How continuous was combat?

[Duncan Head]

Once the lines closed, they stayed closed, at least according to our sources

Begs the question - this is precisely the point at issue.

So - would sir like to provide a few source quotes showing them periodically opening up again?

Perhaps there is at least one:

Thus urged on by animosity and ambition they assailed each other, considering this their own affair rather than that of their generals. Being veterans they raised no battle-cry, since they could not expect to terrify each other, nor in the engagement did they utter a sound, either as victors or vanquished. As there could be neither flanking nor charging amid marshes and ditches, they met together in close order, and since neither could dislodge the other they locked together with their swords as in a wrestling match. No blow missed its mark. There were wounds and slaughter but no cries, only groans; and when one fell he was instantly borne away and another took his place. They needed neither admonition nor encouragement, since experience made each one his own general. When they were overcome by fatigue they drew apart from each other for a brief space to take breath, as in gymnastic games, and then rushed again to the encounter. Amazement took possession of the new levies who had come up, as they beheld such deeds done with such precision and in such silence.

[Patrick Waterson]

Well spotted, Duncan.   Any more?

I looked to see whether such behaviour was replicated in Appian's account of Pharsalus but found only this:

In the rest of the field killing and wounding of all kinds were going on, but no cry came from the scene of carnage, no lamentation from the wounded or the dying, only sighs and groans from those who were falling honorably in their tracks. The allies, who were looking at the battle as at a game, were astonished at the discipline of the combatants. So dumfounded were they that they did not dare attack Cæsar's tents, although they were guarded only by a few old men. Nor did they accomplish anything else, but stood in a kind of stupor.

[80] As Pompey's left wing began to give way his men even still retired step by step and in perfect order, but the allies who had not been in the fight, fled with headlong speed, shouting, "we are vanquished," dashed upon their own tents and fortifications as though they had been the enemy's, and pulled down and plundered whatever they could carry away in their flight. Now the rest of Pompey's legions, perceiving the disaster to the left wing, retired slowly at first, in good order, and still resisting as well as they could; but when the enemy, flushed with victory, pressed upon them they turned in flight.

Maybe praetorians did things differently ... the answer would, however, seem to be that in Appian Civil Wars 3.68, the Forum Gallorum grudge fight between two contingents of praetorians in a marsh, there was nobody in effective command on either side.  One can see the same kind of thing operating at Second Bedriacum in AD 69:

... the enemy's array was now less compact; for, as there was no one to command, it was now contracted, now extended, as the courage or fear of individual soldiers might prompt. - Tacitus, Histories III.25

Note the correlation between effective command and line integrity.  This kind of individual backing and filling is to be expected from leaderless troops (no overall commander, although centurions and perhaps tribunes would still be present), and is why the Romans stipulated that a man may not leave his place in the line of battle upon pain of death and why in Caesar and Ammianus men remain where they fall while in Appian's Forum Gallorum account we have them being "instantly borne away".

On the broader subject of mediaeval infantry combat, should we consider Dupplin Moor and Agincourt?  Both of these funnelled one side into a killing ground operated by the other, feeding in men on a fairly continuous basis and considerably exercising the defenders.  But how intensively and for how long?

[Duncan Head]

Well spotted, Duncan.   Any more?

Not sure. It would be worth going through all Goldsworthy, Sabin, Zhmodikov and Koon's references and checking them, but that's a large task. At one point I thought the "repeated charges" in Livy's version of Zama were conclusive - because someone has to back off without being immediately followed up to make a second charge possible - but it turns out to be a dodgy translation.

[RichT]

So for separations and rests mid battle, we have, spread across a wide period, Duncan's Forum Gallorum (App BC 3.68), Anthony's Neville's Cross, and my Edgehill, quoted earlier. It could be argued (and perhaps will be argued by our resident contrarian), that these are all special cases and not typical, which has a little virtue, though these are just what we've found in a casual search.

given a 3.5 hour battle and 'two or three' breaks suggests that each session of fighting 'strongly and relentlessly' would have lasted - or perhaps averaged, as both sides would have become less fresh as the day wore on - about an hour.

Assuming that the 3.5 hours of the battle is wholly taken up by fighting, which is in dispute; and that the lulls themselves were much much shorter than the periods of fighting, which is in dispute.

However, when (or rather if, as I believe Anthony is not keen) we come to look at Republican Roman battles, we shall tend to find ourselves looking at somewhat different casualty totals and hence one would assume different loss rates.

I don't think so - Philip Sabin (The Face of Roman Battle) finds that the victor's loss rates in Roman battles were generally similar to those in hoplite battles (about 5 percent). He proposes a similar thought experiment to mine:

"For example, even if we assume that just 5 per cent of the troops were in the front rank, and that they struck their adversaries only every five seconds, and that less than 1 per cent of these attacks caused death or mortal injury, each army would suffer 5 per cent fatalities every ten minutes."

I think this is a question which demands and deserves an answer - is it plausible to think either that armies that could inflict 5 percent casualties every ten minutes could remain in contact for hours? Or that men could remain in contact, fighting, for hours yet only cause casualties once every thirty minutes or more? Something is wrong somewhere, and the proposed lulls are a suggested solution to the problem. Just pretending the problem doesn't exist does not seem satisfactory.

So - would sir like to provide a few source quotes showing them periodically opening up again?

Well I have said several times in this thread that lulls aren't explicitly mentioned, or else we wouldn't be having this discussion, would we? Or maybe we would, since they are explicitly mentioned on occasion (see Duncan's example), yet here we still are... But mostly, lulls are inferred.

If this is a serious proposal that such armies historically closed partially and piecemeal, I am going to have to ask for some source evidence.

Really? Do you find it incredible that armies, especially Roman armies, of tens of thousands of men, formed in three lines and divided into independent sub units, closed partially and piecemeal? I'd have thought that was implicit in all battle accounts. As Duncan says, a proper answer needs a lengthy trawl through a lot of literature, which I'm not doing now, and in particular you really do have to read Koon on Livy before we can get any further on this (even if you disagree with his findings, he catalogues all Livy's battle vocabulary). But OK, I'll follow up a few of the references in Goldsworthy. As this will make for a long post, I'll split it, coming up.

[RichT]

Caesar BC 1.43-5 (Ilerda)
"In this hope, he drew out three legions, and having formed them in order of battle, commanded the first ranks [antesignani] of one of them to run before, and gain the place. Afranius perceiving his design, despatched the cohorts that were upon guard before the camp, a nearer way to the same eminence. The contest was sharply maintained on both sides: but Afranius's party, who first got possession of the post, obliged our men to give ground, and being reinforced by fresh supplies, put them at last to rout, and forced them to fly for shelter to the legions... The first ranks therefore being put into disorder, the legion in that wing gave ground, and retired to a neighbouring hill. Caesar, contrary to his expectation, finding the consternation like to spread through the whole army, encouraged his men, and led the ninth legion to their assistance. He soon put a stop to the vigorous and insulting pursuit of the enemy, obliged them to turn their backs, and pushed them to the very walls of Lerida. But the soldiers of the ninth legion, elated with success, and eager to repair the loss we had sustained, followed the runaways with so much heat that they were drawn into a place of disadvantage, and found themselves directly under the hill where the town stood, whence when they endeavoured to retire, the enemy again facing about, charged vigorously from the higher ground. The hill was rough, and steep on each side, extending only so far in breadth as was sufficient for drawing up three cohorts; but they could neither be reinforced in flank, nor sustained by the cavalry. The descent from the town was indeed something easier for about four hundred paces, which furnished our men with the means of extricating themselves from the danger into which their rashness had brought them. Here they bravely maintained the fight, though with great disadvantage to themselves, as well on account of the narrowness of the place, as because being posted at the foot of the hill, none of the enemy's darts fell in vain. Still however they supported themselves by their courage and patience, and were not disheartened by the many wounds they received. The enemy's forces increased every moment, fresh cohorts being sent from the camp through the town, who succeeded in the place of those that were fatigued. Caesar was likewise obliged to detach small parties to maintain the battle, and bring off such as were wounded. The fight had now lasted five hours without intermission, when our men, oppressed by the multitude of the enemy, and having spent all their darts, attacked the mountain sword in hand, and overthrowing such as opposed them, obliged the rest to betake themselves to flight. The pursuit was continued to the very walls of Lerida, and some out of fear took shelter in the town, which gave our men an opportunity of making good their retreat."

Commentary - OK, it's not a pitched battle as such, but it shows all the classic Roman elements of feeding in partial forces as required and relieving those engaged (where possible) - as piecemeal as can be. The fight lasting five hours should bring joy to some, but this isn't five hours of standing banging away with swords (explicitly not). In fact it reads very much as if the whole fight (at least the continuation of the fight in the narrows) was with missiles, and only at the very end, once their missiles were spent, did Caesar's men charge with the sword - at which Afranius' men immediately fled (having adopted the Lusitanian style of skirmish fighting). So we have a five hour fight in which forces engage piecemeal and there is very little hand to hand fighting.

Appian BC 3.68 (Forum Gallorum)
"Thus urged on by animosity and ambition they assailed each other, considering this their own affair rather than that of their generals. Being veterans they raised no battle-cry, since they could not expect to terrify each other, nor in the engagement did they utter a sound, either as victors or vanquished. As there could be neither flanking nor charging amid marshes and ditches, they met together in close order, and since neither could dislodge the other they locked together with their swords as in a wrestling match. No blow missed its mark. There were wounds and slaughter but no cries, only groans; and when one fell he was instantly borne away and another took his place. They needed neither admonition nor encouragement, since experience made each one his own general. When they were overcome by fatigue they drew apart from each other for a brief space to take breath, as in gymnastic games, and then rushed again to the encounter. Amazement took possession of the new levies who had come up, as they beheld such deeds done with such precision and in such silence."

Commentary - the break for a breather we have already noted. But note also the contrast with a 'normal' fight - because of the terrain, there could be no flanking or charging, and so they closed front to front and fought it out with swords. The implication being that in a normal battle, there would be flanking and charging instead.

Appian BC 4.128 (Philippi)
"The day was consumed in preparations till the ninth hour, at which time two eagles fell upon each other and fought in the space between the armies, amid the profoundest silence. When the one on the side of Brutus took flight his enemies raised a great shout and battle was joined. The onset was superb and terrible. They had little need of arrows, stones, or javelins, which are customary in war, for they did not resort to the usual manœuvres and tactics of battles, but, coming to close combat with naked swords, they slew and were slain, seeking to break each other's ranks. On the one side it was a fight for self-preservation rather than victory; on the other for victory and for the satisfaction of the general who had been forced to fight against his will. The slaughter and the groans were terrible. The bodies of the fallen were carried back and others stepped into their places from the rear ranks. The generals flew hither and thither overlooking everything, exciting the men by their ardor, exhorting the toilers to toil on, and relieving those who were exhausted so that there was always fresh courage at the front. Finally, the soldiers of Octavius, either from fear of famine, or by the good fortune of Octavius himself (for the soldiers of Brutus were not blameworthy), pushed back the enemy's line as though they were putting in motion a very heavy machine. The latter were driven back step by step, slowly at first and without loss of courage. Presently their ranks began to dissolve and they retreated more rapidly, and then the second and third ranks in the rear retreated with them, all mingled together in disorder, crowded by each other and by the enemy, who pressed upon them without ceasing until it became plainly a flight. The soldiers of Octavius, then especially mindful of the order they had received, seized the gates of the enemy's fortification, but at great risk to themselves because they were exposed to missiles from above and in front, but they prevented a great many of the enemy from gaining entrance. These fled, some to the sea, and some through the river Zygactes to the mountains."

Commentary - another argument from the negative, but here also the engagement is identified as unusual in being face to face without the 'usual manoeuvres and tactics'. The drive back step by step like a heavy machine, if it had involved hoplites rather than Romans, would have some people reaching for the 'o' key. 

These two accounts, incidentally, remind me of:

Plutarch Pyrrhus 21.6 (Battle of Asculum)
"The Romans, having no opportunity for sidelong shifts and counter-movements, as on the previous day, were obliged to engage on level ground and front to front; and being anxious to repulse the enemy's hoplites before their elephants came up, they fought fiercely with their swords against the Macedonian spears"

Josephus BJ 6.136-147
"However, the soldiers that were sent did not find the guards of the temple asleep, as they hoped to have done; but were obliged to fight with them immediately hand to hand, as they rushed with violence upon them with a great shout. Now as soon as the rest within the temple heard that shout of those that were upon the watch, they ran out in troops upon them. Then did the Romans receive the onset of those that came first upon them; but those that followed them fell upon their own troops, and many of them treated their own soldiers as if they had been enemies; for the great confused noise that was made on both sides hindered them from distinguishing one another's voices, as did the darkness of the night hinder them from the like distinction by the sight, besides that blindness which arose otherwise also from the passion and the fear they were in at the same time; for which reason it was all one to the soldiers who it was they struck at. However, this ignorance did less harm to the Romans than to the Jews, because they were joined together under their shields, and made their sallies more regularly than the others did, and each of them remembered their watch-word; while the Jews were perpetually dispersed abroad, and made their attacks and retreats at random, and so did frequently seem to one another to be enemies; for every one of them received those of their own men that came back in the dark as Romans, and made an assault upon them; so that more of them were wounded by their own men than by the enemy, till, upon the coming on of the day, the nature of the right was discerned by the eye afterward. Then did they stand in battle-array in distinct bodies, and cast their darts regularly, and regularly defended themselves; nor did either side yield or grow weary. The Romans contended with each other who should fight the most strenuously, both single men and entire regiments, as being under the eye of Titus; and every one concluded that this day would begin his promotion if he fought bravely. What were the great encouragements of the Jews to act vigorously were, their fear for themselves and for the temple, and the presence of their tyrant, who exhorted some, and beat and threatened others, to act courageously. Now, it so happened, that this fight was for the most part a stationary one, wherein the soldiers went on and came back in a short time, and suddenly; for there was no long space of ground for either of their flights or pursuits. But still there was a tumultuous noise among the Romans from the tower of Antonia, who loudly cried out upon all occasions to their own men to press on courageously, when they were too hard for the Jews, and to stay when they were retiring backward; so that here was a kind of theater of war; for what was done in this fight could not be concealed either from Titus, or from those that were about him. At length it appeared that this fight, which began at the ninth hour of the night, was not over till past the fifth hour of the day; and that, in the same place where the battle began, neither party could say they had made the other to retire; but both the armies left the victory almost in uncertainty between them"

Commentary - as a running fight in a fortification, also not a pitched battle, but again the references to the 'advances and retreats', especially in the daytime portion of the fight, are strongly suggestive of what the normal manner of fighting was. The duration of the fight is also noteworthy of course, but again, this is quite clearly not time spent in continuous contact.

That's enough I think - I'm not trying to establish a fully worked out thesis here which is obviously a lot more work, but I think it's plain that armies could and did engage piecemeal (which doesn't mean haphazardly, I just mean in small elements, not the whole lot at once), and that fighting did not typically or necessarily mean massive blocks standing face to face for hours on end banging away at each other. I think that everyone would agree with this - it seems uncontroversial and self evident. If everyone except one person agrees, then that is undoubtedly as close as we are ever going to get, so maybe we are done here?

[Patrick Waterson]

So for separations and rests mid battle, we have, spread across a wide period, Duncan's Forum Gallorum (App BC 3.68), Anthony's Neville's Cross, and my Edgehill, quoted earlier. It could be argued (and perhaps will be argued by our resident contrarian), that these are all special cases and not typical, which has a little virtue, though these are just what we've found in a casual search.

Were I to try and draw a straight line of military practice through an isolated engagement in 42 BC to another in AD 1346 and a further engagement in AD 1642, and pass it off as representing mainline classical practice, how credible would this seem and how long would it be before someone took me to task on it?

However, when (or rather if, as I believe Anthony is not keen) we come to look at Republican Roman battles, we shall tend to find ourselves looking at somewhat different casualty totals and hence one would assume different loss rates.

I don't think so - Philip Sabin (The Face of Roman Battle) finds that the victor's loss rates in Roman battles were generally similar to those in hoplite battles (about 5 percent).

But losers' loss rates tended to be rather higher, as we saw at Forum Gallorum.

Antony and Pansa each lost about one-half of their men. The whole of Octavius' prætorian cohort perished. - Appian Civil Wars III.70

He proposes a similar thought experiment to mine:

"For example, even if we assume that just 5 per cent of the troops were in the front rank, and that they struck their adversaries only every five seconds, and that less than 1 per cent of these attacks caused death or mortal injury, each army would suffer 5 per cent fatalities every ten minutes."

I think this is a question which demands and deserves an answer - is it plausible to think either that armies that could inflict 5 percent casualties every ten minutes could remain in contact for hours? Or that men could remain in contact, fighting, for hours yet only cause casualties once every thirty minutes or more? Something is wrong somewhere, and the proposed lulls are a suggested solution to the problem. Just pretending the problem doesn't exist does not seem satisfactory.

One might also puzzle over why on a Napoleonic battlefield (apologies for bringing in a different time era and military system) it took on average a soldier's weight in shot to kill each soldier when target practice tests established that around 50% of shots fired hit a company sized target at 160 yards (see here).  The fact is we get these apparent anomalies and a serious historian simply accepts that these things happen.  In the case of Napoleonic armies we can surmise that the vast majority of troops habitually fired high because this is a frequently attested phenomenon; there is also the matter that 'company-sized targets' do not cater for depth and do not allow for multiple hits on front-rank individuals.  The consequence of people getting hold of this kind of information (e.g. that it took about 4,000 shots to cause each fatality at Vittoria in 1813) is that they start to believe that one could stand in front of a regiment of infantry at 50 yards and not be hit until the four thousandth round was fired.

In the case of classical armies, lulls are not an answer because they do not match other incidental details of battle narratives.  We do not have a clear answer why lethality should customarily be so low throughout history, but remarkably - almost incredibly - low lethality rates are endemic in warfare of just about any period (how many longbow arrows were shot per dead or incapacitated French knight? how many Lee-Enfield rounds were fired per German felled at Mons?) and particular tactical circumstances can have drastic effects on lethality.  If we seek explanations for ther classical results, I would look for them partly in equipment and mainly in technique, e.g. take away shields and see what happens; stop opponents dodging thrusts and see what happens.

I suspect the root of the matter may be a phemonenon noticed in WW2 air fighting: about 20% of the pilots score 80% of the skills, and furthermore about 2% of the pilots score around 50% of the kills.  One can train troops intensively and extensively, but until they themselves work out how to succeed they are going to use rote procedures rather than adapting their activities to their opponents.  Realistic training helps a lot, but it is still the few who acquire the lion's share of the kills - and the majority often fail to score at all in a particular engagement.

If this is a serious proposal that such armies historically closed partially and piecemeal, I am going to have to ask for some source evidence.

Really? Do you find it incredible that armies, especially Roman armies, of tens of thousands of men, formed in three lines and divided into independent sub units, closed partially and piecemeal?

Yes, given the importance of line integrity and what happens to contingents which find themselves with gaps around them (the Thespians at Delium and the Sciritae at First Mantinea, to take two examples from the same culture and time period).  Romans were particular fans of order, and the division into subunits existed to keep formations together at the lowest level, not to keep them apart.  Try to move a crowd and then try to move a regular unit and the difference will be very apparent.

I'd have thought that was implicit in all battle accounts.

But why?

[Patrick Waterson]

Now let us look at a few more Roman battle narratives.

Thapsus, 45 BC
"The battle was long, severe, and doubtful in all parts of the field until toward evening, when victory declared itself on the side of Cæsar, who went straight on and captured Scipio's camp and did not desist, even in the night, from reaping the fruits of his victory until he had made a clean sweep." - Appian Civil Wars II.97

Well ... not a lot of detail but 'long, severe and doubtful in all parts of the field' suggests a certain continuity of fighting.  (Caesar's account is somewhat different, making the whole thing sound like a walkover, albeit one without breaks.)

Munda, 45 BC.
Caesar's army is 'seized by fear' and he only gets them moving with a show of heroics (he manages to avoid or parry 200 missiles aimed at him).
"Then each of the tribunes ran toward him and took position by his side, and the whole army rushed forward and fought the entire day, advancing and retreating by turns until, toward evening, Cæsar with difficulty won the victory." - Appian Civil Wars II.104

The 'whole army' rushed forward, not bits of it here and there.  It then 'fought the entire day' with alternate advances and retreats, indicating continued contact or something very similar to it throughout.

Caesar's - or rather Hirtius' - account of Munda shuffles the early difficulties into the pre-contact phase when Caesar's men were trying to get at their opponents; note this, as Caesar's men had to get up a steep hillside, which would make any sort of breakoff or lull highly problematic.  He then moves onto the action: both armies were equal in 'congressus' and 'clamor', the approach and the war-shout, but when the pila flew Caesar's men took trivial losses whereas a 'multitudo' of Pompey's fell.  This was followed by:

Our right wing, as we have explained, was held by the men of the Tenth legion; and despite their small numbers, their gallantry none the less enabled them by their exertions to inspire no little panic among their opponents. They proceeded, in fact, to exert strong pressure on the enemy, driving him back from his positions, with the result that he began to transfer a legion from his right, to give support and to prevent our men from outflanking him. As soon as this legion had been set in motion Caesar's cavalry began to exert pressure on the enemy left wing, so that, no matter how gallantly the enemy might fight, he was afforded no opportunity of reinforcing his line. And so, as the motley din — shouts, groans, the clash of swords — assailed their ears, it shackled the minds of the inexperienced with fear. Hereupon, as Ennius puts it, "foot forces against foot and weapons grind 'gainst weapons"; and in the teeth of very strong opposition our men began to drive the enemy back. The town, however, stood them in good stead. And so they were routed and put to flight on the very day of the Liberalia; nor would they have survived, had they not fled back to their original starting point. - Spanish War 31

The loser's dead were reckoned to be at least 30,000.  Hirtius gives Caesar's as 1,000 dead, 500 wounded: an unusually low ratio of wounded.

Munda was a long and savage fight with no intermissions noted - and it is hard to see, given the topography and the action, how any could have occurred.

That may be enough for the moment - I am not trying to establish a fully worked out thesis, but I think it is plain that armies did not engage piecemeal (in small elements), but rather the whole lot at once, apart from reserve lines; and that fighting typically meant massive blocks standing face to face for hours on end banging away at each other. I think that everyone would agree with this - it seems uncontroversial and self evident. If everyone except one person agrees, then that is undoubtedly as close as we are ever going to get, so maybe we are done here?

[RichT]

OK Patrick, fine by me. I will take imitation as the sincerest form of flattery.

I'm sure anyone reading all of this can make up their own minds also, and will probably decide that the case is unproven either way, which would be a very sensible conclusion. I certainly have more questions than answers (which I think is a good thing).

The fact is we get these apparent anomalies and a serious historian simply accepts that these things happen.

[Erpingham]

On the broader subject of mediaeval infantry combat, should we consider Dupplin Moor and Agincourt?  Both of these funnelled one side into a killing ground operated by the other, feeding in men on a fairly continuous basis and considerably exercising the defenders.  But how intensively and for how long?

I've been unable to find a solid timescale for Dupplin, short of it starting in th morning and finishing in the afternoon.  Several sources give a length for Agincourt, mostly 2-3 hours but the shortest is 30 minutes  By comparison, Waurin (who was in the middle of the fight) said Verneuil lasted 45 minutes and the fighting was particularly fierce and evenly matched.

[Dave Beatty]

If you care to try your hand at swinging an axe, I have a pile of logs in the back yard that need chopping into firewood... As another modern recreation, there are numerous stories of hacking through the jungles of Vietnam or New Guinea during WWII with machetes - the hacker (no not that kind  ) had to be changed out very frequently due to exhaustion.

You might also read some of the classical accounts of the Battle of Cannae (and others for that matter) where physical exhaustion played a role in moderating the carnage.

[aligern]

The examples of continuous combat are all easily explainable as punctuated combat. If the author lives and writes in a context in which fighting goes on for say fifteen minutes and then there is a break-off, a pause and then a resumption , it will not excite comment that this happens. Someone earlier mentioned that most ancient battle reports were contained within one or two hundred words and thus phases of a combat where nothing decisive, even actually nothing at all happens do not generate a comment,nespecially as the audience all know what happens in combat. The author's comments on a battle likely concentrate on the unusual, the different and the instructive, whether militarily, or on the moral point that the author wishes to get across.
We are using information on how long a fit human can sustain arm swinging and contact with a resistant object, my boxers, Dave's machete wielders, because it marks a fixed point. The concept that a man can only manage ten to fifteen minutes of intense activity, perhaps forty five minutes of activity with rests looks unarguable . It may well be that fighters could go on for hours if the rests were long enough and if bith sides are fatiguing at a similar rate and hence have mutually assured preservation.
Punctuated flurries and withdrawals happily explains how one force can push back another a considerable distance without the loser being rolled over backwards in the first few yards, it explains how a commander can ask for one more step and be heard because his men are not in a continuous pushing contest, but are in a mutual step back. A punctuated combat explains how Romans can take advantage of a pause to retire a line and replace it.  One struggles with how a commander  can send up the third line to support if ahead of them  is a continuous line of fighters engaged in uninterrupted combat with an enemy line , presumably movng backwards if their need for relief is obvious.  Surely  it makes more sense if they are disengaged at the point of relief. 
On casualty rates I did say at the beginning that most of a frontbrank soldiers efforts will go into defensive or cautious actions rather than risky attempts to kill an opponent.  It is in his interest to slash or thrust in a way that might get advantage, but will not over extend him, or unnecessarily exhaust him.  We have raised, several times the record that the winner in an ancient battle suffers very few casualties compared to the loser, even when the armies are similar in weaponry and tactics. Also, if battle normally resulted in the extinction of the front ranks then who would stand there?
A model of battle in which there are pauses for short withdrawal after short but intense activity would expkain well how both sides tire from pushing and fencing and casualties mount.  The  side that is less fit, less well equipped or has suffered more casualties because it suffered initial deaths and de-shielding from a volley of pila begins to notice that things are going against it whlst it is taking a short rest. Its officers  shout and cajole, the braver men step forward,they fight  for another period, but the situation does not improve, at the next step back these men are visibly more tired , the enemy wre encouraged, eventually they break and run and their  opponents  kill whom they catch. That explains well how combat that causes few casualties, but results in differential rates of physical and emotional exhaustion between the two sides results in a withdrawal and then rout in which the liser suffers a high casualty rate.
Roy

[Patrick Waterson]

On the broader subject of mediaeval infantry combat, should we consider Dupplin Moor and Agincourt?  Both of these funnelled one side into a killing ground operated by the other, feeding in men on a fairly continuous basis and considerably exercising the defenders.  But how intensively and for how long?

I've been unable to find a solid timescale for Dupplin, short of it starting in the morning and finishing in the afternoon.  Several sources give a length for Agincourt, mostly 2-3 hours but the shortest is 30 minutes  By comparison, Waurin (who was in the middle of the fight) said Verneuil lasted 45 minutes and the fighting was particularly fierce and evenly matched.

Thanks, Anthony.  If we now examine the nature of the fighting, would either side at Dupplin or Agincourt have had a chance for breaks, given the apparently continual feeding in of additional troops by the attacker?  At Verneuil it looks as if the Scots fought without a break but Bedford's English who were fighting Narbonne's French gained a respite of sorts - if letting the French run and moving over to finish off the Scots can be considered a respite.  Regarding timing, does not Waurin give the action between Bedford's and Narbonne's wings as three quarters of an hour, indicating the fight against the Scots went on somewhat longer?

You might also read some of the classical accounts of the Battle of Cannae (and others for that matter) where physical exhaustion played a role in moderating the carnage.

Our main sources for Cannae are Polybius III.15-6 and Livy XXII.47-49.  The only references to physical exhaustion are in Livy, in XXII.47.9-10:

From this moment the Romans, who had gained one battle to no purpose, gave over the pursuit and slaughter of the Gauls and Spaniards and began a new fight with the Africans. In this they were at a twofold disadvantage: they were shut in, while their enemies ranged on every side of them; they were tired, and faced troops that were fresh and strong.

and in XXII.48.6:

Hasdrubal ... sent in the Spanish and Gallic cavalry to help the Africans, who were now almost exhausted, though more with slaying than with fighting.

The initial Roman exhaustion would have accelerated Roman casualties; the final African exhaustion does not appear to have reduced them.  We may note further that in response to the Roman attack on Hannibal's camp:

Hannibal, as a matter of fact, did leave a sufficient guard in his camp; and as soon as the battle began, the Romans, according to their instructions, assaulted and tried to take those thus left by Hannibal. At first they held their own: but just as they were beginning to waver, Hannibal, who was by this time gaining a victory all along the line, came to their relief, and routing the Romans, shut them up in their own camp; killed two thousand of them; and took all the rest prisoners.

So Hannibal's men were not too exhausted to win a new fight and carry out a fresh pursuit.  That said, I think the troops Hannibal took to help his camp would have been the peltasts and slingers who had done the initial skirmishing and had been at a loose end since then, as everyone else was busy at the time.  They would thus have been fresher then most.

What stands out in Polybius' and Livy's narratives is that there were no 'rest breaks'.  We shall now examine whether this is an omission on account of familiarity.

Returning for a moment to Richard's listing of Ilerda, Forum Gallorum and the questionable inclusion of Titus' assault on the Jerusalem Temple.

The examples of continuous combat are all easily explainable as punctuated combat. If the author lives and writes in a context in which fighting goes on for say fifteen minutes and then there is a break-off, a pause and then a resumption , it will not excite comment that this happens.

The question which arises is: if this is the case, then why does Caesar comment on it at Ilerda and Appian at Forum Gallorum?  If it were the familiar norm, why mention it at all?

Titus' assault: to use an action against Jews and claim it is typical of Roman fighting methods is inappropriate: Jews of the 1st century AD typically used hit-and-run tactics, e.g. during the retreat of Cestius Gallus in Josephus, Jewish War II.19.7-9 (Whiston chapters).  So for that matter did Spanish tribes - and, by Caesar's time, Roman soldiers who had served for a long time in Spain (as encountered at Ilerda).  Such actions do not tell us much, if anything, about standard practice (except of course for Jewish and Spanish standard practice).

The concept that a man can only manage ten to fifteen minutes of intense activity, perhaps forty five minutes of activity with rests looks unarguable.

But this considers only the average unfit 20th century specimen.  There is also - particularly in view of earlier mention of the traditional pace of human activities - the matter of just how 'intense' this supposedly intense activity was.  The rhythm of life back then was slower and easier.  Classical societies encouraged, even taught, economical and energy-efficient fighting styles.  Roman soldiers and mediaeval knights engaged in protracted exercise sessions with double-weight weapons to build up their stamina.  The manpower came from an essentially healthy, generally well-fed stock accustomed to day long hard physical labour.  I do not think we can assign them limitations based on our own.  (They did have limitations, naturally, but we may need to reset our sights on what those were.)

[Erpingham]

I've been unable to find a solid timescale for Dupplin, short of it starting in the morning and finishing in the afternoon.  Several sources give a length for Agincourt, mostly 2-3 hours but the shortest is 30 minutes  By comparison, Waurin (who was in the middle of the fight) said Verneuil lasted 45 minutes and the fighting was particularly fierce and evenly matched.

Thanks, Anthony.  If we now examine the nature of the fighting, would either side at Dupplin or Agincourt have had a chance for breaks, given the apparently continual feeding in of additional troops by the attacker?  At Verneuil it looks as if the Scots fought without a break but Bedford's English who were fighting Narbonne's French gained a respite of sorts - if letting the French run and moving over to finish off the Scots can be considered a respite.  Regarding timing, does not Waurin give the action between Bedford's and Narbonne's wings as three quarters of an hour, indicating the fight against the Scots went on somewhat longer?

[/quote]

The fighting at Agincourt and Dupplin was atypical, so we need to beware of using it as a model for all medieval combat.  In both battles, an attack on a narrow front was stopped then assaulted from the sides.  Withdrawal was impossible because a larger second force was fed into the melee from behind.  The two effects compressed the fighting lines meaning that they didn't have room to fight properly and many died by crush effects. In both cases, piles of bodies developed around the edges of the formation, which would suggest the lines were static.  For another battle with this effect, see Rosebeke (the compression here was caused by cavalry, with the blocking force again dismounted men-at-arms).  Othee also is said to exhibit this effect but I haven't looked at the original sources for that one (perhaps I should).

As to Waurin's timing, he is referring to the melee stage of the battle, after the cavalry charges and the archery exchange, before the rout.  He doesn't distinguish between the fight against the French and against the Scots in this timing.  He himself was probably fighting the French, which may have given rise to the confusion.

Regarding exhaustion, it is worth reading some of the exploits of that medieval superman, Don Pero Nino.  His biographer give a detailed description of his heroics at the siege of Pontevedra in 1397.  Fighting on foot (after an earlier interlude of fighting on horseback), Pero Nino led his men in a melee outside the castle.  The enemy then fell back and he pursued them to the bridge, where he took a cross bow hit.  He then assa  ulted the bridge, took another crossbow bolt in the face (which dazed him for a while) and forced his way across the steps of the bridge, taking sword blows to the helmet and a nasty shield blow to the face (nasty because the crossbow bolt was still lodged there).  Eventually exhaustion forced the two sides apart.  Pero Nino had been fighting for two hours. His biographer notes that normally Pero Nino was left unsatisfied with only an hour of fighting but his was glutted by this.  Leaving aside the heroic element in this (PN is brave, strong, wise, enormously skilled, shrugs off wounds (at one point he extracts an arrow from his leg and cauterises the wound)) there are some parameters here.  The two hour fight is not continuous but is considered one "bout" of combat.  PN is up for more combat after a mere hour but exhausted by two.  Though, as I've said, the man seems to have been extraordinary in his appetite for fighting and most men-at-arms were less fanatical. But we do know that other knights certainly trained for endurance.

[Erpingham]

I've now looked at Monstrelets description of Othee in 1408.  It isn't quite the same as the other three in that there is none of the description of death by crushing, the inability to use weapons and so on.  At Othee, the Liegous are taken in the rear by a Burgundian flanking force, so are trying to fight in two directions.  Monstrelet gives the length of the battle as "upwards of an hour", during half an hour of which, it was impossible to tell which side would win.

[Erpingham]

And, to finish up for now, Aljubarrota 1385.  Lasted by modern estimate not much more than an hour and included two separate phases of hand-to-hand combat.  The second of these was estimated by a chroniclers to have lasted 30 minutes.  Most participants only fought in one of these phases but the Portugese van fought both.  The gap between the two phases was probably only a few minutes - the castilian main division was advancing to the assistance of its van when the latter was overwhelmed.