Published articles

Verhagen, P., J. Joyce & M.R. Groenhuijzen (eds.) 2019. Finding the Limits of the Limes: Modelling Demography, Economy and Transport on the Edge of the Roman Empire. Springer, Cham.

This open access book demonstrates the application of simulation modelling and network analysis techniques in the field of Roman studies. It summarizes and discusses the results of a 5-year research project carried out by the editors that aimed to apply spatial dynamical modelling to reconstruct and understand the socio-economic development of the Dutch part of the Roman frontier (limes) zone, in particular the agrarian economy and the related development of settlement patterns and transport networks in the area. The project papers are accompanied by invited chapters presenting case studies and reflections from other parts of the Roman Empire focusing on the themes of subsistence economy, demography, transport and mobility, and socio-economic networks in the Roman period.

The book shows the added value of state-of-the-art computer modelling techniques and bridges computational and conventional approaches. Topics that will be of particular interest to archaeologists are the question of (forced) surplus production, the demographic and economic effects of the Roman occupation on the local population, and the structuring of transport networks and settlement patterns. For modellers, issues of sensitivity analysis and validation of modelling results are specifically addressed. This book will appeal to students and researchers working in the computational humanities and social sciences, in particular, archaeology and ancient history.

Groenhuijzen, M.R. & P. Verhagen 2017. Comparing network construction techniques in the context of local transport networks in the Dutch part of the Roman limes. Journal of Archaeological Science: Reports 15: 235-251,

This paper aims to compare network construction techniques and evaluate which one achieves the best representation of a local provisioning system that connects rural settlements with military castella in the Dutch part of the Roman limes. Using an existing site dataset and a dataset of least-cost path reconstructions between all sites, a number of network construction techniques are described and applied, including maximum distance networks, proximal point networks, a Delaunay triangulation, a Gabriel graph and efficiency networks. They are evaluated using the network metric of average path length and ‘local’ average path length to reach the castella, along with other indicators. Ultimately the Gabriel graph and proximal point networks with a high number of neighbours proved to be the best representation through a good performance on the evaluated indicators as well as the presence of a number of downsides in the other networks, with the Gabriel graph being slightly better due to a smaller number of links needed. This study thus shows that the choice for a network construction technique in archaeological case studies is important and presents a possible strategy to approach such a problem.

Joyce, J. & P. Verhagen 2016. Simulating the Farm: Computational Modelling of Cattle and Sheep Herd Dynamics for the Analysis of Past Animal Husbandry Practices. Amsterdam, Vrije Universiteit Amsterdam (LAC2014 Proceedings),

This study forms part of a larger project analysing the development of the cultural landscape in the Dutch Roman limes region, including the development of the rural economy. The primary research aim in this study is to investigate changes in pastoral production as a result of differing population dynamics of sheep and cattle herds subjected to different animal husbandry practices. Differences between such practices have not been considered in previous models of pastoral production in the region where herds have been treated as static entities. In this study we simulated different animal husbandry strategies based on trends and developments inferred from the faunal record in a well-documented area of the limes zone. The results of these simulations were analysed for differences in population viability, as well as relative differences in production of meat, wool and manure, and relative requirements for pastureland requirements and labour. Our analysis shows that cattle and sheep herds exploited for products supplied by the animals whilst living (manure and traction from cattle, wool from sheep) exhibit higher growth rates than herds exploited for meat. In addition, they are able to supply larger quantities of calories. However these herds have higher requirements for labour and land, presenting a possibly limiting factor. We conclude that, in addition to complementing intensification and extensification of arable farming observed in the region, agriculturalists would have benefitted from herds with higher growth rates, particularly given their risk-aversion behavior during periods of economic uncertainty. These simulations will be subsequently used to produce more nuanced scenarios of pastoral production within agent-based modelling of the mixed rural economy in the Dutch Roman limes zone.

Verhagen, P., J. Joyce & M.R. Groenhuijzen 2016. Modelling the Dynamics of Demography in the Dutch Roman Limes Zone. Amsterdam, Vrije Universiteit Amsterdam  (LAC2014 Proceedings),

In this paper we present the first results of a study that aims to better understand and model the size and development of population in the Dutch Roman limes zone, as part of a larger research project investigating the development of the cultural landscape in the area. Earlier estimates of population size have only used general assumptions based on settlement density and supposed Roman army recruitment requirements and have not considered in any detail the factors influencing population growth and decline. This paper first presents the existing evidence for estimating population size in the area and then discusses the evidence for mortality and fertility estimates in the Roman period, necessary to better understand the large-scale demographic processes involved. From this evidence, new estimates for population size in the Early and Middle Roman period are calculated, using agent-based modelling to better understand the dynamics of population growth and the effects of recruitment of soldiers by the Roman army. It is concluded that earlier calculations underestimated the potential for population growth as well as the effect of forced recruitment on demography.

Verhagen, P., I. Vossen, M.R. Groenhuijzen & J. Joyce 2016. Now you see them, now you don’t: Defining and using a flexible chronology of sites for spatial analysis of Roman settlement in the Dutch river area. Journal of Archaeological Science: Reports 10, 309-321,

In this paper, we present a method to deal with poorly dated site inventories for purposes of (supra-) regional settlement pattern analysis. We created a site database for the Roman period in the Dutch River Area on the basis of existing digital inventories, and analyzed the quality of dating information provided in those inventories in order to better understand the development of settlement patterns during the Roman period. We did this by applying principles of aoristic analysis, dividing the time spans assigned to each registered artefact over the archaeological periods considered, and then simulating the probability of finds belonging to a specific archaeological period. Using this method, it is possible to judge the quality of dating information per site, and to analyze patterns of settlement density and site location while taking into account the uncertainty of dating information. Our analyses broadly confirm earlier studies, but they also provide a more solid, quantitative basis to previous work. Furthermore, they highlight the limitations of using the currently available information for settlement pattern analysis.

Groenhuijzen, M.R. & P. Verhagen 2016. Testing the Robustness of Local Network Metrics in Research on Archeological Local Transport Networks. Frontiers in Digital Humanities 3:6,

Abstract: With the increased application of network analysis in archeology to form hypotheses, particularly concerning the research on mobility, a need has arisen to validate the network analysis results. This paper presents a case study of a local transport network in the Dutch part of the Roman limes between 70 and 270 AD created using a least-cost approach, and tests the robustness of the local network metric of betweenness centrality and the archeological interpretation thereof. It is demonstrated that while the majority of sites have a robust and thus reliable betweenness centrality, there are still a large number of sites for which the network measurements are very dependent on the precise structure of the network present. Testing robustness of network analysis results thus proves a useful tool for both validating the network modeling results and the archeological interpretations of that network.

Groenhuijzen, M.R. & P. Verhagen 2015. Exploring the dynamics of transport in the Dutch Limes. eTopoi Journal for Ancient Studies, Special Volume 4, pp. 15-47.

Abstract: Very little is known about local scale transport in the Roman period in the Netherlands, and therefore this paper aims to reconstruct and examine local transport networks through an integrative approach, using data and research from palaeogeography, physiology, archaeology and computational archaeological approaches. This study particularly focusses on the role of different modes of transport and the role of the natural environment. Firstly, a palaeogeographic reconstruction is presented encompassing both the natural and the cultural landscape. Transport networks for several modes are then reconstructed by calculating least cost paths that incorporate functions from physiological research. These networks are analysed using standard network analytical procedures. Interesting interpretations can be inferred from the results of these procedures, including some relating to the different characteristics of the transport modes but notably also the relative important role of stone-built settlements in the networks and the relative lack of control that the Roman forts have over the transport network.