Macedonian phalanx: overarm, underarm or both?

[RichT]

Justin:

I'd much rather make sense of the sources if possible.

That is the quest on which we are all engaged. But when it comes to vocabulary you can make more progress by looking at usage than by looking at etymology (as Jim's paratrooper analogy shows).

If you want to speculate about Pydna, it's possible the phalanx (specifically the Chalcaspides - the Peltasts might be doing their own thing out of Aemilius' immediate sight) advanced in 2 cubit spacing, made contact with the legions, then closed up (by doubling) to 1 cubit while the fight was static. That is possible, though speculative, but doesn't shed any light either on the sarissa hold or the ability to advance.

Anthony:

Incidentally, on the Macedonians v. Romans question didn't one of the ancient historians actually do a comparison, or is my memory deciving me?

Your memory is not deceiving you - Polybius (18.29 f). We have discussed it quite extensively on this forum (oh boy, have we discussed it) and in Slingshot. If certain parties were present, who mercifully are not, we would now begin a twenty page, increasingly acrimonious, thread discussing it again.

Maybe if I give a summary, we can avoid going over it all again. So - the problem of intervals:

Nobody knows what file interval hoplites used - but based on the shield size (3 feet), on the literal meaning of words like 'synaspismos' (shields together), and on phenomena like the 'phalanx drift' as each hoplite seeks the protection of his neighbour's shield, it is commonly assumed that hoplites used a 3 foot interval (if they used any established interval at all).

Macedonian phalanxes are said in their origin to use 'synaspismos' (as well as 'probole'), and much is made in various sources of their close order, their density, their weight, and so forth, in various forms and expressions, and they are also thought to have used smaller shields (2 feet). This leads us to expect the Macedonian phalanx to use a smaller file interval than hoplites, thus making them a more close order formation.

But the manuals are explicit that the usual file interval of the Macedonian phalanx was 3 feet (2 cubits), with the 1.5 feet (1 cubit) interval being a defensive formation (which they call synaspismos). Polybius is also clear that the Macedonian phalanx used a 3 foot interval (this is clear in his comparison with the Romans, which is quoted above in this thread, and in his criticism of Callisthenes' account of Issus, in which his calculations require a 3 foot interval), and Polybius also uses synaspismos interchangeably with pyknosis to refer to this 3 foot interval.

So - if the Macedonian phalanx is a closer order formation than the hoplite phalanx, this requires either that hoplites used a wider interval than 3 feet, or that Polybius and the manuals are wrong about 3 feet being the normal interval of the Macedonians.

Now the Romans - Polybius compares the Macedonian phalanx with the Roman formation, and notes that the Romans used an interval double that of the Macedonians (so that each Roman in the front rank faced two Macedonians). Unfortunately he is not totally explicit in giving the Macedonians 3 feet and the Romans 6 feet intervals as we would expect - for the Macedonians, he talks mostly about rank intervals rather than files, and for the Romans, he talks about the space between men, rather than the intervals between files. This leaves some scope for doubt about exactly what he means. Still, if we take this passage, the criticism of Callisthenes, and the manuals together, we conclude that the Macedonians did usually use 3 foot intervals and that the Romans must therefore have used 6 foot intervals.

This has troubled many, as 6 foot intervals seem far too wide for a heavy infantry formation like the Romans.

So various solutions have been proposed to these problems (the Macedonian-hoplite comparison and the Macedonian-Roman comparison). For hoplites, some accept the 3 foot interval for Macedonians and assume therefore that hoplites used a 6 foot spacing, and basically fought in quite open order. Similarly, some take Polybius at his word and assume that Romans also fought with 6 foot intervals. If this seems 'too wide' to us, then that is just a problem with our understanding and imagining, and there is no reason not to go with the Polybian evidence here.

Others though - for reasons given above in the case of hoplites, and because Romans don't seem to be a more open order formation than hoplites - prefer to assign a 3 foot interval to hoplites and Romans alike. This requires rejecting Polybius' and the manuals' normal 3 foot spacing for Macedonians, and assuming that the 1.5 foot spacing was the normal one (or at least, was the one that formed the basis of the comparisons with hoplites and Romans).

Various compromises are possible, basically assuming that Polybius is being overly pedantic, and that hoplites, Romans and Macedonians could all use various intervals in different circumstances, so that comparing them directly using one particular interval, as Polybius does, is not strictly valid. This would mean for example that hoplites and Romans usually fought with a 3 foot interval but could also on occasion open out to 6 feet, while Macedonians usually fought on 3 feet, but could if required close up to 1.5 feet. How mobile they would be at 1.5 feet is open to doubt - probably not very mobile at all (according to the manuals), but in a standing fight, that might not have mattered.

So Polybius is comparing typical or usual formations, but these were not universal. Then we could fall to arguing, for each particular case, which formation is being used (eg for Pydna) - but actual hard evidence is, so far as I know, lacking.

[Mark G]

Pretty good summary there.

[Prufrock]

Yes, very fair summary. We've all had that discussion a few times in various places, but Richard has covered the major points clearly and sympathetically.

[Justin Swanton]

Polybius is clear and unambiguous about the width of the phalangite and Roman files: they are both three feet wide (the Greek checks out  ):

That when the phalanx has its characteristic virtue and strength nothing can sustain its frontal attack or withstand the charge can easily be understood for many reasons. For since, when it has closed up for action, each man, with his arms, occupies a space of three feet in breadth.... when the phalanx has its characteristic close order as regards both depth and breadth, as Homer expresses it in these verses: ............ I have now, for purposes of comparison, to speak of the peculiarities of the Roman equipment and system of formation and the points of difference in both. Now in the case of the Romans also each soldier with his arms occupies a space of three feet in breadth.... The consequence will be that one Roman must stand opposite two men in the first rank of the phalanx, so that he has to face and encounter ten pikes

Any plausible attempts made to resolve this contradiction? And it is a contradiction and a pretty blatant one at that. Polybius is failing preprimary maths here.

[RichT]

Well I could just say GIYF - or in this case the search feature of this forum, upper right of the screen.

Or to help you along a little - you missed out a rather crucial sentence:

"Now, a Roman soldier in full armour also requires a space of three square feet. But as their method of fighting admits of individual motion for each man—because he defends his body with a shield, which he moves about to any point from which a blow is coming, and because he uses his sword both for cutting and stabbing,—it is evident that each man must have a clear space, and an interval of at least three feet both on flank and rear, if he is to do his duty with any effect. The result of this will be that each Roman soldier will face two of the front rank of a phalanx, so that he has to encounter and fight against ten spears."

So - the space occupied by the man and his kit is three feet in both cases - but in addition, the Romans need a gap  between men, resulting in a total of six feet. Like I said, not as clear as we would wish, but not a blatant contradiction either.

[Justin Swanton]

Well I could just say GIYF - or in this case the search feature of this forum, upper right of the screen.

Or to help you along a little - you missed out a rather crucial sentence:

"Now, a Roman soldier in full armour also requires a space of three square feet. But as their method of fighting admits of individual motion for each man—because he defends his body with a shield, which he moves about to any point from which a blow is coming, and because he uses his sword both for cutting and stabbing,—it is evident that each man must have a clear space, and an interval of at least three feet both on flank and rear, if he is to do his duty with any effect. The result of this will be that each Roman soldier will face two of the front rank of a phalanx, so that he has to encounter and fight against ten spears."

So - the space occupied by the man and his kit is three feet in both cases - but in addition, the Romans need a gap  between men, resulting in a total of six feet. Like I said, not as clear as we would wish, but not a blatant contradiction either.

I checked the Greek - again. Literally, Polybius says this:

"three feet according to the man to his rear and the man at his side".

The only way to square this with his previous statement that a Roman soldier occupies a space three feet wide (he either occupies a three-foot wide space or he doesn't) is to take Polybius absolutely literally about the three feet "according to" - kata - the man at his side. In other words the man at his side, either side, determines that distance, which is three feet, i.e. the three feet is between the man on his one flank and the man on his other.

The "three feet according to the man to his rear" can only be measured between himself and the man behind him as there is nobody in front, i.e. he has a clear three feet behind him to enable him to manoeuvre during a mobile sword-fight which, as a fencer knows, is more a case of back-and-forth rather than sideways movement. But he needs at least a three-feet breadth to use his sword and shield effectively

So Polybius is making clear that a Roman absolutely needs that nine square feet of space he occupies. Why bother saying that? I think he might be alluding to something his readers know and we don't: the fact that a phalangite doesn't need that amount of space. Reading through the manuals, I pick up that a phalanx could switch from intermediate to close order fast - really fast. There's one passage in Aelian where files are interleaved, like this:

abababababababab
abababababababab
abababababababab
abababababababab

Doubling, as Aelian describes it, is done simply by all the b's moving into the interval between the a's, creating a doubled unit in a second or two.

A phalanx then, in intermediate order, advances, sarissas lowered at the ready, towards its enemy. Just before combat it slows down and the b men move into the space beside the a men, suddenly presenting the enemy soldiers with twice the number of opponents they originally thought they had. Psychologically a very effective ploy. This compact phalanx moves slowly against the enemy (it isn't completely immobile). If the enemy gives way it can just as quickly resume its intermediate formation and continue marching forward.

If Polybius assumed his readers knew this (and why wouldn't they?) then the one Roman facing two phalangites makes sense, and the contradiction in Polybius is resolved.

Or can anyone show me another way of resolving it?

Erratum: just realised that the midpoint from the man to his side to his midpoint is three feet. That resolves the problem much more simply. It also leaves you with files three feet wide. Work it out. 

[Prufrock]

The only way to square this with his previous statement that a Roman soldier occupies a space three feet wide...

And there we have the disagreement in a nutshell. There are other ways to see it.

[Justin Swanton]

The only way to square this with his previous statement that a Roman soldier occupies a space three feet wide...

And there we have the disagreement in a nutshell. There are other ways to see it.

Go for it. 

[Justin Swanton]

Here's the solution to Polybius' three feet problem.

"three feet according to the man to his rear and the man at his side".

The men stand three feet apart, i.e. there is a three foot distance between the midpoint of one man and the midpoint of the man beside him. This leaves you with 3 foot wide files, not six foot wide ones. Et voila!

If one tries to maintain that the three feet extends from the three foot wide original spacing of the legionary Polybius mentions earlier, then you are left with files looking like this and no, I can't buy it. 

[Erpingham]

While we can indeed sort out the conundrum by ignoring some of Polybios' words and attributing them to a mistake, the difficulty comes in which ones do we ignore?  If it was simple, it wouldn't have been exercising scholars since the 19th century, if not before.

[Justin Swanton]

While we can indeed sort out the conundrum by ignoring some of Polybios' words and attributing them to a mistake, the difficulty comes in which ones do we ignore?  If it was simple, it wouldn't have been exercising scholars since the 19th century, if not before.

Point is, we don't have to ignore any of them.

[Prufrock]

According to Polybius the Macedonians need three feet. The Romans also need three feet, but they need extra space too, three feet between the man and the man next to him and the man behind him, and we have each Roman facing ten pikes.

But you will say that since you've now *conclusively* shown the Romans occupied three feet each, then the Macedonians must've occupied 1.5, (and this also fits neatly with your proposed overhand grip theory), so.... 

Sorry, it's been said before, the same counter arguments apply, and I'm too jaded to go through it all again

[Justin Swanton]

According to Polybius the Macedonians need three feet. The Romans also need three feet, but they need extra space too, three feet between the man and the man next to him and the man behind him, and we have each Roman facing ten pikes.

But you will say that since you've now *conclusively* shown the Romans occupied three feet each, then the Macedonians must've occupied 1.5, (and this also fits neatly with your proposed overhand grip theory), so.... 

Sorry, it's been said before, the same counter arguments apply, and I'm too jaded to go through it all again

Oh, I'm always ready to be shot down in flames, Aaron.              

But it's also nice to make sense of the sources.             

[Erpingham]

While we can indeed sort out the conundrum by ignoring some of Polybios' words and attributing them to a mistake, the difficulty comes in which ones do we ignore?  If it was simple, it wouldn't have been exercising scholars since the 19th century, if not before.

Point is, we don't have to ignore any of them.

Well, you are either ignoring that the Macedonians need three feet or you are ignoring the fact that the Romans need twice as much space as the Macedonians.  It is also possible to propose that Polybios has forgotten to tell us he has switched from a three foot Macedonian frontage to a 1.5 foot one to make his point - you then don't need to ignore words but postulate a linking sentence which tells us that, while three foot was normal for Macedonians, they sometimes closed up to 1.5 feet, which either Polybios forgot to put in or it was lost in transcription down the centuries.  But you can't credibly read it as if it doesn't have any issues in it.

[Prufrock]

Oh, I'm always ready to be shot down in flames, Aaron.              

But it's also nice to make sense of the sources.             

It's not about shooting people down in flames (or being shot down oneself!), it's just ground that's been gone over before (Richard and Anthony have shown the main issues, but there's also Polybius's criticism of Callisthenes to come in yet as well) and it always ends up the same way.