In ancient times, there were at least about twenty trading centers in the Netherlands that supplied Roman army bases. PhD student Mark Groenhuijzen made this discovery using computer models.
To protect their vast empire from incursions from enemy tribes, the Romans maintained a massive defense line between roughly AD 40 and 240. These limes (Latin for 'border') also ran through the Netherlands:from the point where the Oude Rijn flowed into the North Sea to Nijmegen, the then Noviomagus. Soldiers came and went along this line, all of whom, of course, had to eat and drink. “Of course the provisions did not come from Rome, but mainly from the immediate vicinity,” says Mark Groenhuijzen of the Free University, who examined the limes and supplies. He obtained his doctorate for this research on 13 November.
Unpaved paths
Archaeologists have been puzzling over exactly how the soldiers were supplied. A popular and very plausible theory is that there were certain centers from which food and drink went to the Romans. Merchants would bring goods to centrally located points from nearby small villages. And from there it went, according to the theory, to the Roman army bases. “Because that is a lot more efficient,” says Groenhuijzen. “A few assembly points were also found, which were supplied from the hinterland.”
Featured by the editors
MedicineWhat are the microplastics doing in my sunscreen?!
AstronomySun, sea and science
BiologyExpedition to melting land
But there must be more. It is unclear exactly where they were. Computer models offer a solution. This allows you to find out what the most likely places are for those collection points and what the transport of goods looked like. “We can't find anything archaeological from the roads from villages to those centers. Those were unpaved paths. But with computer models we can see what the most logical routes were.”
Using the computer models, Groenhuijzen uncovered those paths. He looked at thirty possible sites for collection centers, which archaeologists have identified as likely sites. Groenhuijzen concludes that it was plausible that twenty of these collection points were present in the Dutch river area.
To arrive at those conclusions, Groenhuijzen entered a lot of data into the models, such as different modes of transport. Goods were probably transported by horse or ox and wagon and small quantities were also carried on the back. Groenhuijzen also checked where places to live had already been found, how long the transport took and what the Netherlands looked like in ancient times. That's quite a difference. “The rivers were not diked then. Along the water you found embankments that were high and dry. People lived there and transport took place. It was swampy in between those ramparts.” After all this was entered into the computer, the model calculated the most logical places for assembly centers.
Predict where to dig
The main goal of this computer modeling research is to provide a new perspective on archaeology. “It is a new perspective on existing archaeological issues that allows us to disprove or confirm hypotheses. I want to contribute to the debate with this.”
Archaeologist and lime expert Jasper de Bruin of Leiden University is enthusiastic about Groenhuijzen's research. “This analysis method is another step forward and adds something to archaeology. For a long time we only looked at archaeological data per site and region. But the large networks that lie behind it, such as transport, roads and collection centers, are then difficult to map. This often involves processing big data, a lot of data. Thanks to computer models, this is now possible. We must be open to this in archeology and not cling to old, ingrained views. We have a beautiful, open field, and this new research fits in well with that.”
It would be very interesting if the computer models had a predictive value, continues De Bruin. That you can deduce exactly where a trading place or harbor was. Groenhuijzen sees that too. “Then we can send archaeologists there to dig. I would like to make the current computer model suitable for this, which will probably yield much more new data and discoveries.”