Native Americans used dice thousands of years before the Bronze Age

The traditional six-sided die has been around since the Bronze Age, with the earliest known pieces from approximately 3000 BC uncovered in Mesopotamia and the Indus Valley. Now, a new study has found evidence that Native Americans were likely using dice for gaming and gambling more than 6,000 years earlier, since the end of the last ice age.

The dice looked different than the polyhedral shapes we’re used to playing with today. The oldest ones identified in the study were known as “binary lots.” These artifacts — found at archaeological sites from the Folsom Period in Wyoming, Colorado and New Mexico — date back roughly 12,800 to 12,200 years. The pieces were flat, two-sided dice crafted from wood or bone and could be tossed, similar to how we flip a coin today. The findings were reported in a paper published April 2 in the journal American Antiquity.

Lead study author Robert J. Madden, a doctoral student of archaeology at Colorado State University in Fort Collins, argued that the evidence has been hiding in plain sight.

“We’ve had a great record of this subject during the historical period after Europeans got here,” Madden said, referring to the past 1,000 to 2,000 years. “But then before that, we really didn’t know, like, how far does it go back before this? And that’s really what this paper adds.”

The study creates new criteria for deciphering old dice and allows archaeologists to further explore how games have evolved over time, researchers say.

Deciphering prehistoric dice

Because the dice were not the traditional cube shape, when they were initially uncovered archaeologists often referred to them as “gaming pieces,” Madden said. He noticed the gaps in the literature, particularly the lack of information about how dice were utilized before Europeans came to America, and wanted to come up with a way of identifying whether the pieces were used as dice.

“I kind of picked up on that there were some of these really old examples — these discs, and these late Pleistocene 12,000-year-old context — that just people were saying, ‘Well, we don’t know what these things are,’” Madden said. “I thought, ‘Boy, if they are dice, that is really significant.’”

Madden used a 1907 analysis by ethnographer Stewart Culin, which identified 293 sets of historic Native American dice, to create a four-part test to evaluate an unidentified object. The test utilized descriptors such as the shape of the object and the markings on each side to determine whether they were binary lots.

The doctoral student identified more than 600 previously unknown sets of Native American dice from 45 prehistoric archaeological sites in the western United States from the Late Pleistocene until after the period of European contact.

Games archaeology “has been overlooked for so long as unimportant or invisible in mainstream archaeology. This paper does quite a bit to demonstrate what is possible if you have the knowledge of traditional gaming practices and look for analogues in the archaeological record,” said Walter Crist, an archaeologist at Leiden University in the Netherlands who specializes in ancient games. Crist was not involved with the new study. “I think this is a crucial study for research on games in the Americas, but also for prehistoric archaeology worldwide.”

“There also is a long tradition, indeed still continuing to this day, in West and South Asia and North Africa of using natural objects such as cowry shells, sheep ankle bones, and unmarked split sticks or reeds as binary dice in exactly the same way he describes in North America. The difference is that these objects can almost never be positively identified as dice in the archaeological record,” Crist added in an email. “Thus, it is very likely that people were using these natural objects much longer before they started manufacturing polyhedral dice around the time of the earliest states. So we cannot really say when dice really started being used before it is visible archaeologically.”

While Madden’s test is only for the purposes of identifying Native American dice, he noted that it also works for many Old World dice, and it’s possible there are other examples across the world of gaming pieces yet to be classified as dice.

Throughout history, dice have been used for many different things, including important decision-making or even divination, such as ancient Roman belief that gods controlled the outcomes of dice. However, Native American dice were almost exclusively used for gambling and games of chance, Madden noted.

The binary lots also are still used today. Madden said he came across tutorials posted on YouTube of how to play the dice games from 2,000 years ago.

Jelmer Eerkens, a professor of anthropology at the University of California, Davis, argued that it is impossible to say with certainty that the objects are dice without including their archaeological context, such as whether they were found in a house or what they were found next to.

“I agree these COULD be dice, but I’m not convinced they necessarily were. I think the study has flawed logic that could easily allow some items to be classified as dice when they actually were not used in such a fashion. In other words, just because it’s rectangular in shape and fits in your hand, doesn’t mean it’s a cell phone,” Eerkens, who was not involved with the new study, said in an email.

“I have little doubt that people globally enjoy, and have long enjoyed, games of chance. That said, I would be super surprised if games of chance were played the same 12,000 years ago as today, and people used the same implements,” Eerkens added. “Material culture evolves, and usually fairly rapidly.”

Madden argued that, based on previous literature and analysis, there is strong evidence that similar artifacts have been used as dice for the last 2,000 years, so it seems likely that the Native American binary lots were also used as dice since their invention at least 12,000 years ago.

While the theory of probability wasn’t identified until 1654 by French mathematicians Blaise Pascal and Pierre de Fermat, Madden said that this test opens new avenues for understanding how humans interacted with the concept before the development of the theory.

“The archaeological evidence of games for me humanizes the past and shows us quite a bit about how we are still the same humans we were back then: we crave human interaction, and find interesting and creative ways to interact with the objects around us to create interesting ways to use them to interact,” Crist said. “Fundamentally, dice games and board games were the same in the past as they are now — the rules may not have been as complex as those you’d buy in a game store, but the basic actions are the same and they remain instinctual to us.”

Taylor Nicioli is a freelance journalist based in New York City.

Sign up for CNN’s Wonder Theory science newsletter. Explore the universe with news on fascinating discoveries, scientific advancements and more.

For more CNN news and newsletters create an account at CNN.com