The mechanism of Roman line relief

[andrew881runner]

if what you said happened, which is enemies going among the centuriae of ha stati after they had positioned the back centuria in the back again, the enemies would be surrounded  on 3 flanks, the principes advancing in front and the ha stati centuriae in the flanks. So this would be even worse for enemies. Enemies would have to rush following the centuriae of ha stati retreating (and  I think very strange that someone seeing the Roman doing some complex maneuver has the initiative to break the line of battle to push forward in that very moment, convincing the other guys to do the same... I think that most probable reaction would be astonishment in watching this sudden manoeuvre during the fight, unless they had already seen it, but I don't think so many survived to many battles against Romans, historically if they saved their life in one battle it was a great result) only to find themselves trapped among the centuriae oh ha stati in the flanks and the fresh heavy principes running fiercely against them.
Anyway I am not saying this was a simple maneuver. It needed a big effort and timing and probably sometimes something went wrong. Since doing it in a game is a bit simpler than doing it in real life. But the general result, putting away tired men and pouring fresh men into the battle when necessary, was what gave a big advantage to Romans.
We all know that Romans liked simple and effective things, not complicated ones (watch for example the structure of their camps, simple, rational). I think they used this tactics because it was the simplest and most effective to achieve this result.
Then you have to consider that mainly enemies of Rome which were not Germans or gauls used tactics based on phalanx or shield walls (even Germans sometimes). A phalanx has not the flexibility to do a fast rush in the gaps created by Romans doing their tactical retreat. As I said, this could happen if someone understood in seconds what was happening and convinced other guys, in seconds, to follow him to make a very dangerous rush with unknown effect. This during a battle where main concern of people over there was generally to stay alive, thus keeping as possible a good order to achieve that goal. [emoji1]

[Justin Swanton]

Each file (except one, on a flank) would pass between two files of principes, who would pick up the combat as it reached them.

This hearkens back to the beginning of this thread where we looked at the precise meaning of 'ordo' for Livy. Mind if I reproduce part of an earlier post?

Let me insert the original latin words into the English translation: [3] The Romans had formerly used small round shields; then, after they began to serve for pay, they made oblong shields instead of round ones; [4] and what had before been a phalanx, like the Macedonian phalanxes, came afterwards to be a line of battle formed by maniples [manipulatim = 'in a maniple-like fashion'], with the rearmost troops drawn up in a number of companies [ordines]. [5] The first line, or hastati, comprised fifteen maniples [manipuli], stationed a short distance apart; the maniple [manipulus] had twenty light—armed soldiers, the rest of their number carried oblong shields; moreover those were called "light—armed" who carried only a spear and javelins. [6] this front line in the battle contained the flower of the young men who were growing ripe for service. behind These came a line of the same number of maniples [manipulorum], made up of men of a more stalwart age; these were called the principes; they carried oblong shields and were the most showily armed of all. [7] this body of thirty maniples [manipulorum] they called antepilani, because behind the standards there were again stationed other fifteen companies [ordines], each of which had three sections [partes], the first section in every company [actually 'of each'] being known as pilus. [8] The company [ordo] consisted of three vexilla [vexillis] or "banners"; a single vexillum had sixty soldiers, two centurions, one vexillarius, or colourbearer; the company  [actually 'They'] numbered a hundred and eighty—six men. The first banner [vexillum] led the triarii, veteran soldiers of proven valour; the second banner [just 'the second'] the rorarii, younger and less distinguished men; the third banner [just 'the third'] the accensi, who were the least dependable, and were, for that reason, assigned to the rear most line. [9] when an army had been marshalled in this fashion, the hastati were the first of all to engage. if the hastati were unable to defeat the enemy, they retreated slowly and were received into the intervals between the companies [ordinum] of the principes. The principes then took up the fighting and the hastati followed them. [10] The triarii knelt beneath their banners, with the left leg advanced, having their shields leaning against their shoulders and their spears thrust into the ground and pointing obliquely upwards, as if their battle—line were fortified with a bristling palisade. [11] if the principes, too, were unsuccessful in their fight, they fell back slowly from the battle—line on the triarii. (From this arose the adage, "to have come to the triarii," when things are going badly.) [12] The triarii, rising up after they had received the principes and hastati into the intervals between their companies [ordinum], would at once draw their companies [ordinibus] together and close the lanes, as it were; then, with no more reserves behind to count on, they would charge the enemy in one compact array. One could say that from [9] onwards Livy uses 'company' to designate a file but it does seem to strain the meaning as he has used it so consistently up to now to designate a much larger body of men. Add to that the mention of intervals between the maniples - which the Triarii companies would have replicated in their own deployment - and the logical conclusion would be that the 'intervals between the companies' of [12] corresponds to gaps between one grouping of vexilla and the next.

I just couldn't convince myself that 'ordo' in Livy's context meant 'file'.

[Mark G]

That's another valid interpretation.

Read Roys from last week again for a third.

[Erpingham]

If the relief system depends upon a break in the fighting, then a) any relief system (not just the Roman one) will work during a break so anyone can do it, and b) if there is no break in the fighting the system cannot work.

I think you are oversimplifying a bit.  Any relief system by a highly trained army would work if it would fit into the sort of timeframe required.  You have only got as long as your opponent takes to remotivate and relaunch the fight.  If they were winning and you were losing, this may not be long.  So, it would need to be an army organised in such a way as to allow relief and trained to do it.  Is anyone else this organised?  I agree on the second point but , then, some of us believe there were normally breaks and others think there weren't any breaks, so how realistic you think such a system is depends where on this continuum you sit. 

[Patrick Waterson]

if what you said happened, which is enemies going among the centuriae of hastati after they had positioned the back centuria in the back again, the enemies would be surrounded  on 3 flanks, the principes advancing in front and the hastati centuriae in the flanks. So this would be even worse for enemies. Enemies would have to rush following the centuriae of hastati retreating (and  I think very strange that someone seeing the Roman doing some complex maneuver has the initiative to break the line of battle to push forward in that very moment, convincing the other guys to do the same... I think that most probable reaction would be astonishment in watching this sudden manoeuvre during the fight, unless they had already seen it, but I don't think so many survived to many battles against Romans, historically if they saved their life in one battle it was a great result) only to find themselves trapped among the centuriae of hastati in the flanks and the fresh heavy principes running fiercely against them.

Let us review what is involved: I do agree that for the relief manoeuvre the principes would have to move up close behind the hastati.  However under the proposed system the hastati have to disengage alternate centuries and then move them laterally behind the still-engaged centuries.  This means that 50% of the fighting line is removed from fighting and as soon as it has turned is also offering its flank to any enemy who follow them up.  The remaining 50% is under attack from the front and both flanks.

Now the centuries who have fallen back and turned to offer their flanks to any enemy who follows them up have to move behind the engaged centuries - and must complete this manoeuvre before the principes can do anything at all.  If they are being pressed in the flank by an enemy who has followed up they can move only by sacrificing the outer file or files.  Meanwhile their movement cuts off the retreat of the engaged centuries, who are still under attack from the front and both flanks.

Anyway I am not saying this was a simple maneuver. It needed a big effort and timing and probably sometimes something went wrong. Since doing it in a game is a bit simpler than doing it in real life. But the general result, putting away tired men and pouring fresh men into the battle when necessary, was what gave a big advantage to Romans.
We all know that Romans liked simple and effective things, not complicated ones (watch for example the structure of their camps, simple, rational). I think they used this tactics because it was the simplest and most effective to achieve this result.

I think that pinpoints the problem with this relief system.    I think you are right about the Roman preferring simple solutions, and would suggest that the actual system they used was much simpler than that proposed.

Then you have to consider that mainly enemies of Rome which were not Germans or Gauls used tactics based on phalanx or shield walls (even Germans sometimes). A phalanx has not the flexibility to do a fast rush in the gaps created by Romans doing their tactical retreat. As I said, this could happen if someone understood in seconds what was happening and convinced other guys, in seconds, to follow him to make a very dangerous rush with unknown effect. This during a battle where main concern of people over there was generally to stay alive, thus keeping as possible a good order to achieve that goal. [emoji1]

Rome's formative campaigns were mainly fought against Sabines, Volsci, Hernici, Aequi and above all Samnites, who seem to have used more flexible formations and whom Livy portrays as being aggressive in their battlefield behaviour.  They also regularly fought Etruscans, who do seem to have used at least a phalanx-related system for part of their history, and Gauls, who used their own system which involved impetuous attacks and shieldwall-type defences (cf. Sentinum 295 BC and Telamon, 225 BC).  Most of Rome's opponents do not appear to have used a phalanx, but rather a more flexible system.  Furthermore, even the traditional hoplite phalanx could rapidly exploit gaps in an enemy line, as at Delium in 424 BC and First Mantinea in 418 BC.

If the relief system depends upon a break in the fighting, then a) any relief system (not just the Roman one) will work during a break so anyone can do it, and b) if there is no break in the fighting the system cannot work.

I think you are oversimplifying a bit.  Any relief system by a highly trained army would work if it would fit into the sort of timeframe required.  You have only got as long as your opponent takes to remotivate and relaunch the fight.  If they were winning and you were losing, this may not be long.  So, it would need to be an army organised in such a way as to allow relief and trained to do it.  Is anyone else this organised?  I agree on the second point but , then, some of us believe there were normally breaks and others think there weren't any breaks, so how realistic you think such a system is depends where on this continuum you sit. 

The Roman line relief system had to be able to operate under adverse pressure.  If opponents were considerate enough to pause just when the Romans wanted to relieve themselves - I shall rephrase that - to exchange lines, then all well and good: one could have a trooping the colour display and it would serve.  If not, then the Roman system would lead to defeat against every opponent who seized upon the simple truth that keeping up the pressure would render the system - and the Roman army - inoperative.

So - in what way can we pass a tired line through a fresh line without loss of continuity in fighting?

[andrew881runner]

Patric you don't get my point. I explained exactly how enemies would find in a worse situation if they tried to follow retreating ha stati, since they would have been surrounded on 3 sides. Read newly my point. so that relief system I have shown can even take into account a very hypothetical enemy who see Romans do a quick manoeuvre and suddenly decides to do a powerful attack in that moment (how this could happen, I have no idea though, since it would imply a loss of cohesion in the battle line even for attackers)

[Erpingham]

If not, then the Roman system would lead to defeat against every opponent who seized upon the simple truth that keeping up the pressure would render the system - and the Roman army - inoperative.

Only if we assume that, without line relief, the Roman army collapsed.   Given the later examples which we have dredged up in another thread where the Romans seem to have been able to use their reserve line to manoeuver to reinforce a flank, we should also be open to the idea that Romans could use their deployment in depth to impovise a functional reserve even if prevented from line relief. It is also true that, not being Roman, the attacker can't recycle troops so the ranks doing the pressuring are growing tired as fast as, if not faster than, the Romans. The fighting would therefore still be more balanced that perhaps you suggest.

[Justin Swanton]

There is another way of making line relief work that takes up Patrick's idea whilst preserving the meaning of Livy's 'ordo' as a company rather than a single file. Let me give the mechanism first and then see if it fits the sources second. (This theory reproduces an idea of Paul Bardunias on the Arkaion Bellum group)

The hastati are drawn up with the standard 3' from the centrepoint of one file and rank to the next, a combat formation. Behind them the principes and triarii are drawn up in open formation, with 6' between the centrepoint of one file and the next, whilst maintaining 3' between the centrepoints of the ranks. This is a standard marching formation which leaves clear gaps between the files. Thus:

The hastati are hammered back by their opponents against the principes. This is the automatic signal for line relief to begin.

The rearmost ranks of the hastati begin to filter down the passageways between the files of the principes.

This process continues until the frontmost ranks of the hastati rest against the principes.

The last of the hastati cover the gaps between the files of the principes. It is time for the handover to take place.

As the last hastati fall back in the gaps, the second man of each file of principes moves up quickly to take his place in the front rank. This switch takes only a moment, not enough time for the enemy to exploit it. The remainder of each principes file moves up.

As the remaining hastati pass down the passageways, the principes that count as even numbers in their files fill the gap left behind whilst the odd-numbered principes move up. The file passageways close rapidly.

A few moments later, the entire line of principes is formed up in close battle array and giving the tired enemy a hard time.

Does this interpretation fit the sources? Looking at Livy:

"If the hastati were unable to defeat the enemy, they retreated slowly
and were received into the intervals between the companies of the
principes. The principes then took up the fighting and the hastati
followed them. The triarii knelt beneath their banners, with the left
leg advanced, having their shields leaning against their shoulders and
their spears thrust into the ground and pointing obliquely upwards, as
if their battle-line were fortified with a bristling palisade. If the
principes, too, were unsuccessful in their fight, they fell back
slowly from the battle-line on the trarii. (From this arose the adage
'to have come to the triarii,' when things are going badly.) The
triarii, rising up after they had received the principes and hastati
into the intervals between their companies, would at once draw their
companies together and close the lanes, as it were; then, with no more
reserves behind to count on, they would charge the enemy in one
compact array."

inter - 'between' - can also have the meaning of 'among(st)'. Amicus inter hostes - ' a friend amongst enemies'. Thus 'intervals between companies' - intervalles inter ordines - can be translated as 'gaps amongst (or within) the companies'. The triarii closing the 4' gaps between their files in the same manner the principes closed theirs makes good sense of this passage.

[Patrick Waterson]

An excellent set of diagrams, Justin.

Patrick you don't get my point. I explained exactly how enemies would find in a worse situation if they tried to follow retreating ha stati, since they would have been surrounded on 3 sides. Read newly my point. so that relief system I have shown can even take into account a very hypothetical enemy who see Romans do a quick manoeuvre and suddenly decides to do a powerful attack in that moment (how this could happen, I have no idea though, since it would imply a loss of cohesion in the battle line even for attackers)

It would not be a matter of deciding whether or not to follow up, because there would be every reason to follow up an opponent who is giving way.  And the hastati's 'prior' centuries would be under attack from three sides (and destroyed or at least much weakened) before the principes could move up to engage.

I did read your explanation.   It just seems to assume that the enemy will let the posterior centuries withdraw and assemble behind the prior centuries and only then push into the gap.  I cannot see any self-respecting Gaul or Samnite waiting that long instead of following up immediately.

If not, then the Roman system would lead to defeat against every opponent who seized upon the simple truth that keeping up the pressure would render the system - and the Roman army - inoperative.

Only if we assume that, without line relief, the Roman army collapsed.   Given the later examples which we have dredged up in another thread where the Romans seem to have been able to use their reserve line to manoeuver to reinforce a flank, we should also be open to the idea that Romans could use their deployment in depth to improvise a functional reserve even if prevented from line relief.

Please bear in mind that the examples of the flexible third line derive from Caesar's time (1st century BC and the 'Marian' legion) whereas the system under specific discussion is the 3rd-2nd century BC 'Polybian' legion, in which the third line consisted of triarii which, until Scipio began innovating at Ilipa and Zama, seem to have remained immobile throughout the battle.

It may incidentally be indicative that we lack any mention (as far as I know) in sources of instances where the Romans were unable to carry out line relief because of frontal pressure - being surrounded at Cannae did seem to wreck the system, and it is not clear whether the legion being pressurised by the phalanx at Cynoscephalae had managed to relieve its first and/or second lines while being pushed back.  What we do know is that Flaminius had given it up as a lost cause and moved to the Roman right to see what he could do there.  He was evidently unable to improvise a functional reserve in the circumstances, even starting with an uncommitted third line (this was 197 BC and still the Polybian legion).

It is also true that, not being Roman, the attacker can't recycle troops so the ranks doing the pressuring are growing tired as fast as, if not faster than, the Romans. The fighting would therefore still be more balanced that perhaps you suggest.

If this were the case the hastati would not feel the need for relief, as in the battle against the Insubres in 223 BC (Polybius II.33).  The need for relief occurs when the opposition are getting the better of the Romans, for whatever reason - perhaps through committing their finest troops against the hastati, who are Rome's less-than-finest.

[andrew881runner]

come on guys I really don't get why you discuss so much about a thing that I have clearly shown as it works in my video, more or less. "ha stati were received into The INTERVALS between the companies of the principes": this is exactly what I have shown in my video and it is not represented in the images you posted.

[Justin Swanton]

come on guys I really don't get why you discuss so much about a thing that I have clearly shown as it works in my video, more or less. "ha stati were received into The INTERVALS between the companies of the principes": this is exactly what I have shown in my video and it is not represented in the images you posted.

'the intervals between the companies' - intervalles inter ordines. The two meanings of inter are 1. 'in the midst of' and 2. 'between'. It's one of the rare cases where Latin is ambiguous. 'Intervals in the midst of the companies' can be perfectly well interpreted as gaps between widely-spaced files of the companies, which allows for a line relief that a) is easy for the soldier to execute (he simply has to move backwards through file gaps), and b) never leaves his flanks or rear exposed: he is not obliged to turn his back to the enemy and he does not find himself fighting more than one enemy soldier at any time. It works for me 

(BTW in adopting this mechanism I was obliged to abandon a cherished theory of my own  )

[andrew881runner]

come on guys I really don't get why you discuss so much about a thing that I have clearly shown as it works in my video, more or less. "ha stati were received into The INTERVALS between the companies of the principes": this is exactly what I have shown in my video and it is not represented in the images you posted.

'the intervals between the companies' - intervalles inter ordines. The two meanings of inter are 1. 'in the midst of' and 2. 'between'. It's one of the rare cases where Latin is ambiguous. 'Intervals in the midst of the companies' can be perfectly well interpreted as gaps between widely-spaced files of the companies, which allows for a line relief that a) is easy for the soldier to execute (he simply has to move backwards through file gaps), and b) never leaves his flanks or rear exposed: he is not obliged to turn his back to the enemy and he does not find himself fighting more than one enemy soldier at any time. It works for me 

(BTW in adopting this mechanism I was obliged to abandon a cherished theory of my own  )

but in that way principes have to change their formation from. close to open and newly to close. That seems not necessary.
Then, you don't have to give the back to the enemy according to the system I have shown. You can slowly go backward while still watching and fighting enemy if necessary. It is matter of seconds anyway. All can happens in 20 seconds. Enemy has no time to react.
Anyway explain me why there were 2 centuriae in each manipulum, one called front centuria the other called back centuria, and how the heck they were one in front of the other and then necessarly one next to the other to form. a continuous line.
Then why the gaps among manipulum in marching order were exactly of the size of a manipulum, which is the size of a centuria, since manipulum in marching order had one front centuria and one back.
All this brings to only one solution. [emoji1]
As I said, I is simple, easy and effective. And Romans loved simple things.

[Erpingham]

it is not clear whether the legion being pressurised by the phalanx at Cynoscephalae had managed to relieve its first and/or second lines while being pushed back.  What we do know is that Flaminius had given it up as a lost cause and moved to the Roman right to see what he could do there.  He was evidently unable to improvise a functional reserve in the circumstances, even starting with an uncommitted third line (this was 197 BC and still the Polybian legion).

Forgive me my poor recollection of Hellenistic battles but isn't Cynoscephalae where an unnamed tribune rescues the situation by improvising and taking troops on a flank attack?

As you have pointed out we don't have many cases where the Romans fail to do line relief (there is the one where they don't leave enough room, but don't they win anyway?).  It may have happened in some of those battles where a consular army is beaten up but we have no details.  But as usual, we are speculating from the evidence we have.

[Justin Swanton]

Yes. The tribune left the hastati to continue pursuing the broken phalanx on the Macedonian left flank and led the principles and triarii around the rear of the right flank phalanx. One of the few cases of an actual battlefield column at work (seen in another thread somewhere).

[Patrick Waterson]

but in that way principes have to change their formation from. close to open and newly to close. That seems not necessary.
Then, you don't have to give the back to the enemy according to the system I have shown. You can slowly go backward while still watching and fighting enemy if necessary. It is matter of seconds anyway. All can happens in 20 seconds. Enemy has no time to react.

You can either go slowly or perform the exercise in 20 seconds: one or the other.  For the withdrawing centuries to fall back slowly to a depth of at least eight men (the assumed depth of the non-withdrawing centuries) requires then to take at least eight paces backwards - with the enemy maintaining contact, because when troops fall back the natural inclination is to press the advantage and follow up, not to let them go.  Then the withdrawing centuries turn 90 degrees, exposing a flank to these opponents in contact: I noticed in the video they offered their right flank - in combat, a death sentence.  Now, with the opponents continually cutting down helpless sideways-facing legionaries, the withdrawing centuries somehow have to move behind the centuries which did not withdraw.  I cannot see this working: the left-hand files might get clear but the rest will be crowded back and slaughtered as the enemy press forward.  Meanwhile, the principes can see nothing except the backs or sides of the hastati, but the enemy are flowing round three sides of each of the hastati centuries which did not withdraw.

So what can the principes do?  If the whole line advances, they will simply crush pockets of hastati against the enemy.  If alternate centuries of principes advance, they will enter combat as the sideways-facing hastati in front of them are massacred, but this is not line relief.

You can slowly go backward while still watching and fighting enemy if necessary.

Yes, good observation, and this seems to be the key to the whole line relief system.  If you can go backwards slowly while fighting the enemy, why not have the line of hastati go backwards through the line of principes?  This is a simple and effective way of handing over the fighting and there are no intricate manoeuvres to go wrong.  Furthermore, this method of relief does not depend upon the enemy giving you a break in which to execute the manoeuvre.

Anyway explain me why there were 2 centuriae in each manipulum, one called front centuria the other called back centuria, and how the heck they were one in front of the other and then necessarly one next to the other to form. a continuous line.

Nobody has entirely explained this point, largely because our sources do not.  However if we look at Polybius VI.40, where he describes the Roman army on the march when contact with the enemy is anticipated, we get a clue:

"The order of march, however, is different at times of unusual danger, if they have open ground enough. For in that case they advance in three parallel columns, consisting of the Hastati, Principes, and Triarii: the beasts of burden belonging to the maniples in the van are placed in front of all, those belonging to the second behind the leading maniples, and those belonging to the third behind the second maniples, thus having the baggage and the maniples in alternate lines. With this order of march, on an alarm being given, the columns face to the right or left according to the quarter on which the enemy appears, and get clear of the baggage. So that in a short space of time, and by one movement, the whole of the heavy infantry are in line of battle." - Polybius VI.40

This would indicate that the 'prior' century marches in front and the 'posterior' century behind, so that when they face right or left as above they are side-by-side and the maniple is ready for action.  'Prior' and 'posterior' would thus seem to designate marching order and not position on the battlefield.  This is a deduction from a clue, but seems to fit.

Then why the gaps among manipulum in marching order were exactly of the size of a manipulum, which is the size of a centuria, since manipulum in marching order had one front centuria and one back.

But where in any of our sources does it state that such gaps between manipuli existed?  Incidentally, we seem to be in agreement that 'prior' and 'posterior' designate the marching order for a maniple's centuries.