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

[Jim Webster]

If there is any evidence it is of the troops falling back in Column (red)

Alas not; translator's artefact.

IX.57.1: "As for Amompharetus, at first he did not really believe that Pausanias would really dare to leave him behind; he therefore remained form in his resolve to keep his men at their post; when however Pausanias and his troops were now some way off, Amompharetus, thinking himself forsaken in good earnest, ordered his band take their arms and led them at a walk towards the main army."

But one does need to read further.

In IX.59, the Achaemenid army takes up the pursuit, with the Persian cavalry and infantry leading, "at their best speed and in great disorder and disarray," a description which indicates that nobody in their own army could possibly have overtaken them.  Mardonius crosses the Asopus in pursuit, but "He could not see the Athenians; for, as they had taken the way of the plain, they were hidden from his sight by the hills; he therefore led his troops on against the Lacedaemonians and Tegeans." - IX.59.2

Observe that while the Spartans were moving through the hills, the Athenians 'had taken the way of the plain'.  These contingents were travelling in parallel, i.e. on a broad front, not in series (as would be the case with a march column).  This is confirmed by what happened next.

Nowhere does this even hint that this was done by troops shambling like locusts on a broad front. It could have been done by contingents forming up into columns two or three wide (or given the size of some of the contingents, one man and his servant wide, and pulling back)

[Jim Webster]

One may also note that when attacked by the Persian cavalry, the Spartans and Tegeans are almost immediately in a defensive formation, not cut up by cavalry attacks on a long marching column.  This in itself would indicate they were moving in a broad front, shallow depth configuration, allowing them to assemble rapidly into combat formation.

No it doesn't. Any army marching and being harassed by cavalry stays in a tight column, probably on a frontage of ten to eight wide, because then you're in line of battle if you turn 90 degrees.

But the whole example is a waste of time from your point of view
1) Now where does it show troops moving locust like across the rough terrain
2) The distances involved were small. If there are something like seven or eight Furlongs to the Km, then nobody probably moved more than 5km and it was nearer 3
3) These movements were made in the face of the enemy so would be made by units marching in a manner that would enable them to form line of battle at an instance.  The speed with which they got into battle formation showed this. None this is evidence for units marching all Higgledy-piggledy over a broad front.

[Erpingham]

I was wondering whether this series on Roman armies on the march might help at all.

From what I can see, it looks very good for Roman armies, misleading for Greek and Macedonian armies and not really representative of Biblical armies at all. 
<snip>
Frankly, if there were the slightest relationship between Roman and Achaemenid practice, it would be a gift for archaeology and the study of history because one could extrapolate so many things.

Rather than dismiss it out of hand,  we could actually consider this quite carefully constructed discussion of how a well-disciplined army marched and look at what some of the various experts thought were the parameters of food supply, march pace, transport capability. 

[Justin Swanton]

On the subject of tidy formations or not tidy formations when in a wide column, Asklepiodotus maintains that the open order of 2 yards frontage per file is a natural formation and doesn't have a name. I would like to suggest that when a crowd of people walk cross-country on a broad frontage they instinctively keep about 2 yards between themselves and those around them, in the same way a flock of birds keeps a certain distance between themselves to avoid midair collisions. So the Persians, preserving neither ranks nor files, and sometimes even mixing up their national groupings, advance along a wide avenue with a spacing that permits getting around any obstacles in their way like rocks without slowing their pace.

Of course this needs to be proved. Any examples of crowds on the march in the open?

I've been fell walking (and walking generally) for many years. People do keep between three and six feet apart, and they tend to spread out into a two or three people wide irregular column as they follow the leader down the easiest line of advance. If it's a path, they'll stick to the path and walk at the width of the path, because if you try and walk next to the person on the path, the person on the path naturally walks much faster than the person on the ground next to the path.

So the implication is that should a lot of people advance on a front several hundred yards wide on ground that doesn't have paths but has been sufficiently cleared to be passable, they would actually form a lot of irregular narrow columns that advance in parallel.

[Erpingham]

It might be worth looking at how Zulus moved, for example.

Ah, so archaeologists have found a connection between them and Achaemenid practice?

Actually, its not to bad an idea.  Here we have a disciplined army not bound by the conventions of the European parade ground.  I haven't really got time to research this at the moment but we know that Zulu armies moved by regiments, each with its supply chain of boys and probably with its own cattle herd.  I'm not aware of the density used or whether they used routes - they may have wandering directly across country as Patrick implies, I just don't know.  One thing we should be aware of is scale.  Zulu armies were in the tens of thousands not the millions.  They also had no cavalry or baggage animals ( just a few ponies for officers and scouts). 

[Flaminpig0]

It might be worth looking at how Zulus moved, for example.

  Zulu armies were in the tens of thousands not the millions. 

One of the many facts that IMHO opinions kills off the multi-million man argument is just how does one go about  commanding such a host? As others have intimated on this thread Patrick and Justin's Persians move more like a swarm of insects than a human army.

[Jim Webster]

On the subject of tidy formations or not tidy formations when in a wide column, Asklepiodotus maintains that the open order of 2 yards frontage per file is a natural formation and doesn't have a name. I would like to suggest that when a crowd of people walk cross-country on a broad frontage they instinctively keep about 2 yards between themselves and those around them, in the same way a flock of birds keeps a certain distance between themselves to avoid midair collisions. So the Persians, preserving neither ranks nor files, and sometimes even mixing up their national groupings, advance along a wide avenue with a spacing that permits getting around any obstacles in their way like rocks without slowing their pace.

Of course this needs to be proved. Any examples of crowds on the march in the open?

I've been fell walking (and walking generally) for many years. People do keep between three and six feet apart, and they tend to spread out into a two or three people wide irregular column as they follow the leader down the easiest line of advance. If it's a path, they'll stick to the path and walk at the width of the path, because if you try and walk next to the person on the path, the person on the path naturally walks much faster than the person on the ground next to the path.

So the implication is that should a lot of people advance on a front several hundred yards wide on ground that doesn't have paths but has been sufficiently cleared to be passable, they would actually form a lot of irregular narrow columns that advance in parallel.

Now two different things are being conflated.
If the ground has been sufficiently cleared (which would involve cutting down brush, levelling it a bit so people don't twist ankles in holes they cannot see etc but not producing a graded road surface) then troops could advance in columns in parallel, or even if the fancy took them, advance in battle formation. After all we know the Persians did prepare battlefields for chariots. But it'll involve a lot of man hours but not as much as genuine road making. Troops probably wouldn't be able to travel as quickly as on a road

However this is an entirely different thing from telling troops to just advance across a broad front to avoid choke points. Where the ground hasn't been cleared men would probably march in single file following whatever paths there are. These paths might be separated by a hundred yards or even wander apparently at random. It is a far slower way of moving men than just having decent march discipline and moving them through the choke points

[Justin Swanton]

On the subject of tidy formations or not tidy formations when in a wide column, Asklepiodotus maintains that the open order of 2 yards frontage per file is a natural formation and doesn't have a name. I would like to suggest that when a crowd of people walk cross-country on a broad frontage they instinctively keep about 2 yards between themselves and those around them, in the same way a flock of birds keeps a certain distance between themselves to avoid midair collisions. So the Persians, preserving neither ranks nor files, and sometimes even mixing up their national groupings, advance along a wide avenue with a spacing that permits getting around any obstacles in their way like rocks without slowing their pace.

Of course this needs to be proved. Any examples of crowds on the march in the open?

I've been fell walking (and walking generally) for many years. People do keep between three and six feet apart, and they tend to spread out into a two or three people wide irregular column as they follow the leader down the easiest line of advance. If it's a path, they'll stick to the path and walk at the width of the path, because if you try and walk next to the person on the path, the person on the path naturally walks much faster than the person on the ground next to the path.

So the implication is that should a lot of people advance on a front several hundred yards wide on ground that doesn't have paths but has been sufficiently cleared to be passable, they would actually form a lot of irregular narrow columns that advance in parallel.

Now two different things are being conflated.
If the ground has been sufficiently cleared (which would involve cutting down brush, levelling it a bit so people don't twist ankles in holes they cannot see etc but not producing a graded road surface) then troops could advance in columns in parallel, or even if the fancy took them, advance in battle formation. After all we know the Persians did prepare battlefields for chariots. But it'll involve a lot of man hours but not as much as genuine road making. Troops probably wouldn't be able to travel as quickly as on a road

However this is an entirely different thing from telling troops to just advance across a broad front to avoid choke points. Where the ground hasn't been cleared men would probably march in single file following whatever paths there are. These paths might be separated by a hundred yards or even wander apparently at random. It is a far slower way of moving men than just having decent march discipline and moving them through the choke points

Fine. The idea is that the the ground is sufficiently cleared on either side of the choke points, i.e. that the column has about 600m of cleared and passable ground at all times. Which is partly what makes Thermopylae a genuine choke point since the ground around the mountain is not cleared of forest, even if it can theoretically be scaled.

[Erpingham]

Time for a visual stimulus, I think.  To me, the most evocative of August 1914 pictures.  A German battalion advancing through knee-deep flowers and clouds of dust.  As we look back, we can probably see a front of about 300m, though the ground is probably less uneven.  Despite being a regular army, moving across country they are not in tight rows and files but more a clumpy mass.  Note too the big gaps between companies - I think it is Delbruck who tells us that keeping units discrete with gaps between was standard German doctrine on the march as it helped keep a steady pace (avoiding stop/start).

As I say, an aid to imagining, not to be taken as literal evidence of Persian practice.

[Jim Webster]

Fine. The idea is that the the ground is sufficiently cleared on either side of the choke points, i.e. that the column has about 600m of cleared and passable ground at all times. Which is partly what makes Thermopylae a genuine choke point since the ground around the mountain is not cleared of forest, even if it can theoretically be scaled.

No it's not fine. Admittedly 600m is an improvement on the figures that have been bandied about like 3000m or even more but not it's not fine.
The only reason that we've having these bizarre discussions is that if you have to move six million people by foot through the area it becomes impossible unless you assume they adopt bizarre expedients like clearing roads hundreds of meters wide.
Actually most of the difficulties disappear if you assume the Persian army plus baggage that had to march down the road was under 500,000 men. There might have been another 100,000 in the area who had been  there for a couple of years, digging canals, guarding stockpiles and suchlike. But the whole fatuous debate stems from the impossibility of accepting the figures Herodotus gave as the actual army marching, as opposed perhaps the theoretical manpower of the army.
It's a bit like taking the figures Polybius gives for the total Roman manpower about 225BC and assuming it was all deployed against the Gauls

[Erpingham]

This one was triggered looking  through images of the Zulu war prompted by Patrick.  Most images of the British columns are of long thin ones but here we have a rather straggling column, made up of roughly parallel sub-columns as envisaged by Justin.  Within the wide column we see companies marching in close order, squadrons of cavalry and lines of wagons.  Width of the advance perhaps 100m ?  In many ways, it reminds me of the typical pictures of medieval armies on the advance - the parallel wagon columns, the clumps of infantry and cavalry.  One of the mistakes Delbruck made in visualising the persian army on the march is he could only envisage a German Corps moving along a road in narrow column.  A wider, sprawling, approach like this, would have seen shorter columns.  But note that while some units are in quite tight order, the overall density of the column is not that great, with intervals between sub-columns and also between units.

[Justin Swanton]

Fine. The idea is that the the ground is sufficiently cleared on either side of the choke points, i.e. that the column has about 600m of cleared and passable ground at all times. Which is partly what makes Thermopylae a genuine choke point since the ground around the mountain is not cleared of forest, even if it can theoretically be scaled.

No it's not fine. Admittedly 600m is an improvement on the figures that have been bandied about like 3000m or even more but not it's not fine.
The only reason that we've having these bizarre discussions is that if you have to move six million people by foot through the area it becomes impossible unless you assume they adopt bizarre expedients like clearing roads hundreds of meters wide.
Actually most of the difficulties disappear if you assume the Persian army plus baggage that had to march down the road was under 500,000 men. There might have been another 100,000 in the area who had been  there for a couple of years, digging canals, guarding stockpiles and suchlike. But the whole fatuous debate stems from the impossibility of accepting the figures Herodotus gave as the actual army marching, as opposed perhaps the theoretical manpower of the army.
It's a bit like taking the figures Polybius gives for the total Roman manpower about 225BC and assuming it was all deployed against the Gauls

OK, let's assume an army of 480 000 men. About the best they could hope for in terms of roads is a track wide enough for a cart, allowing, say, a column 4 men abreast.

That means a column 120 000 men long.

Allow two yards depth per rank and you have a column 240 000 yards or 136,36 miles long.

A Consular army of 22 000 infantry is estimated to have formed a column 6 men wide and about 15 miles long including baggage (a little more than 3 miles for the infantry if the ranks stay fairly close together) and be able to complete a march of about 10 miles from camp to camp in about 9 hours. How does it work for an army 20 times that number?

[Justin Swanton]

This one was triggered looking  through images of the Zulu war prompted by Patrick.  Most images of the British columns are of long thin ones but here we have a rather straggling column, made up of roughly parallel sub-columns as envisaged by Justin.  Within the wide column we see companies marching in close order, squadrons of cavalry and lines of wagons.  Width of the advance perhaps 100m ?  In many ways, it reminds me of the typical pictures of medieval armies on the advance - the parallel wagon columns, the clumps of infantry and cavalry.  One of the mistakes Delbruck made in visualising the persian army on the march is he could only envisage a German Corps moving along a road in narrow column.  A wider, sprawling, approach like this, would have seen shorter columns.  But note that while some units are in quite tight order, the overall density of the column is not that great, with intervals between sub-columns and also between units.

Interesting. The infantry are still thinking of roads with their tight columns but the cavalry and wagons don't seem to give a hoot. Notice how uneven the terrain is and even the wagons aren't bothered by it. That slope in the background looks like a 1 in 7 gradient or so.

[Jim Webster]

OK, let's assume an army of 480 000 men. About the best they could hope for in terms of roads is a track wide enough for a cart, allowing, say, a column 4 men abreast.

stop right there
Firstly Patrick will tell you there is only one Road and that's the Persian Royal road and that doesn't enter Greece

Secondly, Ancient roads or tracks tended to have variable widths. In South African terms roads used to widen immensely when they came to a river. The ford or drift would be far wider than the road to ensure that wagons were not churning up the river bed through very heavy usage.

Thirdly we have Persian engineers apparently working in the area, they dug a canal, they could have tidied the roads up a bit as well. You cannot assume just a column four men wide. The "road"could have been wide enough to pass two parallel columns four men wide down.

Fourthly
On the level bits, moving in parallel you might have had cavalry units out covering the flanks, or light troops out screening the flanks. After all it was a Persian army, it almost certain marched with some semblance of military order

[Jim Webster]

Interesting. The infantry are still thinking of roads with their tight columns but the cavalry and wagons don't seem to give a hoot. Notice how uneven the terrain is and even the wagons aren't bothered by it. That slope in the background looks like a 1 in 7 gradient or so.

Infantry keep formation because they might be called upon to fight. They're soldiers.
As for the wagons note how they're largely following the infantry and are actively avoiding the uneven terrain between the camp and the infantry
Also whatever the slope is, troops have largely avoided it