This week we’re going to look a specific piece of early Roman military equipment, the humble bronze pectoral, which it turns out is surprisingly tricky for us to confidently reconstruct, in part because the period of its use that most interests us (the run from c. 264 to c. 146 where Rome is winning its first big overseas wars) is a relative gap – fancy word, ‘lacuna‘ – in our evidence, making it really difficult to correlate what our literary source (Polybius) is telling us to the physical evidence we have (both preserved examples and artwork). This was, we are told (by Polybius) the armor of the common Roman soldier in the period of their greatest wars, yet on some level we do not really know what it looked like. Not with certainty, in any case.
In particular I am going to argue that the most common reconstruction of this armor, as a single bronze plate suspended usually by leather straps over the chest, is probably wrong and that the armor more likely existed as a complex harness, simplified in brief literary description down to just its core element. But as we’ll see, this is going to be a zone of what I term ‘real uncertainty’ – a situation where without new evidence coming out of the ground, we simply cannot know for sure.
So this is not just an exercise in working through how to reconstruct one specific kind of equipment, but also how historians engage in questions that exist in a zone of really low confidence.
But first, as always, affording a full panoply of heavy infantry equipment as is the duty of any propertied Roman citizen is expensive! If you want to help me waste spend my money on reproduction ancient military equipment, you can support this project over at Patreon. If you want updates whenever a new post appears or want to hear my more bite-sized musings on history, security affairs and current events, you can follow me on Bluesky (@bretdevereaux.bsky.social). I am also active on Threads (bretdevereaux) and maintain a de minimis presence on Twitter (@bretdevereaux).
Dark Blue: Ligures
Brown: Veneti
Pink: Etruscans
Light Blue: Piceni
Light Green: Umbrians
Dark Green: Oscans (including the Samnites, discussed below)
Orange: Messapii
Yellow: Greeks
Gold: Latins (including the Romans)
Polybius
Our first stop is Polybius. Polybius wrote in the mid-second century (that is, the 140s), but his history covers the period from 264 to 146 and his description of the pectoral is placed relatively early in the narrative, in 216, as part of a larger explanation of the Roman military system. There is thus immediately a question as to if the details Polybius is giving are correct for 216 or for the 140s when he wrote. In practice, the answer must be something of a mix: Polybius has sources that reach back and might give him details appropriate to the period (he seems to have the writings of a military tribune to use for this description of the dilectus), but it seems likely that his description of the pectoral comes from observing it. Consequently, while I suspect that Polybius’ description of who is required to wear what may be accurate for 216, he has clearly seen the pectoral and understands it to still be in use in his own day (indeed, at other points in this extended passage, he explicitly notes things that used to be one way but had changed by his own day).1
That’s handy, because Polybius is the only source that describes this armor. Later historians – Livy, Plutarch, etc. – seem broadly unaware of it and it really does seem like the pectoral was in the process of going extinct when Polybius was writing (for reasons below). So we have one description of the armor, but at least it is by an eyewitness. Here it is (Polyb. 6.23.14-15, trans. mine):
The many [hoi polloi, “the common folk”] taking a bronze plate a span [c. 23cm] on all sides, which they place over their chests and call ‘heart protectors’ [καρδιοφύλαξ, very literally ‘heart protector’], finish their armaments. However those worth more than ten thousand drachmas [= the first class of Roman infantry], instead of the heart-protector wear mail coats [αλυσιδωτοί θώρακες, “hooked [or chain] cuirasses” which we know is the Greek way to say ‘mail coats.’]
And…that’s it. From later authors (Varro, De Ling. Lat. 5.116; Plin HN 34.18) we get the Latin name for this armor, pectorale (pectorale, pectoralis (n) for the Latin nerds), thus the English term ‘pectoral’ but no more details of its construction.2
Crucially, this armor doesn’t show up on any highly visible Roman military monuments. The reason is fairly simple: the earliest really visible Roman military monuments are the Pydna Monument (168) and the so-called Altar of Domitius Ahenobarbus (late second century) by which point the pectoral was already on the way out. Ancient artists tend to prefer high status equipment and so with the pectoral on the way out (though likely still very much in use in 168) and the poorer, lower status armor, they didn’t depict it, instead preferring to use mail armor to signal Roman soldiers (specifically, mail is used on the Pydna Monument to signal ‘these are Romans’ in contrast to Macedonians or Gauls).3
As a result, scholars initially didn’t have a lot to go on except Polybius’ description – the archaeology, as we’ll see, doesn’t really get sorted out until the last 40 years or so. So they reconstructed on that basis. A ‘span’ (σπιθαμή) is a ‘natural’ unit, the distance between the thumb and the little finger at full extension, which is conveniently more or less half of a cubit (the length of a forearm out to the end of the middle finger), which eventually becomes formalized in Attic measurements (which Polybius tends to use; other places might have slightly different measures for the same terms) as 23.1cm and 46.2cm respectively.
That leads to the most common thing we see in artistic reconstructions and reenactor kit: the pectoral is reconstructed as a brass or bronze plate, usually about 1-2mm thick (the normal thickness for breastplates), 23cm by 23cm square. Since obviously it needs to be attached to something it is often shown backed in leather, with leather straps around the waist and over the shoulders holding it in place. I am going to call this reconstruction – a single plate, 23cm square, on a leather harness – the ‘traditional’ reconstruction.
That size lets the pectoral cover most of the chest, but it does nothing for the belly, sides or shoulders. On that basis, I have very often heard scholars regard it as a very minimal, almost token defense, unlikely to do much at all to protect the men wearing it. And again, before there was much archaeology to work with (or before finds had been analyzed, arranged chronologically and had their development worked through), you can see how this is the most logical extrapolation of what Polybius is saying.
But I do want to note some things here. Polybius’ description of this armor is extremely brief. He does not even bother to explain what Roman mail armor is like at all – no description, for instance, of its length (to the knees) or shoulder-doubling or the front-closure mechanism. If it weren’t for period depictions of mail, we would probably reconstruct it without these elements. As for the pectoral, all he says is that it is a span square and the Romans have a funny name for it. Which is to say it is entirely possible that Polybius is leaving out some details here. Which brings us to:
The Development of the Italic Pectoral
This, of course, is the point at which we naturally turn to archaeology to provide us both physical examples of this kind of armor and also visual representations of it. And he we run into an immediate problem: the third and second century feature a near total lacuna of Italic armor, in both artwork and preserved examples. The problem is frustrating in its elegant simplicity: the Roman military system – terribly efficient and in its way, anti-aristocratic – coincides as it expands with the end of aristocratic ‘warrior burials’ wherever it goes. Thus as Rome during the fourth and early third century goes about consolidating control of Italy, the amount of nice tomb paintings with aristocratic warrior in procession or burials with arms and armor drop to basically nothing. The Roman army is removing the evidence we might have for the Roman army. Astoundingly frustrating.
The evidentiary record begins to pick up a bit in the second century with more artistic depictions of Roman soldiers as the Roman state engages in more monumental depictions of its soldiers (noted above), but by that point mail rather than the pectoral is the ‘national armor’ of Rome’s armies (even though the pectoral is likely in use) and pectorals never appear. The really strong archaeological record for armor will have to wait until the imperial period, when the permanent stationing of Rome’s armies on the frontier of the empire means they sit in one place long enough for us to recover bits of armor.4 Weapons show up more often than armor (pila more often than any other type of weapon, a testament to their disposability) and we get a lot of helmets (for reasons not entirely clear to me), but functionally no body armor from this period. The best we can do are tiny fragments of metal rings for mail and even those are rare.
Worse yet, as mentioned before, the pectoral was going extinct in this period. Notably, when our evidence improves massively in the first century BC and AD, the pectoral is nowhere to be found. No source mentions it as still in use in that period, no artist depicts it, no finds of it are recovered. Polybius is thus our last source for this armor, suggesting that by the start of the first century, it had been wholly replaced by mail. No shock, mail is awesome (if expensive). But that means we cannot look for later examples to help us understand what Polybius is saying.
But we can can look at earlier ones.5
The Italic pectoral seems to have arrived from the Middle East in the 8th or perhaps 7th centuries (sometime between c. 750 and c. 680). This form of armor, a more or less flat metal place (as opposed to an enclosing breastplate of the sort we see in Greece around this time) has Middle Eastern precedents (we see Assyrian soldiers in artwork wearing similar armor), though how exactly it made it to Italy is unclear – Phoenicians seems most probable, but uncertain. In either case, by the seventh century, these pectoral armors are quite common over all of Italy, including Latium (where Rome is) and Etruria. The armor at this point generally consists of two bronze plates (a front plate and a back plate), which might be rectangular or circular, about 20-25cm wide (or tall; sometimes these are even smaller than this) and which were connected by leather straps. We generally call these ‘rectangular’ and ‘single disc’ pectorals. When decorated (and they very frequently are), they usually feature either geometric designs (often rectangles within a rectangular cuirass) or animal designs, either punched into the plate or embossed.
And you can see how an archaeologist looking at these pectorals from the seventh century might be thinking, “ah, I see exactly what Polybius was talking about: a bronze plate a span square!” Except, of course, the sixth century is not the second century and these pectorals keep evolving.
Now, in significant parts of Italy, especially Etruria, these pectoral armors begin to be replaced in the late sixth century by Greek-style armor, especially for elite, high-status warriors. In particular, the Etruscans love the tube-and-yoke (linothorax) armor when it shows up and it swiftly becomes a marker of elite status, though pectorals do occasionally show up in Etruscan art, albeit less frequently, but they are certainly petering out. Annoyingly, at roughly this point the archaeological record for Rome specifically also dries up, so it isn’t clear exactly what armors are popular in Rome in the very early Republic (our literary sources assume Greek-style armors, which may be right, but they are guessing and deeply anachronistic in their assumptions).
However in central Italy, in the Apennines Mountains, the pectoral persists and undergoes some significant design changes. Around 600, we start to see changes to the strap mechanisms holding the armor together: one shoulder strap is replaced with a pair of bronze plates connected by a hinge. The resulting harness gets pretty complex, as you can see in the figure of the Capestrano Warrior (c. 550), where the harness that holds the pectoral also supplies a scabbard (suspended at the chest) for the sword and there is a clear contrast between the metal hinged plate (over the right shoulder) and the more reddish-colored leather straps (of which there are three, two wide and one narrow) holding the harness and scabbard together.
In the early fifth century, this design is both enhanced and greatly simplified with the emergence of the first ‘triple disc’ pectorals. These are so named because the front plate (and back plate) take the form of three discs in a triangular arrangement, though I must stress this is a single plate with three circular designs on it in a roughly triangular shape, not three individual circular plates. Indeed, earlier archaeologists supposed that the ‘triple disc’ cuirass must have evolved in two stages from the disc pectorals discussed above and posited a ‘double disc’ cuirass, which turns out not to have existed.
These triple-disc breast- and back-plates were joined together not by leather straps but by a simplified version of the hinged plate system used in the sixth century disc pectorals, except now there are four connecting plates: one each over the shoulders (each of them hinged) and one at the sides (without hinges). These plates also get a bit wider, providing relatively fuller coverage over the upper body and the armor is supplemented by a wide bronze belt worn around the waist which protects the lower abdomen. You can see the full armor clearly in artwork:
In the second-half of the fourth century (so 350 onwards), we see these triple-disc cuirasses joined by another type, particularly on the western coast of southern Italy (so the area south of Latium), the ‘rectangular anatomical cuirass.’ This takes the existing triple-disc harness structure, with its bronze belt and connecting side and shoulder plates, but instead of the triangular triple-disc cuirass, it substitutes rectangular breast- and back-plates, with the designs on these invariably mimicking the musculature on Greek muscle cuirasses, although – because these plates are smaller than Greek breastplates (which wrap around the body) – the muscles depicted are visibly smaller-than-lifelike. In short, the artistic form of the muscle cuirass is being copied, but this is not an effort to mimic the actual muscles of the man wearing the armor.
To give a sense of size, recovered triple disc cuirasses range from 27-32.5cm tall and 26-28cm at the widest, while the rectangular anatomical cuirasses range from 29.5 x 37cm tall to 25 x 30cm wide for the front plates.6 Combined with side and shoulder plates that tend to be 5-8cm wide and a wide bronze belt (7-12cm wide, 70-110cm long, ~1mm thick), these really do cover most of the upper body, albeit with gaps, and are something closer to an articulated breastplate than they are to the small ‘heart protector’ of the Capestrano Warrior.
And you may note that a rectangular plate over the chest of c. 30cm by c. 28cm is not very far from Polybius’ description of “a bronze plate a span on all sides” and better yet is far more likely to have actually be in use in the third and second centuries for Polybius to see.
This is, as an aside, a good example – particularly the triple-disc component – of how simple the decoration of these armors could get. The cuirass is cut out of sheet metal, has three simple discs hammered into its shape and is otherwise mostly unadorned. Assuming sufficient bronze, such cuirasses could likely be made relatively quickly and cheaply, compared to something like a muscle cuirass (or certainly compared to later mail armor).
Notably – and this is going to matter in a moment – these fifth and fourth century pectoral harnesses do not appear without bronze belts or connecting plates. You will find these pectorals in museums without those added elements, in many cases because when the first of these armors were excavated (and/or looted) it was done carelessly and so the smaller plates were missed. However, whenever we get these armors with secure provenance or see them depicted in artwork, as Michael Burns notes, without exception, we get the full harness with all seven elements (frontplate, backplate, 2 shoulder plates, 2 side plates, bronze belt). We never to my knowledge ever see them suspected in simple leather harnesses; it surely was possible to do so, but it is unclear that anyone ever did after the introduction of the four-plate harness.
What Michael Burns thinks is happening (revising earlier work by the late, great Peter Connolly), and I think he is right, is that Southern Italic peoples are responding to the increasing presence of Greek muscle cuirasses coming in through Greek colonies in Southern Italy. But rather than just copying the muscle cuirass, they seem to have innovated from their own single-disc pectorals (which didn’t always cover a whole lot of the chest) to the triple-disc to create a kind of ‘exploded’ muscle cuirass. Initially, they do this by taking their own armor form, the single-disc cuirass, and expanding it out into a full ‘exploded’ breastplate, but eventually, in the fourth century, there’s enough artistic crossover that designs that use a rectangular plate and intentionally mirror Greek artistic tropes appear alongside triple-disc styles (which do not go away). It is worth noting that some of these triple-disc and rectangular anatomical armors are wonderfully decorated with complex designs, but many of them are very minimally decorated, especially as we get into the fourth century, suggesting a demand for a cheaper, no-frills version of this protection.
And then in 290 the Romans win the Third Samnite War and take control of the non-Greek parts of Southern Italy. And as noted above, when the Romans incorporate a given part of Italy into their ‘alliance’ system, for reasons that are not entirely clear to us (but the pattern is very strong), warrior burials, ritual weapon depositions and aristocratic artwork of warriors stop. Which means right around the year 300, our evidence for the Italic pectoral tradition simply vanishes. Really, we basically have an expanding bubble of darkness, radiating out from Rome (which is also probably how the Roman conquest felt to the Samnites), blinding our ability to track the development of armor in Italy.
So by the third century, we do not see any pectorals, because we don’t see much of anything (except helmets; we continue to see those) for quite some time.
Except…
The Weird Exception We Need To Dismiss
The one odd exception to this is a pectoral disc found in the siege camps at Numantia.7 It is 17cm wide and circular, with a pattern of concentric circles and a large central knob and for quite some time if you went looking for an actual Roman pectoral this is what you would find.
The problem is that it isn’t Roman, it is very obviously Spanish. This spent a century not getting noticed because archaeologists working on ancient arms and armor tend to be very geographically specialized, so folks working on Roman and Italic arms and armor are not likely to be very familiar with the arms and armor of the fifth century Celtiberian Meseta. But if you are familiar with that, it is very clear that this is not a Roman pectoral at all, but a Spanish one, despite it turning up in a Roman camp.
Right: Via the Museo Arqueologico Nacional, Madrid, a Celtiberian pectoral harness (MAN 1940/27/AA/314, main disc 18cm in diameter, late fifth to early fourth century), showing similar concentric circle motifs and punch-holes around the outer edge.
First, while Italy had single-disc circular pectorals these had been replaced in the archaeological and artistic record in the fifth century by the larger triple-disc pectorals discussed above. Moreover, those earlier single-disc Italic pectorals don’t feature raised concentric circles as part of their normal artistic motifs. The more often have animals on them, or punch-holed simple geometric designs. They were also flat and did not feature central knobs.
But you know who did have pectoral harnesses with circular central plates featuring raised concentric circle designs and prominent central knobs? The Celtiberians, who are the people who lived at Numantia, where these camps were. Now the tricky bit here is that these pectorals are also – as far as we can tell – long out of use in the Iberian Peninsula as well: they persist through the fifth century, but fade out at the beginning of the fourth.
But whereas it is a little difficult to imagine a second-century Roman soldier decided to bring a piece of armor with him to Spain that had been out of use in Italy for something like four centuries, it is a lot easier to imagine the same Roman soldier in Spain might have looted a temple or a tomb (or simply struck a burial while entrenching his camp) that contained a fifth or very early fourth century Celtiberian disc-harness and that this soldier then looted the shiny bronze plate, later to be (for whatever reason) discarded in the camp.
Reconstructing the Roman Pectoral
So that is the shape of our evidence: with the Numantia pectoral removed (because it is not Roman at all, but Celtiberian), we have no examples of this armor from the third or second centuries B.C. What we do have is a tradition of pectoral armors which lead to the emergence of the triple-disc and rectangular anatomical pectoral harnesses in the fourth century, which we lose sight of in the general lacuna for most non-helmet military equipment in the third and second century. When our evidence returns, they are gone but we have this report by Polybius that poorer-but-still-propertied Romans in the heavy infantry (so not the poorest Romans fighting, those are the velites or do not serve at all) wear a bronze pectoral plate about a span square over their chest.
That admittedly quite poor evidence base leaves us with really just two options, both of them somewhat unsatisfactory.
The first option, the one taken – so far as I can tell – by the great majority of modern artistic reconstructions, is to simply read Polybius and reconstruct exactly what he says. That gives these Roman soldiers a single metal plate, typically shown mounted on a leather backing with leather straps, about 23cm square. This is, in a sense, the philologically elegant solution: it assumes nothing not in our text. The problem, from an archaeological perspective, is that this effectively requires arguing one of two cases: either first that sixth century pectoral – with its simple leather suspension – somehow survived in Italy for four centuries to be observed in action on the battlefield by Polybius in the mid-second century without leaving any other evidence at all. Not one piece of artwork, not one surviving example in the intervening period, despite the fact that we have sixty-seven fifth and fourth century examples of the later pectoral types (45 triple-disc and 22 rectangular anatomical cuirass types). That could be right. But it is a heroic assumption.
Alternately, the argument would be that the Romans at some point developed their own version of the pectoral, probably based off of the rectangular anatomical type, which discarded with the wide bronze belt, the shoulder plates and the side plates and so consisted only of a breastplate and a backplate. The problem here is simple: as Michael Burns notes in his survey of Italic pectorals, that configuration never occurs in artwork or in archaeology where site and provenance are secure. We do not have a single example of those later Southern Italic pectorals – the types that emerge after the more complex harness structure discussed above – dispensing with those pieces. Could they have done? Of course. But as of 2005 (and so far as I know, to the present), we have no evidence that anyone ever did. This solution thus requires conjuring into existence an effectively unknown armor-type. That could be right, particularly given how bad our evidence for Roman arms and armor in the Early Republic is. You can even imagine, if we had evidence of it, how we’d explain it: the broadening participation in the Roman army leads to poorer Romans to take up the Samnite cuirasses (that is, triple-disc and rectangular anatomical cuirasses) they have seen, but to jettison the ‘extra bits’ to make it cheaper and more affordable, effectively reversing a few centuries of armor development to create a stripped down breast- and back-plate only version. That’s what we’d posit, if we had some, but we don’t have some and I would argue that it runs against the rules of evidence as practices in archaeology to conjure into existence an unattested variant of an object-class (which does not developmentally link to anything else you can see) simply because it would be convenient. That is not how we assess coins or pots, I do not see why we would do it with armor.
That leaves another option: Polybius is describing the Southern Italian pectoral harness we can see, but doing so incompletely. It is not hard to imagine how the Romans will have picked up this armor: they spent the period from 343 to 290 fighting the Samnites in Campania and the Samnites are the major users of the triple-disc cuirass and Campania is where we most often see them in artwork. If the Romans weren’t already using this armor (and remember, we have no evidence at all of what armor the Romans are using in c. 300), they could certainly pick it up.
Then Polybius comes along in the mid-second century, where this armor is already dying out, largely replaced by mail, but still hanging on here or there – perhaps as hand-me-downs used by poorer Romans. One advantage of the pectoral harness’ seven-part structure is that it is a sort of ‘one-size-fits-no-one’ set that would be reasonably easy to modify or pass down to new users (unlike a Greek-style muscle cuirass, which really needs to be fitted to the wearer). Polybius then, writing about the Roman army as it existed in the Second Punic War (218-201) and, as per Rawson, using perhaps the accounts of some military tribunes, is aware of this armor’s place in the military regulations of that time and so includes it but with only minimal description. As a Greek, Polybius is used to thinking about body armor as a single piece – a breastplate, a tube-and-yoke cuirass, a mail coat – rather than a harness, so looking at a rectangular anatomical cuirass that is, perhaps 30cm by 28cm for its front plate, he describe sit simply as ” bronze plate a span on all sides.” Just as he doesn’t include the details of Roman mail armor’s shoulder doubling, he feels no real need to include the shoulder and side plates of the harness and he may not even be aware that the wide bronze belt has any real armor value at all (early archaeologists made the same error, assessing it as purely decorative, but it would offer some protection).
I think these are the three options we are left with for the pectoral: surprising sixth-century survival in the mid-second century, otherwise un-evidenced recreation of an older form out of the fourth-century rectangular anatomical cuirass or simply that it is the rectangular anatomical cuirass, harness at all, that Polybius has described incompletely. My own instinct is that the latter is probably correct. One interesting thing is that compared to, say, muscle cuirasses, these pectoral cuirasses of both the triple-disc and rectangular anatomical types were probably produced from sheet metal (sheet bronze, in particular), rather than forged from an ingot, which would have made it relatively easier to produce larger numbers of armors – equally if one opted for a style with simple decorations, amply in evidence in the archaeological record. Meanwhile, as noted the design is fairly easy to adjust for size. Jeremy Armstrong and Nicholas Harrison suggest that this in part allowed for “the expansion of warfare in Italy seen in the fourth century and marked by Rome’s wars of conquest” and I think that is right.8
Now in the fourth century, that armor might still be restricted to the fairly well-off. But in the late third or early second century, it is not hard seeing how the introduction of an even better but also substantially more expensive armor – mail – might ‘push’ existing pectoral cuirasses (again, of both types) down the socioeconomic ladder as the Roman first census class was required – as Polybius tells us – to acquire mail. The spare armor might ‘flow downwards’ as it were, making the affluent man’s undecorated but still shiny bronze armor of c. 350 the poor man’s pectoral of c. 150. Indeed, there is no reason it couldn’t be the very same piece of armor.
I do not think the evidence allows us to answer this question with confidence, but I do think that simple inertia has led scholars to continue reproducing the ‘traditional’ pectoral reconstruction long after it stopped being the most likely one. Instead, the most likely solution is that the Romans had continued to use, in some form, the full triple-disc or rectangular anatomical cuirass, including metal connecting plates (and perhaps the wide bronze belt) and that what Polybius was seeing was not, in fact, the small decorative chest-plates of the sixth century but rather this armor.
- On where Polybius is getting all of this from see Rawson, E. “The Literary Sources for the Pre-Marian Army.” PBSR 39 (1971)
- Well, that’s not quite right: Varro states these were made out of rawhide, which is just wrong so far as we can tell – these are always made in bronze.
- On the Pydna Monument, see Taylor, M. J. “The Battle Scene on Aemilius Paullus’s Pydna Monument: A Reevaluation.” Hesperia 83, no. 3 (2016). On the so-called Altar of Domitius Ahenobarbus, see Stilp, F. Mariage et Suovetaurilia (2001).
- Plus the substantial finds from Kalkriese, the site of the Battle of the Teutoberg Forest (9AD).
- The best thing to read on this is actually a dissertation, M. Burns, “The Cultural and Military Significance of the South Italic Warrior’s Panoply from the 5th to the 3rd Centuries BC” (2005) which is available online.
- These figures are direct from Burns, op. cit., which you should read for more details.
- Schulten, Numantia: Die Ergebnisse der Ausgrabungen 1905-1912. Band III. Die Lager des Scipio, (1927), 50, taf. 44.19.
- J. Armstrong and N. Harrison, “The Armorer’s Craft: Using Experimental Archaeology to Explore the Production of Bronze Armour in Italy c. 400-300 BCE. JRMES 22 (2021): 43-68. If you are interested in this armor, do try to turn up this article – it has both production methods, but also a lot of images and illustrations of what the triple-disc cuirass looks like.
Typo?
“This form of armor, a more or less flat metal place”
place or plate?
Concerning the ‘traditional’ reconstruction of the pectoral. Not that I think this is a *good* methodology, but the square plate with four straps holding it to the chest IS the description you get of the חֹשֶׁן of the High Priest in Exodus, especially chapters 28 and 39.
Now again, I want to stress how terrible this is as a comparison point; the חֹשֶׁן is made out of linen inset with gems and the high priest is most definitely not contemplating fighting anyone in this thing. It’s a ceremonial outfit, not armor. But I do wonder if that had some perverse influence on people; at the very least, my brain instantly went there when you started describing the traditional reconstruction. I can imagine later scholars, especially once Europe Christianized, carelessly making a spurious connection there.
Presumably some evidence for how widespread this style was in the mid-first millennium BCE.
Also the High Priest in the time of Nero was probably the last person still wearing a pectoral?
Considering that it’s usually understood that Romans adopted the fighting methods of the Samnites (pilum + sword combination, flexible maniples, etc…), it wouldn’t be too surprising that they also adopted their armour. Otherwise I’m slightly wary of assuming too much standardisation in what Roman armour was like; it seems entirely plausible to me that the sparsity of evidence hides what might be more variation than is assumed.
Today I learned what I cubit actually is, after reading about it all the time in history books. Thanks for the post, Bret.
Surely part of the reason Roman expansion eliminates the warrior burial is that the huge demand for military mobilization as a socioeconomic factor causes significant pressure to pass your used arms and armor down to relatives, vs other societies where showing off wealth via consumption was somewhat more important?
If one assumes that, of course, it would be a strong argument for the Roman pectoral to be basically the same thing their neighbors used, especially if it’s an unusually-easily-passed-on type of armor.
(which would line up pretty well with “Polybius just incompletely described it”…)
There’s a class of people for whom ‘passing down the worn armor’ is not going to happen, because they have the income to provide their sons with new armor, and new armor is preferable to repaired armor. This is the class the Romans did incidentally eliminate a lot in making socii, taking enough land from defeated enemies that non-working landlord families were reduced to including themselves in farm work and distributing that land to create more family-worked farms.
It is a commonplace that during their expansion in Italy, the Roman Republic more or less maximized the possible number of heavy infantry they could levy, by chopping up conquered land into farms prosperous enough to support heavy infantryman, but usually not prosperous enough to support aristocratic cavalry.
If so, you’d probably see a lot of Roman farmers of this class that can afford armor but struggle to do so, thus tending to perpetuate the cheaper and older types of armor, because at this point in their history not even the Romans are rich enough to easily outfit their entire (huge) army in mail.
I would have posited the same explanation against the existence of warrior burials in general. Why waste good gear, whose bill of material alone is measured in man-years? And yet people would still bury those precious iron swords.
Winning at war is a matter of survival for any society, but that doesn’t stop individuals from partying or preparing for the afterlife. It can’t be coercive because even modern murderously totalitarian regimes fail to stop elite status displays. So all we’re left with is vague gesticulating at the superior Roman culture that abhors waste to create a supremely efficient war engine, folded a million times over and destined to conquer the world. But to vaguely invoke “culture” is to fail to answer the question.
The argument is that if you have to choose between a warrior burial or arming the next generation of that specific family, you will choose to arm them – but if you don’t have to choose, the elite status display makes sense as well.
Thus, then, the Roman ‘terraforming’ split up wealth that could support the elite status displays without impinging on the heirs and effectively stamped it out by raising the demand for said arms and creating more households in an economic band where secondhand armor is not so expensive as to be unaffordable, but also where new armor is expensive enough that skimping on the price is worth considering despite the loss in protection.
I think a proposition such as that needs to account for why it would apply to this society more than it would to other ones in similar conditions (the statement that they would become more militarised under Roman suzerainty might beg the question a little), as well as consideration that burial of weapons has deeper cultural connotations than just showing off wealth.
I think part of this has to do with how Romans viewed the army. The army was, at least as I understand it and before it became professional, something boys did during the transition to manhood–a finishing school, sort of. Then most of the boys came back with a few scars, a few stories, some new friends (not an inconsiderable consideration given how Rome worked), and maybe some loot, and take part in civic life as a full man. But the point is, they came back and took up farming again. The army was a stage in life, not life itself.
There’s that whole story of Rome being threatened with disaster that speaks to this. The Oracle said that the most valuable thing needed to be sacrificed to avert the disaster, so a dude in full combat kit sacrificed himself, saying that the most valuable thing in Rome was a citizen willing to take up arms to defend it. The key is that the guy was NOT a warrior. He was a farmer who’d served in the army.
Different societies viewed military service differently. Sparta is one extreme–only the warrior-elite were considered citizens. In the Middle Ages the warrior-elite were focused on being warriors, while the peasants viewed being in the army as a thing that occasionally occurred (note that nobles may be buried with arms and armor, but peasant levies rarely were, unless it was mass graves). Rome viewed military service as an important duty of a citizen, but it was, for a long time anyway, a temporary one, not the defining feature of a person’s life.
What kind of armor would have been used during the extreme mobilizations in the civil wars? They couldn’t make so much fresh mail. (Did they keep a stockpile of armor several times larger than they expected to use? Surely not, given how expensive armor was.)
Also, a variant of e.g. the triple-circle pectoral made not from expensive bronze but comparatively cheap rawhide would make perfect sense, no? Less protective, less expensive, suitable for the lower classes still fighting as heavy infantry (rather than velites), therefore less likely to be depicted, and unlikely to survive rather than rotting away. And as mail moves down from the top of the hierarchy, pushing bronze pectorals, those push rawhide pectorals in turn.
They wouldn’t have had a centralized stockpile, but most families subject to the draft would own their own set of whatever was required for their wealth class (or possibly better, because their lives are on the line) and bring it when drafted. That of course leads to the question of how everyone liable for service is able to afford all that armor, which is the subject of our gracious host’s book project.
I didn’t mean a central stockpile; I know no such thing existed.
However, let’s compare the dilectus to the Carolingian select levy.
Mail wouldn’t have been order-of-magnitude cheaper to the Romans, for every Gaius and Titus to have their own armor (especially since Roman farmers remarkably tended toward having smaller plots). Economically it should have looked quite similar, even if the administrative process was turned inside out. In whatever fashion, in summation the people of a village would own several sets of armor, and whoever happened to be drafted that year, would ~borrow one (and return it when arriving home from a successful campaign). The utilization rate of sets of armor could be fairly high (you don’t want to leave such expensive capital lying unused), with different soldiers from the village rotating through it over time. But this raises the question that, if the total number of men under arms grows steeply (since the process allows that, both in raising more than the usual number of legions, and in not demobilizing armies stood up in previous years), where the extra equipment (particularly armor, the most expensive item) would come from. And since it couldn’t be mail, what it would be.
I don’t know about an order of magnitude cheaper, but the Roman social structure does give more disposable income to the middle class soldiers. Could be because Romans have higher productivity overall, could be because the income is spread more equitably, or both.
A mail shirt is expensive, but mail links are small and easy to accumulate over time. If a sword or breastplate breaks, you have to reforge the entire thing. If mail breaks, you just need a few links to patch it back together. Making extra links and patches is something a blacksmith can do whenever they’ve got a free hour, and eventually they’ll have enough for a shirt. Making a breastplate demands a fixed block of time.
So I would expect a steady accumulation of mail shirts. The total number of men under arms is growing steeply, but over a period of decades, not months. I can imagine supply keeping up with demand.
Yes, by this point the Romans would have been more marketized and more urbanized than Carolingians. (On the other hand, they had worse technology: they still use scratch-plows (ard->ardous) rather than what we imagine under “plow” (caruca).) It seems plausible to me that there may have been twice as many non-peasants per peasant in this period than among Carolingians, so also twice as much metalwork per peasant. So, the Romans get to play around with a knee-length mail hauberk-equivalent per two objective mansi, not four? Good start, but still far from every household having one.
Gradual accumulation is possible, but it applies equally to the Romans and to other mail-using cultures.
And this is why I’m specifically asking about sudden increases in men under arms. E.g. Caesar/Pompeians, then half a decade later Caesarians/Liberatores.
These societies are resource constrained, but they still have choices about what they’re going to spend resources and wealth on.
For a modern example, cars. These are common in any largely urbanised industrial country. They don’t vary by much across the world, the same brands and models with minor variations are available everywhere. And they get used for the same purpose, moving people around. (I’m leaving out the non-urbanised non-industrial countries where you get, say, technicals.) And cars are a major investment with a long working life for the average family, like a suit of armour.
Wikipedia has a page “List of countries and territories by motor vehicles per capita” and while there are lots of similarities, there are some dramatic differences. Taiwan is near the top, almost 1000 cars per 1000 people, just on twice as many as South Korea at 501. Pakistan has about three times as many cars per 1000 people as India or Bangladesh. New Zealand has 817 cars per 1000 people, while Denmark has only 478.
There would be differences at smaller scales too. Across the USA I believe there are cities where “everyone” owns a car, even if poor, and cities where lots of people don’t.
All these differences are no doubt explainable. The point is that different societies with similar levels of availability and wealth can “decide” what they spend it on.
So I don’t think there is anything inevitable about Roman society having a lot of personally owned armour. We can look for factors that might have made it impossible, for example why legionaries couldn’t all be wearing heat treated high carbon steel; but some other society not having similar levels of ownership does not sufficient. Sometimes “that’s just what happened” is the explanation, unsatisfying though it may be.
> Also, a variant of e.g. the triple-circle pectoral made not from expensive bronze but comparatively cheap rawhide would make perfect sense, no?
This is a questionable assumption IMO – there is very little evidence for large scale use of leather as ‘cheap’ armour in pre-modern Europe. There just isn’t that much land being dedicated to raising cattle, and you don’t get that much thick leather from a single cow anyway. This seems to substantially change at some point in the mid to late 16th century, and might well be primarily driven by the rise of ranching in Spanish imperial possessions in North America, but obviously that’s 2000 years in the future for the Romans.
And, of course, there is no textual or archaeological evidence for such. As our host puts it:
> Could they have done? Of course. But as of 2005 (and so far as I know, to the present), we have no evidence that anyone ever did. This solution thus requires conjuring into existence an effectively unknown armor-type.
And
> That’s what we’d posit, if we had some, but we don’t have some and I would argue that it runs against the rules of evidence as practices in archaeology to conjure into existence an unattested variant of an object-class (which does not developmentally link to anything else you can see) simply because it would be convenient.
Which I would argue apply just as strongly – if not moreso, tbh – to a hypothetical leather pectoral as they do to alternative forms of the bronze pectoral.
A strong point for “expensive” bronze is also that it accumulates. If you never loose a soldier wearing such an armor, it will be good for centuries, given a little maintance. And after 5 generation you might have 3 sets of the stuff floating around in an rather unasuming household. Same is even true to an extend for mail.
Leather armor when worn, even when properly maintained will not hold more than a generation.
They don’t have many cattle, but they do keep a reasonable amount of pigs, sheep, and goats. (Are the skins of those species less suitable somehow?) Since these are kept mainly for their wool, meat, milk, etc., the leather is almost a free albeit quantity-limited byproduct, and processing it into armor takes comparatively little labor, unlike a linothorax, assuming that is made of several layers of spun and woven textile. (Since an N-layered linothorax is basically N tunics glued together if you squint, one would have as much or more spun yarn in it as the rich-farmer household’s clothing combined. But so would a fishing boat’s sail.)
I understood those parts about reasoning to apply to, well, reasoning, as in “please for the love of Saturn, do some! — Would it have made sense for someone to newly create an item along the lines of an obsolete version of that item, when the up-to-date version is abundant at that time and place? No. A writer being less than completely detailed, or a school of modern readers misinterpreting the text, is a more likely hypothesis than the alternative, that ancient people were stupid.”
And we know that patterns were quite often carried across into other materials. Close to this time, the Romans take a Celtic helmet shape made in iron, and adopt it but make it in bronze for a period before they start making it in iron. This doesn’t make it necessary that they would analogize patterns, of course. The most interesting question here is economics: how much leather (of all species of domestic animal) did they have, and what else did they use it for?
When it comes to making armour from it, thickness matters a _lot_. Sheepskin, goatskin, pigskin are all generally pretty thin. Even with cows, only a small proportion of the overall hide is thick enough to make useful armour.
I’m not aware of any culture anywhere in the world which has used thinner leathers like this as a protective element of armour (they get used as facing layers on fabric garments, or as foundation layers that protective elements are attached to, but those are different problems). It’s always thick leather from cows or something bigger (e.g. buffalo).
And again, regardless of feasibility we have an absolute lack of evidence. If we want to presume this was a common way to armour poorer Romans, we need a better justification than “well it makes sense to me”. There’s no textual evidence, no artistic evidence, no archaeological evidence.
Most ancient soldiers had no body armour whatsoever. A passage in Plutarch’s Lives assumes about one breastplate per ten Greek infantry and the ratios of shields to breastplates in Archaic sanctuaries suggests that there were several fancy shields with bronze parts for every breastplate (most men were lucky to afford a shield with some iron nails in it, as late as the Viking Age most Slavic shields seem to have had wooden bosses). If you don’t count the token heart-protectors of Polybius’ Romans, the first and only army I know where most soldiers had iron or bronze body armour was the army of the Roman Principate. The assumption that they must have worn something is a fundamental error of empathy like assuming that the powers can’t have sent their men to war in 1914 in red trousers and spiky leather hats.
You don’t need to empathize with your soldiers to think that it’s better for the effectiveness of the army that their chance of being killed is as low as possible, and it certainly doesn’t need to be a matter of empathy in scenarios where the soldiers provide their own equipment and can be assumed to have a vested interest in making it as hard to kill them as possible. That’s apart from social contexts in which the leader of a war doesn’t have sufficient command authority to just expect people to turn up, and there need to be positive incentives for the soldiers.
I’m curious if your assertion of most soldiers having no armour at all is predicated specifically on metal, and does not account for how textile armour is certainly not useless (properly made it can be quite effective), and definitely better than nothing.
Textile armour is cheaper than metal, but it’s still not cheap. The series on fabric production outlines pretty clearly just how much work goes into making a suit of clothing for everyone in the family per year. Making a piece of armour with twenty or more layers of fabric would be a very substantial amount of extra work on top of that. It’s certainly not impossible – obviously they became a fairly common element of armour in medieval Europe – but it’s a hell of a lot more expensive than a shield.
Isator, it appears that a minority of warriors in some ancient cultures had linen or rawhide or felt armour. A smaller minority in some cultures had bronze breastplates, disc cuirasses, coats of mail, or coats of scales. If you think that most soldiers in the Macedonian phalanx had linen or rawhide body armour AND bronze helmets AND bronze greaves AND shields with some bronze parts (and there is pretty good evidence for that supposition), that is lavish by ancient standards! And behind that phalanx were probably about as many men who could put together a goatskin shield, two spears, and a sword or a helmet and who the kings stopped calling out because the phalanx was so much more useful.
This post does not seem to be maintaining consistent standards. It responds to a point of “most soldiers had no armour” to “the kind of armour of this particular group of soldiers was quite expensive” to “an army would reduce itself down to the number of warriors who would have the good armour”. Was the implication intended by the first comment a suggestion that armies would stop calling upon most warriors in their society?
Setting aside that the comment of “a Macedonian style army would stop calling upon anybody besides those who could form a sarissa phalanx” seems quite unfamiliar with how Hellenestic armies fought. The usefulness of a phalanx drops precipitously without supporting medium and light infantry and decisive cavalry.
Does anyone have a good picture of what the anatomical version looked like? The article has some illustrations of the three disc version, but I’m having trouble imagining the other style.
[Something like this.](https://www.reddit.com/media?url=https%3A%2F%2Fi.redd.it%2F7w27ecipkt6g1.jpeg)
There is an original in the Royal Armouries collection in the UK: https://royalarmouries.org/collection/object/object-9535
Why didn’t they just use solid iron for armor?
They had iron weapons by then, so why bother with bronze plate or mail? Hammering a piece of iron into a vague torso-shape seems like would be both better and less labor intensive than either, so why did it take about a millennium after this for them to become common?
Metal was expensive. The bronze armor is being handed down because it was made in the past and is still available, so it’s not costing a fortune to make from raw materials in the present. Even bronze armor is still pretty valuable, but it’s cheaper in the sense of it not requiring more work right now.
Like seriously, any metal armor is a big expense in a society that is mostly agrarian and it requires surplus collected from hundreds of farmers to support a specialist supply chain for non-agricultural products.
Also while steel was certainly superior to bronze, I’m not sure plain iron necessarily was – it would have depended on the method used to work it, FWIH.
Iron plate is a lot harder to do than bronze. I forget why now, but there are engineering concerns in play.
Armor needs to be really quite thin to avoid being unreasonably heavy – around 1mm. Getting an iron plate hammered to a consistent 1mm thickness is difficult. Possible, of course, but hard. Much easier in copper-alloys, so bronze persists for finicky shapes that either need to be cast (rams) or need to be really thin (armor, helmets). When the Romans begin to reliably produce sheet iron, they make the switch over to iron helmets (by that point, their armor is mostly mail, with some scale and segmented armor on the side).
Iron is a lot harder to work than bronze. (it gets complicated by different types of iron, etc.) for starters, bronze can be cast (easier), it’s also relatively soft and ductile so it can be hammered into shape fairly easily.
Basically if you need finicky shapes bronze is a lot easier to work with tahn iron at this stage. And that includes something like “Make a relatively thin sheet of metal”.
Making large iron plates is genuinely quite hard – solving this is one of the high medieval metallurgical advances that “unlocks” plate armour to really develop (and to come down in cost below mail).
I’d also add that the difference in mechanical properties between iron (and steel, because there is no clear divide between them) and bronze matter less for armour than for weapons. A key advantage of iron over bronze is that it can be made very hard, such that it can be sharpened to a fine edge that cuts/stabs better than, and lasts longer. But hardness is less desirable for armour, because hardness also means brittle; much better for armour to deform plastically and dissipate impact energy than shatter when struck. The forms of iron (in terms of carbon content, heat treatment, work hardening, etc…) that have the right properties for armour are only slightly better than bronze at being armour.
> The forms of iron (in terms of carbon content, heat treatment, work hardening, etc…) that have the right properties for armour are only slightly better than bronze at being armour.
This isn’t particularly true – heat treated medium carbon steel is substantially better. But it’s a lot harder to make.
Because copper-tin bronze is much better! Its far easier to work into thin curved three-dimensional shapes than low-grade iron from a bloomery furnace, and its harder and tougher and prettier. Ferrous metals only start to perform better than bronze when you get medium carbon contents and heat treatment, and in the first millennium BCE almost all nice simple edged tools like swords or axes were never quenched. Quenching a broad, thin sheet hammered into three dimensions without cracking it or distorting it or making it too brittle is even harder. This is why so many early iron helmets are made in several parts riveted or welded together, and why so many European, Indian, and Japanese cuirasses have a flatish front, a flatish back, and two sides strapped or riveted in place rather than enclosing the whole torso in just two pieces of steel.
That being said, a big problem with understanding Central Italian armour is that copper alloys are much more archaeologically visible than iron alloys. Its possible that there was more Greek, Thracian, and Italian iron armour than we know which was reduced to a few rusty stains or incophrehensible scraps in sanctuaries and burials. But bronze armour was Homeric and Greek armour was fairly clearly not designed with protection as the main goal: looking beautiful and shiny and imposing was at least as important. If they had wanted protection above all, it would not have been the bare minimum thickness to stop some thrusts. So wearing bronze (or gilt bronze!) was more desirable than newfangled iron!
Since armour thinness has already been accounted for as something necessary to prevent it from being too heavy, do you have a counterexample of people wearing thicker armour in the interests of placing the highest premium on protection?
There’s medieval jousting tournament armour, which just has to protect the rider against lance impact from straight ahead. Very thick and heavy, also very hot and restrictive, but an acceptable tradeoff for a piece of specialised sporting equipment.
In 16th C Europe it was still possible for heavy steel plate armour to provide some worthwhile protection against bullets. The usual battlefield tradeoff was helmet and torso armour only, giving up plate arm and leg protection because that would have immobilised the wearer.
Does it really need to be said outright that I was asking about armour intended for warfare rather than a sporting event?
Medieval and sixteenth-century armour tends to be thicker and stronger than ancient armour. Ancient armour sits at the bare minimum viable thickness against weak weapons, even the highly ornate armour for wealthy and powerful people. The medieval stuff is not annoying to wear, and notably, as it became common, shields became less universal.
To be fair to the Etruscans, the linothorax seems more protective than those early forms of pectoral.
Whatever style the Romans manufactured themselves, the pectoral armor of the other Italian cultures would have almost certainty have been looted by victorious Roman armies. Given how it would have been passed down and the sheer number of victories, that could have added up to a lot of armor over the years (even assuming the pectoral remained elite armor in non-Roman Italy).
My guess is that this has already been considered, but I do have a bit of a guess as to why all the helmets.
Basically, humans get more attached to helmets than to other pieces of armor- they feel more personal, more ‘me’. I even have a (very, *very* shaky, but checkable) bit of evidence. I heard a football coach (high school) describe an occasional issue, where a player would grow attached to a helmet when they started out and insist on the same one every year, even as they grew until it was no longer the correct size.
Between a stronger emotional connection and being much cheaper than a full set of armor, I can see helmets being popular to hang on to, even if the rest of the armor moves on to a new user.
So, y’know. To the extent that you’re willing to take clues on warriors a couple thousand years ago from modern high schoolers, that’s a potential explanation.
There’s an interesting bit because viking era scandinavia famously has a dearth of helmets. We have earlier ones (from the roman/migration era iron age) but basically nothing from the actual viking period. (whereas we do have armour, weapons, etc.)
Plural of anecdotes is not data…but I’ll share we see the same thing in the construction industry. Gloves, vest, glasses, harness etc are all disposable by the guys, but hardhats get customized and carried around. To the point safety has to forcibly retire them due to age and sun damage.
Personalized decoration of military helmets seems to begin in the mythological age and persist in fighter pilots. A helmet may be decorated in some cultures to make the individual recognizable, which would discourage passing the helmet on. It is also likely for a helmet to be the first or only part of the armor to be fitted to the wearer, and re-fitting armor without ruining it gets harder each time (except for chain mail).
“Rawson, E. “The Literary Sources for the Pre-Marian Army.” PBSR 39 (1971)”
Congrats, Dr Devereaux: I read that and immediately thought “ooooh, that’s a periodization that’s been superceded in more recent scholarship.” This blog apparently is successfully teaching me things, not just telling me things that drain right out of my head immediately afterwards.
Huzzah!
When it comes to the profusion of helmets left behind in Roman camps, I really have to wonder if that’s a sign of them being relatively difficult to repair (beyond replacing cheekplates) compared to body armor, to the point of not being worth repairing and becoming more disposable by comparison. After all, this is thin metal in a roughly hemispherical shape that needs to have dents hammered out, but you wouldn’t want the steel to become too work-hardened (or else the hammered dent becomes a brittle point of weakness and doesn’t flex under impact to help the inner lining spread the force out), so you’re looking at more heats than putting replacement rings on mail and potentially additional heat-treating on top of that…
“and we get a lot of helmets (for reasons not entirely clear to me)”
Obelix was collecting them.
I was going to say, this was obviously because of the Gauls.
Many times when you see a Roman legionnaire sent flying, their helmet gets knocked off, and they presumably don’t often bother to pick them back up given that there are a bunch of dangerous Gauls in the neighborhood.
For the same reason, you might see a lot of caligae as well, except that those are mostly made out of leather, and so would quickly decompose.
You mention artists bias for high status arms.
Could the same artist bias have made single disc pectorals invisible?
That is, when in 6th century, new fancy armour types appear in Italy – Greek type armour in Tuscany, multidisc armour with hinged plates in rest of Italy – the old single disc armour is no longer visible: it is not buried in graves, it is not deposited in gifts to gods, it is not depicted on grave monuments and it is also not depicted in pottery paintings.
Does it mean that it does not exist?
But in 3rd century BC, no Italian armour is visible in the above manners (which depend on the contemporaries regarding it prestigious to give it away or depict). And from late 6th to 4th century, while Italians elsewhere left evidence of their armour, the Romans of Rome on left bank of Rome seem to have not done so. Yet the Romans of late 3rd century BC are clearly attested to have had armour even though they left no evidence.
Is it possible that from 6th century BC, even as the rich Italians developed the multidisc armour with bronze hinges and preserved it by burial and art, the poorer Italians continued to use the single disc armour that was no longer fashionable for their officers – and were invisible: left their armour as inheritance rather than bury it or donate to gods, could not afford to commission artworks etc.? Does this dependence of preservation on prestige mean that item types that were no longer prestigious could remain in use for centuries while invisible?
Including simple bronze plates supported by leather/rawhide straps, and maybe for the even poorer a breast covering of rawhide with no metal parts as Varro says? The latter would rot even if it were buried.
Also about comparison with Kalkriese…
There is actually a nice waterlogged site for the precise Middle Republic where the prestigious intentional deposits and artwork are absent.
Trasimeno.
On 21st of June, 217 BC, Hannibal ambushed a huge Roman army on the north shore of Lake Trasimeno.
Most Romans fought and died on dry land. And Hannibal stripped their armour and did a good job distributing the armour to his men.
But we are explicitly told that many Romans did reach the shore and tried swimming the lake but drowned. And others waded only to neck deep, but Carthagineans on horseback waded after them and hit their heads.
How successfully could Hannibal have fished out the armour of Romans who drowned and sank in deep waters of Lake Trasimeno?
How many Romans who drowned on 21st of June 217 have been excavated recently? What armour are they wearing?
> You mention artists bias for high status arms.
Could the same artist bias have made single disc pectorals invisible?
I guess it could, but we also have physical evidence for the three disc type, but not for the single disc type.
Yes. My point is that our physical evidence ALSO depends on the prestige. Burying arms or depositing them as sacrifice to gods.
Whereas Roman republic, of 3rd century across Italy and 4th…5th century in Rome herself, did not generate this type of physical evidence – arms were neither buried with dead nor deposited as gifts to gods. Presumably they were passed down, sold or recycled for scrap metal. Even an actual Roman warrior grave looks like this:
https://en.wikipedia.org/wiki/Sarcophagus_of_Lucius_Cornelius_Scipio_Barbatus#/media/File:Sarcofago_di_scipione_barbato,_a_forma_di_altare_con_volute_ioniche,_in_marmo_peperino,_280_ac.,_dal_sepolcro_degli_scipioni_sull’appia_antica.jpg
His feats of arms are described only verbally:
“He captured Taurasia Cisauna in Samnium – he subdued all of Lucania and led off hostages.”
but the decorations of the grave are purely architectural, no arms shown and no arms inside the grave with his bones.
An option would be if the out-of-fashion arms continued in use as “humble” arms which were worth neither depiction in art nor appropriate for intentional depiction. The only context in which these might be preserved would then be unintentional and unscavengable battlefield deposits – like Lake Trasimeno. Have these been explored?
“worth neither depiction in art nor appropriate for intentional depiction.”
More correctly
“neither worth depiction in art nor appropriate for intentional deposition”.
Thank you, interesting post that tells us how little we sometimes know. And thank you for answering Joseph Jarosch’s question. I had been wondering about more or less the same thing.
Hm, now I wonder how filled with rage the statue pfp types might get at the suggestion that the glorious warrior cultures of Ancient Italy got some of their ideas on how to protect themselves in battle from the Middle East.
Oh, right!, the thing that actually caught my eye in this was the idea that mail armor was ideographic of Romans, when of course, these days we use some weird pauldron iconography on basically incoherent base armor.
” bronze plate a span on all sides.” -> “bronze plate a span on all sides.” (I can’t type the proper
as practices in archaeology -> as practiced in archaeology
I choose to believe that the disappearance of armour in the archaeological and artistic record as the Roman Republic expanded occured because the Romans didn’t use armour (helmets excepted; every man needs some shiny headwear). The Romans fought like the manly men they were, daring the Etruscans and Samnites and whohaveyou to try and stab their naked, muscular, glistening from the sweat of marching bodies, only to be stabbed in turn if they tried.
This solution is elegant because it doesn’t make any assumptions at all – it only works with the evidence we have, which is none, as far as armour is concerned. Thus we satisfy Occam’s razor.
And we all know that’s how the Spartans fought.
Maybe not as far as not wearing armor, but this is my wild theory too. Especially considering our host’ armor sequence article, maybe most landed roman (barely not poor) would prioritize helm and shield. And then they scavenge armor from everywhere else. So it becomes the opposite of uniform. Maybe Polybius is describing the only single thing similar for any of them.
The Ksour Essef Cuirass honestly looks pretty badass.
“We never to my knowledge ever see them suspected in simple leather harnesses”
->suspended
There have been hints in other posts that the key to Roman conquest was a really good recycling program, and I think that fits well here too. The Republic was most successful when it didn’t bury any metal, mail is no harder to make from captured scrap than from new ingots, and bronze armor was processed out of decorated forms into undecorated plates as it was being phased out.
My tentative reaction is that this is a reasonable statement of the problem. I would emphasize that even the minimal Polybian single plate for the breast is more armour than most ancient soldiers had, and a more complete and decorated assembly for breast and back seems far too much to expect every man in a mass army to provide himself with. But Polybius certainly does not describe the kardiophylax in detail. The Indian and Persian Four Mirrors armours and the 12th-14th century European plate (singular) for the breast were usually worn over linen armour or mail to cover the gaps.
One detail: “probably produced from sheet metal (sheet bronze, in particular), rather than forged from an ingot.” Obviously this is about 2000 years before rolling mills, so if you want sheet metal, you have a team of burly men with hammers beat it out of a bar or ingot. You might pour liquid bronze onto a flat stone to get started on the spreading out but it won’t come out a consistent thickness or with both sides nice and smooth.
Are you aware of David Sims’ work on likely methods for producing sheet metal (iron) in the imperial period?
Sim and Kaminski had no archaeological or textual evidence for their rolling mills and were focused on the more industrialized and technologically sophisticated Roman Principate. In general, approaching low-tech problems with ‘they must have had specialized tooling’ is a worse approach than ‘they must have had helpers and practiced a lot with a few simple tools and materials. ‘ As a very well known specialist in reproducing ancient armour put it:
Roman Imperial Armour: The Production of Early Imperial Military Armour overreaches in basing so many of its assertions on the limited experience of a hobbyist blacksmith. It would be presumptuous for a metal-detectorist to write a treatise on archaeological practice, but for some reason academics think that any experience at all qualifies them to extrapolate on ancient craft industries.
For context, the Middle Republic is when Italians are probably adopting the first crank in the world (on the Iberian rotary quern, invented sometime around the 5th century BCE). The world of Pompeii with many rotary donkey mills was centuries in the future. So even if you think that rolling mills might have been used for some purposes in the Roman Principate, postulating them as a common technology in 3rd/2nd century BCE Italy is a big ask.
The Capestrano warrior is wearing a sombrero?!?!
It’s probably a representation of a petasus.
People working in areas with warm, bright summers probably have a tendency to independently develop wide-brimmed hats.