Lorica hamata / segmentata

[Caratacus]

His estimations are very inaccurate I'm sorry to say.

In what way are they innaccurate, Erik? David Sim is normally very careful about recording the times it takes for him to perform every manufacturing process - he is, after all, a very experienced blacksmith. His explanation of the method for producing butted mail seems entirely logical to me - but then I have no experience at all at working with iron!

I agree, of course, that the size of the links makes a great difference to the length of time to produce a suit of armour. The only thing that I can see is a possible error is his summation for the time it would have taken to equip an entire 5,000-man legion with a hamata suit, where he assumes that there would be a suit for every man and a spare suit in the stores - and he comes up with an estimated time of some 29,000,000 man-hours for the entire job! Five thousand spare suits of lorica hamata would have taken some storing!

Are you perhaps saying that he is over-estimating the number of links of, say, 7mm OD, in the suit? That would make a certain difference to the completion time, I would think.

Caratacus/Mike Thomas

[markusaurelius]

Tarbicus,
I wonder if Seg's had the bad nack of not being as "durable" over the long run. The leather attachments (as well as the brass ones) quickly collected moisture which would have start the metal rusting even if well taken care of.

I don't think so...leather can be re-oiled and bronze can be tinned easily enough.

I was referring to the fact that water& vapor etc. gets trapped under the brass fitting where it can't be oiled, polished etc. Leather as well attracts moisture because of its porosity, even when oiled there will IMO always be some water trapped in between the leather and the iron. This will cause corrosion that can't easily be cleaned or maintained. Furthermore " The electrolytic reaction between the copper alloy and the iron of the plates would have led to these being the first place that corrosion would start" Bishop & Coulston, Roman Military Equipment 2006.

Mail doesn't have these durability issues. Don't get me wrong, I have a Newstead replica displayed at home, and not a mail suit. Segs are just more "fun"

[markusaurelius]

His estimations are very inaccurate I'm sorry to say.

In what way are they innaccurate, Erik? David Sim is normally very careful about recording the times it takes for him to perform every manufacturing process - he is, after all, a very experienced blacksmith. His explanation of the method for producing butted mail seems entirely logical to me - but then I have no experience at all at working with iron!

Do you think that perhaps he was also factoring in actually making the sheet iron as opposed to the thin wires required to make the mail rings? I would think that flattening a chunk of iron, either by hammer or mechanical would be quicker than making enough wire to complete a mail armor suit.

[Erik D. Schmid]

In what way are they innaccurate, Erik? David Sim is normally very careful about recording the times it takes for him to perform every manufacturing process - he is, after all, a very experienced blacksmith.

In your post you stated that he said it would take one man between 12 and 15 months to make one suit from butted links. I simply said it wouldn't based on my own calculations. Blacksmithing is a far cry from mail-making. Mail-making is more closely related to jewelry making than anything else. It requires years of constant practice in order to get a decent rhythym down with which to draw conclusions relating to construction time from.

I'm not trying to downplay his work, but his calculations just do not jive with what I have found in my own experiments.

The only thing that I can see is a possible error is his summation for the time it would have taken to equip an entire 5,000-man legion with a hamata suit, where he assumes that there would be a suit for every man and a spare suit in the stores - and he comes up with an estimated time of some 29,000,000 man-hours for the entire job! Five thousand spare suits of lorica hamata would have taken some storing!

That time is not far off from what it would have taken based on my calculations. I figure it would have taken around 20,000,000 man hours. This is based on a shirt having 20,000 links and a person being able to weave around 1,000 per day. So, in a little over half a century the legion would be ready to go. Now, add another person and the time gets cut in half and so on. It doesn't take many people to really pump out product.

100 people should be able to put out around 100,000 links per day. This is approximately five shirts per day. Increase the link size and you increase the output of finished garments. Bear in mind that this is just an estimate. Actual production may easily have been double this figure.

He does tend to overestimate links I have found though. It was something I brought up to him several years ago. He calculated a ratio of solid to riveted links of 4/1. This is incorrect. An alternating row garment will almost always have more riveted than solid links. Generally it is a 1/1 ratio. He was not very enthusiastic to hear this.

Do you think that perhaps he was also factoring in actually making the sheet iron as opposed to the thin wires required to make the mail rings? I would think that flattening a chunk of iron, either by hammer or mechanical would be quicker than making enough wire to complete a mail armor suit.

Possibly. But, the wire drawing would not really add that much more time to be honest. I have a friend who draws me wrought iron wire by hand periodically and it doesn't take him long at all to create a decent amount. But, this is all idle speculation. This is an area that needs more research.

[Caratacus]

He does tend to overestimate links I have found though. It was something I brought up to him several years ago. He calculated a ratio of solid to riveted links of 4/1. This is incorrect. An alternating row garment will almost always have more riveted than solid links. Generally it is a 1/1 ratio. He was not very enthusiastic to hear this.

Sorry if you took my remarks to be some sort of a criticism of your original comment - this wasn't my intention, I was simply asking for some information there!

The quoted time of 12 to 15 months came from Sim's paper in the Britannia journal - nothing to do with me! Yes, I can see that blacksmithing is rather a crude term for what we are talking about here. My remark was rather more about the methodology David Sim uses to establish a possible/likely time period for any particular task.

My reason for putting the quote above is that, if the links were to be riveted, would this not further increase the time to complete a suit, as it would involve several more operations that just butt-joining the connecting links? I am sure (it's been a while since I looked at the paper) that Sim was trying to produce an estimate of the shortest time that would be required to produce a ring-mail suit.

I would have thought that it was obvious that in an alternate row system, the ratio of solid to riveted/butted rings was bound to be 1:1? However, if you examine the actual links, then an open ring will be joined to two solid ones in the row above and two in the one below, hence the 4:1 ratio. Maybe we are talking about the same thing from two different viewpoints here?

I can quite see where there would be a decrease in time required if there were to be a whole group of armourers working flat out on the project! Sounds to me like the perfect job for a bunch of soldiers who were on "jankers". :lol:

Caratacus/Mike Thomas

[D B Campbell]

However, if you examine the actual links, then an open ring will be joined to two solid ones in the row above and two in the one below, hence the 4:1 ratio.

I may be out-of-line here -- I'm neither an armourer nor a re-enactor -- but your "1 open ring/4 solid rings" doesn't seem (to me) to be the "module" of the shirt, Mike.

It seems to me (feel free to correct me!) that it's not the 5-ring pattern that's replicated, but a 3-ring pattern, because the 2 in the row below already belong to another open ring lower down! In fact, it may even be a 2-ring pattern, if the 2 on the right already belong to another open ring to the side -- do they?

You're right to point out the extra fiddly work in linking the five rings, but most of them will already be linked to another open ring. So (from a logical viewpoint) Erik's 1-to-1 ratio seems more likely. (Don't all shout at me at once!!)

[Sardaukar]

Just wondering about some aspects:

1. Blunt trauma.

Chain mail wearer is going to get a lot of blunt trauma compared to plate or "lorica segmentata" from hits. And padded undergarment might make it as hot as plate...and still not as resistant.

2. Comfort wearing the armour.

Chain mail main weight is going to rest on your shoulders, even when using tight belts to transfer some of it on hips. Is "segmentata" different ?


To me, "lorica segmentata" would offer superior protection to incapacitating hits (blunt and penetrating to torso), even when chain mail offers more protection area. Neither really protect the areas vulnerable fighting behind scutum.

Militarily, scutum and helmet would be most important defences. Type of body armour would be only 3rd in importance, meaning attack has avoided shield and helmet.

[Tarbicus]

1. Blunt trauma.

Chain mail wearer is going to get a lot of blunt trauma compared to plate or "lorica segmentata" from hits. And padded undergarment might make it as hot as plate...and still not as resistant.

Chainmail makes the best protection on a motorbike apparently, and I did read there's a school of thought that, while mail is excellent against blunt trauma, it's only a secondary defence to the padded garment beneath it.

[Gaius Julius Caesar]

I can just see my mates skidding down the road or track after laying down the bike, a shower of sparks flying off their new 'Chain-mail' leathers.... :lol:

[Erik D. Schmid]

Sorry if you took my remarks to be some sort of a criticism of your original comment - this wasn't my intention, I was simply asking for some information there!

hehe... No worries.

The quoted time of 12 to 15 months came from Sim's paper in the Britannia journal - nothing to do with me! Yes, I can see that blacksmithing is rather a crude term for what we are talking about here. My remark was rather more about the methodology David Sim uses to establish a possible/likely time period for any particular task.

Oh I understand that. His methodology is flawed however as I will attempt to explain. I am looking at his paper while writing this response in fact.

My reason for putting the quote above is that, if the links were to be riveted, would this not further increase the time to complete a suit, as it would involve several more operations that just butt-joining the connecting links? I am sure (it's been a while since I looked at the paper) that Sim was trying to produce an estimate of the shortest time that would be required to produce a ring-mail suit.

Yes, riveting does increase the time required to produce a garment, but not as long as he states.

I would have thought that it was obvious that in an alternate row system, the ratio of solid to riveted/butted rings was bound to be 1:1? However, if you examine the actual links, then an open ring will be joined to two solid ones in the row above and two in the one below, hence the 4:1 ratio. Maybe we are talking about the same thing from two different viewpoints here?

No matter how you look at it, you are only going to have one riveted link for every solid link. Period. Think of a simple tube of mail made of alternating rows of riveted and solid links ten rows wide. You will have five solid and five riveted, which equals the same number of links per row. Sim assumed that there were four solid links to every riveted link. I quote:

If we consider a mail coat made of a series of solid rings, with four riveted rings passing through it where the rings are 6 mm in diameter, then the whole coat, will require 170,353 rings (42,588 riveted rings
and 127,764 solid rings).If each solid ring takes I minute then
I (min) x 127,764 (rings) = 2,129 hours
If it takes 3 minutes 47 seconds to make a riveted ring (this includes making the copper rivet) then, 42,588 (riveted rings) x 3 minutes 47 seconds (per ring) = 2,684 hours 41 minutes
Total time 2,129 hours + 2,684 hours = 4,813 hours (1.3 years, given a working day of approximately 10 hours).

My own calculations for a shirt having a 50" chest and a 30" length with no sleeves and no doubling would require around 60K links with his dimensions. Half of them being solid and the other half riveted.

What I consider his biggest mistake is using the idea that one person would be doing every step. This seems to be a very modern viewpoint. The lone armourer slaving away making his wares is a popular idea that has been romanticised over the years. However, it has hardly any basis in reality when it comes to mail making. If it were true, then there would have hardly been any mail made as it would have taken far too long to produce it.

The production of solid links would not have been as arduous of a procedure as he makes out. One minute of agregate labour per solid link is far too long. His tools are interesting, but the Romans may have had something far more sophisticated. You never know. Judging by the amount of mail made during this period I would hazard that it was very likely they were doing something to produce these links that we have not thought of yet. I would also get rid of the touching up on a tapered mandrel because there simply is no need for this. He only states that it seemed a possibility. His production time of riveted links is way off as well.

The point I am trying to make is that mail was a mass produced item. There was a very large division of labour. Knowing this then pretty much negates any of his conclusions relating to production times. My previous estimates on how much mail could be produced were rather low. In fact they were very low. In all honesty, one hundred workers should be able to produce approximately 400K links per day. That means roughly six shirts per day based on the idea of 4mm ID links. Up the link size even a small amount and the number of finished shirts increases.

To give an example, when making mail was the only thing I had to do during the day I was able to produce upwards of 1,000 riveted links per day. This included lapping, flattening, piercing, weaving and riveting. If all I had to do was weave and rivet, the amount would have been closer to around 5K to 6K links per day using the alternating row system meaning I could produce around 2.5K to 3K woven riveted links per day.

So, using this data one person working alone could produce a mail shirt in about a month and a half. The extra time figured in for wire drawing and solid link manufacture. This is a very far cry from the 1.3 years quoted above.

This is not meant to be derogatory towards David or his research, but more along the lines of a simple reminder that a good bit of practical experience can give a person an entirely new outlook on a subject.

[Gaius Julius Caesar]

wouldn't you have people doing every step?

People punching out solid rings,
people cutting the wires to length
people flattening ends, people making the little holes for the rivets
and people assembling the shirts......

probably my following of the actual methods of construction is wrong, but is that what you are sayin Erik.....a sort of assembly line of sorts.

[Caratacus]

Ah yes - I can see where you are coming from on this one now! I can also see the source of the error that DS made, i.e. the assumption that the 4:1 ratio of wholeplit rings was continued throughout the whole suit. Of course, each split ring is joined to four whole rings, but only if you consider each split ring in isolation from the row in which it finds itself.

However (as you say), the time period quoted by DS was for one man working. I'm not sure that he intended this to be taken literally. It would surely be the case that there would be quite a few working on this, each one performing a different task (and possibly changing around so as not to get too bored with the whole thing and start making mistakes?)

Your estimate of around 5 to 6 weeks for a complete suit (made by one 'man') is an interesting figure. I've seen another estimate of 3 to 4 weeks for a complete suit of segmentata armour (presumably this assumes that the required plates of metal are already available, together with the sheet brass for the fitments). If so, this means that the times to produce either type of armour were roughly comparable - which removes this as a possible reason for prefering one type of armour over the other! The true differnce might come down to the level of skill required for the two production techniques. Hmmm?

Caratacus/Mike Thomas

[Erik D. Schmid]

Exactly Byron! This was the method used for centuries. Documents show that in Milan circa 1400 AD, there were 100 different mail-makers. This is not including the link-makers that were a distinct group. Now figure each shop has about four to five workers and it will give you an idea as to the amount of mail that was being produced.

I would not doubt that the Roman era saw much the same type of production quantities.

Mike, whether Sim intended that figure to be taken literally, it did come across as mail taking an infinitely long period of time to make, which we know is certainly not the case.

Why one seemed to be preferred over the other is definately a very interesting avenue of research that would most likely yield some very intriguing information.

[john m roberts]

Maybe it was just the soldiers' preference. The segmentata was lighter. Give any soldier a choice between a llighter piece of equipment or a heavier one, and he'll opt for the lighter just about every time. I was in the US Army when it was changing over from the M14 to the M16. In many ways the M14 was a superior weapon, but the M16 weighed about half as much and soldiers rejoiced upon turning in the old logs for the Mattel toy. They knew they might never be in a real firefight, but they would have to carry a rifle all day long, and they preferred the lighter one. It may have been the same with the segmentata.

[Wander]

OK -- I'm going to ask a really stupid question here, and betray all my ignorance, so be gentle in your answers:

I've done a fair bit of blacksmithing, including some blades. I know swords pretty well. I know knives, and bows and arrows. I've made mail, and I've worn mail, and I've worn leather armor and some pieces of plate (legs, pauldrons, elbow cops, etc.).

But... I don't have a lorica segmentata, and I've never worn one. It just looks to me like it wouldn't be too hard for a blade tip (or spearhead, or arrowhead, etc.) to get between the lames and find the flesh underneath. From what I can see, mail provides much more complete coverage -- even if it is heavier and harder to make.

Am I wrong? Am I missing something? Or is that probably one of the reasons why mail, in the form of the lorica hamata, never really went away, even after centuries of segmentata usage in the legions?