Computational Approaches for Recognising and Reconstructing Ancient Games: The Case of Ludus Latrunculorum

The study of games in the ancient world has long been hindered by the loss of rules because they were rarely written down. The Digital Ludeme Project aims to apply computational methods to the available evidence for games throughout history to expand the scope of how games can be studied. This involves a two-pronged approach based on documenting evidence of known game rules in specific times and places, and simulation of candidate rule sets borrowed from similar games to calculate gameplay metrics. As an example, this methodology is applied to the Roman game Ludus Latrunculorum, which involves the relatively uncommon custodial capture rule. Adapting documented rule sets featuring this rule and applying them to known Roman era boards, various game metrics are examined to engage with how computational methods can shed light on potential rule differences or preference in game experience. Finally, future applications and improvements of our methodology are discussed.