The main goal of this project was to learn something about how the Romans might have selected the terrain and position of their temporary camps. Although it is quite hard for us to know how important the Romans thought the selection of a camp location was — let alone how important it actually was to the success of the Roman military — it is clear that Vegetius and Pseudo-Hyginus, among other writers, thought that it was important enough to write down guidelines (and in the case of Vegetius, probably send those guidelines to the emperor1), and as such I think it is interesting and important to study the guidelines we have and the decisions that clearly must have been made on the ground by commanders regarding the placement of camps.
These decisions are quite alien to me, and I imagine that many people in the modern era might feel the same. However, we do make decisions that are at least thematically similar quite frequently — choosing a path on a wet or icy sidewalk, maneuvering through muddy areas in the spring, and deciding where our footsteps fall, while clearly on a far different scale of risk, involves some of the same basic assessing of a situation. Studying the places where the Roman army decided to rest is almost like studying the footsteps of a giant or a colossal machine navigating a marsh or an area of quicksand — watching out for dangers along the way, carefully (or perhaps not) placing one foot down after another.
The questions I wanted to answer are summed are fairly well in my primary research question: At what distance are Roman marching camps typically located from land features, such as roads, water sources, and elevation, that were considered important by sources from the time? An important part of this question is understanding what the ancient sources considered important — although I go into much more detail in the Temporary Camps and Camp Placement section, I will attempt to sum up their thoughts as follows.
The Roman sources appeared to think that access to water was a very important quality of a potential site, as both Vegetius2 and Pseudo-Hyginus3 specifically outline this, although they also both warn against getting too close to water because of flood risks. From a logistical standpoint, this makes a lot of sense — a Roman army that could include thousands of men as well as baggage animals and camp followers4 would certainly need a large quantity of water for drinking, not to mention other uses like washing or cooking. In fact, historian Jonathan P. Roth estimates that a Roman army of 40,000 men and the necessary pack animals for such a force would require over 80,000 and 600,000 liters of water per day, respectively5.
They also considered elevation, although it seems that they did not want to be overlooked rather than simply wanting the highest position, as Pseudo-Hyginus ranks flat plains above hills and mountains as suitable terrain6. They also both warn against various land features that in general would help an enemy approach the camp an a stealthy fashion, like wooded areas and gullies7 — however, these would be hard to analyze directly and would have greatly increased the scope of this project. Although neither source directly mentions roads, I feel that they are a worthy inclusion for this study — the army often built the roads at this time8, and it would be strange for them not to use their own creations.
In analyzing the positions of the temporary camps in this study, it seems clear that the Romans did not deviate significantly from the general maxims laid out by Vegetius and Pseudo-Hyginus. The most striking instance of this is the relationship between temporary camp locations and water sources. Of the 179 camps studied, all but 2 fell within 2 kilometers of a water source, and 75% of them were within 500 meters.
One might also note that there is evidence that flood risk was taken into account as well — less than 6% of the camps were within 100 meters of freshwater sources, but 25% are within 250 meters 75% are within 500 as stated above. However, this method does not take into account the difference in elevation between the river and the camp site or some sort of calculated flood risk for the area, and as such this evidence is partial at best. Conducting a more intricate study with those things in mind would be very interesting and useful in further study of this topic, but fell out of the range of my abilities in the time I had.
Elevation is another area where there is a correlation between the writings of our ancient sources and the actual placements of camps. The camps are not particularly high — in both the northern and southern sections of the SRTM topographic data, the average camp falls within the lowest 30% of elevations found in that area of the island.
A visual inspection also shows that the camp locations do not favor height so much as a lack of significantly higher surroundings. There are some exceptions to this rule, but these are typically either in mountain passes where the “high ground” would naturally be hard for an army to navigate or wider valleys where the high ground is somewhat far away. In many cases however, in many cases the locations chosen resemble Psuedo-Hyginus’ ideal camp or his second choice — a small rise that dominates a flat region and a flat plain, respectively9. One important thing to note is that the camps are not always found at the exact peak of the rise — my best guess is that in these cases the need for flat enough ground forced the camp planners to compromise, but without more contemporary data and a lack of imagery to study directly it is somewhat hard to say.
As for roads, there seems to be significant (but lesser) correlation with camp locations similar to my observations on Scotland’s water features. Over 45% of the temporary camp sites studied were within 2 kilometers of a road, which is significant, especially given the low amount of Roman road coverage in Scotland. Where there are roads, there are often large clusters or lines of camps that follow the path of the road — which is probably to be expected, because it makes little sense not to use the roads built at least in part for military use, and even in cases where the camps predated the construction of the road, the area where the road was built is likely a commonly used or advantageous path. I believe that it would be wise to expect a more pronounced relationship in areas with more major Roman roads.
Outliers and Interesting Cases
One of the most common situations in which the sites of temporary camps strayed from the guidelines laid out by the sources is mountainous areas. The camps tend to be located in mountain passes, which by nature are overlooked by the mountains they pass through. It would obviously be quite difficult for an army to march straight up a mountainside without taking advantage of these passes, so it is quite understandable that these situations were an “exception to the rules” or something to that effect.
It seems that even in these cases the Romans sometimes attempted to make the best of a bad geographic situation. There are several situations where a camp is on elevated ground, within the mountain pass but seemingly on the slope of one of the mountains. My best guess is that these camps are located at the highest point within the pass at which there was ground suitable enough to site a camp, so that they overlook the rest of the pass. However, I hesitate to put this at anything beyond conjecture without a more in-depth inspection of these sites and the landscape around them, via in-person viewing or more detailed photographic evidence.
There are also cases where a site was chosen in a low valley despite there being hills or rises somewhat close by. Again, it is hard to do more than conjecture as there are a variety of possible reasons for this. It could be that these are the relics of bad commanders or surveyors, but I think that it is more likely that sometimes the advantages of elevation were outweighed by some other pressing need or made irrelevant by the circumstances. For example, it could be that commanders wanting to move quickly or not expecting attacks would forgo the maxims of our sources in order to move more quickly, or perhaps they were simply the last-resort “unavoidable camps” that Pseudo-Hyginus describes10.
Perhaps the strangest and most interesting outlying case to examine is the two camps more than 2 kilometers away from a water source — Muirhouses and Kinglass Park. Interestingly, these two camps are actually adjacent to each other, located along the southern shore of the Firth of Forth (see photo on right). The camps seem to break from the advice of Vegetius and Psuedo-Hyginus in other ways as well — although they seem to be situated above the coast, they are both overlooked by an even bigger rise to the south.
The reasoning behind the precise positioning of these camps in not completely clear, but there is a reason to be in the general area — the path of the Antonine Wall is generally understood to have been just to the north of these camps, although the precise path is not well understood due to a lack of archaeological and visual evidence11. The category that these camps fall into is not speculated on by Rebecca Jones in Roman Camps in Scotland, but whether these were construction camps used during the building of the Antonine Wall, a natural stopping point before marching beyond the wall, or something else entirely, their positioning here makes much more sense with the added context of the nearby wall.
Overall, I found it very interesting that in most instances, the Roman military seems to have had methods of siting camps that were similar to those outlined in the works of Vegetius and Pseudo-Hyginus. Especially since Vegetius is unlikely to have had any military experince12, I expected that there would be more deviation from his guidelines or something that he had not picked up from the sources he examined. However, although he might not emphasize the necessity of a water source to the level that Pseudo-Hyginus or the data would indicate, everything else that I examined seems to indicate that Vegetius was not far from the traditional military wisdom of the day.
Another aspect of the results that I found interesting is the level to which these camp locations are correlated with nearby water sources. When I initially started looking into the placement of ancient fortifications, I thought that high elevation might be the most important factor. This may have stemmed from the fact that I was initially considering fortifications in general rather than specifically temporary camps, but even so I was surprised how important water seems in the selection of these camp sites, especially since high elevation does not appear to have been considered important in the way I had initially thought.
This research was fascinating, and much more nuanced than I had ever expected at the outset. However, I learned a lot during this project, and I look forward to learning more in the future, either through research of my own or that of others. I think that there is a lot more that could be done in this area and I hope to see more of that work soon.
- Stelten, Leo F, translator, Epitoma Rei Militaris By Flavius Vegetius Renatus (New York : Peter Lang, 1990, Print): XIII-XVI.
- Stelten, Epitoma Rei Militaris: 45-47.
- Gilliver, Catherine M., The Roman Art of War: Theory and Practice. A Study of the Roman Military Writers (London: University of London, University College London (United Kingdom), 1993, ProQuest, Web): 244-245.
- Breeze, David J., The Roman Army (New York: Bloomsbury Academic, 2016, Print): 42-47.
- Roth, Jonathan P., The Logistics of the Roman Army at War (264 B.C.-A.D. 235) (Leiden: Brill, 1999, Print): 119.
- Gilliver, The Roman Art of War: 244-245.
- See notes 2 and 3.
- Breeze, The Roman Army: 121-122.
- See note 3.
- See note 3.
- Jones, Rebecca H., Roman Camps In Scotland (Edinburgh: Society of Antiquaries of Scotland, 2011, Print): 244, 280.
- See note 2.