this post was submitted on 08 Feb 2026
52 points (100.0% liked)

HistoryArtifacts

770 readers
179 users here now

Just a community for everyone to share artifacts, reconstructions, or replicas for the historically-inclined to admire!

Generally, an artifact should be 100+ years old, but this is a flexible requirement if you find something rare and suitably linked to an era of history, not a strict rule. Anything over 100 is fair game regardless of rarity.

OTHER COMMS IN THE HISTORYVERSE:

founded 9 months ago
MODERATORS
Deceptichum@quokk.au
PugJesus@piefed.social
52
Roman Dodecahedron (lemmy.world)
submitted 1 month ago by FauxPseudo@lemmy.world to c/historyartifacts@piefed.social
22 comments fedilink hide all child comments
 

Rules say reconstructions are allowed but doesn't specify if illustration counts.

--

Those are objects from the Roman Empire. About 130 have been found in total.

They are typically made of brass and fit in the hand but can vary from 35 grams to a kilo. Each side has holes of different sizes with rings around them. On each corner you will find a sphere protruding.

But there are three things you won't find.

  1. you won't find them in or near the capital. They are found all over the European part of the empire including Britain, but never once in the Near East or African portions of the empire and also never in what we call Italy today.
  2. you won't find why they exist. There is no known use for them that survives scrutiny. But you will find lots of speculation.
  3. you won't find any markings. No labeles, no symbols, no gauges, or numbers. Just holes with rings.

Are they part of a mystery colt, some rather expensive game piece, blacksmith training? No one knows.

Find a 3D print file and make one yourself and ponder it for a while. Or draw one. I haven't found any woodworking plans. Maybe I could fix that.

Micron, A5.

top 22 comments
sorted by: hot top controversial new old
[–] lvxferre@mander.xyz 7 points 1 month ago* (last edited 1 month ago) (1 children)

So, here's a map of the distribution of those dodecahedra:

Wikipedia claims they're dated from the 2nd to the 4th centuries CE, the main body was casted in bronze, and the knobs were soldered afterwards. It also mentions that "The archaeological context in which the dodecahedra have been found is often uncertain. Where it is known, the context is commonly military or funerary; other discoveries have been in baths, a theatre, a coin hoard, and on a riverbed." And the holes of each dodecahedron are different from all the others.

Parsing all this info together leads me to believe that:

  1. Military in nature or purpose. This explains why they're found in a military context, and their distribution in the map; I think that cluster is in Germania Superior.
  2. Individual in ownership and usage. As in, a dodecahedron wouldn't be shared across multiple people; it would belong to someone in specific. This might explain why they were found in a coin hoard, and in funerary contexts.
  3. Relatively important for the owner. Ditto as above.
  4. Something only elites were expected to have. Bronze wasn't exactly cheap; and if those objects were more popular, we would've found a lot more of them.
  5. Roman, not Celtic. I'd believe them to be Celtic if they were either way more local (say, all of them were found exclusively in Aquitania or in Britannia), or if they were considerably older. But given Imperial times I'm fairly certain the ones spreading them were the Romans.

Based on those, the most reasonable explanation I've seen is that they were used with wooden cipher discs. Here's a video explaining the hypothesis in details; I find it way more convincing than the alternative ones (knitting, candle holder, religious significance, etc.)

[–] FauxPseudo@lemmy.world 4 points 1 month ago (2 children)

Excellent breakdown.

Any ideas on if it was used for ciphering that they wouldn't be located in/near the capital too? Or why they are completely absent from Africa or the East?

The problem I have with them being used for ciphers is that it would require each person using it for that to possess one of the same size. But there is a 30x difference in size from the smallest to the largest.

[–] lvxferre@mander.xyz 3 points 1 month ago

I think the biggest appeal of the dodecahedron was that it allowed a person to keep track of a dozen ciphers at the same time. That would be only useful in a region where there was constant conflict with outsiders (unlike Africa or Italy), and where there was a good chance of the enemy knowing some basic Latin (unlike in the East).

The problem I have with them being used for ciphers is that it would require each person using it for that to possess one of the same size. But there is a 30x difference in size from the smallest to the largest.

If the hypothesis from the video is accurate, you'd only need compatibility between dodecahedra of two people if they were in communication with each other; for example, between a legatus and his immediate subordinates. And even then, you don't need both dodecahedra to be identical, only a single side being identical is enough.

[–] JohnnyEnzyme@piefed.social 2 points 1 month ago (2 children)

...that they wouldn’t be located in/near the capital too? Or why they are completely absent from Africa or the East?

I watched a recent video hypothesis on these, based on latest findings, and it made the case that these dodecahedra were quite likely specifically suppressed from all Roman records, being... Celtic or Celtic-Germanic in origin, hence their geographical distribution. Also, it seems they were most commonly found in grave and funerary sites, and furthermore, many were coated internally with human bone dust, further making a case on that front.

Actually, here's the video: (glad I saved it)
https://www.youtube.com/watch?v=OIt6oLFoH64

I forget if the video covered it, but I guess there are a couple hangups with the above: 1) was there really much of note that would have been shared by both Germans and Celts of this period? 2) why were these (evidently) only found where Romans had a mixing / fringe presence, and not deeper in to Celt / Germanic territory?

@lvxferre@mander.xyz @PugJesus@piefed.social

[–] lvxferre@mander.xyz 2 points 1 month ago (1 children)
  1. was there really much of note that would have been shared by both Germans and Celts of this period?

By those times I don't think even the Celts shared a lot with each other, besides the "core" of their religion and speaking related languages. And they often merged with local peoples, and the influence of those in their culture was a big deal, I don't expect e.g. some Celtiberian (nowadays a chunk of Spain) to share a lot of their culture with someone from Noricum (nowadays Austria).

That's one of the things that make me specially sceptic on the idea those were Celtic religious items — which Celts would be they from? And why were other Celts using those too?

  1. why were these (evidently) only found where Romans had a mixing / fringe presence, and not deeper in to Celt / Germanic territory?

If my hypothesis (Roman ciphering devices) is correct, they were only found in those regions because they're where the Romans were in constant conflict with other peoples, and there was a need for the Roman military to "juggle" dozens of ciphers. So you wouldn't see them in "core" areas of the empire (like Italia, Hispania or Africa), but you wouldn't see them in areas with no Roman activity whatsoever.

@FauxPseudo@lemmy.world mentioned the East. Due to the Roman-Sassanid wars, there would be grounds to use those there too, but I think in practice they were not necessary, as the odds the enemy spoke some basic Latin were smaller.

Plus it just clicked me another relevant detail: those ciphers were specially common in Gaulish-speaking areas, and Gaulish and Latin grammars resemble each other quite a bit. To the point I wouldn't be surprised if Gauls were able to actually decode plaintext Latin, or Latin using a simple cipher; developing a more complex system of ciphers seems like a sensible answer to that.

[–] JohnnyEnzyme@piefed.social 1 points 1 month ago (1 children)

By those times I don’t think even the Celts shared a lot with each other, besides the “core” of their religion and speaking related languages.

Right. My layman's understanding is that "Celtic culture" had some very basic commonalities (language, I think?), but was in fact incredibly diverse and not nearly as uniform as we moderns sometimes assume.

ciphering device

I'm still having a hard time understanding how that would work given the size of the holes varying with almost every piece. Also, why would so many contain human bone dust, and be found in graves and funerary areas?

Let me know if you happen to catch the video. It's pretty common for such things to be a lot more wildly speculative than they let on, but that one seemed pretty reasonable... to me, a layman. 😅

[–] lvxferre@mander.xyz 1 points 1 month ago* (last edited 1 month ago) (1 children)

I think the presence of human ashes can be explained through Roman funerary practices.

The Romans sometimes buried their dead with consumable and belongings, and mixed/smeared the ashes of the dead on them. Here's an example of that: in 2024 they found a burial urn in Carmona (Spain), from Roman times. The urn contained white wine mixed with human ashes, and there was a gold ring in it.

So, if the dodecahedra were personal items, and they signalled a high military officer, it made a lot of sense to bury them alongside their owners. The ashes, thyme, frankincense etc. were likely there because of that. And, if they were "encryption devices", this also symbolised that the secret was being buried alongside the dead.

That's just my hypothesis, mind you. It makes sense for me, but it's possible it's completely bollocks. Take it with a grain of salt.

I’m still having a hard time understanding how that would work given the size of the holes varying with almost every piece.

If they were used exactly like in the video I linked, the dodecahedra of the sender and receiver wouldn't need to be completely identical; only a single pair of faces (and the associated discs).

It's also possible one of the participants didn't even use a dodecahedron. The thing wouldn't be required to encode/decode text; it's only a convenient way to keep track of which cipher to use for which message, you know?

For example. Let's say you got a high officer and four subordinates. The subordinates barely contact each other, so they don't need ciphered text between themselves; but they all need to contact the higher officer, so each would use their own cipher.

For the subordinates it isn't a big deal; it's just a single cipher. They could write it down, memorise it, or even use a pair of discs (with no dodecahedron). But the high officer would need to keep track of four ciphers; then the dodecahedron becomes useful.

[–] JohnnyEnzyme@piefed.social 2 points 1 month ago (1 children)

One thing that seems to throw a 'spanner in to the works' of the idea of these objects being Roman, is that evidently there's little if any record of dodecahedra in all of recorded Roman documents, artifacts, histories, and whatever.

Also, there's the fact that they're seemingly *not* found in traditional Roman-held areas, but moreso in outer, fringe areas in which the Romans and Celto-Germans had some overlap.

At least, that's my understanding so far.

[–] lvxferre@mander.xyz 2 points 1 month ago (1 children)

The hypothesis they're military cripto stuff explains neatly both things: they'd be useless in traditional Roman areas (there was military in those, just not fighting "outsiders"), and the military probably didn't want too many people to know how to use them*.

If they were a Celtic development I'd expect them to be way more local. For example, most of them were found in France, right? If they were Gaulish you'd find some in Belgium and England, but you wouldn't be finding them in Austria or Croatia.

*note: if that's correct, it wouldn't be the first case of us losing info about the Romans because they silenced it themselves. Or the worst. The worst are the 28 books written by emperor Claudius, on the Etruscans and the Carthaginians.

[–] JohnnyEnzyme@piefed.social 1 points 1 month ago (1 children)

I agree with some of that, and in terms of other stuff... I don't have the slightest idea. I'm curious about your 'cipher theory,' in any case. I'm looking forward to seeing how the research and theory goes...

Thank you, my fellow ape, for sparing me your time upon these matters!

[–] lvxferre@mander.xyz 1 points 1 month ago

The theory isn't even "mine", to be honest. And although it's the one I find the most likely to be true, I low-key want it to be false (and the true explanation to make me go "...duh, this is so obvious, why didn't I think about this???").

And you're welcome — it's nice to chat with you!

[–] PugJesus@piefed.social 2 points 1 month ago (1 children)

: 1) was there really much of note that would have been shared by both Germans and Celts of this period?

Oh, yes, certainly. One of the things often noted in the modern day about Caesar's Commentarii is that he draws a much 'stronger' line between the peoples than actually existed. He might have wanted to contrast the "Noble, civilizable" Celts with the "Barbaric, dangerous" Germanics, or he simply might have been calling matters as he saw them through the ethnographic lens of an ancient Roman. Like most regions before concentrated state institutions, it's not really a boolean "Celt/Germanic" so much as it is a gradient, with a lot of outliers. Not only that, but "German" is sometimes thought to be a Celtic word meaning, roughly, "neighbor" - the two cultures were geographically close and interacted on a regular basis.

  1. why were these (evidently) only found where Romans had a mixing / fringe presence, and not deeper in to Celt / Germanic territory?

No clue, I'm afraid!

[–] JohnnyEnzyme@piefed.social 2 points 1 month ago

...it’s not really a boolean “Celt/Germanic” so much as it is a gradient, with a lot of outliers.

Thanks for explaining... makes a lot of sense!

[–] Kolanaki@pawb.social 6 points 1 month ago* (last edited 1 month ago)

After reading a lot about these things a few years ago once made aware of them, I firmly believe these were basically just decorative doodads like one might keep on their desk or a shelf today. Why should ancient people not have useless bullshit, too?

[–] voxthefox@lemmy.blahaj.zone 5 points 1 month ago (1 children)

I thought they figured out these were some kind of tool for weaving/sewing

[–] FauxPseudo@lemmy.world 10 points 1 month ago

That doesn't check out. Too rare for such a common activity, doesn't explain the rings around the holes, why they are never found in the company of weaving or sewing support, why never found with women, why they aren't found in Italy where weaving and sewing is done on grand scales, etc.

[–] Depress_Mode@lemmy.world 4 points 1 month ago (1 children)

Dick girth measuring gauge

[–] FauxPseudo@lemmy.world 2 points 1 month ago (1 children)

The average Roman must have been tiny.

[–] Depress_Mode@lemmy.world 3 points 1 month ago* (last edited 1 month ago)

Well, that would be keeping with Roman male beauty standards. If your dick doesn't fit in any of the holes, you're an inelegant barbarian. Only the classiest of microdicks would have made it into the smallest hole.

[–] leriotdelac@lemmy.zip 2 points 1 month ago

Wow, this is incredibly interesting!

[–] Etterra@discuss.online 1 points 1 month ago (1 children)

My personal theory is that they were an art fad.

[–] FauxPseudo@lemmy.world 3 points 1 month ago

I hesitate to call anything lasting centuries a fad. When you and your grandfather 10 generations back are into the same thing that's tradition, not a fad.