The Fascinating History of Noughts and Crosses

Origins in Ancient Civilizations

Tic Tac Toe may seem like a simple game invented for school children with a pencil and paper, but its roots stretch back deep into ancient history. Archaeologists have discovered boards resembling the grids used today carved into clay, stone, and wood dating back thousands of years. The most famous early ancestor of the game is Terni Lapilli, a game played in the Roman Empire around the 1st century BC.

Unlike continuous modern play, Terni Lapilli often only gave each player three movable pieces. You had to place your three pebbles and then, assuming no one won, slide them into empty grid spots — an early form of a shifting combination game. Boards matching this description have been found scratched into the floors and surfaces of buildings across Rome, suggesting it was a genuinely popular pastime among all social classes, from soldiers garrisoning Hadrian's Wall to citizens in the marketplaces of the Forum Romanum.

Some historians have also pointed to even older precursors. Ancient Egyptian boards resembling noughts-and-crosses grids have been discovered carved near temple foundations. While the exact rules of these ancient games are lost, the geometric similarity to modern Tic Tac Toe grids strongly suggests that the human desire to claim and defend territory on a structured grid is one of our most ancient and intuitive gaming impulses.

Etymology: Noughts, Crosses, and "Tic Tac Toe"

In the United Kingdom, Ireland, and Australia, the game is commonly referred to as "Noughts and Crosses", where "nought" refers to the 'O' symbol and "cross" refers to the 'X'. This name is a beautifully literal description of the game's components. The modern American name, "Tic-Tac-Toe", emerged in the late 19th century and has a more oblique origin.

Historical records suggest that "tick-tack-toe" was originally a blindfolded pencil game where players would tap a slate with a pencil to guess numbers. Gradually, the name was transplanted onto noughts and crosses. The rhyming, playful quality of the name made it instantly memorable among children in the early 1900s, cementing it in American vernacular culture. The word "toe" is sometimes theorized to come from the sound the pencil made when it hit the slate — a "tac" or "toe" — though this etymology remains debated by linguists.

Across the world, the game goes by dozens of different names. In some parts of South Asia it is called "X and O," in Germany it is called "Drei gewinnt" (Three wins), and in parts of Latin America it is called "El Gato" (The Cat), a reference to the old English phrase "cat's game," which denotes a draw — because a cat always catches its own tail in a futile, endless loop, just as a tied Tic Tac Toe game ends in a circular, unavoidable stalemate.

The Medieval and Renaissance Period

After antiquity, clear records of the game become sparse until the medieval period. However, scholars believe that grid-based placement games were a continuous part of European folk culture throughout the Middle Ages. Many medieval manuscripts and illustrations depict scholars and students playing logic games on ruled grids, and the simplicity of drawing a 3×3 grid with a stylus on a wax tablet made it a natural classroom diversion.

During the 16th and 17th centuries, as printing became widespread, references to "Crosses and Noughts" began to appear more consistently in English-language texts. The game was considered so fundamental and universal that it was rarely described in detail — writers assumed their readers already knew it implicitly, much like how a modern author might reference "rock, paper, scissors" without providing an explanation.

The Game That Launched Artificial Intelligence

What sets Tic Tac Toe apart from mere historical trivia is its pivotal role in the birth of early computing and Artificial Intelligence. In 1952, a British computer scientist named Alexander S. Douglas programmed OXO as part of his PhD thesis at the University of Cambridge.

OXO was a program written for the EDSAC (Electronic Delay Storage Automatic Calculator), allowing a human player to compete against the machine using a rotary telephone dial to select their moves. The game's result would be displayed on a cathode ray tube — a primitive monitor. This was one of the very first interactive video games ever created, predating Pong by over twenty years. Douglas used his program to demonstrate how machines could successfully parse human input and simulate decision-making, laying an early theoretical groundwork for interactive computing.

The choice of Tic Tac Toe was deliberate. Because its state space is small enough to enumerate computationally, it was the perfect proving ground for demonstrating machine intelligence. OXO could play a perfectly rational game — it would never make a strategically unsound move — which was astonishing to observers in 1952 who were accustomed to machines that could only perform fixed sequential calculations.

Later Implementations: From BASIC to AI Research

Throughout the 1970s and 1980s, Tic Tac Toe was a staple exercise for budding programmers learning languages like BASIC and Assembly. Personal computer magazines of the era routinely published Tic Tac Toe programs as introductory code listings, and many programmers trace their first successful AI implementation back to this game. Its simplicity makes it ideal for teaching the concepts of loops, conditional logic, arrays, and basic search algorithms.

The game was also immortalized in popular culture in the 1983 film WarGames, where a military supercomputer named WOPR learns the concept of mutual assured destruction by repeatedly simulating Tic Tac Toe — concluding that "the only winning move is not to play." This scene resonated deeply with audiences because it used the universal familiarity of Tic Tac Toe to communicate a profound insight about nuclear deterrence theory and the futility of unwinnable conflicts.

In academic computer science, Tic Tac Toe remains one of the most commonly used examples for teaching graph search algorithms, game tree analysis, recursive programming, and the design of evaluation functions. University courses around the world — from introductory programming to advanced AI — use it as a pedagogical tool precisely because its complete game tree fits on a single slide.

The Mathematics of a Solved Game

Mathematically, Tic Tac Toe is classified as a "solved game" — meaning that from any position, the optimal outcome can be determined with certainty, assuming perfect play. The game has exactly 255,168 possible completed games. Of these, 131,184 are first-player wins, 77,904 are second-player wins (from non-optimal first-player moves), and 46,080 are draws.

Due to rotational and reflective symmetry of the 3×3 grid, the number of strategically distinct first moves collapses to just three: the center square, a corner square, and an edge square. The center is the strongest, leading to the fewest losses and most draws when played correctly. This symmetry analysis is a beautiful application of group theory to combinatorial game analysis.

Conclusion

Today, while no longer computationally challenging — even a smartwatch can calculate the entire game tree using the Minimax algorithm in microseconds — Tic Tac Toe remains an important pedagogical tool. It's often described as the "Hello World" of AI game programming: the first meaningful program you write that actually makes strategic decisions. Its journey from Roman pebbles scratched on stone, through the medieval schoolroom, into Alexander Douglas's pioneering AI research, and finally to a JavaScript implementation in a modern web browser is a testament to the enduring power of pure logic as one of humanity's oldest and most fundamental forms of play.