Did the Romans have the composite bow?

[Eleatic Guest]

The penetrative power of the composite bow can be overrated. The latest edition of 'Antike Welt' has an experiment report of shooting a composite bow at a stationary armour sample at 5 metres carried out by LEG XXI Primigenia.

The complete test report can be found here:
http://www.milites-bedenses.de/
Forum
-- Schutzwirkung roemischer Panzerung gegen Pfeilbeschuss

In the mean-time, they have made another test, this time on the penetrative power of bulbous-nosed arrowheads. More anecdotal than the composite bow test, but well illustrated with pics:

http://www.milites-bedenses.de/
Forum
-- Die sogenannte Zwiebelnasen-Pfeilspitze

[Endre Fodstad]

I really have to point out that "composite bows" strictly put means any bow made from several materials. It also includes "straight" bows such as those ash-yew composites found in the Nydam bog. It is quite possible to make a recurved/reflex bow as a self-bow (i.e. made from only one material). Some people have touched on this, but many bow discussions tend to drift toward the idea that self=straight and composite=recurved/reflexed

I also think people tend toward all too modern thinking when comparing bow types, and a tendency toward thinking of bows as some sort of muscle-powered gun. It is quite possible to make a very strong or a very weak straight self bow or a very strong or very weak recurved composite bow - the bow is made for the archer's strength.

A very powerful bow is not always and advantage. It requires a powerful man to draw a powerful bow not matter how it is made, and if the bow is under a lot of stress when strung (which typically also means it is time-consuming or hard to string, not always a military advantage) it also tends to require a lot of repair or even replacement frequently (not a military advantage either, especially if the bow is of a complex construction that needs specialists knowledge to repair) - this is one of the reasons so many later horse archers (and likely also earlier horse archers although the sources are more scanty) carried multiple sets of bows. The problem is enhanced in wetter climates and composite bows made from many components - even the best fishglue weakens over time when exposed to moisture, as most of us conservators know only too well.

We know from 13th century Mameluke training manuals that the preferred horse archer tactic was to unleach arrows with 75 meters or so as the maximum range, and to prefer quantity over power (at least during training) - in an age when armour often was far more covering than it was during the roman period. Was it really necessary to use very powerful bows? As Goldsworthy notes in The Roman army at war (p67, paperback 1998 version) the parthians spent ages whittling down the roman foot at Carrhae; it doesn't really seem like they were carrying the heaviest warbow available, even if the romans were permanently "turtling". The romans sthemselves hardly faced all that many heavily armoured and covered opponents compared to themselves (and when they did, they often handled them in close combat); it strikes me as uneccessary to bring a 180lbs monster composite-reflex horn-wood-sinew bow, tough to string and a highly stressed structure, if you are fighting lightly armoured parthian horse archers that you probably outrange even with a weaker bow or unarmoured german tribesmen along the Rhine.

I don't think any real study of the geographical preferences of archery equipment employed by the imperial army (is it even possible, given the source situation?) but I guess the answer to the questions posed by the original poster would differ greatly from region to region, even during the same time period.

I realise some of these points have been previously touched upon here.

When it comes to arrow vs armour testing, my group has some experience with that. We started out with covering a wooden block with textile armour and maille and eventually went on to securing a pig's haunch between two ropes and dressing it up in a specially made aketon, maille and coat-of-plates ("undressing" it as we went; we ended up doing this as a public show in the end, also using swords, axes and spears against it - fun, but the aketon has to be remade frequently). We tend to get similar results under similar conditions, but we also see how many variables matter in this sort of test - the range is very important, as is the "give" in the target, the power of the bow, the arrowhead used, the mass of the arrow, the quality of the armour and arrowhead, the angle of the shot, et cetera et cetera. After several attempts at doing the armour "properly" (getting 15th century steel from an english supplier, among other things) we more or less decided the best thing one could do for accuracy was to attempt to balance the testing - firing an arrow at 5 paces provides a very powerful shot, but what archer would want to be (as noted by others here) 5 paces from someone who wants to kill him and is carrying the means to do so? Sadly, this meant that with our resources and time, a more realistic 30+ meters test was beyond us. At best, such tests give a general idea of the effectiveness of archery against armour. At the worst, they can give results that probably hurt our understanding of how these things works.

The one thing that never fails to provide solid and deep penetration is when the armour is entirely removed and we're shooting at a haunch protected only by its "underwear" (a linen "shirt") - no matter the bow, arrowhead or range; I would definitely want to be wearing some sort of protection, were I a roman or medieval soldier and getting shot at by an archer.