Feanor

I think it's worth mentioning here that while the individuality of combat styles such as pankration does mean that there would've been little benefit in training legionaries in them, as their strength was their ability to fight as a unit, the same can be said of the Greek Phalanx. Much has been discussed of the apparent contradiction between the Greek love of single combat and their wartime fighting style which relied on teamwork to keep the phalanx formation together. Since both the Greek and Roman fighting formations would have been undermined by soldiers indulging in one-on-one combat, which would presumably mean that such feats were discouraged, both cultures had the same love of individual combat and competition and so experienced the same contradictions. So while I agree that any individualistic "martial art" or ancient fighting style that you might be able to trace back to ancient times doesn't really tell us much about how the Roman soldier was trained, I think the same thing goes for the Greek soldier.

I just spent a great day at a Roman museum in Aalen, Baden-Wuettemberg in Germany, which was the front line of the Roman Empire for several centuries. They have some Roman legionary equipment that you can put on and mess around with for fun. I was amazed to find that the handle on the rectangular "scutum" shield is just a horizonal one-handed grip in the middle of the shield. While this makes sense for carrying the shield around easily, it makes no sense whatsoever to me for combat. The postition of the handle is awkward - you have to twist your wrist unnaturally to hold it, which not only makes it hard to hold but also it means that the impact of any charge would be absorbed by the wrist. The wrist is a weak joint at the best of times but when twisted it's not at all strong. Here is a photo to see what I mean: I'd have thought that the shield would be held along the forearm, with a loop on the left-hand third of the shield for the elbow end of the forearm and a handle on the right hand third for the hand to grip naturally. It would allow the impact of receiving a charge to be much more easily absorbed by the shoulder, and it would also make it much easier for the soldier to push his shield back into the enemy. I believe modern riot shields (at least, the rectangular perspex Roman-style ones) are of a similar design. I spoke to the museum curators about this and they agreed with me that it didn't seem practical but said that all the evidence they know about points to the shield handles being designed in the same way as the replicas. Sure enough, a search online confirms that all replica Scutums, and all diagrams of Scutums, indicate that the handle was designed in this apparently awkward manner. Do you guys have any idea what historians and replica-makers are basing their design on? Are we sure that the central handle isn't just a carrying handle, and that there weren't other combat handles located elsewhere on the shield?