> No written rules for this game survived antiquity. To reconstruct how the game may have been played, researchers turned to the Ludii General Game System — a comprehensive digital platform developed at Maastricht University that can model and simulate thousands of historic board games. The results were published in the journal Antiquity (Volume 100, Issue 409, 2025).
> Using Alpha-Beta search agents — the same class of algorithm that powered early chess computers — the team ran 1,000 simulated rounds for each candidate ruleset, allowing one second of processing time per move. The AI tracked which lines on the board were used most frequently during play, generating detailed edge-usage statistics....
> Nine game configurations matched the wear criteria. All of them were blocking games, and the most frequently matching format was a four-versus-two game in which pieces start on the board. This site faithfully reproduces one of these AI-validated configurations.
I would say this is more "inspired by" Ancient Rome.
This feels like the thing that makes me deeply skeptical of swaths of archaeology and palaeontology as a science rather than being a kind of fandom.
I imagine the incentives of having a crisp story for media consumption don’t help. I’d hope to read a lot more: “we’re missing the majority of the pieces to this puzzle. This represents our best guess given current evidence and methods.”
Archaeology is an unusual discipline in that it incorporates so many others as tools. Chemistry, physics, geology, CPSC, you name it. It's difficult enough to figure out what people were doing based on ruins and trash pits. It's harder still when there are so many disciplines involved that, each, introduce their own uncertainties.
That being said, "We asked an AI..." is a special kind of uncertainty that goes above and beyond anything else Archaeologists do.
--------
"No written rules for this game survived antiquity. To reconstruct how the game may have been played, researchers turned to the Ludii General Game System — a comprehensive digital platform developed at Maastricht University that can model and simulate thousands of historic board games. The results were published in the journal Antiquity (Volume 100, Issue 409, 2025).
Using Alpha-Beta search agents — the same class of algorithm that powered early chess computers — the team ran 1,000 simulated rounds for each candidate ruleset, allowing one second of processing time per move. The AI tracked which lines on the board were used most frequently during play, generating detailed edge-usage statistics.
These statistics were then compared to the physical wear patterns on Object 04433. To account for human cognitive biases — such as right-handed players preferring to play on the right side of the board — the researchers applied symmetry transformations to the simulation results, maximising consistency between AI-generated play and the actual marks left by ancient players.
Nine game configurations matched the wear criteria. All of them were blocking games, and the most frequently matching format was a four-versus-two game in which pieces start on the board. This site faithfully reproduces one of these AI-validated configurations."
--------
It's interesting that they considered use-wear on found pieces as input for their AI. Still, this study made a lot of assumptions. I wouldn't be surprised if a different team could use the same methods and come up with a completely different result.
I think the study would be more palatable if it was presented as an exploration of AI-aided methods and very strongly impressed that any one result was not the point.
The actual paper is written by archaeologists for archaeologists - just like all papers are written by people in a feild for people in a field.
Your complaint is like complaining that a CS paper doesn't even mention that P = NP is still unknown, and that it just assumes that the best sorting methods are O(n log n). Some things are considered general knowledge in a field, and in archaeology one of those things is: all of this is a best guess given current evidence and methods.
The whole paper is full of citations to other work, which is full of citations to other work - all this work linked includes detailed reports of what was found where and what-else was there, people who do statistical analyisis of similar findings, people who ask questions like "what explanations can there be for it?", more importantly "what else would we find if X was true, if Y was true?". When new evidence arrives, people can and do go back and re-examine these things. Your random skepticism and all the questions you may ask have already been asked and addressed, and frankly: these archaeologist's conclusions carry far far far more believable weight than your half-assed skepticism.
> No written rules for this game survived antiquity… a comprehensive digital platform… that can model and simulate thousands of historic board games.
Model and simulate based on what?
> Nine game configurations matched the wear criteria.
So their idea was to generate candidate rulesets, have AI try to figure out rational play, then see which pieces would be moved most often and match that to the forensic evidence?
This seems impossible. If you started with chess pieces and a chessboard or even a go board and go pieces, it is absolutely impossible to reconstruct the game as they’re currently played.
Similar in that it is asymmetrical, very old, and we don't quite know what the rules were (though with tafl we have slightly better historical evidence, including an account from Carl Linnaeus in 1732, which has allowed us to produce a few educated guesses). In fact tafl is sometimes speculated to derive from the Roman game of ludus latrunculorum - I'm not sure if that is the same game described here.
I was thinking of them yesterday because I noticed one for sale in the antiquities shop in Andor, which brought up all sorts of Earth/Rome/Star Wars cannon questions.
Since the rules are entirely reconstructed, I have to say I'm a bit skeptical of a game where a central point is keeping a count of moves up to 150. It seems too unpractical for a casual game.
Are there other known ancient games that work like this?
I won second try with hares on medium, I think the trick is to:
A) make sure you get at least one hare to one of the central two spots (will always be possible) and camp it there. Keep your other hare moving around on your side.
B) the opponent will have to bring two hounds over to catch your other hare. To do this they will have to create a gap between the first and second hound in the middle spot above/below your camped hare. As soon as this happens, move your camped hare upwards.
It's pretty easy after that TBH. The AI has usual burnt a load of moves by that point and you have so many options from that position
The symmetry is restored if you say you're playing a two game match and the winner is the one that survives longer as the hares. But the game seems simple enough that I imagine it shouldn't be too hard to play optimally.
Maybe the concept of desiring balance is modern. Maybe they wanted to encode asymmetry in the game to mirror nature, or the Master / Slave dynamic which they likely considered to be a natural law.
The rules seem like a rather boring pre-decessor to the mills game/nine men's morris. I guess this is what I would draw inspiration from though, to make the game more interesting, because the board is actually kind of neat.
I always like to hear about ancient board game reconstructions. Like music and religion, games are something every culture creates. Another recent example is the case of Liubu, a game from ancient China for which the rules were lost. Still, a reconstruction is being attempted by Carnegie Melon:
The AI could have constructed any number of rules seeing as how it's literally a "best guess" and it chose rules that aren't very fun. Congratulations, I guess.
Edit: Or, put a different way: Sometimes the rules to games are lost for a reason; they're not very good.
The videos on the game (and all his other videos) with Irving Finkel, a curator at the British museum, are spellbinding. He has the looks, manners and enthusiasm of an eccentric museum curator from central casting!
> A partial description in cuneiform of the rules of the Game of Ur as played in the second century BC has been preserved on a Babylonian clay tablet written by the scribe Itti-Marduk-balāṭu.
This is much more palatable and cool since they’re not just randomly guessing what the game can be like this article is
35 comments:
> No written rules for this game survived antiquity. To reconstruct how the game may have been played, researchers turned to the Ludii General Game System — a comprehensive digital platform developed at Maastricht University that can model and simulate thousands of historic board games. The results were published in the journal Antiquity (Volume 100, Issue 409, 2025).
> Using Alpha-Beta search agents — the same class of algorithm that powered early chess computers — the team ran 1,000 simulated rounds for each candidate ruleset, allowing one second of processing time per move. The AI tracked which lines on the board were used most frequently during play, generating detailed edge-usage statistics....
> Nine game configurations matched the wear criteria. All of them were blocking games, and the most frequently matching format was a four-versus-two game in which pieces start on the board. This site faithfully reproduces one of these AI-validated configurations.
I would say this is more "inspired by" Ancient Rome.
This feels like the thing that makes me deeply skeptical of swaths of archaeology and palaeontology as a science rather than being a kind of fandom.
I imagine the incentives of having a crisp story for media consumption don’t help. I’d hope to read a lot more: “we’re missing the majority of the pieces to this puzzle. This represents our best guess given current evidence and methods.”
Archaeology is an unusual discipline in that it incorporates so many others as tools. Chemistry, physics, geology, CPSC, you name it. It's difficult enough to figure out what people were doing based on ruins and trash pits. It's harder still when there are so many disciplines involved that, each, introduce their own uncertainties.
That being said, "We asked an AI..." is a special kind of uncertainty that goes above and beyond anything else Archaeologists do.
--------
"No written rules for this game survived antiquity. To reconstruct how the game may have been played, researchers turned to the Ludii General Game System — a comprehensive digital platform developed at Maastricht University that can model and simulate thousands of historic board games. The results were published in the journal Antiquity (Volume 100, Issue 409, 2025).
Using Alpha-Beta search agents — the same class of algorithm that powered early chess computers — the team ran 1,000 simulated rounds for each candidate ruleset, allowing one second of processing time per move. The AI tracked which lines on the board were used most frequently during play, generating detailed edge-usage statistics.
These statistics were then compared to the physical wear patterns on Object 04433. To account for human cognitive biases — such as right-handed players preferring to play on the right side of the board — the researchers applied symmetry transformations to the simulation results, maximising consistency between AI-generated play and the actual marks left by ancient players.
Nine game configurations matched the wear criteria. All of them were blocking games, and the most frequently matching format was a four-versus-two game in which pieces start on the board. This site faithfully reproduces one of these AI-validated configurations."
--------
It's interesting that they considered use-wear on found pieces as input for their AI. Still, this study made a lot of assumptions. I wouldn't be surprised if a different team could use the same methods and come up with a completely different result.
I think the study would be more palatable if it was presented as an exploration of AI-aided methods and very strongly impressed that any one result was not the point.
I have often wondered how to make this clear wrt science communication
Eg much is not known about dinosaurs. Many things cannot be found in fossils (obligatory xkcd: https://xkcd.com/1747/)
How do you communicate what is unknown or what can't be known?
My favorite dinosaur theory is they looked far more like a plump turkey than a Komodo dragon.
We tend to just drape a bounding skin volume over the bones and give it a color, MAYBE some feathers.
But a whale skeleton looks nothing like the mass of flesh of a whale. Skeletons give no clue about an Elephant’s head shape.
So imagine a huge fluffy owl or yellow chick with a trex skeleton deep inside.
The actual paper is written by archaeologists for archaeologists - just like all papers are written by people in a feild for people in a field.
Your complaint is like complaining that a CS paper doesn't even mention that P = NP is still unknown, and that it just assumes that the best sorting methods are O(n log n). Some things are considered general knowledge in a field, and in archaeology one of those things is: all of this is a best guess given current evidence and methods.
The whole paper is full of citations to other work, which is full of citations to other work - all this work linked includes detailed reports of what was found where and what-else was there, people who do statistical analyisis of similar findings, people who ask questions like "what explanations can there be for it?", more importantly "what else would we find if X was true, if Y was true?". When new evidence arrives, people can and do go back and re-examine these things. Your random skepticism and all the questions you may ask have already been asked and addressed, and frankly: these archaeologist's conclusions carry far far far more believable weight than your half-assed skepticism.
> No written rules for this game survived antiquity… a comprehensive digital platform… that can model and simulate thousands of historic board games.
Model and simulate based on what?
> Nine game configurations matched the wear criteria.
So their idea was to generate candidate rulesets, have AI try to figure out rational play, then see which pieces would be moved most often and match that to the forensic evidence?
This seems impossible. If you started with chess pieces and a chessboard or even a go board and go pieces, it is absolutely impossible to reconstruct the game as they’re currently played.
Somewhat similar to Tafl games: https://en.wikipedia.org/wiki/Tafl_games
Similar in that it is asymmetrical, very old, and we don't quite know what the rules were (though with tafl we have slightly better historical evidence, including an account from Carl Linnaeus in 1732, which has allowed us to produce a few educated guesses). In fact tafl is sometimes speculated to derive from the Roman game of ludus latrunculorum - I'm not sure if that is the same game described here.
Reminds me of the mystery of the Roman dodecahedrons - https://en.wikipedia.org/wiki/Roman_dodecahedron
I was thinking of them yesterday because I noticed one for sale in the antiquities shop in Andor, which brought up all sorts of Earth/Rome/Star Wars cannon questions.
This reminds me of bear games: https://en.wikipedia.org/wiki/Bear_games
Here's a tablebase analysis for a simple bear game I constructed a while back: https://emarzion.github.io/coqtbgen/
Since the rules are entirely reconstructed, I have to say I'm a bit skeptical of a game where a central point is keeping a count of moves up to 150. It seems too unpractical for a casual game.
Are there other known ancient games that work like this?
Too assymetric IMO. I have no problems winning with hounds even on hard, but I have no chance of winning except against easy.
I won second try with hares on medium, I think the trick is to:
A) make sure you get at least one hare to one of the central two spots (will always be possible) and camp it there. Keep your other hare moving around on your side.
B) the opponent will have to bring two hounds over to catch your other hare. To do this they will have to create a gap between the first and second hound in the middle spot above/below your camped hare. As soon as this happens, move your camped hare upwards.
It's pretty easy after that TBH. The AI has usual burnt a load of moves by that point and you have so many options from that position
I guess you have to put it in context. There are old tic-tac-toe boards also sculpted in stone (https://legioxiiisl.blogspot.com/2013/02/roman-board-game-te...), which is a very bad game. Anything would be better.
Maybe they were playing Gobblet Gobblers instead which is a much more fun game on the same size board.
The symmetry is restored if you say you're playing a two game match and the winner is the one that survives longer as the hares. But the game seems simple enough that I imagine it shouldn't be too hard to play optimally.
Ancient example of an unbalanced meta. Maybe we are missing some of the patch notes for balance changes
Maybe the concept of desiring balance is modern. Maybe they wanted to encode asymmetry in the game to mirror nature, or the Master / Slave dynamic which they likely considered to be a natural law.
The Netrunner card game is very asymmetric and very modern. I guess you could say it's balanced but nothing asymmetric can be perfectly balanced.
Every casino game is asymmetric. I believe the casinos are trying to encode the Charlatan/Chump dynamic.
Or they just switched sides and played again, which is less silly.
The rules seem like a rather boring pre-decessor to the mills game/nine men's morris. I guess this is what I would draw inspiration from though, to make the game more interesting, because the board is actually kind of neat.
Neat, but this game seems too simple. I think AI has missed a core gameplay mechanic.
There needs to be something to stop a deadlock. Like an element of luck for the hares?
I always like to hear about ancient board game reconstructions. Like music and religion, games are something every culture creates. Another recent example is the case of Liubu, a game from ancient China for which the rules were lost. Still, a reconstruction is being attempted by Carnegie Melon:
https://projects.etc.cmu.edu/liubo-lab/
It should be "novus ludus" not "nova ludus". (now go and write 100 times...)
The AI could have constructed any number of rules seeing as how it's literally a "best guess" and it chose rules that aren't very fun. Congratulations, I guess.
Edit: Or, put a different way: Sometimes the rules to games are lost for a reason; they're not very good.
I was curious to learn more about how they discovered and reconstructed the rules. They explain that at: https://ludus-coriovalli.web.app/about
Another ancient board game to be decoded is https://en.wikipedia.org/wiki/Royal_Game_of_Ur. This is from 2600BC so is older to Christ than Christ is to us.
The videos on the game (and all his other videos) with Irving Finkel, a curator at the British museum, are spellbinding. He has the looks, manners and enthusiasm of an eccentric museum curator from central casting!
There's an online version to play that I find surprisingly addictive:
https://royalur.net/
You mean Irving Finkel: https://en.wikipedia.org/wiki/Irving_Finkel
> A partial description in cuneiform of the rules of the Game of Ur as played in the second century BC has been preserved on a Babylonian clay tablet written by the scribe Itti-Marduk-balāṭu.
This is much more palatable and cool since they’re not just randomly guessing what the game can be like this article is
Playing a lost 2000 year old game is awesome.
Cool however no idea how to play this!