The Economic Impact of the Roman Army: Supplying Hungry Soldiers

In his Epitoma de Rei Militaris Vegetius states that "Armies are more often destroyed by starvation than battle and hunger is more savage than the sword".1 Vegetius's statement is based on the axiom that without proper supplies armies will not be able to function or even survive.2 It is this aspect of military organisation that has gained in importance in the academic studies of the Roman army over the last few years.3 In these studies it has been assumed that as a result of this military need for supplies the growth in demand for agricultural goods was extensive. Moreover, it has been assumed this increase in demand had an impact on local economies.4 However, the question remains, due to many uncertainties in the current knowledge of the exact process of supply, how significant this economic impact was.5 This paper focuses on this important topic and will attempt to analyse how significant the economic impact of the Roman army was on local economies of the province of Germania Inferior during the period of the first and second century AD as a result of its need for supplies to feed its soldiers?6 
 Before commenting on this research question, this paper will first discuss in three sections three essential research topics concerning the supply of the Roman army and its economic impact. The first section will assess which supplies were needed and in which quantities these were acquired. The second section will investigate which of these supplies were produced locally. The third section will discuss whether or not there is evidence for changes in the production of these local supplies that can be ascribed to the demand of the Roman army.
 In order to be able to comment on the significance of the economic impact of the army as a result of the supplies it needed, it is important to know which supplies were needed and in which quantities these supplies were acquired. The daily rations of soldiers, which are known from several literary sources, consisted mainly of grain, meat, vegetables, olive oil, wine or vinegar and salt.7 The main part of these rations was formed by grain of which the soldiers received circa 880 grams per day.8 The meat distributed amongst the soldiers was mostly beef, pork or mutton and soldiers probably received circa 160 grams of it a day.9 Basing his assumption on late Roman literary sources, Roth assumed that roman soldiers received 40 grams of olive oil and circa  a quarter of a litre of wine or vinegar a day.10 Beans and lentils constituted the most important species of the vegetables consumed by soldiers.11 Unfortunately, there is no reliable data on the daily portions of these vegetables or salt in literary sources. 
According to estimates of Alföldy the number of soldiers stationed in Germania Inferior in the first century AD ranged from circa 40.000 soldiers during the first half of the century to 35.000 at the end of this century. At the beginning of the second century AD the number of soldiers declined drastically due to the relocation of several army units. Consequently, the number of soldiers dropped according to Alföldy to circa 21.000 soldiers in the first two decades of the second century AD. This number probably did not decrease much during the remainder of this century.12 We can thus assume that the number of soldiers stationed in Germania Inferior in the first two centuries AD ranged between the 40.000 and 20.000 soldiers.  
Combining these numerical data gives an indication of the amount of supplies that were needed to supply the hungry soldiers of Germany Inferior: circa 13.000 metric tonnes of grain, 2.300 metric tonnes of meat, 584 metric tonnes of olive oil and 3.650.000 litres of wine or vinegar would have been needed on a yearly basis during the first century AD to supply the soldiers with rations. The amount needed for the supply of the army units of the second century AD would have been roughly half this amount. These amounts give an indication of the enormous demand for supplies needed by the army. For example,  for the supply of grain alone an area of circa 130 square kilometres would have been needed to produce the 13.000 metric tonnes needed.13
 As this paper aims to investigate the economic impact on local economies of Germania Inferior, it is important to distinguish between those supplies that could have been acquired from producers in the province of Germania Inferior and those that were acquired from other regions of the Roman empire. 
Wine and olive oil would have been imported from southern regions since the climate of this region was unsuitable for the production of grapes and olives.14  
Until recently the assumption was made on the basis of literary sources and environmental conditions that most grain acquired by the Roman army in Germania Inferior was produced mostly in areas to the south of the DRA, such as the province of Gallia Belgica and a small area of the most southern part of Germania Inferior.15 These assumptions are corroborated by the large absence in local settlements of spelt and bread wheat, which were the main grain species used by the army, and by finds of exotic weeds amongst supplies of grain in army camps.16 According to Kooistra it is likely that from the beginning of the second century AD most spelt and bread wheat was acquired from the most southern parts of Germania Inferior, this period seeing the development of large villas in this fertile loess area.17 However, recent archaeobotanic studies of camp sites and rural settlements in the DRA have given indications that the need for grain was partially provisioned by local economies of the DRA. These indications consist of finds in army camps of local grain and weed species.18 It thus seems likely that the grain acquired by the army was produced locally in large parts of Germania Inferior and not only in the most southern regions as was assumed until recently. 
There are several indications for the use of locally produced meat and animals. In the vicinity of castrum of Nijmegen slaughter facilities were found, confirming that living animals were transported to the vicinity of the camp and were slaughtered there. Davies also reports indications for the slaughter of animals in army camps along the Rhine.19  This makes it likely that animals were acquired from local sources, as transporting animals over long distances would have been problematic.20 A second indication is the correspondence between the relative importance of animal species in rural and military sites. The local production of meat centred mostly on the production of cattle and sheep, with the production of pigs forming only a small percentage.21 The same relative importance of cattle and sheep can be seen in the archaeozoological remains at army sites.22 This correspondence indicates  that the army acquired animals from the herds produced by local economies. Moreover, t would be logical if the army acquired meat from local sources since the transport of living or dead animals was problematic.23 It is thus very likely that local animals were used to supply the soldiers with meat.
 Having established which parts of the rations could have been produced locally, it is important to investigate if there are indications for the production of surplus or for the specialisation of production as these can give indications of the relative significance of the economic impact. 
Specialisation of the production of animals has been investigated in several studies. 24 Both Filean and Groot conclude that there are no indications for the development of new breeds, as had been thought previously. Groot also concludes that the mortality profiles of the animals show that animals supplied to the army were slaughtered at later ages, indicating that the army acquired animals that were no longer useful to local economies. If the demand for meat would have been sufficiently large one would expect local economies to start breeding larger cattle which would have supplied more meat. One would also expect to see mortality profiles in which cattle are slaughtered at a younger age when they would have been nearly full grown, making the economic yield highest.25 These facts might mean that the demand for meat by the army was not large enough to stimulate local economies to specialise their production. Groot concludes that it is quite possible that the demand of the army could have been supplied without the need of specialised production. In this case only extension of production would have been needed.26 Pollen data and developments in capacity of stables in local settlements indicate that such an extension of production probably took place from the second half of the first century AD.27 
 Pollen data and developments of the storage capacity also indicate that the production of grain was extended during the second century AD.28 Especially the developments in storage capacity show that surplus production could be quite large.29 The fact that the grain species found in these settlements are the same as in the army camps indicates that this surplus production could well have been intended for the army. The fact that the local economies produced a surplus which they sold on the market is furthermore indicated by the appearance of Roman goods in settlements, which could not have been produced by local economies itself. Groot concludes that the appearance of these goods increases from the beginning of the second half of the first century.30 One has to wonder, however, if a large part of the surplus production perhaps was intended for the growing urban community, which was also reliant on local economies for its supply of grain, and thus could not have been used by the army. A second caveat one has to keep in mind is the fact that it is unsure whether the total storage capacity of the investigated settlements was used for storing grain. 31 If not this could mean that the surplus production of grain was lower than estimated up until now. However, it seems likely that at least a part of this surplus production was intended for the military market.  
 After having discussed three essential topics it is possible to comment on the economic impact of the army. As numerical data about then number of soldiers and rations of these soldiers show, it can be concluded that the demand of the army must have constituted a large economic factor. It has to be concluded that the army did use grain and meat produced locally in the province of Germania Inferior. Even more important is that grain used by the army was not only produced in the most southern parts of the region, but in large parts of Germania Inferior. It thus seems likely that the demand for goods affected the whole economy of the province. Moreover,  it seems very likely that that the demand of the army was large enough to stimulate local economies to extend their production in order to produce surplus.  It is only from the second century on, however, that local economies really started to increase their production. It remains to be questioned how this fact can be explained when the beginning of the second century saw the departure of half the troops stationed in this province. It might be that the absolute production of surplus was never that high in most parts of the province and that the depart of this many troops did not decrease the significance of the army for local economies and the need for extension of production, or it might be that even though the demand of the army formed an initial impulse for extension of production the resulting surplus was consumed partially by the growing civilian population.


Bibliography

Alföldy, G. 1968, Die Hilfstruppen der römischen Provinz Germania inferior (Epigraphische Studien 6), Düsseldorf.

Cavallo, C. / Kooistra, L.I. / Dütting, M.K. 2008, 'Food supply to the Roman army in the Rhine delta in the first century A.D.', in: S. Stallibrass/R. Thomas (eds.), Feeding The Roman Army. The Archaeology of Production and Supply in NW Europe, Oxford, 69-81.

Davies, R.W. 1971, 'The Roman Military Diet', Britannia 2, 122-142.

Derks, T. / Roymans, N.  2011, 'Studying Roman Villa Landscapes in the 21st century. A multi-dimensional approach', in: T. Derks / N. Roymans (eds.), Villa Landscapes in the Roman North: economy , culture and lifestyle, Amsterdam, 1-44.

Filean, E.P. 2008, 'A biometric perspective on the size of cattle in Roman Nijmegen, The Netherlands: implications for the supply of urban consumers and the Roman army', in: S. Stallibrass/R. Thomas (eds.), Feeding The Roman Army. The Archaeology of Production and Supply in NW Europe, Oxbow, 99-115.

Groot, M. / Heeren, S. / L.I. Kooistra / W.K. Vos 2009, 'Surplus production for the market? The agrarian economy in the non-villa landscapes of Germania Inferior', Journal for Roman Archaeology 22, 231-252.

Groot, M. 2008a, Animals in ritual and economy in a Roman frontier community. Excavations in Tiel-Passewaaij (Amsterdam Archaeological Studies 12), Amsterdam.

Groot, M. 2008b, 'Surplus production of animal products for the Roman army in a rural settlement in the Dutch River Area', in: S. Stallibrass/R. Thomas (eds.), Feeding The Roman Army. The Archaeology of Production and Supply in NW Europe, Oxford, 83-98.

Herz, P. 2011, 'Finances and Costs of the Roman Army', in: P. Erdkamp (ed.), A Companion to the Roman Army, Chichester, 307-322.

Kehne, P. 2011, 'War- and Peacetime Logistics', in: P. Erdkamp (ed.), A Companion to the Roman Army, Chichester, 323-338.

Kooistra, L.I. 2009, 'The Provenance of Cereals for the Roman Army in the Rhine Delta. Based on Archaeobotanical Evidence', in: S. Zimmer (ed.), Kelten am Rhein: Akten des dreizehnten Internationalen Keltologiekongresses (Beihefte der Bonner Jahrbücher 58), 219-237.

Roth, J.P 1999, The Logistics of the Roman Army at War, Leiden.

Thomas, R. / Stallibrass, S. 2008, 'For starters: producing and supplying food to the army in the Roman north-west provinces', in: S. Stallibrass/R. Thomas (eds.), Feeding The Roman Army. The Archaeology of Production and Supply in NW Europe, Oxford, 1-17.



1 Vegetius Epitoma de Rei Militaris 3.3, translation by Monfort 2002, 70.
2 See also Davies 1971, 122.
3 Thomas/Stallibrass, 2008, 1.
4 Davies 1971, 122-123; Bender 2010, 165.
5 Roth 1999, 142; Whittaker 2004, 88.
6 Especially the Dutch River Area (DRA), which is a part of Germania Inferior, will function as a focal point, because different archaeozoological and archaeobotanical studies of this area have discussed the theme of supply to the army
7 Roth 1999, 14, 24-26, 33, 35, 37-39, 41; Herz 2011, 310-311, 315; Kehne 2011, 324; Davies 1971, 125. 
8 Roth 1999, 18-19; 21-24; Kehne 2011, 324, 334 (note 2).
9 Roth 1999, 27-32; Kehne 324.
10 Roth 1999, 35, 39-40, 43.
11 Davies 1971, 132-133; Roth 1999, 33
12 Alföldy 1968, 136-163.
13 Groot et al 2009, 235.
14 Cavallo et al 2008, 70; Kooistra 2009, 222.
15 Kooistra 2009, 219-220, 230-231; Cavallo et al 2008, 70, 77; Groot 2008b, 85; Groot et al 2009, 231-232, 247; Derks/Roymans 2008, 18-19.
16 Cavallo et al 2008, 75; Kooistra 2009, 222, 226-227; Derks/Roymans 2008, 19. 
17 Kooistra 2009, 230; see also Cavallo et al 2008, 77; Groot et al 2009, 247.
18 Kooistra 2009, 220, 223, 226-227; Cavallo et al 2008, 72-75, 77; Groot et al 2009, 248.
19 Davies 1971, 126-127.
20 Filean 2008, 100; Cavallo et al 2008, 76.
21 Groot 2009 et al. 
22 Filean 2008, 72-73.
23 Cavallo et al 2008, 76: Groot 2008a, 74-75.
24 Filean 2008, Groot 2008a.
25 Groot 2008a, 76; Groot 2008b, 89.
26 Groot 2008a, 76; Groot 2008b, 90.
27 Groot et al 2009. 
28 Groot et al 2009.
29 Groot et al 2009, 248.
30 Groot 2008a, 25.
31 Groot et al 2009, 248.