Could the Persian Empire logistically support an army several million strong?

[Justin Swanton]

We wouldn't need 3000 transports. Presuming the navy has the job of entirely supplying 3 million men throughout the campaign. That's 3000 tons of grain a day deposited on the shore by 60 50-tonner ships. Assume an average round trip of 10 days from the Balkan/Macedonian/Thessalian shore to Asia Minor and you need 600 not-very-big ships. Include the supply depots and what is furnished by the locals and you probably need considerably less.

A reminder it is Patrick who claims 3000 transports, not me.  Also, the figure of 3,000 tons calculated at the beginning was wrong on the figures being used at the time (It should be 3,500 tonnes at 1 kg per man per day) and that was before we realised there are at least 4.5 million people to feed according to Herodotus, so we need 4,500 tonnes.  Then we placed all the grain in amphorae, which cut the weight of grain a 50 ton ship could carry by half (the other half being ceramic).  So you need the equivalent of 9,000 tonnes of stores offloaded per day.  That is assuming you can source your fodder locally.  So, the true figure is around 1,800 50 tonne ships.  I leave the plausibility of that in the context of the time to others.

I propose that the grain taken to the storage depots along the Balkan and Asia Minor coastlines would be in amphorae since there's plenty of time to get it there so weight is not an issue. The grain taken from Asia Minor to the army travelling along the Balkan coast would probably be in sacks to save weight and facilitate loading on mules. That gets your vessels back to 50 tons grain per ship which supplies only a percentage of the army's needs whilst it is on the march. How much could the depots store and the locals supply?

[Erpingham]

It would be an idea to try and determine what percentage of the army's needs would be met by a) local contributions, b) the supply depots, and c) the navy.

I tried a thought experiment based on the idea of "Feed the Army for a day".  Imagine a community of 10,000 people - on the big side for the area but lets try it.  Assume it feeds itself under normal conditions.  It is given notice to boost production over three harvests and quickly creates perfect grain stores.  It's basic annual production is 3,650,000 person days a year.  A ten percent increase in production will store up 365,000 person days a year, 1,095,000 in 3 years.  So a sustained production increase of 40% would give us the sort of scale we need for Herodotus' figures for a day.  We can speculate on how many largish polities like this there were, how viable the resources needed for building that huge single use granary, how viable the increase in production, especially when a lot of the male population may have been conscripted to build the great road and associated infrastructure.  But in the end, it is speculation.  As I've said several times, just calculating numbers like this only takes us so far unless they are placed in the context of an army on a mission.

[Mark G]

That would be 10 000 working people, for your maths to work, and a quite unusual 365 day a year growing period, uninterrupted for three years, from a standing start.

[Flaminpig0]

One characteristic that both Thracian Tribes and Greek City states share is that hey are not command economies. It is not that simple to get a lot of independent small farmers to increase production so dramatically so as to feed an army of several million,even supposing they had the ability to do so.

[Erpingham]

That would be 10 000 working people, for your maths to work, and a quite unusual 365 day a year growing period, uninterrupted for three years, from a standing start.

Actually, no.  It is based on mouths to feed and the number of days on which they are fed.  So the entire community, on a break even basis, must produce 10,000 rations a day for a whole year.  It is deliberately abstracted, so as not to have to fiddle with yield rates, growing periods, ration scales, the balance of crops v. animal products etc.

With little refinement, we can see some obvious flaws that mean productivity is lower.  Half the population would have been children and ate less.  Half the adults would be women, who would also eat less.  We assume constant harvests and no spoilage.  We assume that full scale up could be achieved in a year (where did the extra seed come from, have you got that much spare growing land under your control that you can clear and plant it?)  The point was to give a ball park idea of what the "each city could feed the army for a day" theory represents. 

[Patrick Waterson]

Can we have some numbers please as to how large an increase in production you imagine Xerxes demanded? How many men they were supposed to feed as opposed to being fed from imports.
By this time Greece was almost certainly dependent on grain imports anyway

Disagree about Greece being dependent upon grain imports: that did not happen for another century, when Egypt sent corn to whichever cities were prepared to send troops to help it out.  Even Athens was self-sufficient in 480 BC; fuss about Black Sea corn does not begin until Thucydides makes mention of it.

As for numbers, we make our own guesses from Herodotus VII.119:

"Similar accounts were returned by the officers in the other towns. Now the dinner, about which a great deal of fuss had been made and for the preparation of which orders had been given long ago, proceeded as I will tell. [2] As soon as the townsmen had word from the herald's proclamation, they divided corn among themselves in their cities and all of them for many months ground it to wheat and barley meal; moreover, they fed the finest beasts that money could buy, and kept landfowl and waterfowl in cages and ponds, for the entertaining of the army. They also made gold and silver cups and bowls and all manner of service for the table. [3] These things were provided for the king himself and those that ate with him. For the rest of the army they provided only food. At the coming of the army, there was always a tent ready for Xerxes to take his rest in, while the men camped out in the open air. [4] When the hour came for dinner, the real trouble for the hosts began. When they had eaten their fill and passed the night there, the army tore down the tent on the next day and marched off with all the movables, leaving nothing but carrying all with them."

What the above indicates is that there were no imports but a lot of drain on local resources.  One notes incidentally the corn being 'in their cities' - the above (Perseus' Godley) translation incidentally seems to have a phrase out of place: "they divided corn among themselves in their cities and all of them for many months ground it to wheat and barley meal" is elsewhere translated as: "the inhabitants made a division of their stores of corn and proceeded to grind flour of wheat and of barley for many months together".  It seems more sensible that they ground many months' grain allowance at once than that they spent months grinding grain ... either way, that was the equivalent of their winter store gone to feed Xerxes' army for a day.

What you have to beware of here is that Germanicus and Chronicles may well have got their information from the same, flawed, source. They don't need to be independent.

Germanicus had Egyptian inscriptions translated; would the writer of Chronicles have bothered to acquire information from such a source?  The Chronicler in any event did not take his numbers for chariots or cavalry from that source.  I think we can safely conclude that they were independent.

What you must remember is that the Bible does have a penchant for large numbers, the number of Israelites which Moses is supposed to have led may have been larger than the population of Egypt.

This was exactly what worried the pharaoh who decided to end the Hebrews' work-free, tax-free status, and his successor, who wanted their male children out of the way.

The question about persistent large numbers in the Bible is whether they were persistently large because of several generation of writers' constantly overheated imagination or because the numbers really were large.  If it were only the Bible which mentions armies of hundreds of thousands, we migth be justified in assuming exaggeration.  But all major cultures of the period record themselves and/or their opponents as fielding massive armies.

[Patrick Waterson]

Also, the figure of 3,000 tons calculated at the beginning was wrong on the figures being used at the time (It should be 3,500 tonnes at 1 kg per man per day) and that was before we realised there are at least 4.5 million people to feed according to Herodotus, so we need 4,500 tonnes.  Then we placed all the grain in amphorae, which cut the weight of grain a 50 ton ship could carry by half (the other half being ceramic).  So you need the equivalent of 9,000 tonnes of stores offloaded per day.  That is assuming you can source your fodder locally.  So, the true figure is around 1,800 50 tonne ships.  I leave the plausibility of that in the context of the time to others.

Some curious arithmetic here.

1,800 ships for 9,000 tons is 5 tons per ship; on Anthony's figures it should be 25.  The number of ships is thus 1,800/5 = 360.

Tonnage in a merchantman is generally reckoned in gross registered tonnage rather than deadweight tonage and thus is not so much weight as volume; putting grain in amphorae may double the weight but it does not double the volume.  Hence the 50-tonner could be carrying 50 tons of grain in 50 tons of amphorae, or more likely 40 tons of grain in 40 tons of amphorae if we consider amphorae to add 20% to overall volume.  This would take the daily complement of ships down to 225.

With little refinement, we can see some obvious flaws that mean productivity is lower.  Half the population would have been children and ate less.  Half the adults would be women, who would also eat less.  We assume constant harvests and no spoilage.  We assume that full scale up could be achieved in a year (where did the extra seed come from, have you got that much spare growing land under your control that you can clear and plant it?)  The point was to give a ball park idea of what the "each city could feed the army for a day" theory represents. 

A good idea.  The impression Herodotus gives is that the cities routinely kept in store something like six months' supply (1,800,000 person-days or so), which suggests (along with a lot of other incidental information) that yields were habitually well above subsistence.  In Herodotus VII.118 we get the information that Antipater son of Orges reckoned that it cost the Thasians 400 talents to feed Xerxes' army for a day and that this was pretty much the experience of other communities.  This hints at (without proving) a certain amount of buying-in (as opposed to importing per se) to make up what was needed.

[Patrick Waterson]

One characteristic that both Thracian Tribes and Greek City states share is that hey are not command economies. It is not that simple to get a lot of independent small farmers to increase production so dramatically so as to feed an army of several million,even supposing they had the ability to do so.

But they are not independent small farmers; they are part of a highly cohesive community, a tribe or a polis.

We are not necessarily looking at a dramatic production increase; we are looking at an effort to make up margins in the face of a massive obligation.  The main obligation was met through drastic depletion of stocks.  The additional animal and avian increase was presumably to ensure no shortfall in the Persian nobility receiving only the finest meats and their followers receiving at least something containing protein.

[Justin Swanton]

In Herodotus VII.118 we get the information that Antipater son of Orges reckoned that it cost the Thasians 400 talents to feed Xerxes' army for a day and that this was pretty much the experience of other communities.

How much grain could a talent buy? Any way of arriving at an approximate figure?

After a quick look-around I get an Attic talent equivalent to 6000 drachmae, and a half-drachma enough to feed a family of three for one day. So, doing the sums, 400 talents supplies a day's minimum sustenance to 400 x 6000 x 2 x 3 people = 14 400 000 men. That number can be brought down somewhat by assuming the soldiers were fed more than the bare minimum and by the cost of feeding the Persian VIPs with fancier food along with the gold and silver cups and bowls, but we are definitely not in the 200 000 man category (a talent BTW consists of 26kg of silver so most of the 400 talents - 10,4 tons of silver - did not go into the tableware).

According to this source a loaf of barley bread cost 1 obol (6 obols to the drachma). I can't find out how much the loaf weighed.

The meal BTW would have cost the Thasians about US$ 8,000,000.00 in contemporary terms.

[Flaminpig0]

But they are not independent small farmers; they are part of a highly cohesive community, a tribe or a polis.

I am afraid I am going to have to politely disagree with that statement.

[Jim Webster]

It would be an idea to try and determine what percentage of the army's needs would be met by a) local contributions, b) the supply depots, and c) the navy.

Given that some estimates of the population of Ancient Greece and the Greeks in Asia minor are 12.5 million and 2.5 million (500BC)
https://en.wikipedia.org/wiki/Demographic_history_of_Greece
Michael Grant gives 4,000,000 for the Macedonian kingdom, 7,000,000 for the Ptolemaic kingdom, and 30,000,000 for the Seleucid empire in the the third century BC

So the 6 million that Xerxes dumped on their shores would be a very significant part of the population. I cannot see them feeding a meaningful proportion of that number

[Jim Webster]

 
This was exactly what worried the pharaoh who decided to end the Hebrews' work-free, tax-free status, and his successor, who wanted their male children out of the way.

The question about persistent large numbers in the Bible is whether they were persistently large because of several generation of writers' constantly overheated imagination or because the numbers really were large.  If it were only the Bible which mentions armies of hundreds of thousands, we migth be justified in assuming exaggeration.  But all major cultures of the period record themselves and/or their opponents as fielding massive armies.

The Israelites journeyed from Rameses to Sukkoth. There were about six hundred thousand men on foot, besides women and children. 38 Many other people went up with them, and also large droves of livestock, both flocks and herds.

The first census gave 603,550 men over the age of twenty

Pretty damned good for people wandering in the wilderness. Makes old Xerxes look like a complete waste of space

[Jim Webster]

Can we have some numbers please as to how large an increase in production you imagine Xerxes demanded? How many men they were supposed to feed as opposed to being fed from imports.
By this time Greece was almost certainly dependent on grain imports anyway

Disagree about Greece being dependent upon grain imports:

then you are at odds with most historians. Why do you think the Greeks sent out so many colonists? The main reason historians assure is was because they couldn't feed them at home

[Patrick Waterson]

But they are not independent small farmers; they are part of a highly cohesive community, a tribe or a polis.

I am afraid I am going to have to politely disagree with that statement.

Are you entirely sure you wish to disagree with mainstream history?

[Jim Webster]

   The impression Herodotus gives is that the cities routinely kept in store something like six months' supply

of course people store six months supply. They cannot rely on the southern hemisphere for an extra harvest!
That six month's supply is to get them through to the next harvest, if somebody else eats it, they die