Macedonian cavalry success against close order infantry

[Duncan Head]

I'm sure someone would have provided a shield (do we still believe Alexander's cavalry were shieldless - just checking?). 

It seems to have varied: Diodorus XVII.20.3 in his description of the Granicus has 'Spithrobates' (probably Spithridates) hurl a javelin with such force that it passes through Alexander's shield, but Arrian's account of the same battle has no such mention (and Spithridates is mentioned only as being struck down by Black Cleitus when about to strike with a kopis - a heavy blade - at Alexander).

I had always thought that the "shield" in this case was the sacred shield taken from the Temple of Athene at Troy, and carried in front of Alexander by an attendant (Arrian 1.11.7‑8). That seems the only likely explanation when it is clear that the cavalry did not normally carry shields in mounted combat. (Or else the incident is simply an embroidery by Plutarch, which Patrick's contrast with Arrian might suggest, in which case there may have been no shield at all.) Whatever the explanation, it is an anomaly: the artistic evidence consistently shows Macedonian cavalry fighting without shields (the Mosaic, the Sarcophagus, the Porus Medallion, the Boscoreale statue, and others). And Arrian 1.6.5 is clear that the cavalry took their shields with them specifically when they were going to fight on foot.

The xyston is generally (among classical authors) considered to be a weapon wielded with two hands but in the Alexander Mosaic it is wielded in one hand and still goes right through a presumably armoured(?) opponent's body (the opponent is one of Darius' personal guard cavalry, some of whom show indications of wearing armour - Duncan is good on the fine details of this).

Sorry, what "classical authors" generally regard the xyston as wielded with two hands? I can't think of an author who says that; the tacticians seem to equate xystophoroi and kontophoroi, if I remember the terminology correctly, but I don't think that's enough to press details like how many hands you hold the spear with. On the Mosaic, the Sarcophagus (despite themissing weapons, you can tell where they went!), the Porus Medallion, and later works such as the Marissa fresco and as late as Ptolemy on the Raphia Decree, Hellenistic art consistently shows the cavalry spear held in one hand.

[Patrick Waterson]

The xyston is generally (among classical authors) considered to be a weapon wielded with two hands but in the Alexander Mosaic it is wielded in one hand and still goes right through a presumably armoured(?) opponent's body (the opponent is one of Darius' personal guard cavalry, some of whom show indications of wearing armour - Duncan is good on the fine details of this).

Sorry, what "classical authors" generally regard the xyston as wielded with two hands? I can't think of an author who says that; the tacticians seem to equate xystophoroi and kontophoroi, if I remember the terminology correctly, but I don't think that's enough to press details like how many hands you hold the spear with. On the Mosaic, the Sarcophagus (despite themissing weapons, you can tell where they went!), the Porus Medallion, and later works such as the Marissa fresco and as late as Ptolemy on the Raphia Decree, Hellenistic art consistently shows the cavalry spear held in one hand.

Dionysius of Halicarnassus (XX.11.2) refers to "cavalry spears grasped by the middle with both hands," which seems unsuitable as a description of any Roman cavalry weapon we know, even the cuspis.  As in earlier discussions, he seems to mean a pilum, which in his source may have been given as 'xyston', as in Josephus.

However as you mention the artistic evidence seems universally in favour of single-handed use, which puts the two-handed extension of the centre-of-gravity argument backstage: it may actually be a redundant point as on the Alexander Mosaic we see Alexander employing plenty of reach without needing a two-handed grip.

Hence I shall concede that the idea of a two-handed grip on the xyston is probably a misapprehension or at best an optional extra, but nevertheless maintain that the xyston would outreach a standard hoplite doru when used single-handed. 

[Erpingham]

Hence I shall concede that the idea of a two-handed grip on the xyston is probably a misapprehension or at best an optional extra, but nevertheless maintain that the xyston would outreach a standard hoplite doru when used single-handed. 

Maybe so, but how important is this in the scenario you have outlined, where someone smashes through on the file junction?  In very short order, a long weapon will become irrelevant as the rider tussles with the first couple of ranks (although hoplites in the later ranks would still be able to use their reach to participate against the intruder).

[Duncan Head]

Dionysius of Halicarnassus (XX.11.2) refers to "cavalry spears grasped by the middle with both hands," which seems unsuitable as a description of any Roman cavalry weapon we know, even the cuspis.

Ah yes, I'd forgotten that one. He's writing late enough that he could - just - be thinking of a Parthian kontos - the long spears of the Iranian cavalry tradition do seem to have been used two-handed.

As in earlier discussions, he seems to mean a pilum, which in his source may have been given as 'xyston', as in Josephus.

I still regard this as improbable speculation, of course, and don't think he means "pilum".

[Patrick Waterson]

Hence I shall concede that the idea of a two-handed grip on the xyston is probably a misapprehension or at best an optional extra, but nevertheless maintain that the xyston would outreach a standard hoplite doru when used single-handed. 

Maybe so, but how important is this in the scenario you have outlined, where someone smashes through on the file junction?  In very short order, a long weapon will become irrelevant as the rider tussles with the first couple of ranks (although hoplites in the later ranks would still be able to use their reach to participate against the intruder).

If the Macedonian cavalry wedge and its weaponry operated as I think it did, there would not be a tussle with the first couple of ranks: the point man of the wedge would knock down his target (ahead right) before the latter could respond, his horse would shoulder aside the ahead left man (whose shield would be in the way of any counteraction against horse or rider, and whose face would be in line for a lance-point from the Companion on the point-man's left).  Now we consider the next man in each file: from his point of view, the most immediate concern filling his future and field of view is the body of the man in the rank ahead, which is heading towards him helmet-first if propelled by a xyston point or midriff-first if propelled by a horse.  Either way he is not going to be in a position to do much other than try to keep his balance, and probably stagger back on the man behind him.

Meanwhile, the wedge point man will have shaken off or dropped his first victim and be looking for the first face in his immediate forward field of view, which will receive a bloody lance point unless the recipient's shield can be interposed in time, in which case momentum and angles will send him backwards, perhaps with his own shield rim between the eyes.  As this happens, the men at the edges of the wedge will be putting points into their own targets, spreading the unbalancing and disorder with a ripple effect that steadily eats through the hoplite formation.  As long as the cavalry can keep moving and keep putting their lance-tips where they will do most good, they will not have to slow down and can steadily ride through their opponents, keeping ahead of the 'reaction loop' of any rear-rankers trying to get into position to do them harm.

In essence, the Macedonian wedge would have been a continuously moving body whose spearpoints cleared a path for it, in concept not too unlike the Macedonian phalanx.  It would not have worked as a charge in-stop-draw swords-hack away system, or at least not very well.

There is really only one counter to this form of attack, or only one that I can see being effective on level ground, and that is to have longer weapons than the attackers.  The sarissa of the Macedonian phalanx fulfils this condition, and it is possible that the longer spear of the Iphicratic inventory might also do so, although such Iphicratic spears do not seem to have been used by hoplites in Greece.

[Patrick Waterson]

Dionysius of Halicarnassus (XX.11.2) refers to "cavalry spears grasped by the middle with both hands," which seems unsuitable as a description of any Roman cavalry weapon we know, even the cuspis.

Ah yes, I'd forgotten that one. He's writing late enough that he could - just - be thinking of a Parthian kontos - the long spears of the Iranian cavalry tradition do seem to have been used two-handed.

As in earlier discussions, he seems to mean a pilum, which in his source may have been given as 'xyston', as in Josephus.

I still regard this as improbable speculation, of course, and don't think he means "pilum".

I appreciate the reservations, but wonder exactly how we would rationalise Roman principes being armed with anything other than a pilum as their primary weapon.  Triarii with the hasta fit form to function, but principes seem to be equipped similarly to, and have a role almost identical to, hastati (who despite their name seem to have centred their activities and armoury on the pilum).

If we develop this point it will probably want its own thread.

[Jim Webster]

As long as the cavalry can keep moving and keep putting their lance-tips where they will do most good, they will not have to slow down and can steadily ride through their opponents

You forget one small technical problem, if they're actually killing people and they're going down, the floor is now a mess of writhing bodies, shields and sharp objects. The horses might not panic but they will have to slow down to keep their footing.
If the infantry stand and die, the horses will be slowed. It's if the infantry break that the horses can keep going.
What makes the infantry stand and die is the back rank who are NCOs and ideally lock their shields and keep everybody at it.

Jim

[Patrick Waterson]

You forget one small technical problem, if they're actually killing people and they're going down, the floor is now a mess of writhing bodies, shields and sharp objects. The horses might not panic but they will have to slow down to keep their footing.
If the infantry stand and die, the horses will be slowed. It's if the infantry break that the horses can keep going.
What makes the infantry stand and die is the back rank who are NCOs and ideally lock their shields and keep everybody at it.

I think most of the infantry will be knocked back (or down) and the infantry rear ranks will tend to fall back out of the way as friendly bodies hurtle towards them for one reason or another, which partly does the cavalry's work for them.  The formation will loosen up and the Companions will keep on driving through.  Slowing down for bodies, equipment etc. would be relative rather than absolute, i.e. I envisage the wedge continuing to move at a trot or better as it surges through the infantry formation (rather faster if up against Kardakes at Issus, who did not stand, let alone die in place).  The Alexander Mosaic probably captures the nature of such a drive quite well: there are bodies underfoot but the Macedonian cavalry are nevertheless surging forward (though one does wonder how Alex is going to get that Persian off his xyston).

There was one case when the infantry very definitely did stand and die, and that was at the Granicus: being surrounded and presumably in a solid rectangular formation, the hoplites would have had minimal space to 'recoil' individually or collectively, so when Alex and his wedge charged in they would have hit with their usual initial impact but the stricken hoplites would have become jammed in among their crowded comrades and the wedge would have got stuck - which is entirely consistent with Alex's horse being struck through the ribs by a sword (not so easy to do against a briskly moving animal).

The Sacred Band at Chaeronea received their wounds 'all in front', but whether the file closers could have held their files together against repeated shocks by increasing numbers of thousand-pound horses is another matter.

[Jim Webster]

You forget one small technical problem, if they're actually killing people and they're going down, the floor is now a mess of writhing bodies, shields and sharp objects. The horses might not panic but they will have to slow down to keep their footing.
If the infantry stand and die, the horses will be slowed. It's if the infantry break that the horses can keep going.
What makes the infantry stand and die is the back rank who are NCOs and ideally lock their shields and keep everybody at it.

I think most of the infantry will be knocked back (or down) and the infantry rear ranks will tend to fall back out of the way as friendly bodies hurtle towards them for one reason or another, which partly does the cavalry's work for them.  The formation will loosen up and the Companions will keep on driving through.  Slowing down for bodies, equipment etc. would be relative rather than absolute, i.e. I envisage the wedge continuing to move at a trot or better as it surges through the infantry formation (rather faster if up against Kardakes at Issus, who did not stand, let alone die in place).  The Alexander Mosaic probably captures the nature of such a drive quite well: there are bodies underfoot but the Macedonian cavalry are nevertheless surging forward (though one does wonder how Alex is going to get that Persian off his xyston).

Hang on, moving at a trot is not compatible with infantry bodies hurtling. I've been struck by half ton livestock moving at a trot or thereabouts. You don't hurtle. You are more likely to be turned and pushed aside. They don't aim AT you, they try to go to one side, hence you're more likely to be spun.
Another thing is that horses and cattle do not charge walls and that is what they see with the shields. There are no gaps, the animal will go for the gap and to make a gap you've got to kill someone.
To kill someone you either exploit a gap caused by a missile weapon (Probably the Persian technique) or you've got to kill him with your spear. Remember unless you can convince your horse to charge a solid obstacle (the wall of shields) you've got to stab one of the infantry and move your horse into his space.

There was one case when the infantry very definitely did stand and die, and that was at the Granicus: being surrounded and presumably in a solid rectangular formation, the hoplites would have had minimal space to 'recoil' individually or collectively, so when Alex and his wedge charged in they would have hit with their usual initial impact but the stricken hoplites would have become jammed in among their crowded comrades and the wedge would have got stuck - which is entirely consistent with Alex's horse being struck through the ribs by a sword (not so easy to do against a briskly moving animal).

The Sacred Band at Chaeronea received their wounds 'all in front', but whether the file closers could have held their files together against repeated shocks by increasing numbers of thousand-pound horses is another matter.

Hang on again, you now have briskly moving animals. Before they were only trotting and whilst I've never had to hit a briskly moving animal with a sword, I've had to turn them using a length of blue alkathene and it is perfectly simple to do, from the side they're the size of a barn door, trust me in this, you would struggle to miss.
Also at Granicus, when the Hoplites were killed, did the Cavalry even hit them frontally, or did they wait for light infantry and pikes to break them up a bit?

Jim

[Mark G]

don't forget also, the length of the lance is effectively reduced by the length of the horse.

whilst the hoplite can spear the horse's head , the rider has to wait until the horse has pushed far enough past the shield to get his lance into range.

that is assuming the horse can find somewhere to push into, as Jim says.

[Jim Webster]

don't forget also, the length of the lance is effectively reduced by the length of the horse.

whilst the hoplite can spear the horse's head , the rider has to wait until the horse has pushed far enough past the shield to get his lance into range.

that is assuming the horse can find somewhere to push into, as Jim says.

You don't even need to spear the head, catch the face with the edge of the spearhead  and you're going to deter the horse. Indeed a spear point waving about uncertainly near its eyes is going to freak it out (Which is why horse armour included eye protection)

Jim

[Erpingham]

As I understand it, the advantage of a wedged charge, as oppossed to a linear one, is it concentrates all the "break in" effort in one place.  Most of the linear formation being charged is out of the equation (at least to begin with), whereas a small group of cavalry are in contact with their support close at hand.  It is still, however, going to be a hard fought thing to go through a close order infantry formation maybe 8 deep.  We are in speculative territory about what actually happens but it seems doubtful to me at least that any length advantage of cavalry weaponry will make all the spears of the hoplite formation redundant.  It might provide the first hit but less than a second later the rider is among the spear points of the second and third ranks, who he hasn't hit.  Likewise, I think it is unlikely that anyone could just trot through poking at targets of opportunity with a twelve foot lance.  It is more likely that it would be ditched when it jams into its first victim and the sword would be drawn.

So, I'd speculate that in a successful attack, the wedge leader will push several ranks deep into the phalanx, with half a dozen comrades trampling around widening the gap behind.  Momentum will have slowed, horses and men would be wounded but much of the wedge is still behind and not in contact ready to exploit the pushing apart of the ranks, yet preventing any rushing out to envelop.  Thereafter, it becomes a battle as to whether the wedge can push apart the formation and break through or whether the cavalry point are brought down or pushed out again.  In an unsuccessful attack, the point mans horse runs onto a spear or trips over a falled infantryman and goes down, creating a barrier to exploitation by those behind and the hoplites relock their shields behind the obstacle.

[Duncan Head]

As in earlier discussions, he seems to mean a pilum, which in his source may have been given as 'xyston', as in Josephus.

I still regard this as improbable speculation, of course, and don't think he means "pilum".

I appreciate the reservations, but wonder exactly how we would rationalise Roman principes being armed with anything other than a pilum as their primary weapon.  Triarii with the hasta fit form to function, but principes seem to be equipped similarly to, and have a role almost identical to, hastati (who despite their name seem to have centred their activities and armoury on the pilum).

If we develop this point it will probably want its own thread.

I can't say that I see a problem. Once, everyone had a thrusting-spear. The thrown pilum is initially introduced for the hastati; it is a success, and so it later spreads to the principes. The idea that the principes should be the same as the hastati is probably anachronistic.

I don't see how we would get a resolution to this, though, so continuing the discussion may not be worth it. 

[Jim Webster]

As I understand it, the advantage of a wedged charge, as oppossed to a linear one, is it concentrates all the "break in" effort in one place.  Most of the linear formation being charged is out of the equation (at least to begin with), whereas a small group of cavalry are in contact with their support close at hand.  It is still, however, going to be a hard fought thing to go through a close order infantry formation maybe 8 deep.  We are in speculative territory about what actually happens but it seems doubtful to me at least that any length advantage of cavalry weaponry will make all the spears of the hoplite formation redundant.  It might provide the first hit but less than a second later the rider is among the spear points of the second and third ranks, who he hasn't hit.  Likewise, I think it is unlikely that anyone could just trot through poking at targets of opportunity with a twelve foot lance.  It is more likely that it would be ditched when it jams into its first victim and the sword would be drawn.

So, I'd speculate that in a successful attack, the wedge leader will push several ranks deep into the phalanx, with half a dozen comrades trampling around widening the gap behind.  Momentum will have slowed, horses and men would be wounded but much of the wedge is still behind and not in contact ready to exploit the pushing apart of the ranks, yet preventing any rushing out to envelop.  Thereafter, it becomes a battle as to whether the wedge can push apart the formation and break through or whether the cavalry point are brought down or pushed out again.  In an unsuccessful attack, the point mans horse runs onto a spear or trips over a falled infantryman and goes down, creating a barrier to exploitation by those behind and the hoplites relock their shields behind the obstacle.

Just remember that a wedge of horsemen cannot 'push' in the same way a wedge of infantrymen can. Horses react badly to being pushed from behind, it unleashes a whole heap of dominance behaviours and they tend to bite and kick each other.
I suppose you could get the horses so tightly packed that they couldn't do this, but the riders aren't going to be a lot of use either

Jim

[Patrick Waterson]

As I understand it, the advantage of a wedged charge, as oppossed to a linear one, is it concentrates all the "break in" effort in one place.  Most of the linear formation being charged is out of the equation (at least to begin with), whereas a small group of cavalry are in contact with their support close at hand.  It is still, however, going to be a hard fought thing to go through a close order infantry formation maybe 8 deep.  We are in speculative territory about what actually happens but it seems doubtful to me at least that any length advantage of cavalry weaponry will make all the spears of the hoplite formation redundant.  It might provide the first hit but less than a second later the rider is among the spear points of the second and third ranks, who he hasn't hit.  Likewise, I think it is unlikely that anyone could just trot through poking at targets of opportunity with a twelve foot lance.  It is more likely that it would be ditched when it jams into its first victim and the sword would be drawn.

I would reiterate that in this era there seems to be no evidence of any hoplite infantry actually having an anti-cavalry drill: the earliest I can think of is Arrian's ektaxis against the Alans, although Antony's legionaries improvised a testudo-like technique against Parthians (Plutarch, Life of Antony, 45).  Hence any idea of massed ranks presenting massed spearpoints, or even presenting spearpoints at horse eye level, should be discarded.  The Macedonian Companions were new to a system that was quite unprepared for them, and this lack of previous experience would go a long way to explaining their earlier successes.  By the time their hoplite opponents could have expected to develop a counter-technique, Alexander's opponents had run out of hoplites.

If the xyston did penetrate a victim (as opposed to dealing a point-in-and-rip-out wound or a stroke that became self-extracting as the opponent fell) then it might well break, but the spiked butt would serve for at least one more stroke, by which time one might well be through the enemy formation.  As far as I can establish from the accounts of Alexander's battles, Companions resorted to sword-work only in cavalry battles (the record is not sufficiently complete to be certain on this point, but it is the impression one receives).

So, I'd speculate that in a successful attack, the wedge leader will push several ranks deep into the phalanx, with half a dozen comrades trampling around widening the gap behind.  Momentum will have slowed, horses and men would be wounded but much of the wedge is still behind and not in contact ready to exploit the pushing apart of the ranks, yet preventing any rushing out to envelop.  Thereafter, it becomes a battle as to whether the wedge can push apart the formation and break through or whether the cavalry point are brought down or pushed out again.  In an unsuccessful attack, the point mans horse runs onto a spear or trips over a fallen infantryman and goes down, creating a barrier to exploitation by those behind and the hoplites relock their shields behind the obstacle.

This seems not unreasonable, although I would not see the point man's horse running onto a spear.  Hoplites habitually used their spears overarm and do not seem to have grounded the butt except as part of kneeling/standing and waiting to move, e.g. the Spartans at Plataea.  Curiously enough, the one time when Alexander's Companions were held up and got into a shoving match was on the riverbank at the Granicus - against Persian cavalry.

Hang on, moving at a trot is not compatible with infantry bodies hurtling.

I did say 'at a trot or better', indicating that a trot would be the minimum rather than the average speed.

I've been struck by half ton livestock moving at a trot or thereabouts. You don't hurtle. You are more likely to be turned and pushed aside. They don't aim AT you, they try to go to one side, hence you're more likely to be spun.

Yes, this is roughly what I would anticipate at such a speed (and I do appreciate first-hand experience in the matter ), although I would see the initial impact of the cavalry wedge and much if not all of the follow-through as moving somewhat faster.

Another thing is that horses and cattle do not charge walls and that is what they see with the shields. There are no gaps, the animal will go for the gap and to make a gap you've got to kill someone.
To kill someone you either exploit a gap caused by a missile weapon (Probably the Persian technique) or you've got to kill him with your spear. Remember unless you can convince your horse to charge a solid obstacle (the wall of shields) you've got to stab one of the infantry and move your horse into his space.

The 'wild card' here is training: both the training of the cavalry (including their mounts) and the training of the infantry they are up against.  A wall of shields is not 'solid' so much as continuous.  One can conjecture about what shields-on-poles arrangements the Macedonians may have set up for training purposes, but I do not see training a warhorse to push between shields (aim for the bit where they join or overlap) as being particularly challenging.  Besides, the point man will have removed the owner of one of those shields with his lance by the time the horse gets there and the said owner will be (involuntarily) doing his bit to open up the rank behind.

Hang on again, you now have briskly moving animals. Before they were only trotting* and whilst I've never had to hit a briskly moving animal with a sword, I've had to turn them using a length of blue alkathene and it is perfectly simple to do, from the side they're the size of a barn door, trust me in this, you would struggle to miss.

*Not quite; 'at a trot or better'.

The pertinent question would seem to be: how easy is it to do this while the next animal's rider is putting a lance point into your neck or face?  This is why I think that as long as the cavalry kept moving they would be inside the 'reaction loop' of both individual infantrymen and the infantry formation as a whole.

Also at Granicus, when the Hoplites were killed, did the Cavalry even hit them frontally, or did they wait for light infantry and pikes to break them up a bit?

Nobody tells us explicitly, but the impression one gets is of a simultaneous attack from all sides.  It would of course make sense to use archery to prepare the sector to be hit by the wedge, and we do note that at the Granicus the Macedonian cross-river cavalry attack is accompanied (though not prepared) by Agrianes and archers, while at Issus the archers are deployed so that they seem able to 'shoot in' the Companions.  That said, the Companions did not wait at Issus because of the 'incoming mail' being sent by the Persians, but at Chaeronea or when dealing with the mercenaries at the Granicus I would expect Alexander to have taken advantage of a preliminary shower of arrows from the archers - just enough to cause an upset rather than a barrage to make a path of dead bodies - a brief 'hurricane bombardment'.  So yes, the clues suggest that the defenders were probably 'prepped' by a bit of archery in each case when the Companions triumphed over hoplites, or Patrick thinks they did.