A peer-reviewed study published in American Antiquity has established that Native American hunter-gatherers were crafting and using dice more than 12,000 years ago, thousands of years before any comparable practice appeared in Europe or Asia. The research, which analyzed more than 600 archaeological sets of two-sided dice from sites across North America, rewrites the timeline of human engagement with probability and structured games of chance. The finding challenges a long-standing assumption that gambling and mathematical reasoning about randomness originated in the Old World.
Bone Dice and Binary Lots in the Pleistocene
The study, authored by archaeologist John L. Madden and released through American Antiquity, centers on what researchers call “binary lots,” or two-sided dice. These are flat objects, often made from bone, antler, or split cane, with distinct markings or textures on each face. A player tosses a handful and scores based on how many land on a given side. The game mechanic is simple, but the underlying engagement with probability is real: players repeatedly confront random outcomes governed by fixed odds.
Madden developed an objective morphological test to distinguish genuine dice from similarly shaped non-gaming artifacts. He built that test from ethnographic reference sets, then applied it systematically to the published archaeological record. The result was the identification of more than 600 sets spanning thousands of years and dozens of regions. Some of the oldest specimens date to the late Pleistocene, placing their creation around 10,000 BCE or earlier. That timeline puts Indigenous dice-making roughly 6,000 years ahead of the earliest known dice use in the Old World, according to reporting on the study.
A Century-Old Ethnographic Record as Scientific Baseline
Madden did not build his reference framework from scratch. His morphological test draws heavily on a foundational 1903 compilation by Stewart Culin, published in the Twenty-fourth Annual Report of the Bureau of American Ethnology. That 806-page volume documented games and gaming implements from Indigenous groups across the continent, with extensive illustrations and descriptions of two-sided dice. Culin’s illustrated corpus alone contains 293 unique dice sets, drawn from dozens of tribal nations.
By grounding his identification criteria in Culin’s well-documented ethnographic sample, Madden created a bridge between the historical record and the deep archaeological past. If an excavated artifact matches the morphological profile of objects that ethnographic sources explicitly identify as dice, the case for classification strengthens considerably. This approach sidesteps a common problem in archaeology: the temptation to project modern assumptions onto ambiguous objects. Instead of guessing, the method compares physical traits against a verified baseline.
The study also highlights the value of long-term curation and documentation. Culin’s work preserved detailed observations from Indigenous communities at a moment when many traditional practices were under pressure from colonial policies. By treating those descriptions as primary scientific data rather than quaint anecdotes, Madden demonstrates how early ethnography can inform present-day quantitative analysis. The result is a rare continuity: the same types of objects appear first in living cultural contexts, then in historic museum collections, and finally in deeply buried archaeological layers.
What the Artifacts Look Like in Practice
Museum collections help illustrate what these gaming kits actually contained. The Smithsonian’s National Museum of the American Indian holds a complete gaming set that includes a bowl, dice, and counters, representing the full material setup for a dice game as played in the 19th century. Separate catalog entries for individual bone dice show how museums classify and curate these objects as distinct material culture, complete with collection and expedition metadata.
These are not abstract curiosities locked in storage. They are physical evidence that dice games were structured social activities with standardized equipment. A bowl provided a consistent throwing surface. Counters tracked scores across rounds. The games had rules, stakes, and repeated play, all features that distinguish gambling from idle tossing of random objects. In some documented cases, the stakes included valuables such as clothing, tools, or food, underscoring that these were consequential contests embedded in everyday economic life.
Because the artifacts are so standardized, they lend themselves to formal analysis. The size, thickness, and surface treatment of the dice affect how they tumble and land. Even without written rulebooks, archaeologists can infer aspects of gameplay from wear patterns and from how many pieces appear together in a single context. Madden’s compilation shows that these traits recur over vast distances, suggesting that dice gaming was not an isolated curiosity but a widespread and enduring tradition.
Split Cane Dice and Social Life in the 13th Century
While the oldest specimens push the timeline back to the Pleistocene, younger archaeological finds fill in the long middle stretch of the tradition. At Promontory Caves in Utah, researchers recovered split cane dice dating to the 13th century. These artifacts are associated with Athabaskan-speaking groups who were moving into the region, and the Natural History Museum of Utah has interpreted the gaming materials as evidence of social interaction during a period of migration and resettlement.
That interpretation carries weight because dice games are inherently communal. They require at least two participants, shared rules, and some form of agreed-upon stakes, whether material goods, social standing, or simple entertainment. Researchers have also noted gendered dimensions to participation, with some games historically associated with women and others with men, suggesting that gaming served distinct social functions within communities rather than operating as a single undifferentiated pastime.
In the Promontory Caves context, dice appear alongside footwear, tools, and household debris, painting a picture of daily life rather than ritual exclusivity. The presence of gaming equipment in what appears to be a domestic setting hints that chance-based play could ease tensions, reinforce alliances, or provide structured diversion during long winters and uncertain migrations. Far from being peripheral, gambling may have been one of the ways people negotiated risk in societies that already lived with environmental unpredictability.
Why 6,000 Years Earlier Matters
The conventional story of probability begins in the ancient Near East and Mediterranean, with knucklebones and marked dice appearing in Mesopotamian and Egyptian contexts several thousand years before the common era. Madden’s findings do not erase that history, but they displace it from the origin point. If Indigenous North Americans were using binary lots more than 12,000 years ago, then the human impulse to engage with randomness through structured play is both older and more geographically dispersed than previous scholarship recognized.
This has consequences beyond academic chronology. For decades, popular accounts of probability’s history have treated European mathematical traditions, particularly the 17th-century work of Pascal and Fermat, as the starting line. That framing implicitly positions Indigenous cultures as recipients rather than originators of ideas about chance. The archaeological evidence compiled in this study directly contradicts that assumption. Hunter-gatherer societies in North America were not waiting for contact with European thought to develop structured games of risk and reward; they had been refining such practices for millennia.
At the same time, Madden is careful not to claim that prehistoric players were doing formal mathematics in the modern sense. The dice games documented in the ethnographic and archaeological record do not come with written probability tables or algebraic proofs. Instead, they reveal an intuitive, experiential engagement with randomness. Players learned odds from repeated play, adjusted their strategies, and understood that some outcomes were more likely than others even if they lacked symbolic notation to express those patterns.
Recognizing this older and broader history of chance has ethical as well as intellectual stakes. It challenges narratives that treat Indigenous technologies and knowledge systems as static or purely “traditional.” Dice gaming, as traced through bone, cane, and antler, shows experimentation, regional variation, and long-distance continuity. It also underscores the importance of collaboration between archaeologists, ethnographers, museums, and descendant communities in interpreting what these objects meant in the past and what they continue to signify today.
Accessing the Research and Its Context
For readers who want to explore the study in more depth, the article is hosted within Cambridge’s digital platform, and general guidance on using that system is available through the Cambridge Core help pages. Those pages explain how to navigate journal content, manage institutional access, and troubleshoot common technical issues that can arise when viewing PDFs or supplementary material.
Questions about subscriptions, permissions, or institutional logins can be directed to Cambridge’s support team using the published contact details, which list regional email addresses and other options for reaching staff. Individual readers who encounter problems accessing the American Antiquity article can also submit a request through the online support form, providing information about their institution and the specific journal issue.
Together, the archaeological synthesis, the century-old ethnographic record, and the curated museum collections show that dice games in North America are not a marginal curiosity but a deeply rooted human practice. Long before probability became a field of formal study, people were already throwing marked pieces of bone and cane, watching them tumble, and building social worlds around the uncertain outcomes. The new evidence does not just push back a date on a timeline; it invites a reassessment of who gets credited with some of humanity’s oldest experiments in understanding chance.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
