The Carriage of Weapons Crispvs (Paul Geddes)
This article is intended to address a number of points, some of which many readers will have come across before, so it should not contain any great surprises but I will also attempt to call attention to some things which may not be so well known.

    For any soldier or person depicting a soldier, the method by which his weapons are carried is an important consideration. This would be no less true for the Romans as it is for soldiers today. It follows then that if we see (for example) one of the Rhineland sculptures showing a weapon being carried in a particular way and we take this depiction to be accurate, we are looking at a method of carriage that the Roman army employed, and moreover, a method which suited the way they did things. From my point of view, if they did something a particular way, we should try our best to do it the same way. We may not always understand the true purpose of what we are emulating but in some cases we will and in others, doing so may help advance our knowledge and understanding.


    Most readers will be aware that swords can be worn either on a baldric or directly attached to a belt. What is less well known, and in some cases less well understood, is exactly how sword scabbards were attached to their belts and baldrics and how they hung when attached. Note: for ease of writing I will use the term 'belt' only to describe a belt worn around the waist throughout this article. 'Baldric' indicates a belt or strap passed diagonally across the body.

    Also, when I speak of sculpture I will be referring to the provincial sculptures from the Rhineland and Adamklissi, which are generally held to be accurate representations.

   Looking first at the baldric method most of us use, this method is less well represented in first century AD sculpture, but there are still a number of examples. These show two things in common which we should be aware of.

   Firstly, in all cases the baldric passes OVER rather than under the soldier's belt (fig.a and fig.b), meaning that it was not felt necessary to hold the scabbard in place with the belt as most of us currently do. This accords with the evidence of one of the Mainz column bases (fig.c) and a metope from Adamklissi showing three auxiliaries (fig.d), both of which show swords suspended on baldrics but do not show belts.

    Secondly, these sculptures also show that the baldrics were reasonably short, as they consistently show the pommels of the swords at a point above the elbow and just below the armpit, meaning that the mouths of the scabbards must be at or somewhat above the level of the military belt. The wearing of swords in a high position is also paralleled on cavalry stelae, showing that this position is also achievable when the sword is worn on a belt. If it was felt to be useful to carry the sword so high, this may in fact be one of the reasons why the upper pairs of suspension rings on Roman scabbards were some way below the scabbard mouths. As some readers may already be thinking, wearing a scabbard on a baldric short enough to allow it to sit this high means that it might not be practical when wearing armour unless the baldric could be undone somehow and then refastened. I will return to this shortly.

Fig. a

Fig. b

Fig. c

Fig. d

So how do we approach emulating these two things? In the first example, that of passing the baldric over, rather than under the belt, since the second half of the season 2006 I have been experimenting with this myself, as some readers will know. I had noticed the unbelted swords some time before but had avoided doing it myself on the basis of being worried that my sword would jump out of its scabbard as I moved. In practice, when I bit the bullet (so to speak) and did it, I found no significant problems. The only problem I encountered was that following a charge, I would find that my scabbard had worked its way around to my back. Once I had realised this I simply had to remember to pull the scabbard back into position before replacing my sword. Thus it ceased to be a problem after the first time. Since then I have been wearing my sword in this way and have become very comfortable with it.

    It should also be noted here that a sword in a vertical position on a short baldric which has not been passed under a belt will sometime hang at a slight angle. This is normal and should not be thought of as unusual or wrong.

   To the second point, I shortened my own baldric some time ago to about the length indicated on some of the sculptures and found that I did not find it any harder to draw the sword. A shorter baldric also lessens the extent to which the sword will swing about and thus reduces the danger of the sword being lost from its scabbard.

   This brings us to the problem of putting a very short baldric on over armour. Many people, including a number of those reading this will have overcome the problem by attaching a fastener which allows them to fasten and unfasten the baldric. This would be all well and good were it not for two things, namely that the fasteners consistently used by re-enactors are actually fasteners for the straps on horse harnesses, and that the sculptures do not show fasteners in the position they are always placed in. It follows then, that this method of closure is incorrect and should be discontinued. With most re-enactors' swords still being on long baldrics discontinuing the use of fasteners would not present a problem, as the two ends could be sewn together to form a single baldric strap, but if baldrics were to be shortened to the length indicated by sculpture it could present a problem for many re-enactors.

    At this point three useful items of evidence can be cited, which many people may have been unaware of.

    Firstly: the scabbard of a sword found on the island of Delos. This scabbard had only two suspension rings, rather than the normal four, and both were on the same side. Found next to each ring was a small buckle

    Secondly: a sword and scabbard found with the remains of a military belt at the site of the Roman fortress at Vindonissa. Found with the sword and belt plates were the remains of a small buckle similar to the type used on Corbridge type lorica segmentata. This buckle was not big enough for it to have been the buckle of the belt the plates were from.

    Thirdly: the 'dagger' found with the Herculaneum soldier. This dagger (not to be confused with the sword he was also found with) is unlike any other known Roman dagger and in most respects appears to be a very short Mainz type gladius, and it is possible that it may have been a gladius which had broken and which was sharpened to a new point and reused as a pugio. This much is speculation, but significantly here, a small buckle resembling the buckles from Delos was found adhering to its scabbard.

    Taken individually, none of these examples is conclusive but taken together, the three can lend themselves as reasonably compelling evidence for the idea of scabbards having been buckled to their belts or baldrics. This has implications which are relevant for the short baldric. If the baldric was fastened to the scabbard with a buckle which was attached next to the suspension ring of the scabbard, this would allow the baldric to be done up over armour, and it would also make it difficult to see in a sculpture, even if it was depicted in the first place.

     A number of sculptures show swords worn on belts. Often these hang more or less vertically, as one might expect, but on a few of the Rhineland stelae the sword appears to be hanging horizontally or at a shallow angle (fig.e). Why swords might be carried this way is something I have no explanation for but nevertheless, this is how they are sometimes shown. If the scabbard hangs vertically we should expect it to hang high enough that the sword pommel will be at more or less the same height as it would be if it was suspended from a short baldric, as described above.

    To attach the sword scabbard to a belt in a vertical position, the obvious method would be to pass a leather strap or thong through each ring and over the belt to form an X pattern. This could be secured by a knot or a small buckle such as the example from Vindonissa. In order to make a scabbard hang at a shallow angle, as indicated by several of the Rhineland stones, the example of the two buckles from Delos may be instructive.

   At this point, before moving on, it seems pertinent to mention the ease with which a sword can be replaced in its scabbard. Whilst the relevance of how easily a sword can be drawn from its scabbard is obvious, there exists a belief amongst many in the re-enactment community that it is equally relevant how easily a sword can be replaced in its scabbard. Why should this be the case though? In a combat situation the soldier would need to be able to draw his sword rapidly. However, he would hardly be likely to sheath it again whilst engaged in combat. It follows then that in a combat situation the sword would only be re-sheathed once the soldier had been withdrawn from the actual front line of battle, by which time he would have slightly more opportunity to make any necessary adjustments before replacing the sword in its scabbard. Following on from this, as we are told that the Romans thought of training as a bloodless battle and a battle as a bloody training session, we are forced to wonder whether they saw any need to effect the smart, simultaneous re-sheathing of swords that many re-enactment societies pride themselves on even after training, rather than a more leisurely (although presumably not tardy) and careful replacement of swords in sheaths

fig. e

Moving to daggers, there is only one problem with the way most re-enactors wear their daggers. All daggers currently carried by members of the society, as well as those of most other Roman re-enactment societies, are hung from the belt by two straps which pass through the upper suspension rings and which are then hooked over the frogs on the belt. However, a closer look at the sculptural record reveals that these straps do not appear in any known depiction of a Roman soldier. Instead, the Rhineland stelae consistently show the dagger to be suspended from the belt in such a way that the upper suspension rings (the only rings actually used) appear next to and in some cases overlapping the buttons on the frogs, indicating that it is most likely that sheaths were attached to the frogs by tying the upper suspension rings tightly to the frogs with thronging (fig.f and fig.g). This position would limit unwanted movement and is almost certainly more stable than a situation where the pugio hangs from straps some way below the frogs. Both frogs are consistently shown as being virtually in contact with the upper suspension rings on both sides.

fig. f

fig. g.

The method of carrying the pilum, as most re-enactors do, with it gripped in the right hand with the thumb pointing down the shaft and resting in the crook of the arm is attested in the sculptural record but many people appear to be unaware that it may be just as correct to carry it with the thumb uppermost and the arm sharply bent, a method of carriage with is also attested in contemporary sculpture (fig.h). Presumably both methods were used and it may be that it was found useful to change the position of the hand and arm from time to time during route marches and that therefore two or more methods of gripping the pilum were employed on the march

Now is the time to make the minor modifications to your baldric!

   When I shortened my baldric I unpicked an existing sewn join (but I would have cut it if it had been made from a continuous strap) and then donned my armour and taped the baldric together again at various lengths until I had a length which seemed about right. Once I had arrived at the desired length I simply sewed the baldric together again. If you currently have a horse harness fastener on your baldric, it really is now time to remove it, flashy though it is, and give it its own place on your mantelpiece. The baldric should be sewn up as described above. If you then undo the single strap which passes through the forward suspension ring of your sword scabbard you can draw this tighter to shorten the baldric somewhat.

    Similarly, if you have time, check the position of your dagger frogs. If they are positioned closely enough to each other that the sheath can sit comfortably between them, with its upper suspension rings virtually touching them then they are fine already, but if they are positioned wider apart so that the sheath's suspension rings cannot touch them on both sides then you should consider moving the frogs so that they both 'hug' the sheath. It should also be noted that the straps currently commonly used to suspend daggers should be discontinued and replaced with thronging tied tightly between the frogs and the upper suspension rings. It might also be worth trying out the alternative grip for the pilum as well.

Please note that all images except figure 'd' are reproduced here courtesy of the imagebase. Figure 'd' is copyrite to Dr JCN Coulston.

Addendum - Baldric fastening

In line with my thoughts on baldric buckles, I would suggest that this could be achieved by fixing a small buckle to the baldric strap two or three inches from the end and bringing it close to the scabbard. The end of the baldric would be passed through the suspension ring and back to be secured onto the buckle. This would solve the problem of putting a short baldric on over armour but I would point out that there is as yet no conclusive evidence that this was the method employed. It is merely a potential solution to the problem, supported by three potential pieces of evidence and the absence of evidence for the normally assumed alternative.