Saturday 26 May 2012

Greek Logistics

There really is not much written about logistics in the literature of classical Greece.  To be honest, there is not an awful lot written about logistics in the entire ancient world. It is not the sort of thing that chroniclers and historians were particularly interested in. On that basis, I suppose, wargamers are their heirs.

I am aware of one or two items of scholarship on the subject. The main resource, which everyone refers to, is D. W. Engels, Alexander the Great and the Logistics of the Macedonian Army (Berkeley, UCP, 1978). The problem here is that I have not read it (yet). I will get around to it; the last time I checked it was still in print.

After that, Adrian Goldsworthy has an appendix to The Roman Army at War 100 BC – 200 AD, wherein he discusses how Roman armies managed to survive in the field. I think there are a couple more scholarly (and thus hideously expensive) books available on the subject as well.

Recently, I ran across a paper by J. F. Lazenby entitled ‘Logistics in Classical Greek Warfare (War in History 1 (1) 3-18). He makes a number of interesting observations about logistics and, I suggest, his points are germane until the horse and sail became less important in the supply of armies. In that sense, we can probably treat logistics as more or less the same until, say, the American Civil War when steam power, both on railways and rivers, became useful.

Now, during the Peloponnesian wars, it seems to me, most troop transport over any distance was by ship. Greece is a fairly difficult land to move large bodies of troops across, and if you read Thucydides and Xenophon I think you will find that armies marching over distance was rather a rare thing.

As an aside, I do have a sneaking suspicion that, apart from the one or two big battles, the Peloponnesian wars were largely a process of raiding, devastating crops and nipping off home again before the enemy had time to react. Most of the actions of which so much is made seem to me to have been pointless scraps in muddy fields between a few hundred men. But perhaps Thucydides’ cynicism has infected me.

Anyway, the standard distance for a day’s march was between 10 – 20 miles, although on good roads more could be achieved. Xenophon, on his way to Cunaxa, along good Persian roads and not needing to scout or use guides, managed 18 – 25 miles a day, and could possibly do more for a short period.

Baggage, by which we mean the soldier’s kit, weapons, armour, food, bedding, tents and so on, was carried, some by the soldiers themselves, some by servants and some by pack animals or waggons. The maximum load for a porter would be about 100 pounds. Pack animals could carry about four times as much and waggons, well, it depends on the waggon, but say about 1000 pounds.

Now, of course, this gives us some problems. Firstly, all these porters, pack animals and draft animals have to be fed (and paid, too, but that is a bit of a different problem). Each man additional to the fighting force would still consume the same quantity of food, need bedding and so on. So if, as Goldsworthy suggests, the minimum feed was three pounds of bread and one of meat per day, a single porter can carry sufficient for himself for 25 days, or for himself and the soldier he serves for 12 days.

Of course, pack animals and draft animals could carry much more. A single draft horse could manage food for a man for 100 days, assuming that it could be preserved for that long, or for one hundred men for a day. A waggon, of course, could supply 250 men for a day.

Now, armies tended to be around the ten thousand mark. This would require 40 waggons, or 100 pack horses or 400 servant porters for a day, just for the food. When we look at the problem from this point of view, some of the slowness of movement and the lengthy rests start to make sense. It is, in fact, easier to supply an army in a fixed position from a magazine somewhere in the rear. Teams of 40 waggons or 100 pack horses a day are, in that case, perfectly reasonable. Armies in motion are more difficult to keep in supply.

At this level we can start to believe that, for example, at Plataea, the Spartans had 7 servants per soldier. This, of course, may be reported as it is unusual, but supplying the Greek army on station, as it were, could require significant amounts of servant, slave or helot labour.

Frequently, of course, local markets were set up where the troops could buy food. This, however, required some friendly locals with food to spare, or who were too frightened to resist. This could lead to further problems. An army on the march, or a navy passing by, could not necessarily afford the time to storm cities because they refused to sell food. Furthermore, as the Spartans used to their advantage, an enemy based in a country with no markets had to scatter over a wide area and was, therefore, vulnerable to a surprise attack. (Thucydides 8.95 for an example).

If we add up the time it takes for individuals to buy, prepare, cook and eat food, it is not that surprising that armies tended to move rather slowly. Indeed, in the English Civil War 10 miles a day was regarded as being the average; not much had changed since Xenophon’s day. Even simple breads take an hour or two to make and bake. Only so much can be prepared in advance.

So, while, yes, I am banging on again about logistics, to some extent I do not think I am emphasising them excessively. They had a massive influence on the routes armies took, the speed of movement, the security of the camps and even, as I have observed, on the chances of being surprised and defeated, all as a consequence of the basic need to feed people.

To some extent, then, as wargamers, we ignore logistics at the peril of  making our armies live in fantasy worlds where no-one eats.


  1. Hi,

    I do not think that this interferes with your general point, but I suppose the calculations may be a little faulty. 4 lb. per person per day is somewhat high standard. According to US standards during the ACW the daily marching ration was around 2lb. (sometimes less if the meat was fresh). Comparing with necessary energy supply for adult men which work hard, 3000-4500 kcal is enough and this may be adequately achieved with around 2lb. of bread and meat. So I think Goldsworthy is not right with his assumptions.

    Another point is 100lb. as porters load. As this may actually be close to the limit for a strong person, you have to take into consideration that the porter has to take care of himself (or he will need another porter). One has to carry his own blanket or cloak, some possessions - maybe a spoon, a knife, a cup or jug, some simple clothes and other personal items. I think allowing 10lb. for all these is not much, so the free load lowers a bit. From my personal experience I also think 100lb. overall load is a high limit. Making 20-30km with a backpack of 50-80lb. is tiring enough. 80-100lb. visibly limits speed and drains on one's health, which has to be countered with regular longer breaks in march. If you do not, people will be prone to become sick or straggle. Actually, the Romans seem to be marching for 2-3 days and than made one day rest. According to some mountain trekkers one can go around with 1/3 his bodyweight and 1/2 at the maximum, unless he wants to break his spine eventually.
    Summing up, I think 70lb. "free load capability" for a porter is high enough and possibly less if we plan longer marching.

    Yet another point - if draft animales were used (which was probably quite common), they need to be fed, too. Again according to ACW estimates, one animal needs 26lb. of hay and grain per day to stay healthy and productive. If the 10000 army has to make, say, 10 days march, the soldiers need around 200000 lb. of rations. A single animal needs 260lb. for 10 days, which leaves ca.200lb. free carrying capacity (I added a little to your estimation). So the above army needs 1000 animals to carry its food for 10 days. Fortunately we do not have to count the people attending the animals for they may carry all they need themselves (30lb. estimate; they have to tend to animals, so they cannot take much more or would become ineffective). There are no tents, tools, spare parts for weapon repair and many other things which the army will need. There is also no additional load for richer / higher grade soldiers, which may become very burdening (on one occasion some persian general took 200 women harem with him to name but one). There are no merchants with their goods. There is no water in this calculations - although water can be found during march, this is not always possible in adequate amount and sometimes (think deserts) may be a major issue.

    P.S. I ran out of sings, read on below. I warned you before, did not I?

    Best regards,

  2. Forgetting all apart from food and fodder, we get a daily need for 100 animal load cargo to feed the army. This is what we have to constantly supply when the starting supplies were eaten up. If we deliver cargo overland, this mean we have a sharp limit to the distance the army can march from its base of supply. Increasing an animal capacity to 720lb. (which is heavy, I suppose, but the animal will eat this during march, so it will gradually become less), if the route is 10 days of march apart, a single animal can carry only 200lb. to the army (720lb. - 520lb. of fodder to go there and back). There is no break for rest in this estimate, so I do not know how long the animals and their carers will be able to go around and stay effective. They will probably need several days off at the end of journey or the distance will have to be shorter. As the animals march will probably be shorter than soldiers march (aparat from other problems, there will be no generals to hurry them up, at least in majority of situtaions) and the estimate is rather problem-free, the distance from the base overland, at which one can supply an army, will normally be less than 10 days away. Of course, we can use more animals, have smaller army etc. etc., but this is just an estimate to help imagine the problem. The army operations during the SYW, which relied on overland transport, were somewhere between 7-12 days away from nearest base, so the estimates may be close to real thing.

    I think that one important point comes out of all this estimations. One can realize why supply routes were (and still are) so important. If an army gets cut off, it will become a large mass of weakened, hungry people in several days. So the army has to react immediatelly, better be defeated in battle than starve to death.

    P.S. Concerning the last point you made, this is true in most occassions, but not when you play a campaign game. If you play this without logistics, it ends in armies marching anywhere anytime which may be fun, but not realistic at all.

    Best regards,

  3. Hi,

    Thank you for our comments. One of the problems with these sorts of calculations is that no-one thought to write down how much a porter, draft animal or wagon could carry.

    Some argue that the ancients could carry more, being fitter, some less due to poorer diet. It is all a bit hand-wavy and guess work.

    Two things are germane, though. Firstly, on any calculation, a LOT of food and fodder was required, even for a fairly modest sized army. Secondly, somehow most of these armies managed not to starve.

    In Xenophon, a few day's march is followed by rest days, presumably when the soldiers would buy and cook food, restocking for the next march. But this was through (relatively) friendly country. On the retreat to the sea, foraging becomes a lot more important and the rate of march much slower.

    So, yes, in campaign games logistics are much more important as a process to be included. Without, the campaigns become fairly meaningless.

  4. Hi,

    Living of the land to some extent explains why armies could operate with limited transported supplies. Yet it also shows why an army cannot stay in one place for long period without supply from outside. When all in the vicinity is gathered and eaten, you simply have to move somewhere else. It also explains why some tribes / people resisted for long time even much stronger enemies. When the attacking army had to leave of forests and mountains, it had to go relatively quickly. Hiding away and waiting for them to go seems to be efficient tactics in such situation.

    Another implication is that people living in the area where an army stays will quickly become poor and starving. But this is something we do not want to trouble us in our wargames, as you have already written before if I remember. However, surviving the onslaught and devastation could be an interesting game in itself, but belongs better to some RPG, I suppose.

    Best regards,

  5. Hi,

    Living off the land is a possibility, but it does, I think, have two potential ill effects.

    Firstly, it is really going to upset the locals. the French Revolutionary armies came to liberate, but as they didn't have much in the way of logistics, lost support quite rapidly outside the French borders. So it does depend quite heavily on what you are trying to do; living off the land is no way to win hearts and minds.

    Secondly, I think that living off the land slows the speed of movement of a force and makes it much more vulnerable to surprise. Several times Greek forces were surprised while foraging, and the surprised force usually subsequently lost. Foraging is not, I think, a quick business and so the hours that could be assigned to marching would be severely restricted.

    I suppose the other issue might be equipment. an ancient army could, probably, get new equipment from the areas it found itself in. Trieremes could be built, so spears, swords and shields were probably no issue either (armour too, I guess).

    More specialist munitions could not be resupplied locally with such ease. Musket balls probably could be (the Royalists melted down lead roofing from Devizes before Roundaway Down, for example), but shells for mortars or howitzers would be a problem by the 18th Century.

    So, there are lots of issues here. Complexity mounts quite quickly hen you look at a campaign game.

  6. Hi,

    I generally agree with you, but I will point out one thing - equipment of ancient armies. There were many pieces of equipment which could be procured from local resources, but some or most of them would take time to make. Building a triereme is possible, but you need sufficient numbers of specialists and at least several weeks to do this. As with most technological processes, when you build a ship in a hurry, it will be of lower quality, which may seriously affect its efficiency. Making a strong shield probably takes less than triereme, but it cannot be done overnight anyway. Probably the easiest thing is making an arrow or javelin, but when we think of supplying an army there are thousands to be made, so it takes time again. There are also technological features of materials which may or may not be available. For example, you could get wood in most places, but this particular kind of wood you need may not be readily available. Some forges for smithing, together with smiths, would be of assistance as well.

    There are also some places, where many materials and supplies are not available in large quantities. If the commander of an army is not sure where the fighting will occur or if the standard issue will be enough, it is wise to have some spare equipment and weapons with the army (I am not sure about numbers, but Surena had some 1000? camels with arrows for resupply at Carrhae). This may be specific to army with high numbers of archers, but as some argue that most fighting was done with missiles, be them pila or javelins for example, some surplus would be surely welcome.

    And the last which comes to my mind is siege equipment. You need an awful lot of tools to carry on a siege. Some siege engines would be useful as well - these were usually brought with the besieger, at least some crucial parts. For some reason an army could built walls, ramps and other large installations with materials available around, but it seems harder to properly built machines. And siege warfare was quite common and important, so an army without siege equipment was at a disadvantage.

    The one piece of relevant data I remember is Roman fabri, which constituted up to several percent of army numbers. Even though many of them could be taken for particular tasks from legions themselves, these people had to have tools and equipment to do their job properly. These tools had to be carried with an army.
    Early imperial legions routinely took with them around 50 scorpions and several onagers, which makes for several scores of pack animals, which in turn need additional fodder.

    Summing up, an ancient army could do without much additional equipment (and smaller logistic train), but an army with some (most importantly siege machines and tools) could do better or much better in specific situations.

    Best regards,

  7. Hi,

    I suspect the most problematic issue for triremes and some siege equipment was seasoned wood, which cannot be produced at all quickly.

    On the other hand, with slave labour around, unit labour costs would be low, so churning out large quantities of some item would be possible. I think about one million nails were recovered from an abandoned fort in Scotland (OK, Roman not Greek, but still...).

    One thing I have started to ponder a bit is the relation between the technology of the day and the armies. As technology improves, siege engines become more sophisticated and the need for a large logistic effort, not to mention trained engineers grows.

    But most wargamers are not interested in sieges, nor logistics. I guess the difficulty is making them sufficiently interesting, or at least part of the game, to matter.

  8. Hi,

    I do not think slaves would be of much help in campaign. An army has a lot of workforce itself and slaves usually are not efficient workers. Their work/products would be of low quality and this is not what a commander would like to get. Also, an army would need a lot of them and this means more food eaten. So it would probably be better to take more spare javelins, arrows and hoplons then feed people who could produce them.

    I truly agree most wargamers are not interested in logistics at all. It is somewhat better with sieges, under condition that it means storming a city / citadel. One just have to like this sort of gaming.

    Best regards,

  9. Hi,

    I guess it depends on your sort of slave. In the ancient world, most were agricultural drudges, but some were craftsmen, scribes and such like. The idea of chattel slavery is depressingly modern. Ancient slaves could earn and keep money and with it buy their freedom.

    On the other hand, a surprising number of slaves in the south of the US during slavery were also paid employees, although of course their wages went to the owners.


  10. Hi,

    I know slaves were different in skills and motivation. Some even got to become rulers of Rome, so things must have been quite diversified. My point is that one could not get sufficient number of slaves in close vicinity during campaign just at the moment they were needed. At least in majority of situations. If so, they should be taken with an army for campaign and we end as in my previous post - you have to feed (and possibly pay) them, so it could be more convenient to take ready goods instead.

    Slaves could be great as a support of logistics working well back at home, which I suppose was common. But this is a different thing.

    Best regards,

  11. Hi,

    I forgot to say that an army could, of course, rely on slavework during campaign, but for expense of time and I assumed we talk about short term incidents, when we need something more or less urgently. Maybe this was wrong assumption.

    Best regards,

  12. Hi,

    I think that some slaves did go on campaign - Herodotus reports that 7 helots per Spartan hoplite were present at Plataea, but that is a bit controversial, for the reasons you have outlined.

    Given that citizen soldiers of Greek cities were wealthy enough to be hoplites, I'd guess that they did have some servant / slave help in carrying stuff. But I don't think the sources are very clear on this.


  13. Hi,

    I do not dare to say that slaves were not taken with an army for campaign. They sure were. I just suppose (although I do not have any hard data for this) that most of them were porters, cooks and other "general helpers". Only a minority were trained craftsmen, so they could prepare necessary equipment but not that fast. Additional could be gathered from surrounding countryside, but it probably took some time.

    Thinking again, I suppose different armies had their specific situations. Surena got his arrows from ready-to-use stock. Romans (at least from a certain point in time) had their fabri and used legionnaires as workforce. Greeks probably had to rely on slaves more as they were part-time citien soldiers (mostly) and were probably not accustomed to working themselves as Romans did.

    Taking into consideration that army's base could be near or far, supply could be by land or sea, countryside could be abundant in food and fodder or the opposite, sieges were to be conducted or not etc. etc. I now see the complexity of options. Which helped me to grasp an idea of what problems existed and arrange them a little bit. I thank you for that. I am also sorry for deviating from Greeks from time to time, but I do not know the subject deeply as I am more interested in Rome, at least right now.

    Best regards,

  14. Hi,

    Greeks or Romans - it doesn't matter. I think that all pre-railway armies must have had similar sorts of issues and problems to deal with, quite aside from pre-map problems of getting lost.

    I think that possibly I've conflated two different sorts of 'slaves' - the agricultural laborer / porter type and the craftsman type. The latter probably would not go on campaign, but be turning out arrows etc to order. The former would be the servant carrier to the wealthier troops.

    Incidentally, didn't Marius get complaints when he made the legionaries carry their own kit? Previously they had had servants to do that, if I recall correctly.

    So yes, even for relatively simple armies of the ancient world, logistics, even in a relatively restricted form, were complex. It is a wonder they managed to get into battle at all, sometime!