The limitations with experimental archaeology

Last week I wrote a post called The Importance of Written Sources, and in that article, I mentioned using experimental archaeology in the context of HEMA to help recreate fighting systems for which we do not have any written sources. Shortly after I wrote this, an article was published on ScienceNordic[1]. This article includes a video from a group called Combat Archaeology, as well as a summary and quotes about the experiment, and what they found.

The article is of course somewhat sensationally titled. It claims that an ‘Archaeologist discovers a new style of Viking combat’, when the scope was really much more limited than this. The description of the video states:

The experiment attempted to determine what body techniques Viking Age round shields are inclined to facilitate and which they restrict or otherwise discourage. More specifically, the aim was to critically assess body techniques in terms of deflection and to obtain empirical data outlining the effects associated with an aggressive as well as relatively passive use of the shield.

In essence, the experiment is designed to compare the effectiveness of passive vs. active uses of the shield, although this is of course not the same as discovering an entire style, as I’ll mention below.

The experiment comprised of eight trials, where the attacker swung a sword at the defender, and the defender responded with either a passive or active defence. According to the video:

The terms “aggressive” and “passive” are used here to describe the extent to which the shield is actively thrusted forward to meet the attackers blow.

This definition is not particularly precise, and we can see that lack of precision in the trials.

The first passive trial is trial two. Rather than holding the shield still though, it is raised up to meet the incoming attack, and it is raised at a slight angle, so the attacker’s sword slides down the rim of the shield and comes into very light contact with the defender’s helmet.

Given the shield is pushed forward to meet the incoming attack; I would be hesitant to call this a passive defence. The contact made with the defender was extremely light and unlikely to do anything. In an armoured context, might a Viking with mail and a helmet ignore contact like this entirely?

Even if we do accept that this trial was a failure, the question remains of why did the defence fail? Did it fail because it was a passive defence, or because the shield was held at slightly the wrong angle, or because the shield wasn’t held far enough forward, or maybe it simply failed because the defender did not have the correct set of body mechanics to able to make a block like that work?

They note that blow on the shield was improperly placed on the shield, because the attacker’s glove slipped on the shield boss, so they re-did the trial.

In trial three (which was a re-do of trial two), the defender now does something different. They again move the shield up forwards into the path of the sword, and the moment contact happens, they twist their shield to push the sword out of the way. Rather than remove the glove, they instead just changed what the defender was doing, which means that this trial does not actually address the problem with trial two.

The next passive trial is trial five, except again the shield is pushed out to meet the incoming blow at an angle to help deflect the sword, and the defender steps in during this block as well. Aside from the shorter step, this is almost exactly the same movement as that shown in trial four, which had been termed as aggressive shield-use.

The same problem can seen in trial eight, which is also classified as passive shield use, but again is almost identical to the preceding trial, which had been classified as aggressive shield use.

The definitions of aggressive and passive therefore need to be far clearer. Had I been shown this video without any of the explanatory text, I would not have called any of those defences passive.

It also worth noting that only a single trial resulted in failure, which the researchers themselves admitted may have been due to the defender’s glove, an issue that was then never addressed. Several instances of so-called passive defences were successful. Despite this, the article states:

During the experiment, Warming took a hit on the head. Fortunately he was wearing a helmet and the event only served to strengthen his theory.

“It happened when I was was using the shield in a passive way. This illustrates the futility of passive shield use and suggests that they didn’t use the shield in that way,” says Warming.’

The issue of the attacker’s glove is larger than simply identifying a flaw with the second trial. In many of the trials, the attacker’s hand punches into or towards the shield. Is the attacker’s mechanics likely to be accurate, or are they attacking in a way that would be unsafe for their hand if they were not wearing a highly padded glove?

Warming also notes that the shield took more damage when he used it passively rather than aggressively.

“The deep cut here on the edge is from when I used the shield as passive protection. The shield clearly worked better when I used it more actively,” he says.

This observation led to one of the main conclusions of Warming’s thesis: the Vikings may have used their shields to actively fend off blows. Otherwise, they would have quickly broken, he says.

If the difference between a passive and aggressive defence is the length of the accompanying step, as implied by the differences between trials 4&5 and 7&8, then this comment is somewhat nonsensical. If what he actually means is that parrying with the edge of the shield results in more damage to the shield than parrying with the face or boss, this is conflating two separate issues together. It is possible to parry both “passively” and “aggressively” with either the edge or face of a shield.

I also am dubious about the argument that they must not have used the shield passively because their shields would have broken quickly. I would imagine that it would take quite a lot of blows to reduce the shield to a state where it can no longer be used effectively, and this may not have necessarily have been a major consideration.

If we look at disciplines for which we have written sources, we can see the problem with this sort of argument. John Clements wrote an article called The Myth of Edge-On-Edge Parrying in Medieval Swordplay[2], and in that article he argued that one of the reasons you should not parry with the edge of a sword is that swords are valuable and that they would be damaged if used in edge on edge parries.

At first, this sounds like a reasonable argument, except that is a little simplistic and that we have medieval and Renaissance sources that explicitly tell us to parry with the edge of the blade.[3] [4]

If we were relying on an experimental archaeological approach and didn’t have access to written sources, then it would be easy to conclude that edge on edge parries shouldn’t be used, but being able to view the written sources of the time would tell us this was a mistaken conclusion. Similarly, using experimental archaeology might make us believe that parrying with the edge of the shield is incorrect, but this could not be guaranteed to be a correct assumption without being able to view a written treatise on Viking era sword and shield combat.

Even if we conclude that this experiment did in fact give us useful results, then we would still be very far off being able to reconstruct any sort of system from this. If the shield is to be used actively, what does this mean, and how do these aggressive shield defences fit in to an actual system? Should we wait for an opponent to attack, and then use an active defence against them with the shield? Should we use our shield in order to create openings to attack? What situations determine whether we should parry with the sword or with the shield? These are all important questions to answer, and I can’t see how experimental archaeology could provide those answers to be honest.

If enough researchers produced high quality, rigorous studies, then perhaps the resulting body of work could be used to make some tentative conclusions. Until then, my view remains the same, that experimental archaeology is going to produce results of limited usefulness, and that many attempts I have seen at experimental archaeology are not of sufficient quality, and cannot give us the depth of information we can gain from written sources and that we need to be able to reconstruct an actual system of fighting, rather than a handful of potential techniques.




[1] Johanne Uhrenholt Kusnitzoff. “Archaeologist discovers a new style of Viking combat.” ScienceNordic, 30th October 2016, accessed 3rd November 2016.

[2] John Clements. “The Myth of Edge-On-Edge Parrying in Medieval Swordplay.” ARMA, 2002, accessed 3rd November 2016.

[3] Andrea Morini. “About the Flat Parry.” Hroarr, 24th May 2012, accessed 3rd November 2016.

[4] Gregory Mele. “Much Ado About Nothing: Or, About the Cutting Edge of Flat Parries.” In: Stephen Hand (ed.) Spada. Union City:  The Chivalry Bookshelf, 2002. Pages 32-47.

Website Pin Facebook Twitter Myspace Friendfeed Technorati Digg Google StumbleUpon Premium Responsive