The effects of missiles - how important were armour and shields?

[Erpingham]

This discussion started under the Slings thread but has widened to include other missile weapons.

Several points have been discussed.  The actual number of casualties caused by shooting, whether shooting at long range is effective, whether shields have any value against missiles and the effects of different armour types against missiles.

To take to start the issue of shields, among the discussions are whether it is worth having a shield except for close combat.  There has been some discussion about how shields might be used - essentially dynamically to deflect missiles or more statically in a formation presenting seried shields.  Nick Harbud in particular has presented some challenging ideas on the effectiveness and otherwise of shields, which it is hoped he will develop more fully here.

[Mark G]

shields were always a balance between an effective barrier against all threats, and a confidence building tool for the fighting man, and a specialist part of a specific weapons system.

is the shield designed to stop missile weapons, or help parry a hand to hand blow?
is the shield expected to be used as part of an attack, punching the opponent with it
is it a shield used to deflect light weapons, rather than stop the missile dead?
what is the main arm for the shield bearer?
what are the expected weapons of the opponents?

all this matters far more than whether one specific enemy with one type of weapon could penetrate the shield.

not to mention that the ancients took great care to protect the shield less side of men, which in itself must indicate the importance they placed on them - or at least did until armour reached a certain point - and penetrative weapons reached a certain point - that the weight an encumbrance became unwelcome.

[Nick Harbud]

Thanks Tony,

For those who would like to catch up on the discussion so far, the link below should send you to the "Slings...." thread.  One's own humble contributions and the ensuing digressions start around post #16.

http://soa.org.uk/sm/index.php?topic=984.0

[Patrick Waterson]

A key comment to note before proceeding is:

I thought I had made clear that I am not so much interested in casualties as effect on morale, etc, - particularly the ability of the target unit to carry through on an attack or hold its ground in the face of enemy missile fire.  In this respect, the shield is a minor factor compared to the ability of the shooters to hit their target or the opportunities they have to do so.  Research by modern archers indicates that a bowman might get off 3 shots at a charging cavalryman and up to 5 at a foot target, but the chances of a hit at the longer distances is minimal.  Pretty much all the effect comes from the closest shot, which as the Blenheim example earlier in this thread indicates, might only hit 20% of the time.  That sort of casualty rate did not discourage the attack at Blenheim and therefore, irrespective of whether one has a shield or not, it would probably not do so in antiquity.

Hence the essential point (Nick, please correct me if I have misunderstood) is that of the factors to detail when considering whether an attack succeeds or fails in the face of shooting, being shielded or otherwise is not high up on the list.

I would however question two of the above assumptions:

Pretty much all the effect comes from the closest shot

Actually my impression is the reverse: the first volley is the one that has most effect, with subsequent volleys maintaining and reinforcing such disorder and disadvantages as the first volley imposes.  If we consider Agincourt, the French attack started to go to pieces from the first serious English volley (I do not count the earlier tickling with flight arrows to get the French to move).  The closest volley, when indirect shooting would no longer have been possible, would have been comparatively ineffectual as many of the archers would not have had a clear shot - and the French did carry on to engage in melee (albeit at a significant disadvantage).  The pincushioning on the way in ruined their order, upset their orientation and turned the attack into a choked scramble.

which as the Blenheim example earlier in this thread indicates, might only hit 20% of the time.  That sort of casualty rate did not discourage the attack at Blenheim and therefore, irrespective of whether one has a shield or not, it would probably not do so in antiquity.

Most of our examples of massed missile use in the pre-gunpowder period come from mediaeval engagements, e.g. Crecy, Wisby, Poitiers, Agincourt.  While missile use rarely if ever stops an attack outright, it does considerably degrade and attenuate the attacker/defender, which admittedly is a separate question to discouragement.

I would however suggest that in a battle like Hastings, where the shield wall was an integral part of the English defence, the shields made a significant difference.  The Norman archery was never going to win the battle by itself, but was intended to soften up the English to the degree where a cavalry charge would be successful against them.  The archers duly delivered the required weight of shot - but to no avail or effect, apart from one that got in a rather important eye - and it was only later in the battle, when the angle of shooting was changed to circumvent the shields, that the archers were able to make an appreciable contribution and make the 'softening' work.

Hence, although I tend to agree that shieldedness or shieldlessness may not be critical in determining whether or not attackers close or defenders hold when being shot at, it can be a major factor in determining the overall effectiveness of shooting, especially regarding disorder or morale thresholds being reached.

[Nick Harbud]

Pretty much all the effect comes from the closest shot

Actually my impression is the reverse: the first volley is the one that has most effect, with subsequent volleys maintaining and reinforcing such disorder and disadvantages as the first volley imposes. 

Not if you follow the curve derived for musket fire, which gives for a typical charging cavalry target:

20yds      84% hits
110yds    50% hits
220yds    26% hits

As can be seen more hits are scored in the closest volley than the preceding ones put together. 

Of course, these are ideal hits against a canvas screen.  As per earlier posts, one should probably divide by at least 4 to arrive at hits against a real target.  Indeed, Hughes quotes one contemporary estimate that no more than 15% of rounds fired were effective and this is without taking into account any protection from armour, shields, etc.

[Patrick Waterson]

I would point out that musketry is a completely different weapon system from muscle-powered missile weapons.  To begin with, it is direct fire only (well, you can try shooting indirectly with muskets, but it tends to be just a waste of ammunition) whereas archery is mainly indirect, allowing many more ranks to participate effectively until the enemy is too close for an angled trajectory.  Second, it has a murderous recoil which is absent in bows, javelins, slings etc. and which tends to make musket shots go high, the more so if troops are not trained to fire low.  Third, muskets had a misfire rate variously estimated but I have seen 25% frequently touted, whereas the misfire rate for muscle-powered missile weapons was very much lower and essentially negligible.

Finally, though not necessarily exhaustively, formations were usually considerably deeper in the pre-gunpowder era, giving massed missile shooting more depth of target.  This would have increased the number of hits obtained by long-range volleys, or at any rate those using an indirect trajectory.

For these reasons I would not make comparisons between gunpowder weapons and pre-gunpowder missile weapons.  Personally, I would ditch the musketry statistics until the musket era comes round. 

[Mark G]

Pat is correct that there is nothing worth comparing between our period and the musket period.  he and I have discussed this on other levels in the past.

Agincourt's first volley offers far more fruitful comparison

- remembering that it was a steeply plunging shot to achieve the range, and was therefore not against the armoured knight, but against his unarmoured horse - and that the response of the French knights in later battles was to dismount, does indeed suggest that even against the very best bows in the very best trained hands, good armour worked.

[Erpingham]

Using Agincourt as an exemplar for longbow/cavalry interaction has its problems.  The French were both badly organised and, because of the fact that the cavalry units were a mixture of factions who until the previous year were at war with each other, difficult to control.  They were also charging across a wet ploughed field.  Unlike most cavalry attacks in the HYW, therefore, they didn't contact (almost - three men apparently died in the archer lines).  It seems the primary cause was pulling up rather than being shot down (in wargames parlance, failed the charge home test). 

The other thing to remember is that Agincourt is rare in that the cavalry didn't charge home.  At Crecy the French cavalry charged home numerous times.  Mauron, Poitiers and others before Agincourt, Verneuil and Patay afterwards are other examples.  So even with a missile weapon as effective as a longbow, cavalry had a good chance (the English didn't trail stakes around for nothing).

On the devastating volley front, there is an anecdote in Montluc'smemoirs where, as a young officer in a crossbow company, he has his men hold their fire until short range then mows down some Italian cavalry.  If I have a moment, I'll track it down.

[Imperial Dave]

Purely from a reenacting perspective, I used to be a (long) bowman in an archer unit. We went to many shows from the small (Tintagel) to the enormous (Tewkesbury) and size does matter with regards to archery!

In the small engagements, there would be a few archers acting as fire support but no massed effect. Shooting into combat at these shows was almost pick your target and fire. Unfortunately the opposition would peel off individuals to chase us down and away. I guess you could say we acted as skirmishers. We could fire up to about 20 yards away (then no closer - partly for safety reasons and unless we had specific authority never "flat" shot) but mostly for the fact we were always on the cusp of running away! Shooting reasonably closely was accurate (ish) but normally you only got to fire off one or two rounds before running away

In the big engagements, we would be brigaded together with other archer companies and organised into mass shooting/volleys. In these occaisions, mostly without hills to help, we were almost exclusively shooting over our own troop blocks into the enemy until they engaged in hand to hand. This meant that the ranges we shot at were normally no closer than 30-40 yards. Generally we could get off quite a few volleys from long range to the time it was "danger close" for our own troops.

In terms of effect (this is purely reenactment and not a real battle context), the small engagements had a higher percentage hit ratio per archer than the large scale engagements although the shooters normally got mauled quicker. Also because the large engagements normally gave several rounds of fire more than the smaller engagements, the actual number of casualties was often higher.

Shields are remarkably effective against 45 degree shot "dropping" arrows if you have them, as they tend to have a circumference larger than the target's body. Flat or near flat shooting when skirmishing, one on one, clear target with no intervening troops, is easier to get a hit as the shield only covers a proportion of the targets front body mass.

One thing I will say is that even in reenacting where no one is trying to actually kill you, as an archer it is very unnerving to see the enemy up front and close and when they charge at you with a dirty great pointed stick or horrible chopping weapon....you run 

[Erpingham]

Here we go.  Montluc is commanding a company of crossbowmen at the skirmish of St Jean de Luz.  he has set out to try to relieve some gendarmes who are having a problem with a stronger force of Spaniards.  Montluc has about 200 men.

Captain, said I, take you only care to save
your self and your Gens-d'armes, at the
same instant crying out. Shoot, Comrades,
at the head of these Horse. I was not
above a dozen paces distant from the Enemy
when I gave them this Volley, by which
(as it appear'd by the testimony of the
Prisoners who were taken a few days after)
above fifty Horses were kill'd and wounded,
and two Troopers slain, an execution that
a little cool'd their courage, and caus'd their
Troops to make a halt.

The Spanish are halted long enough for the Gendarmes to withdraw but the crossbow company is now exposed and Montluc has to extracate his men - a hairy adventure.

[Patrick Waterson]

Dave, your information about overhead shooting is interesting and useful to show what is possible, and also the usefulness of shields against 'dropping shot'.  Did you notice attackers being slowed by your shooting?  I have seen the occasional re-enactment with attackers slowing, bunching and holding up shields when shot at, and wonder if your 'opponents' noticeably slowed and bunched - or just put up the shields and carried on regardless.

Anthony, the point about Agincourt was not to hold it up as an exemplar of longbow/target interaction but rather to indicate that with massed archery the first, more distant volleys would be effective - as opposed to musketry or Montluc's crossbow volley where close means effective.  It is interesting that his 200 men downed, incapacitated or injured two men and 50 horses with the one volley - a rough rule-of-thumb casualty rate of 25% of the number shooting for this range and target.  Again, I suspect that some unfortunate targets took multiple hits rather than that 75% of the crossbowmen missed a side-of-a-barn-sized target at twelve paces.  As with musketry, hitting the horses is what stops the charge - at Agincourt, did the mounted French pull up because (as Mark suggests) they were fast disintegrating under the arrow storm or because their appetite for Englishmen had suddenly failed them and charging home no longer seemed appealing?

I think Nick may have a point about shieldedness not being a major determinant of whether or not a charge makes contact: I would however suggest that it can make a significant difference as to whether the charge arrives in a condition to do anything meaningful, and that this will further depend upon the degree of protection provided by the shield.

[Erpingham]

Anthony, the point about Agincourt was not to hold it up as an exemplar of longbow/target interaction but rather to indicate that with massed archery the first, more distant volleys would be effective - as opposed to musketry or Montluc's crossbow volley where close means effective.

Fair enough.  What I think I'm trying to unravel is that Agincourt isn't a great examplar.  The French cavalry, knowledgeable as they were, were reluctant to get involved in this charge - because it lacked glory , because the Armangnac's wouldn't follow an accursed Burgundian or because everyone could see that the ground would slow them down so much they'd get shot to bits - your choice.  I therefore think, having made a show, they pulled up early before archery could have done much at all.  If you take the opinion of Mike Loades in the latest Osprey, the English didn't waste many arrows in "dropping shots" - they waited till the range was short enough to use a flat trajectory.  I understand his reasoning and while agreeing that most shooting at an attacking enemy probably belonged in a "mad minute" under 100 yds range , it seems to be that a lot of exchanges between archers occured at longer ranges (see the famous anecdote from Towton, for example) and I think a lot of provoking archery against a defender wouldn't be close in either.

It is interesting that his 200 men downed, incapacitated or injured two men and 50 horses with the one volley - a rough rule-of-thumb casualty rate of 25% of the number shooting for this range and target. 

Yes, it's why I remembered it.  It fits with Nick's musketry charts broadly speaking.  Other interesting points are that they clearly aimed at the horses and that, if the Spanish had been charging seriously, they couldn't have been stopped at 12 paces IMO (even dying horses would have covered that distance at a gallop before falling.  Perhaps the Spanish intended to intimidate them rather than get tangled up with them, as they were really wanting to fight with the French Gendarmes?  The Spanish may have been in more than one rank, so the front rank probably took a higher casualty rate percentage wise - this may have had an effect in checking their advance.

[Imperial Dave]

Dave, your information about overhead shooting is interesting and useful to show what is possible, and also the usefulness of shields against 'dropping shot'.  Did you notice attackers being slowed by your shooting?  I have seen the occasional re-enactment with attackers slowing, bunching and holding up shields when shot at, and wonder if your 'opponents' noticeably slowed and bunched - or just put up the shields and carried on regardless.

In the big battles, the infantry blocks normally slowed and bunched just before charging to contact. Partly for protection against archery using collective shields and partly in anticipation of the shock of impact to the fore.

[Nick Harbud]

Many thanks to all for their varied and enthusiastic contributions.  I shall offer a few further thoughts for your consideration.

I contend there are a number of similarities between musketry and human powered missile weapons.  Firstly the probability of hitting the target broadly follows an inverse square relationship with distance and this would be true for both types of weapon.  The rationale behind this should be obvious and one can obtain a good curve fit use this formula to the observed results.

With respect to plunging or high angle fire, I suspect (but once I have had a chance to do a few sums, I shall share them with this forum) that inherently this is less accurate than flat trajectory fire.  That is, there is even less chance of hitting at long range than with direct fire alone.

Direct fire is, of course, necessary if one wishes to penetrate any protection on the target.  Partly this is because at any given range the high angle shot must have less force behind it than the flat trajectory.  Also, trials have shown that at more than 20-40 degrees from the perpendicular, an arrow will tend to glance off shields, plate armour, etc.

With regard to rates of fire, longbows and crossbows were not wildly better than the Napoleonic musket (5 rounds/minute).  In The Great Warbow Simon Stanley reckons the maximum rate of sustained longbow fire to be 6 shots/minute.  In Secrets of the English War Bow Mark Stretton carries out what is possibly a more realistic trial involving a moving target representing a charging horseman.  He concludes that 7s is required between shots for this short period.  In The Medieval Archer Jim Bradbury reckons it takes 12s to reload a windlass crossbow.  Incidentally, horsemen charge at around 10yds/sec and infantry at about half this speed.

With regard to Towton, this is a slightly different situation as the opposing sides did not immediately seek to contact one another, but stood off and engaged in a missile duel.  (Dare one say it, something not totally unfamiliar to the Duke of Marlborough.)  Instead of popping off a handful of volleys, they were able to empty their quivers (20-30 arrows per man).  Of course, all was confusion due to the weather, but one can speculate that both sides saw individuals on the other side hit from time to time and therefore concluded they were winning the duel without being able to judge the overall effect of their fire.

Hope this provokes further stimulating discussion.

[Erpingham]

We can add a few more data points to our longbow shooting.  Loades quotes Mark Stretton (again) as saying he could deliver 10 shots in a minute with a 140lb longbow but not 20 in two minutes.  It should also be remembered that the sixteenth century professionals reckoned six shots a minute.  I think it is solidly a modern myth that longbowmen would rain clouds of death on anything within 350 yds.  Almost certainly one of the things that sorted the men from the bows was dropping a few livery shafts on the enemy at range and seeing what he came back with.  At Towton, the Lancastrians emptied their quivers at long range - poor judgement, rubbish fire control.  For another example, look at the Genoese at Crecy.  Professionals, they tested the range a couple of times but the English held fire until they were well in range then let them have it.  Controlled shooting, giving heed to effect but also to sustaining the fight, was a hidden skill in the longbow story.  We just don't know how longbowmen were commanded in the field.  Indeed, when it got up close and personal were they shooting controlled volleys or shooting as individuals?  In wargames rules, it is below the level of abstraction for everything but one-to-one skirmishing I'd guess.    Incidentally, in Villani's opinion, longbowmen could shoot three arrows for every one a crossbowman got off.