When I was studying the metallurgy of ancient Rome, I saw a rather controversial study done by Michael Fulford, David Sim and Alistair Doig called "The production of Roman ferrous armour:a metallographic survey of material from Britain,Denmark and Germany, and its implications".Here is what it says about the quality of the armor of ancient Rome:

In the material itself there is perhaps a greater variation, from martensitic steel to pure ferrite, but the hardness is consistently greater than about 180Hv. Of our samples, 80% are of ferrite, most showing equi-axed grains from hot-working, and with hardness ranging from 124 to 325HV; the latter (no. 20, Vindolanda helmet) was achieved by heavy cold-working and showed elongated ferrite grains.In all, 64 measurements of different sheets of ferrite give an average hardness of 195Hv and a median of 190Hv. Given that annealed iron is normally 90-100Hv, the extent of cold-working to achieve hardnesses in excess of 180Hv can be well appreciated. Although the samples are small, none of the iron helmets was softer than 210Hv, and none of the shield-bosses was harder than 210Hv. Except for the 3 quenched-steel chain-mails (see below),hardnesses of chain-mail and lorica were between 125Hv and 275Hv(one steel lorica registered 295Hv). enter image description here

But on the other hand, the hardness of armor in different centuries of the Middle Ages will be as follows (according to data from Alan Williams' book "The knight and the blast furnance"): enter image description here Here are some examples: enter image description here

enter image description here

By the way, on page 16 of "The Sword and the Crucible", Alan Williams describes Roman metallurgy as "fairly mediocre" in quality.There is another feature, which is that increasing the carbon content increases the hardness of the steel, which seems to be good. However, there is another important parameter - impact strength or the ability of a material to absorb mechanical energy released at high speed. The problem is that by increasing the carbon content, we increase the hardness of the steel, reducing the impact strength and vice versa.

A pure iron Roman plate hardened to 240 Hv will have an impact strength 3 to 4 times that of a medium carbon steel breastplate of the same hardness. By this I want to say that the Roman armor, in theory, should have been even stronger than the medieval one.

So can we then say that Roman armor (and metallurgy), such as the lorica segmentanta, is better than medieval armor (yes,of course, except that medieval armor covers most of the body)?

  • 2
    This may be a better fit for engineering.stackexchange.com
    – Jan
    May 1, 2022 at 20:37
  • 7
    But what I can say is that you are comparing the hardness of armour, when hardness does not seem entirely relevant here. IMHO the more important parameter would be brittleness (not "how quickly does the steel bend?" but "how quickly does the steel shatter?"). The two are correlated, but I do not think you can automatically infer "steel A is more brittle than steel B" from "steel A is harder than steel B". Of course other parameters may be important too, e.g. during production of the armour.
    – Jan
    May 1, 2022 at 20:42
  • @Jan The expectation is that harder steel will be stronger. Flex is important but if the steel/iron is too soft it will crumple and potentially deform. The better quality armour should be the one that retains its function better after being hit. The definition of quality would help here. Is it superior metallurgy, superior design or both? Should intended use matter in the definition? May 7, 2022 at 3:41
  • Looks interesting but inconsistences, It would be interesting to see the original document. Toughness should not be a problem for these relatively thin parts with moderate strength . The implication is that medium carbon steels could be produced in medieval times ; that would be something new in my experience. Iron was pretty much wrought < 0,1 C , or cast iron > 2.0 C. That is what made Damascus blades impressive , a fusion to give medium carbon. But they also talk of cold worked low carbon steel which makes more sense. Hope I can find the originals. Jun 1, 2022 at 20:12

1 Answer 1


I'm not a metallurgist, that's your field. I'm looking at your question from a historic and economic angle:

Was the quality of Roman armour really worse than the armour of knights of the XII-XIV centuries?

Of course. There is a full millennium between the Roman Empire and the XIV century. Even if technology developed very slowly during the dark ages (early medieval period), progress was there. Every generation added a little to it. Some generations more than others. 40 generations adds a lot.

  • What did they (the smithies and armourers) accomplish over time?

Armour became progressively cheaper, lighter and stronger. Look at an early medieval hauberk: it was very expensive to make (many months), worn only by well-to-do military. The sleeves were short, the hauberk reached slightly over the groin area.

Frankish warrior

It doesn't differ much, if at all, from any Roman armour. Also notice the fairly large shield. Roman shields (republic, early & mid-empire) were even larger. Those relatively large shields were needed to make up for what the armour didn't protect.

As technology progressed, we see more armour worn by more common people. Also, armour would protect more of the body. The hauberk of a Frankish knight wouldn't look overtly strange on an English or a Burgundian footsoldier. (Yes, it would, but the armour worn by Burgundian or English footsoldiers was at least as good if not better as that worn by Frankish knights.)

A knight would of course get the best he could afford, which was a good deal more, being custom designed full plate armour, if he was really rich.

I don't doubt Romans could cast plates, but not cheap, light and strong enough. A Roman cuirass is fairly heavy and seriously more expensive compared with a late medieval cuirass. I've seen Roman gladiators equipped with full plate armour, almost like a late medieval knight. (Sorry, no links.) The type was very rare, but it was possible. Or better phrased: not completely impossible.

I mentioned the shield. I noticed - but can't proof - shields tend to get smaller when the technology of the armour allowed for it. A late XIV century knight would wear a shield more for decorative than practical purpose. An early knight wore a large kite shield. I don't think that was purely coincidental. However, plenty of warriors wore a small shield, so it's just something I noticed.

The Romans themselves developed Lorica Segementata, but dropped it at the end of the second century. The reason was probably very high maintenance and repair costs. The different metals reacted to each other, and it rusted, as we nowadays would say: "in the catalogue". Roman soldiers spend a good deal of their time polishing and fixing their lorica segmentata. Reminds me of my army days when we spend many hours polishing our white canvas belts absolutely white. Was a great way to avoid PT!

  • The sizes of the various bits and pieces are mainly dictated by the way soldiers/knights are expected to fight, and only partially by the available technology. So, one could imagine a head-to-toe covered-in-maille legionary as the relevant technology was availabel and widely used, i.e. making chain mail. However, there was no need to cover a legionary completely in maille in a fashion as the medieval knights did.
    – Dohn Joe
    May 2, 2022 at 8:19
  • 2
    Most iron produced in both these periods was wrought iron which was forged. The outer surface could then be case hardened (extra carbon added) to convert it to steel. You then get a layered armour with the hard steel outside and the malleable iron supporting it and stopping it fracturing.
    – user55099
    May 2, 2022 at 8:36
  • 1
    On the subject of plate armor, people tend to forget bronze, which was used to good effect a thousand years before Rome came to be...
    – DevSolar
    May 2, 2022 at 9:16
  • 1
    I flinch a bit at the mention of "steel" in these contexts. Wrought iron does not equal steel, but those finer points of metallurgy are usually lost on most people. (No offense intended whatsoever, it is a difficult field.)
    – DevSolar
    May 2, 2022 at 11:10
  • 1
    I'm curious about the date when lorica segementata was discarded. While this seems to have happened before the 4th century, I've yet to find an authority to provide a more precise date. If I had to guess, I believe it would have been after the death of Septimus Severus, perhaps after AD 235. (There are no military innovations recorded for him.)
    – llywrch
    Jun 2, 2022 at 21:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.