While we are so familiar with Europe's contribution to modern machine learning, let's explore how native cultures have contributed to our understanding of present day systems.
The Native American Tribes have made invaluable contributions in Geometry and Astronomy.
The Ancestral Puebloans, especially those who lived in what is now the Four Corners area (where Colorado, Arizona, Utah, and New Mexico meet), built structures such as Chaco Canyon that are known for their precise alignments with the sun, moon, and stars. The Great Kiva and the Sun Dagger at Fajada Butte in Chaco Canyon show impressive knowledge of solar and lunar cycles.
The Sun Dagger is a petroglyph at Fajada Butte where light passes through rock slabs and marks the solstices and equinoxes with astonishing precision. This structure shows how the Ancestral Puebloans understood geometric principles related to the alignment of light and shadow with astronomical events, much like how we look for patterns in data using clustering and geometric algorithms.
The structures at Chaco are aligned to capture the cycles of the sun and moon, and the Puebloans used this knowledge for agricultural and ceremonial purposes. This is analogous to how we track seasonal patterns or cycles in data, using tools like Fourier transforms to capture periodicity.
Though the Mayans are often seen as a separate culture from North American tribes, they shared significant ties with the Mississippian Culture, particularly in their focus on astronomical cycles. The Mississippians, who built Cahokia (near modern-day St. Louis, Missouri), also constructed large earthworks known as mounds, many of which are aligned with the solstices and equinoxes.
Cahokia had a structure called Woodhenge, a circle of wooden posts that marked the equinoxes and solstices, demonstrating the advanced geometric understanding of the builders. This is similar to the use of geometric shapes in ancient Greece to track celestial movements and correlates to our use of geometric models today to predict patterns in data.
The Lakota people had a deep cosmological understanding rooted in both the earth and sky. The Black Hills of South Dakota are seen as a reflection of the Milky Way, and Lakota star maps show constellations that are integrated with the natural geography of their lands.
The Lakota’s understanding of star constellations was not just for navigation, but for spiritual and practical purposes. They used a form of circular geometry to map the stars, similar to how we map high-dimensional data onto low-dimensional spaces using PCA. The notion of overlaying cosmic knowledge onto the physical landscape is reminiscent of how modern data visualization overlays data points onto a plane to reveal patterns.
The Hopi tribe also had a deep understanding of astronomy and geometry. The Hopi calendar, for instance, is closely tied to the movements of the sun. The Hopi Snake Dance, a ceremonial ritual, is timed based on the movements of the constellations and the solstices.
The Hopi often use spirals, zigzags, and circular motifs in their artwork and spiritual practices. These geometric patterns are reminiscent of fractal geometry, which is becoming increasingly important in understanding the natural world and complex systems. Fractals can model everything from stock market fluctuations to the branching of trees, demonstrating the connection between the Hopi’s geometric art and modern data science techniques.
The Iroquois Confederacy is known for its highly structured governance system, but they also had impressive geometric and architectural skills. Their longhouses were built using principles of symmetry and balance, which are key concepts in both engineering and data modeling today.
The Iroquois practiced balanced decision-making, which is mathematically tied to ideas like symmetry and equilibrium. Similarly, in machine learning, we often strive for balanced models that treat all features equally or assign weights symmetrically. The Iroquois system of governance, where each tribe in the confederacy has a voice, can be thought of as an early application of distributed systems, a concept important in modern machine learning algorithms like ensemble models.
Just as many Native American tribes tracked the cycles of the stars, sun, and moon to make sense of time and seasons, modern data science tracks cycles in data. Whether it’s stock prices, customer behavior, or weather patterns, we often use cyclical analysis and time-series forecasting to predict the future. The circular and radial patterns found in many Native American designs and architecture (like the medicine wheel used by various tribes) relate directly to clustering techniques in data science, where we look for geometric groupings in multi-dimensional space. The emphasis on symmetry, balance, and distributed decision-making found in the Iroquois system is conceptually similar to how we aim for balanced, well-regularized models that don’t overfit or underfit the data. The spiral motifs found in the Hopi and other Native American artworks mirror fractal patterns in nature, which modern mathematics uses to model complex, self-repeating structures like coastlines, clouds, and even neural networks.
By integrating the mathematical, geometric, and astronomical knowledge of Native American tribes, we gain a deeper respect for how their wisdom parallels our modern understanding of data, cycles, and the universe. Just as they used geometry to align their lives with natural cycles, we use mathematical tools like PCA, clustering, and cyclical analysis to understand patterns in data and predict the future. These cultural contributions, combined with those from the Greeks, Egyptians, and Mayans, form a tapestry of knowledge that spans millennia and continents, all of which informs and enriches our work today. This interconnectedness can be expressed in three key ways: patterns in nature, relationships in data, and unity of knowledge.
Across cultures, there has been a deep respect and understanding of natural cycles—whether in the movement of the stars and planets or the rhythms of life on Earth. The ancient cultures we discussed (the Greeks, Egyptians, Mayans, Native Americans) all recognized the cyclical nature of time, seasons, and life itself. These cycles were essential not only for survival but also for spiritual understanding.
The Chaco Canyon alignments, Mayan temples, and Egyptian pyramids were all carefully constructed to align with the solstices and equinoxes, much like how we align our modern datasets to recognize periodic trends. Today, we use Fourier transforms and seasonal decomposition in time-series analysis to recognize the cycles that repeat in our data—whether it’s the behavior of the stock market, weather patterns, or consumer purchasing habits.
The spiral motifs found in Hopi and Mayan art echo the fractal patterns that emerge naturally in tree branches, coastlines, and even the human nervous system. Modern data structures, like hierarchical clustering and decision trees, are built using similar principles. The fractal patterns inherent in data structures illustrate that from small to large, the patterns are self-similar—just as we observe in nature.
These patterns, when observed in data, give us insight into how deeply connected everything is. The cycles in nature are reflected in the cycles in data, and our ability to model and predict them is a continuation of the ancient practice of observing the cosmos.
The core of data science, just like ancient wisdom, is about discovering relationships. Whether it’s the relationship between the sun and the seasons or between variables in a dataset, everything is connected in some way.
In linear regression, we model the relationship between independent variables and a dependent variable, akin to how ancient cultures understood the relationship between celestial movements and earthly events. The dependent variable could be seen as the Earth reacting to the pull of the moon (causing tides), and the independent variables are the celestial bodies. Ancient wisdom teaches us that even if the connection isn’t always visible, it exists, and our models are built on discovering those relationships.
When we use PCA (Principal Component Analysis) or other dimensionality reduction techniques, we are attempting to unveil hidden patterns in high-dimensional data. This is conceptually similar to how the Mayan calendar or Native American star maps uncovered hidden celestial patterns that weren’t immediately obvious to the naked eye. These techniques help us understand how seemingly unrelated data points (stars, weather, or features in a dataset) are interconnected in a higher-dimensional space.
Regularization techniques like Ridge and Lasso regression help us balance our models, ensuring that no single variable dominates. This concept of balance and harmony mirrors the principles of many indigenous cultures, like the Iroquois’ focus on consensus and balance in decision-making or the Hopi’s understanding of the balance between the earth and sky. Achieving this balance in our models ensures that we are not “overfitting” or losing sight of the larger picture—much like maintaining balance in life ensures that no single part overwhelms the whole.
One of the most profound aspects of data science, like many other fields of knowledge, is that it builds on insights from many different cultures and time periods. The Greeks may have formalized geometry, but the Maya, Egyptians, and Native Americans used it for practical purposes long before. This unity of knowledge reminds us that truth can be discovered in different ways, by different people, but it is always connected.
Just as Euclid laid down the foundations for geometry, Egyptian surveyors used geometry to measure their land after the floods of the Nile, and the Maya used it to track the movements of the planets. Each culture contributed to a broader understanding of geometry, astronomy, and mathematics. Today, when we build complex models to analyze data, we are standing on the shoulders of all these giants, blending their contributions into a unified body of knowledge.
The artifacts left behind by these cultures—whether they be pyramids, star maps, or spiral motifs—are in many ways data visualizations. They are a way of encoding knowledge and understanding in a form that can be passed down through generations. Today, our data visualizations—whether they are scatter plots, heat maps, or network diagrams—are the modern equivalent. They allow us to encode the patterns we see in data and share them with others, revealing truths that may not be immediately obvious without that visual aid.
Just as ancient cultures understood the cycles and relationships that governed the natural world, we, too, seek to understand the relationships between variables in our datasets and the hidden patterns that beneath the surface. What we are doing today with data science is a continuation of that same journey—a journey of discovery that spans continents, centuries, and cultures.
By recognizing the contributions of different cultures—from the Mayan mathematicians to the Iroquois decision-makers, to the Native American astronomers—we deepen our respect for the universal truths that connect us all. The tools we use today may be more sophisticated, but they serve the same purpose: to help us make sense of the interconnectedness of the universe.
Everything is connected — whether it’s the movement of the planets, the cycles in our data, or the wisdom passed down from ancient cultures to modern times. The more we explore, the more we realize that all of our data points are part of a larger, interconnected whole. Just as the circle represents unity and completeness, our understanding of the world, through data and mathematics, brings us closer to a unified vision of reality.
Our work with data and models, from PCA to clustering to regression, is a microcosm of what diplomacy is at its core: finding patterns, relationships, and solutions that serve the greater good. Just as we seek harmony in data, we strive for harmony in the world.
Together, we’re setting the foundation for a future where science, technology, and empathy come together to solve the world’s most pressing problems, creating an environment of mutual respect, understanding, and progress.
Note: Insights supported by conversations with AI.
Top comments (0)