identification in Pompeii' [1] presents a significant contribution to the field of archaeological network research, particularly in the challenging task of identifying urban neighborhoods within the context of Pompeii. This study focuses on the relational dynamics within urban neighborhoods and examines their indistinct boundaries through advanced analytical methods. The methodology employed provides a comprehensive analysis of community detection, including the Louvain and Leiden algorithms, and introduces a novel Convex Hull of Admissible Modularity Partitions (CHAMP) algorithm. The incorporation of a network approach into this domain is both innovative and timely.

The potential impact of this research is substantial, offering new perspectives and analytical tools. This opens new avenues for understanding social structures in ancient urban settings, which can be applied to other archaeological contexts beyond Pompeii. Moreover, the manuscript is not only methodologically solid but also well-written and structured, making complex concepts accessible to a broad audience.

In conclusion, this study represents a valuable contribution to the field of archaeology, particularly for archaeological network research. Their results not only enhance our knowledge of Pompeii but also provide a robust framework for future studies in similar historical contexts. Therefore, this publication advances our understanding of social dynamics in historical urban environments. The rigorous analysis, combined with the innovative application of network algorithms, makes this study a noteworthy addition to the existing body of network science literature. It is recommended for a wide range of scholars interested in the intersection of archaeology, history, and network science.

Reference

[1] Notarian, Matthew. 2024. Social Network Analysis, Community Detection Algorithms, and Neighbourhood Identification in Pompeii. https://doi.org/10.5281/zenodo.8305968